Full text of "كتب في علم الميكروبيولوجي"

See other formats

Diseases

S l Ptrch^l

It M Chalmcis
M Hmtacy

PR Hunter

J Sdhvood
P\fyn-Joncs

Microbiology of Waterborne
Diseases

Steven Percival

Department of Microbiology, Leeds General Infirmary, Leeds, UK

Rachel Chalmers

Cryptosporidium Reference Unit, Singleton Hospital, Swansea, UK

Martha Embrey

The George Washington School of Public Health and Health Services,
Washington DC, USA

Paul Hunter

Chester Public Health Laboratory, Chester, UK

Jane Sellwood

Environmental Virology Unit, UK Health Protection Agency,
Reading, UK

Peter Wyn-Jones

Institute of Pharmacy, Chemistry and Biomedical Sciences,
University of Sunderland, UK

ELSEVIER

ACADEMIC

PRESS

Amsterdam • Boston • Heidelberg • London • New York • Oxford
Paris • San Diego • San Francisco • Singapore • Sydney • Tokyo

This book is printed on acid-free paper

No part of this publication may be reproduced, stored in a retrieval system, or
transmitted in any form or by any means electronic, mechanical, photocopying,
recording or otherwise, without the prior written permission of the publisher

Permissions may be sought directly from Elsevier's Science & Technology Rights
Department in Oxford, UK: phone: ( + 44) 1865 843830, fax: ( + 44) 1865 853333,
e-mail: permissions@elsevier.co.uk. You may also complete your request on-line via
the Elsevier homepage (http://www.elsevier.com), by selecting 'Customer Support'
and then 'Obtaining Permissions'

Elsevier Academic Press

525 B Street, Suite 1900, San Diego, California 92101-4495, USA

http ://www.elsevier.com

Elsevier Academic Press

84 Theobald's Road, London WC1X 8RR, UK

http ://www.elsevier.com

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library

Library of Congress Catalog Number: 2003115959

ISBN 0-12-551570-7

Typeset by Charon Tec Pvt. Ltd, Chennai, India
Printed and bound in Great Britain
04 05 06 07 08 09 9 8 7 6 5 4 3 2 1

Contents

Preface vii

Part 1 Introduction

1 Risk assessment and drinking water 3

Part 2 Bacteriology

Acinetobacter

Aeromonas

Arcobacter

Campylobacter

Cyanobacteria

Escherichia coli

Helicobacter pylori

Other heterotrophic plate count bacteria (Flavobacterium,

Klebsiella, Pseudomonas, Serratia, Staphylococcus)

125

Legionella

145

The Mycobacterium avium complex

155

Salmonella

173

Shigella

185

Vibrio cholerae

197

Yersinia

209

Part 3 Protozoa

16 Acanth amoeba spp. 221

17 Balantidium coli 231

18 Cryptosporidium spp. 237

19 Cyclospora cay et an en sis 267

20 Entamoeba histolytica 285

21 Giardia duodenalis 299

22 Naegleria fowleri 319

23 Toxoplasma gondii 325

Part 4 Viruses

24 Common themes 339

25 The survival and persistence of viruses in water 345

Contents

26 Methods for the detection of waterborne viruses 349

27 Adenovirus 379

28 Astrovirus 387

29 Enterovirus (poliovirus, coxsackievirus, echovirus) 401

30 Hepatitis A virus (HAV) 419

3 1 Hepatitis E virus (HEV) 427

32 Norovirus and sapovirus 433

33 Rotavirus 445

Part 5 Helminths

34 Dracunculiasis 455

Part 6 Future

35 Emerging waterborne infectious diseases 463

Index 469

Preface

The microbiological quality of drinking water varies widely and constitutes a
constantly fluctuating ecosystem composed of a dynamic complexity of niches.
With these multifacted ecological niches in water comes man's ambition to
detect and ultimately control microorganisms that represent a concern to public
health. The health consequences of exposures to waterborne pathogens are sig-
nificant and constitute a grave concern to both developed and developing coun-
tries. However, due to the deficiencies of current knowledge in both identifying
and understanding the ecology of waterborne pathogens, the determination of
safe exposure levels of these pathogens to man still represents an area of inten-
sive investigation. With poor methods available for the detection and monitor-
ing of pathogens, it is presently impractical and costly to try to determine all the
types of pathogens found in drinking water.

Water is a chaotic ecosystem and through every drinking water outlet comes
a multitude of problems when considering health concerns due to waterborne
pathogens. Surely, therefore, in water when it comes to knowing and under-
standing the microbes of human significance we are only touching the 'tip of
the iceberg' when it comes to highlighting microbes that have an impact on
public health. However, the emphasis in drinking water research is placed
upon understanding conditions likely to ensure the safety of drinking-water
supplies and monitoring their fulfilment more directly.

Historically we still adhere to the concept that with the absence of E. colt
from drinking water comes the reassurance that pathogens of human signifi-
cance are also absent. However, what we must consider in water is not just those
microbes that produce short-term suffering, but the ones that may have long-
term health implications not yet discovered. Have we therefore to leave well
alone or to strive to understand pathogens and their behaviour in water and try
to determine other microbes that may well compound other human diseases?

This book looks at the pathogens that are known to be associated with
drinking water and acknowledged to constitute a concern to 'the health of the
nation'. The book brings together the current knowledge of drinking water
pathogens highlighting their basic microbiology, clinical features, survival in
the environment and, in particular, risk assessment. We hope you enjoy read-
ing this book.

Steven Percival

VII

Parti

Introduction

Risk assessment and
drinking water

What is risk?

The provision of safe drinking water was one of the triumphs of 20th century
public health. The use of chlorine disinfection, especially, has resulted in a dras-
tic decrease in exposure to waterborne pathogens, and therefore, waterborne
disease. In the 1970s, water pollution from chemical waste received special
attention following incidents such as the Cuyahoga River in the USA catching
fire. This prominence of chemical risk incidents accelerated the practice of
environmental risk assessment. Consequently, early risk assessment paradigms
focused exclusively on chemical contamination. When the 1993 outbreak of
cryptosporidiosis from drinking water made some 400000 Milwaukee resi-
dents ill and killed over 100, the focus of risk in drinking water shifted to the
microbial. The influences of agricultural and municipal wastewater, as well as
the ageing of water treatment and distribution systems, continue to raise con-
cerns about the microbiological quality of drinking water.

Introduction

In 1994, the US Congress directed the formation of a Commission to inves-
tigate the appropriate uses of risk assessment and risk management in feder-
ally mandated programmes. Their Presidential/Congressional Commission on
Risk Assessment and Risk Management - Final Report (1997) defines risk as
c ... the probability that a substance or situation will produce harm under speci-
fied conditions. Risk is a combination of two factors: the probability that an
adverse event will occur (such as a specific disease or type of injury) and the
consequences of the adverse event. Risk encompasses impacts on public
health and on the environment, and arises from exposure and hazard. Risk
does not exist if exposure to a harmful substance or situation does not or will
not occur. Hazard is determined by whether a particular substance or situation
has the potential to cause harmful effects.'

Risk assessment

Risk assessment is carried out by international agencies (e.g. International
Agency for Research on Cancer and the World Health Organization), national
governments, state and local governments and private industry. This process
uses the broadest sort of regulatory decision making on a national and inter-
national level - from a national government regulating drinking water treat-
ment protocols to the local public health department making a decision about
issuing a boil-water advisory.

The practice of quantitative risk assessment gathered momentum in the
1960s with scientists' attempts to estimate risks of small exposures to car-
cinogens. Quantitative risk assessment would improve decision makers' abil-
ity to set both research and regulatory priorities. During the 1970s and 1980s,
methods of risk assessment began to evolve, and better data were available to
support the assessments and their conclusions. At this point in history, risk
assessment was used primarily to estimate chemical risks by extrapolating
toxicological data from animals to humans.

In 1983, the process of risk assessment was formalized in a report from the
US National Research Council (NRC) called Risk Assessment in the Federal
Government: Managing the Process. This pivotal publication still serves as
the basis for most risk assessment frameworks. The NRC formula comprises
four steps that lead to the evaluation of data on the hazards of the agent in
question, on the extent of human exposure to it, and on the characterization
of the possible risk. This framework was the first to present a systematic
approach to analysing scientific information about substances that may pose
health risks under given conditions.

This four-step framework is now a universally recognized method to charac-
terize the likelihood of adverse health effects from (mainly chemical) exposures:

1. Hazard identification

• determines the types and quantities of the contaminants under study

• identifies the nature of hazards they pose to health

Risk assessment and drinking water

• determines if exposure to the agents can cause an increased incidence of
adverse health effects and

• characterizes the nature and strength of the evidence of causation.

2. Dose-response assessment

• determines the relationship between concentrations (dose) and adverse
effects from exposure (response)

• characterizes the relationship between exposure or dose and the inci-
dence and severity of the adverse health effect

• considers factors that influence dose-response relationships, such as
intensity and pattern of exposure and age and lifestyle variables that
could affect susceptibility

• involves extrapolation of high-dose responses to low-dose responses and
from animal responses to human responses.

3. Exposure assessment

• describes the conditions under which people could be exposed to
contaminants

• determines the intensity, frequency and duration of exposures of humans
to the agents in question

• estimates concentrations of the substances at various points from their
sources through the environment.

4. Risk characterization

• describes the nature of adverse effects that can be attributed to contam-
inants, estimates the likelihood of exposures and evaluates the strength
of evidence and uncertainty

• combines the assessments of exposure and response under various expos-
ure conditions to estimate the probability of specific harm to an exposed
individual or population

• should include the distribution of risk in the population.

Risk assessment can be controversial because risk evaluation is based on
both scientific data and judgement. So much uncertainty exists because our
current knowledge regarding exposure and effects of agents, and their rela-
tionship to humans and the environment, is unclear. Within this framework,
there continues to be active debate about the most appropriate risk assess-
ment approaches, the impact of various kinds of data on risk projections,
and the level and appropriateness of conservatism to use in estimates. The
most frequently debated issues within risk assessment include the use of
default options, which are mostly conservative and more likely to overesti-
mate risk; the validation of methods and models; and the variability in those
exposed - within individuals, among individuals, and among populations.
The question of how to accommodate susceptible subpopulations in the
framework is crucial, and the uncertainty in data and models cannot always
be quantified.

Risk assessment uses research and data from a variety of disciplines -
epidemiology, toxicology, statistics, molecular biology, clinical medicine,
exposure modelling, dosimetry and others. Determining what constitutes

Introduction

sound science requires time to gather information and consider opposing
views. The risk assessor must evaluate the context, source, presumptions, and
biases of scientific studies that will be used as evidence. The policies and pro-
cedures for conducting risk assessment are not static; they have evolved over
time in the face of new demands and the availability of new and better kinds
of information. A spectrum of beliefs and data typically must be deciphered
and evaluated to extract those scientific facts, assumptions and beliefs that
will stand the test of time.

Microbial risk assessment in drinking water

Introduction

While techniques for assessing chemical risks have been in widespread use for
over 40 years, we have less experience performing risk assessments for micro-
bial pathogens. Differences between chemicals and pathogens in environmen-
tal fate and transport and pathogenesis require new approaches to quantitative
risk assessment (Gibson et al. 9 1998). Quantitative risk assessment methodo-
logy has not been thoroughly applied to environmentally-transmitted infectious
diseases. The use of risk assessment to evaluate infectious microbes in drinking
water, shellfish, recreational water, and foods is now being reported, but not in
a systematic way.

Regulatory history

Risk assessment in general is closely tied to governmental policies focusing on
the control of contaminants. In the US Safe Drinking Water Act (SDWA), risk
assessment was used to set maximum contaminant levels (MCLs) for chemicals
that may have the potential to cause health problems in humans. Risk assess-
ment has not been used to develop microbial standards for drinking water in
the USA, and microbial contaminants in drinking water have not been individu-
ally prioritized for regulation (Rose and Gerba, 1991). The maximum contam-
inant level goal (MCLG) for pathogens is zero, but this has been accomplished
through specifying treatment methods for various types of source water and
monitoring for faecal coliform bacteria, which may indicate contamination.
The US Environmental Protection Agency's (EPA) Surface Water Treatment
Rule requires a safety goal of 99.9% reduction in Giardia and viruses
(Environmental Protection Agency, 1998). EPA believes that this level of
removal would result in an annual risk of no more than one infection/10 000
people exposed to drinking water (Rose and Gerba, 1991). The European
Union standard also does not permit any pathogen level above zero in drinking
water (Barrell et ai 9 2000). Their regulatory guidance relies on total coliforms,

Risk assessment and drinking water

faecal coliforms and total colony counts as pollution indicators. Although this
combination of pollution indicators and treatment controls has worked fairly
well over the years, major deficiencies have become known because of emerg-
ing pathogens such as Cryptosporidium and Mycobacterium avium complex
(MAC); in fact, the UK has introduced precedent-setting legislation that singles
out Cryptosporidium oocysts as a detection parameter for drinking water qual-
ity (Barrell et al. 9 2000). Problems have been identified with indicator organ-
isms (e.g. members of the Enterobacteriaceae), such as the fact that viruses and
protozoa can be present and viable when indicator organisms are inactive.
Also, coliforms and other indicator bacteria may be more sensitive to chlorine
than some pathogenic organisms, so the resulting treated water quality assess-
ment can be inadequate. Many communities have experienced waterborne dis-
ease outbreaks even though their water supplies have met mandated coliform
standards (Craun et aL 9 1997).

Application of data

The role of toxicology is large in chemical risk assessment. For many reasons -
the greatest of which are ethical considerations - animal studies are relied on
extensively in chemical evaluation. Usually, the only available type of data on
human exposure to chemicals is occupational in origin or the result of an acci-
dental spill or release. In the absence of human data, toxicological evaluation
usually begins with the simplest, fastest, and most economical tests and pro-
ceeds to complexity only as warranted by the initial results (National Research
Council, 1994). In the case of microbial infection, it is difficult to develop
animal models for most pathogens because of their specificity to humans, but
some do exist. On the other hand, the high incidence of many waterborne
diseases gives us data on human exposure and health effects in the form of
surveillance and outbreak reports that would not typically be available for
chemicals.

Although toxicology, epidemiology, clinical medicine and exposure assess-
ment all contribute data to risk assessment, epidemiology is important because
it provides evidence on human subjects in real-world situations. This is espe-
cially important in microbial risk assessment. The WHO formed a working
group to look at evaluating and using epidemiological data for health risk
assessment. Their guidelines on using epidemiological information have been
published (World Health Organization Working Group, 2000). Because most
epidemiology is observational, drawing conclusions about causation is prob-
lematic. People engage in many different behaviours that affect their exposures:
choice of food, drinking water, recreational activities, workplace and contact
with others. All of these affect a person's susceptibility to becoming infected and
to developing illness once infected. Epidemiologists must try to sort through
these many factors in order to draw inferences about their hypotheses. Because
epidemiology is observational and inductive, no single study can provide a def-
inite answer about cause and effect, even if bias is minimal (Pontius, 2000).

Introduction

Use of surveillance

In the USA and the UK, formal surveillance data are collected through collab-
oration between federal, state and local government entities. The UK system
also includes a central automated screening process that looks for unusual pat-
terns of microbial infection in weekly monitoring reports (Tillett et aL, 1998).
The goals of a surveillance programme consist of characterizing the epidemi-
ology of outbreaks, identifying the agents causing outbreaks, and identifying
how an outbreak occurred (Barwick et al., 2000). Information that can be
gleaned from surveillance includes magnitude of community impact, attack
rates, hospitalization and mortality, demographics, sensitive populations, level
of contamination, duration, medical costs, community costs and secondary
transmission.

The US Centers for Disease Control and Prevention (CDC) defines a
waterborne outbreak as two people experiencing a similar illness after
exposure to drinking or recreational water, and evidence must indicate that
water is the probable source of the outbreak (Barwick et al. 9 2000). However,
the role of waterborne exposure is difficult to estimate for most pathogens.
Those that are transmitted via the faecal-oral route can contaminate food
as well as water or be spread by person-to-person contact. Waterborne
outbreaks of caliciviruses and microsporidia, for example, have been
documented, but although other viruses, such as echovirus and adenovirus,
are present in sewage, they have not been associated with drinking-water
transmission.

Outbreak surveillance uses both epidemiological data and water-quality
data, when available. Epidemiology, which is the study of occurrence
and causes of diseases in a population, identifies associations between expos-
ures and health outcomes using statistical methods, and has been the main
science used to study the relationship between drinking water and infective
disease. Because infectious diseases are caused by many exposures besides just
drinking water, epidemiology can help define the contribution of drinking
water. Water quality data are helpful, but frequently unavailable because
of the transient nature of contamination and the analytical difficulties in
detection (Tillett et aL, 1998). However, the UK has begun regular monitoring
for the presence of cryptosporidial oocysts in municipal water supplies
(Hunter, 2000).

Waterborne disease surveillance data are useful in evaluating our treatment
and supply of safe drinking and recreational water; however, quality and com-
pleteness of investigations vary from place to place, and risks are not fully
identified. The extent of that underestimation varies by locale and is unknown
overall. The ability to recognize an outbreak relies on a number of factors
including public awareness, ill people seeking health care, the extent of
laboratory testing, and local infrastructure available for investigating and
reporting outbreaks. Interventions to increase waterborne illness surveillance
programmes have included the use of 'sentinel' populations, like nursing
home residents, pharmacy sales of anti-diarrhoeal drugs, and school absentee
logs (Proctor et al. 9 1998).

Risk assessment and drinking water

Standardizing the way data are collected, managed, transmitted, analysed,
accessed and disseminated is critical in evaluating the adequacy of cur-
rent regulations for water treatment and monitoring, as well as produc-
ing important information on the pathology and epidemiology of waterborne
diseases. Timely and accurate surveillance data are crucial to identifying
an outbreak's aetiology and therefore methods for mitigation. Surveillance
studies are perhaps the most important source of information we have on
waterborne disease, and improving the infrastructure for surveillance
programmes would enhance our knowledge on waterborne disease (Gostin
et al, 2000).

Microbial risk assessment frameworks

A number of frameworks to address some of the issues unique to microbial
risk assessment have been developed. Some have taken the traditional four-
step chemical assessment paradigm and altered it; others have used a different
basis for analysis. Rose and Gerba (1991) propose the following alteration of
the traditional four steps for use in microbial risk assessment:

1 Hazard identification

• identifies the microorganisms of concern with food or water through
scientific literature such as clinical studies, epidemiological studies and
surveillance, animal toxicology studies, and in vitro toxicology studies.

2. Dose response

• describes the severity and duration of adverse effects that may result
from the ingestion of a microorganism in water through qualitative or
quantitative methods

• relates to the microbe and to the human host: virulence and infectivity,
secondary transmission, asymptomatic infection, low doses of some
organisms resulting in severe effects.

3. Exposure assessment

• assesses the extent of human exposure, such as the occurrence and
concentration of microbial contaminants in raw and treated drinking
water or recreational water, and the actual consumption of drinking
water

• relates to socioeconomic and cultural backgrounds, ethnicity, seasonal-
ity, population demographics, geographical differences, and consumer
preferences.

4. Risk characterization

• depends on the variability, uncertainty and assumptions used in the pre-
vious steps. Data uncertainties include those that might arise in the
evaluation and interpretation of epidemiological, microbiological and
animal studies. Biological variation includes the differences in virulence
that exist within a microbiological population and variability in the sus-
ceptibility within the human population.

Introduction

The Codex Committee on Food Hygiene (1999) has published guidelines
regarding microbial risk assessment. Some of the general principles the commit-
tee has defined include:

• separation between risk assessment and risk management

• inclusion of hazard identification, hazard characterization, exposure assess-
ment and risk characterization in the approach

• use of a transparent and iterative process

• use of precise, best-available data

• consideration of the dynamics unique to microbes, such as growth, sur-
vival, death and the interaction between host and pathogen.

An extensive risk framework has been developed that attempts to address
the issues unique to waterborne microbial risk assessment (Neumann and
Foran, 1997). US EPA has adopted this approach to evaluate future contamin-
ants for regulation (Risk Policy Report, 2001). This three-part framework
comprises:

1. Problem formulation

2. Analysis phase

3. Risk characterization.

1) Problem formulation. The purpose of problem formulation is to identify
the purpose, goals and extent of the risk assessment and consider the major
factors that will come into play. An initial risk model will be formulated in this
step. This conceptual model describes the interactions of a particular pathogen
and a defined population within a defined exposure scenario. An initial char-
acterization of exposure and health effects is conducted. This step considers
pathogen characteristics, host characteristics and host-pathogen interactions.
The exposure scenarios and health effects are defined, including the media,
exposure routes, endpoints and essential assessment variables.

2) Analysis phase. The analysis phase of the microbial risk assessment con-
sists of the technical evaluation of data concerning the potential exposure and
associated health effects and is based on the conceptual model developed dur-
ing problem formulation. This phase consists of two elements: characteriza-
tion of exposure and characterization of human health effects. The quality
and quantity of data available affect analyses - many uncertainties exist. This
analysis characterizes the relationship between dose, infectivity, and the mani-
festation and magnitude of health effects in an exposed population. The rela-
tionship is complex and, in many cases, a complete understanding will not be
possible. Data obtained from animal studies, human clinical studies and out-
breaks are used to generate a curve or model for the quantitative relationship
between dose and response. Animal models may be useful for determining
these relationships, but should be interpreted carefully because of the host
specificity of most pathogens. Another difficulty that may be encountered in a
dose-response analysis is the availability of data regarding infection. In many

Risk assessment and drinking water

cases, infection data will not be available, so the analysis may only be able to
describe the relationship between dose and clinical illness, rather than dose,
infection and clinical illness.

Characterizing exposure involves the evaluation of the interaction between
the pathogen, the environment and the human population and results in the
development of an exposure profile that quantitatively or qualitatively evalu-
ates the magnitude, frequency and pattern of human exposure for the scenarios
developed during problem formulation. Some of the questions asked during
this process have to do with the nature of the pathogen, such as its ability to be
transmitted and cause disease in the host, its occurrence and distribution in the
environment and its ability to survive and multiply. Human factors to be con-
sidered include the demographic characteristics of the exposed population,
how the population is exposed and how long the population is exposed.

Characterization of human health effects involves the interactive analysis of
three critical components: host characterization, evaluation of human health
effects and quantification of the dose-response relationship. Elements include
individual susceptibilities like age, immune status, nutritional status, etc.; clin-
ical manifestations regarding infection, illness and sequelae and characteriz-
ing the nature of the relationship between exposure, symptomatic and
asymptomatic infection and the duration and severity of illness.

3) Risk characterization. Risk characterization is the final phase of the
microbial risk assessment and is the product of combining the information
from the exposure profile and the host-pathogen profile. Risk characterization
consists of two major steps: risk estimation and risk description. Risk estima-
tion describes the types and magnitude of effects anticipated from exposure to
the microbe and the likelihood of those effects occurring. It can be qualitative
or quantitative depending on the data and methods used. The second compon-
ent of risk characterization - risk description - involves describing the event
according to its nature, severity and consequences. Uncertainties associated
with all phases - problem formulation, analysis and risk characterization - are
identified and quantified when possible in this section.

Risk management is the process by which the results of risk assessment are
integrated with other variables, such as political, social, economic and engin-
eering factors, to arrive at decisions about the needs and methods for risk
reduction. Although not included as part of the original four-part paradigm of
risk assessment designed by the NRC, risk management entails the important
process of creating actions and strategies to reduce risks. Examples of how
microbial risk assessment data can be applied to practical risk management
approaches include:

• predicting endemic rates of drinking-water-related infections

• estimating pathogen densities that drive standard-setting and treatment
protocols

• determining the effectiveness of water treatment

• forecasting the risk increase during water-treatment failure

Introduction

balancing microbial risks with risks of disinfection by-products
identifying the most cost-effective options to reduce microbial health risks
(Gale, 1996).

Susceptible subpopulations

Risk assessments are generally used to guide regulatory decisions regarding a
level of exposure resulting in 'acceptable' risks. When the exposure exceeds
the defined risk level, action is taken to reduce the concentration of pollutants
to an acceptable level. Defining what kind of vulnerability merits exposure
protection and where to draw the regulatory line on whom to protect and
how best to protect them are the issues with which policy makers struggle.
Even the definition of 'susceptible' in this context has been a moving target,
with scientists and policy makers approaching it from different directions
(Parkin and Balbus, 2000).

A greater emphasis on susceptible subpopulations has resulted in a more
complex risk assessment process. People are concerned about vulnerable
members of the population, but characterizing who is most sensitive to which
particular risk is an impossible task. Therefore, concepts of susceptibility have
focused on:

1. the probability that an individual will be exposed to a questionable agent
and then react to it

2. the comparison of an individual's susceptibility to that of the majority of the
population

3. the variation of individual states of vulnerability within a population.
Important factors in overall susceptibility include immune status, preg-
nancy, underlying illness and lifestyle (e.g. smoking and drinking habits).
Due to environmental effects on expression, genetic factors may either
result in lifelong or periodic susceptibility.

Host characterization contains susceptible subpopulation factors, for example
age, immune status, genetic background, pregnancy, etc. The elderly, children
and the immunocompromised are perceived as most susceptible to infection and
illness from waterborne pathogens, however, that assumption is not always
founded in fact. Certain illnesses are closely associated with the immunocom-
promised, such as MAC and microsporidia in AIDS patients. Other pathogens,
such as Helicobacter pylori do not appear to affect the immunocompromised
any differently than the immunocompetent (Edwards et aL 9 1991) and, though
H. pylori infection occurs primarily in childhood, its sequelae do not manifest
themselves until adulthood. Viruses have a higher incidence in childhood - prob-
ably because of a child's naive immune system - but it is common for adults to
have a higher degree of morbidity once infected. MAC manifests itself completely
differently in the host. In children, it causes lymphadenitis. In adults with under-
lying lung damage, it causes pulmonary MAC, which is a chronic disorder, like
tuberculosis. Pulmonary MAC has also been showing up in elderly women who

Risk assessment and drinking water

are without predisposing factors (Prince etaL 9 1989; Kennedy and Weber, 1994).
In AIDS patients, MAC manifests itself as a disseminated, end-stage disease.
Up to 40% of AIDS patients develop disseminated MAC in their lifetimes,
though the incidence has dramatically decreased with the advent of retroviral
drug therapy (Horsburgh, 1991; Havlir et aL, 2000).

There are different ways a population can be categorized as susceptible in
terms of the frequency and severity of infection. A pathogen may affect a cer-
tain population by:

1. having a higher rate of infection than the general population, but the same
disease response

2. having a higher rate of infection than the general population, but a less-
ened disease response

3. having the same incidence of infection as the general population, but a
greater level of morbidity and/or mortality

4. both - a higher rate of infection and a more serious outcome. For example,
children can be more susceptible to infection from certain pathogens than
adults because of their inherently reduced immunity, which is a natural state.
Although they are more likely to become infected, many children's outcomes
are mild or asymptomatic and infection usually confers some level of protec-
tion from later exposures. In this sense, children fall into the category of
having a higher rate of infection, but decreased morbidity and/or mortality
compared with adults. This is also a transient category; children are not
automatically in this state at birth, though they may acquire temporary
immunity from their mothers from breastfeeding. Also, once a child has been
infected by a particular pathogen and develops immunity, he or she leaves
that category and enters either the 'general' population or another category
based on some other susceptibility. This cycling in and out of categories of
susceptibility varies for each pathogen (see Table 1.1).

Risk communication

Initially, risk communication was added to the risk assessment process as an
afterthought, along with risk management. Over time, risk managers began to
experience the need for help in communicating environmental risks to the
public and there has been recognition that public input into the goals and
mechanics of risk assessment helps create trust in the overall process. In their
1994 publication, the NRC said, 'Public confidence that risk managers are
addressing real concerns, as opposed to going through a process perfunctorily,
is critical to the future of risk assessment as an activity capable of improving
the quality of life.'

Current risk communication research efforts aim to help risk analysis by
providing a basis for understanding and anticipating public responses to
hazards and improving the communication of risk information among lay
people, technical experts and decision makers. Risk managers who uphold

Introduction

Table 1.1 Variation in infection rate and degree of morbidity and mortality among
susceptible subpopulations

Pathogen

Susceptible
population

Incidence

Morbidity/
mortality

References

MAC

Adenovirus

Helicobacter
pylori

Calicivirus

AIDS/elderly

Immuno-
compromised
(e.g. transplant)

Immuno-
compromised
(e.g. AIDS)

Unidentified
genetic factor

Decreasing
in AIDS;
increasing
in elderly

Same as

general

population

Same as

general

population

Higher than

general

population

One of many
end-stage
infections in
AIDS; chronic
pulmonary
disease in
elderly

Greater risk of

severe/fatal

outcome

Same as

general

population

Higher rate
of illness/
reinfection
than general
population

Prince era/., 1989;
Reich and
Johnson, 1991 ;
Kennedy and
Weber, 1994;
Palellaefa/., 1998;
Havlir era/., 2000

Hierholzer, 1992;
Saad era/., 1997

Edwards era/.,
1991; Battan era/.,
1990

Moe era/., 1999

and regulate health and safety need to understand how people think about
and respond to risk. Experience shows that merely disseminating information
without reliance on communication principles can lead to ineffective health
messages and public health actions (Angulo et al. 9 1997; Owen et aL, 1999;
Harding and Anadu, 2000).

When the focus of risk communication to the public is to warn against a par-
ticular environmental health risk, the task is to notify as well as motivate people
to act to mitigate the risk. However, people tend to underestimate these sorts
of risks and fail to take any protective actions (Wiedemann and Schutz, 1999).
Also, targeting the messages to the right audience is challenging: one must both
identify the populations at risk and design appropriate outreach campaigns
(usually through brochures or media outlets); however, these methods may
reach a very small percentage of the intended audience. Even if the right audi-
ence attains the risk information, they may see themselves as personally not at
risk, or they may not understand what action to take. A California study (in
Harding and Anadu, 2000) of 900 consumers found that 80% of the respon-
dents did not take any action in response to public notification regarding
drinking water hazards; the author speculated that this resulted from the lack
of preventive measures detailed in the notification. In addition, resources are
often not available to those responsible for public outreach to construct and
carry out a carefully researched risk communication strategy.

There are many ways that formal risk communication is used as a part of
water use risk management, including fish consumption advisories, boil-water
notices and other regulatory provisions, which require public notification

Risk assessment and drinking water

when treatment violations occur. In the case of a boil-water advisory or a fish
consumption warning, there may be little opportunity for officials to interact
with the intended audience, and especially with a boil-water advisory, the tim-
ing of the message can be essential to the public's perception of trust in the
source. In Milwaukee, health officials delayed the release of information,
including the issuance of a boil-water order, even though the outbreak
appeared to be waterborne. This delay resulted in the public's outrage and
long-lasting distrust of government officials (Griffin et aL 9 1998; Sly, 2000).
Although quick action to alert communities may enhance trust in the source,
it still may not have any effect on people's willingness to adopt the risk-
reduction behaviour (O'Donnell et al., 2000); 25% of people failed to boil
their water after they had learned about an alert because they did not believe
it to be true (Angulo et aL, 1997).

Most monitoring for specific pathogens in the water is done after there is evi-
dence suggesting an outbreak. The UK has instituted standard monitoring for
cryptosporidial oocysts in the drinking water supply, which has raised the ques-
tion of what to do when oocysts are detected, but no cases of illness are indi-
cated. The relationship between pathogens found in the course of regular water
monitoring and actual health risks is still very unclear, but health officials had to
develop guidelines on what sort of actions to take under those circumstances.
Depending on the follow-up information available, under these guidelines,
health officials have a spectrum of options available for public notification rang-
ing from taking no action to issuing a boil-water advisory (Hunter, 2000).

People given the same environmental risk information will come to differ-
ent conclusions based on their perceptions and values. Risk communication,
then, is challenging because of everyone's different interpretations. Risk miti-
gation usually depends on voluntary compliance and the ability to communi-
cate to all stakeholders is important to obtain compliance. Two challenges of
environmental risk communication are to identify specific groups that may be
at greater risk and to understand the information needs of all stakeholders.
Delivering the details in a way that people understand is important, but not
absolute. To be able to communicate the complexity of the different issues
effectively is more than challenging for the risk communicator and the temp-
tation to disseminate information without stakeholder interaction and assume
that people are satisfied still lingers. However, the risk assessment community
recognizes that the more interaction there is among the interested parties, the
better chance for success in the entire risk process.

References

Angulo, F.J., Tippen, S., Sharp, D.J. et al. (1997). A community waterborne outbreak of
salmonellosis and the effectiveness of a boil water order. Am J Public Hlth, 87(4):
580-584.

Barrell, R.A.E., Hunter, P.R. and Nichols, G. (2000). Microbiological standards for water
and their relationship to health risk. Commun Dis Public Hltb, 3(1): 8-13.

Introduction

Barwick, R.S., Levy, D., Craun, G.F. et al. (2000). Surveillance for waterborne-disease out-
breaks - United States, 1997-1998. MMWR, 49(4): 1-21.

Battan, R., Raviglione, M.C., Palagiano, A. et al. (1990). Helicobacter pylori infection in
patients with acquired immune deficiency syndrome. Am] G astro enter 'ol, 85: 1576-1579.

Codex Committee on Food Hygiene. (1999). Principles and guidelines for the conduct of
microbiological risk assessment. Rome: Codex Alimentarius Commission. CAC/GL-30
(1999).

Craun, G.F., Berger, P.S. and Calderon, R.L. (1997). Coliform bacteria and waterborne
disease outbreaks. JAWWA, 89(3): 96-100.

Edwards, P.D., Carrick, J., Turner, J. et al. (1991). Helicobacter pylori-associated gastritis is
rare in AIDS: antibiotic effect or a consequence of immunodeficiency? Am J Gastroenterol,
86: 1761-1764.

Environmental Protection Agency. (1998). National Primary Drinking Water Regulations:
Interim Enhanced Surface Water Treatment; Final Rule. Federal Register, 63(241):
69477-69521.

Gale, P. (1996). Developments in microbiological risk assessment models for drinking
water - a short review. / Appl Bacteriol, 81(4): 403-410.

Gibson, C.J., Haas, C.N. and Rose, J.B. (1998). Risk assessment of waterborne protozoa:
current status and future trends. Parasitology, 117(Suppl.): S205-S212.

Gostin, L.O., Lazzarini, Z., Neslund, V.S. et al. (2000). Water quality laws and waterborne
diseases: Cryptosporidium and other emerging pathogens. Am J Public Hlth, 90: 847-853.

Griffin, R.J., Dunwoody, S. and Zabala, F. (1998). Public reliance on risk communication
channels in the wake of a Cryptosporidium outbreak. Risk Anal, 18(4): 367-375.

Harding, A.K. and Anadu, E.C. (2000). Consumer response to public notification.
JAWWA, 92(8): 32-41.

Havlir, D.V., Schrier, R.D., Torriani, F.J. et al. (2000). Effect of potent antiretroviral ther-
apy on immune responses to Mycobacterium avium in human immunodeficiency virus-
infected subjects./ Infect Dis, 182(6): 1658-1663.

Hierholzer, J.C. (1992). Adenovirus in the immunocompromised host. Clin Microbiol Rev,
5: 262-274.

Horsburgh, C.R. (1991). Mycobacterium avium complex infection in the acquired
immunodeficiency syndrome. New Engl J Med, 324: 1332-1338.

Hunter, P.R. (2000). Advice on the response from public and environmental health to the
detection of cryptosporidial oocysts in treated drinking water. Commun Dis Public
Hlth, 3(1): 24-27.

Kennedy, T.P. and Weber, D.J. (1994). Nontuberculous mycobacteria. An underappreci-
ated cause of geriatric lung disease. Am J Respir Crit Care Med, 149: 1654-1658.

Moe, C, Rhodes, D., Pusek, S. et al. (1999). Determination of Norwalk virus dose-
response in human volunteers. Presented at health Effects Stakeholder Meeting for the
Stage 2 DBPR and LT2ESWTR, February 12, 1999. Washington, DC.

National Research Council (1983). Risk Assessment in the Federal Government: Managing
the Process. Washington, DC: National Academy Press.

National Research Council (1994). Science and judgement in Risk Assessment.
Washington, DC: National Academy Press.

Neumann, D.A. and Foran, J. (1997). Assessing the risks associated with exposure to water-
borne pathogens: An expert panel's report on risk assessment. / Food Prot, 60(11):
1426-1431.

O'Donnell, M., Piatt, C. and Aston, R. (2000). Effect of a boil water notice on behaviour in the
management of a water contamination incident. Commun Dis Public Hlth, 3(1): 56-59.

Owen, A.J., Colbourne, J.S., Clayton, C.R.I, etal. (1999). Risk communication of hazardous
processes associated with drinking water quality - a mental models approach to customer
perception, Part 1 - a methodology. Water Sci Technol, 39(10-11): 183-188.

Palella, F.J. Jr, Delaney, K.M., Moorman, A.C. et al. (1998). Declining morbidity and mor-
tality among patients with advanced human immunodeficiency virus infection. HIV
Outpatient Study Investigators. New Engl J Med, 338: 853-860.

Parkin, R.T. and Balbus, J.M. (2000). Variations in concepts of 'susceptibility' in risk assess-
ment. Risk Anal, 20(5): 603-611.

Risk assessment and drinking water

Pontius, F.W. (2000). Denning sound science. JA WWA, 92(10): 16-20, 92.
Presidential/Congressional Commission on Risk Assessment and Management. (1997).

The Presidential/Congressional Commission on Risk Assessment and Risk

Management. Vol. 1. Washington, DC.
Prince, D.S., Peterson, D.D., Steiner, R.M. et al. (1989). Infection with Mycobacterium avium

complex in patients without predisposing conditions. New Engl J Med, 321: 863-868.
Proctor, M.E., Blair, K.A. and Davis, J.P. (1998). Surveillance data for waterborne illness

detection: an assessment following a massive waterborne outbreak of Cryptosporidium

infection. Epidemiol Infect, 120: 43-54.
Reich, J.M. and Johnson, R. (1991). Mycobacterium avium complex pulmonary disease.

Incidence, presentation, and response to therapy in a community setting. Am Rev

RespirDis, 143: 1381-1385.
Risk Policy Report. (2001). EPA Adopts Risk Approach to Potential Drinking Water

Contaminants. January 22.
Rose, J.B. and Gerba, C.P. (1991). Use of risk assessment for development of microbial

standards. Water Sci Tecbnol, 24(2): 29-34.
Saad, R.S., Demetris, A.J., Lee, R.G. et al. (1997). Adenovirus hepatitis in the adult allograft

liver. Transplantation, 64: 1483-1485.
Sly, T. (2000). The perception and communication of risk: a guide for the local health

agency. Can J Public Health, 91(2): 153-156.
Tillett, H.E., deLouvois, J. and Wall, P.G. (1998). Surveillance of outbreaks of waterborne

infectious disease: categorizing levels of evidence. Epidemiol Infect, 120: 27-42.
Wiedmann, P.M. and Schutz, H. (1999). Risk communication for environmental health

hazards. Zentralbl Hyg Umweltmed, 202: 345-359.
World Health Organization Working Group. (2000). Evaluation and use of epidemiological

evidence for environmental health risk assessment: WHO guideline document. Environ

Hltb Perspect, 108(10): 997-1002.

Part 2

Bacteriology

Acinetobacter

Basic microbiology

Acinetobacters are strictly aerobic, short and plump rod-shaped bacteria, often
capsulated and classified as Gram-negative but often typified as being 'Gram-
variable' when present in a pure culture. Morphologically, Acinetobacter rods
are 1-1.5 |xm in diameter and 1.5-2.5 jjim in length, becoming coccoid
(0.6-0.8 jxm X 1.0-1.5 |xm) in shape during the stationary phase of growth.
Acinetobacter are catalase-positive, oxidase-negative and non-spore forming.
While Acinetobacter exhibit so called 'twitching motility' as a result of the
presence of fimbriae (polar), as a group they are non-motile. Regardless of
many medical strains of Acinetobacter having optimum growth conditions at a
temperature of 35 °C, environmental strains are able to grow over a wide tem-
perature range.

Acinetobacter occur frequently as part of the commensal flora of animals
and humans and as such are regularly contaminating patients in hospitals,
particularly those with bronchopneumonia and septicaemia.

Origin and taxonomy

The genus Acinetobacter, first identified in 1911 by Beijerinck, was originally
classified as Micrococcus calcoaceticus following its initial isolation from soil

Bacteriology

(Baumann et al. 9 1968). After 1911 at least 15 other 'generic' names had been
used to describe the organisms now classified as members of the genus. The
most documented ones have included Bacterium anitratum, Herellea vagini-
cola, Mima polymorpha, Achromobacter, Alcaligenes, C B5W, Moraxella
glucidolytica and Moraxella Iwoffii (Henriksen, 1973). It was a group of French
microbiologists who first proposed the genus Acinetobacter, which comprised
a collection of non-motile, Gram-negative, oxidase-positive (Moraxella) and
oxidase-negative saprophytes which could be distinguished from other bacteria
by their lack of pigmentation when grown on agar plates (Brisou and Prevot,
1954). It was not until 1971 that the Subcommittee on the Taxonomy of
Moraxella and Allied Bacteria recommended that the genus Acinetobacter
should include only the oxidase-negative strains (Lessel, 1971).

Up until recently the genus Acinetobacter was included in the family Neisseri-
aceae (Juni, 1984). After extensive taxonomic developments it was proposed
that members of the genus Acinetobacter should be classified in the new family
Moraxellaceae. To date the family includes Moraxella, Acinetobacter, Psy-
chrobacter and related organisms (Rossau et ah, 1991) which constitutes a dis-
crete phylometric branch in superfamily II of the Proteobacteria on the basis of
16S rRNA studies and rRNA-DNA hybridization assays (Rossau etal, 1989).

Historically, the genus Acinetobacter has undergone an extensive amount of
taxonomic reclassification. This reclassification of Acinetobacter has been based
on DNA-DNA homology studies. Presently, based on this method of classifica-
tion, there is evidence of at least 19 DNA-DNA homology groups. These have
been accepted as A. baumannii, A. calcoaceticus, A. haemolyticus, A. Iwoffii,
A. radioresistens, A. johnsonii and at present 13 unnamed genomic species.

In taxonomical terms the DNA G + C content of Acinetobacter has been cal-
culated at between 39 and 47 mol%.

Metabolism and physiology

The majority of the strains of Acinetobacter have no major growth factor, being
able to use a large number of organic carbon and energy sources. However,
Acinetobacter spp. fail to utilize carbohydrates, with most documented strains
being unable to use glucose as a carbon source (Juni, 1978). However, acid-
ification of certain sugars, including glucose, arabinose, cellobiose, galactose,
lactose, maltose, mannose, ribose and xylose, via an aldose dehydrogenase, has
been extensively documented. Most acinetobacters are unable to reduce nitrate
to nitrite but some strains can use both nitrate and nitrite as nitrogen sources
by means of an assimilatory nitrate reductase. Numerous Acinetobacter strains
are also documented as being able to metabolize many diverse compounds,
including aliphatic alcohols, some amino acids, decarboxylic and fatty acids,
unbranched hydrocarbons and many relatively recalcitrant aromatic compounds
such as benzoate, mandelate, n-hexadecane, cyclohexanol and 2,3-butanediol
(Towner et ah, 1991).

Acinetobacter

Acinetobacter calcoaceticus are oxidase-negative, catalase-positive and
indole-negative with some strains able to produce urease.

Clinical features

Acinetobacter are ubiquitous but there are significant population differences
between the genomic species found as part of the normal human flora in clin-
ical and other environments. In the general population, Acinetobacter appears
to be characterized by predominant groups of genomic species. These include
A. baumannii and Acinetobacter spp. 3, which can be isolated from the skin
and numerous body sites of infected or colonized patients, particularly in hos-
pital. Rarely they have been isolated from the non-hospitalized population. In
the non-clinical environments A. Iwoffii, A. johnsonii and Acinetobacter spp.
12 seem to be the predominant natural inhabitants of human skin.

Acinetobacter s are more increasingly being associated with nosocomial infec-
tions. These include septicaemia, urinary tract infections, eye infections, menin-
gitis, skin and wound infections, brain abscesses, lung abscesses, pneumonia and
endocarditis. A survey of nosocomial infections indicated that this organism
might be responsible for over 0.5% of endemic nosocomial infections, particu-
larly in critically ill patients, with 3-24% of pneumonia patients using mechan-
ical ventilation devices becoming infected with at least one Acinetobacter spp.
Mortality rates associated with nosocomial Acinetobacter infections are higher
than those for other bacterial species, apart from Pseudomonas aeruginosa.

In recent years, antibiotic-resistant strains of Acinetobacter have been recog-
nized as important pathogens involved in outbreaks of hospital infection, par-
ticularly in high-dependency or intensive care units. As they are being isolated
from a wide range of clinical specimens, including tracheal aspirates, blood cul-
tures, cerebrospinal fluid and pus, this constitutes a major public health concern.
As with all nosocomial-acquired organisms the control of antibiotic usage, to
minimize development of resistant strains, and good housekeeping practices
and effective isolation procedures of infected patients, are important control
factors for acinetobacters in hospitals. The most documented species which is
associated with nosocomial outbreaks is A. baumannii, although other genomic
species, particularly Acinetobacter spp. 3, A. johnsonii and A. Iwoffii, have
also been reported.

Pathogenicity and virulence

Acinetobacter are classified as low-grade primary pathogens stereotyped into the
broad expression 'opportunistic pathogen'. Comprising a number of virulence
mechanisms, Acinetobacter spp. have become involved in a vast array of clin-
ically acquired infections. The majority of these can be located above.

Bacteriology

The virulence mechanisms inherent to Acinetobacter include: the presence
of a polysaccharide capsule; adhesions (fimbriae and capsular polysaccharide),
used to adhere to human epithelial cells; the production of cytotoxic enzymes;
and the lipopolysaccharide (LPS) component of the cell wall and the presence
of lipid A. The production of an endotoxin in vivo has been acknowledged
and this may be responsible for the disease symptoms observed particularly
during acinetobacter septicaemia. Acinetobacter have been shown to obtain
iron from the human body, an important virulence determinant, aided by side-
rophores such as aerobactin, and iron-repressible outer-membrane receptor
proteins (Smith et aL, 1990; Actis et ai, 1993).

Treatment

Historical documented data have shown that all strains of Acinetobacter are
resistant to penicillin, ampicillin, first-generation cephalosporins and chloram-
phenicol, whereas carbenicillin, tetracyclines (particularly minocycline, oxy-
tetracycline and chlortetracycline) and aminoglycosides seemed to be effective
at killing this organism. This still generally remains the situation to date with
environmental isolates. However, with clinical isolates, it is now evident
that Acinetobacter is able to develop or acquire resistance to any new anti-
biotics it may be challenged with including broad-spectrum cephalosporins
and 4-fluoroquinolones. Because of this fact it has become essential with
Acinetobacter infections to guide antimicrobial management by antimicrobial
sensitivity tests.

Survival in the environment

Acinetobacters are ubiquitous, free-living saprophytes. They have been isolated
in soil, seawater, freshwater, estuaries, sewage, contaminated food and mucosal
and outer surfaces of animals and humans (Towner et al., 1991). Also included
in this list is the clinical environment where Acinetobacter have been isolated
in hospital sink traps, hospital floor swab cultures and in air samples in wards
(Towner et ai, 1991).

Acinetobacters have frequently been isolated on granular activated carbon
(GAC) and sand filters, in biofilms and point-of-use devices suggesting water
as an important mode of transmission. In groundwater Acinetobacter has been
detected in large numbers, approximating to a mean of 8 colony-forming
units (cfu)/100ml (range, <1-178). In one study, however, it was found that
in groundwater Acinetobacter constituted less than 0.1% of the heterotrophic
plate count (HPC) population (AWWA, 1999). Some studies, though have
reported that 54% of HPC isolates obtained from groundwater have been

Acinetobacter

acinetobacters. Despite this, of particular concern in water, particularly drink-
ing water, is the fact that acinetobacters have been isolated in a large number
of sites containing no coliforms, which suggests the inadequacy of coliforms
as an indicator of these organisms. Conventional treatment of water by coagu-
lation and filtration is known to remove between 80 and 99% (0.6-2 logs)
and 50 and 99.5% (0.3-2.3 logs), respectively, of bacteria. The effectiveness
of disinfection processes during the water-treatment process can vary apprecia-
bly depending on the disinfectant used and the relative resistance of the
organism; however, typical removals range from 99 to 99.99% (AWWA,
1999). In a number of studies of a chlorinated distribution system, Acinetobacter
has been found to be the most commonly isolated organism often comprising
over 5% of the total organisms identified. In studies conducted in Canada,
approximately 1-2% of organisms isolated from the distribution system were
identified as Acinetobacter.

Naturally-occurring Acinetobacter spp. have been observed to have inactiva-
tion rates similar to other heterotrophic bacteria, such as Moraxella, Aeromonas,
Pseudomonas, and Alcaligenes, when exposed to chloramines. However, some
studies have indicated that acinetobacters can develop increased resistance to
chlorine, chloramines and chlorine dioxide when grown under conditions
favouring cell aggregation (AWWA, 1999). Despite these problems to date there
are no regulatory guidelines for acinetobacters in drinking water.

Methods of detection

Acinetobacters can be readily isolated and cultivated on conventional ordinary
laboratory media without necessary growth factor requirements. Commonly
used laboratory media used for the isolation of Acinetobacter have included,
organic medium 79, mineral medium with crude oil, peptone yeast extract
medium, Trypticase soy agar with glycerol and trypticase phytone medium. The
selective and differential media, such as Sellers agar, Herella agar (contains
bile salts, sugars and bromocresol purple) (Mandel et al. 9 1964) and MacConkey
agar, have also been used for the isolation of Acinetobacter (AWWA, 1999). A
novel antibiotic-containing selective medium, Leeds Acinetobacter medium,
that combines selectivity with differential characteristics has also been described
for the improved isolation of Acinetobacter spp. (Jawad etal. 9 1994) from both
clinical and environmental sources.

In potable water environments, in order to differentiate Acinetobacter from
other normal heterotrophic organisms, Eosin-Methylene Blue Agar and mAC
agar has been used (AWWA, 1999). For the improved recovery of Acineto-
bacter spp. from the environment, samples can be enriched with the addition
of 20 ml of an acetate-mineral medium with 5 ml of a water sample or a fil-
tered 10% soil suspension followed by vigorous aeration at 30°C or at room
temperature (AWWA, 1999).

Bacteriology

Clinical isolates of Acinetobacter grow at 37°C with some being documented
as growing often up to 42°C. However, a temperature of 30°C has often been
recommended for the growth of Acinetobacter. On media such as nutrient agar
and trypticase soya agar, Acinetobacter are known to form smooth, sometimes
mucoid, pale yellow to greyish white colonies, about 1-2 mm in diameter.

Much research is now being focused upon the use of molecular fingerprint-
ing techniques, including analysis of plasmid profiles, restriction endonuclease
digestion and pulsed-field gel electrophoresis of total chromosomal DNA,
random amplified polymorphic DNA profiles, ribotyping, cell envelope and
outer-membrane protein profiles or multilocus enzyme electrophoretic typing
(Thurm and Ritter, 1993) for the profiling of Acinetobacter. In terms of epi-
demiology these methods have all been used successfully to investigate out-
breaks of infection associated with Acinetobacter, but no single system has so
far gained overall acceptance for typing Acinetobacter spp.

Epidemiology of waterborne outbreaks

There have been numerous hospital outbreaks of Acineto b act er-r elated infec-
tions reported in the literature. These have tended to occur on certain wards such
as neurosurgery, burns units and intensive therapy units. There have been a small
number of outbreaks where the strain isolated from clinical specimens was also
isolated from taps, though it was not quite clear whether the tap water was the
source of the outbreak or represented environmental contamination from col-
onized patients (Debast et al. 9 1996; Pina et aL, 1998). There have also been
outbreaks that have been associated with contamination and overgrowth of
humidifiers and aerators (McDonald et aL, 1998; Kappstein et aL, 2000).

In treated drinking water no outbreaks due to acinetobacters have been
acknowledged. However, outbreaks in hospital settings are well documented
and do constitute a cause for concern (Ritter et al., 1993; Pina et al., 1998).

Risk assessment

Health effects: occurrence of illness, degree of morbidity and mortality, prob-
ability of illness based on infection:

• Acinetobacter spp. can cause infection in virtually every organ system: sep-
ticaemia, urinary tract infections, eye infections, meningitis, skin infections,
pneumonia and endocarditis.

• Although Acinetobacter causes mainly opportunist infections in hospitalized
patients, community-acquired infections have been reported. It has been
estimated that 1% of nosocomial infections are known to be caused by
acinetobacters.

Acinetobacter

• Without disruption of normal host defence mechanisms, the role of Acine-
tobacter in human infection remains limited.

Exposure assessment: occurrence in source water, environmental fate, routes
of exposure and transmission:

• The acinetobacters are found ubiquitously in the environment and have been
isolated from fresh water, estuaries, sewage, seawater, drinking water and
biofllms in drinking-water distribution systems. It is a very hardy organism -
especially on fomites and surfaces.

• Acinetobacter is primarily spread from person to person and from fomites,
such as medical equipment. It is possible that aerosolization is a route of
exposure.

• Drinking water has not been shown to be a transmission pathway for Acine-
tobacter, though it has been isolated from drinking water.

Risk mitigation: drinking-water treatment, medical treatment:

• Data indicate that Acinetobacter spp. are inactivated with chlorine disin-
fectant at similar rates to other heterotrophic bacteria, such as Moraxella,
Aeromonas and Pseudomonas. However, some studies have indicated that
acinetobacters can develop increased resistance to chlorine disinfectants.

• The impact of the presence of Acinetobacter in disinfection system biofilm
is unknown.

• Antibiotic resistance has hindered the medical management of Acinetobac-
ter infections, and the overall trend is one of increasing resistance. However,
mild to moderate infections can usually be treated successfully with monother-
apy and severe infections with multiple-antibiotic therapy.

References

Actis, J A., Tomasky, M.E. et al. (1993). Effect of iron-limiting conditions on growth of

clinical isolates of Acinetobacter baumannii.] Clin Microbiol, 31: 2812-2815.
AWWA (1999). Manual of Water Supply Practices: Waterborne Pathogens. Washington,

DC: Amercican Waterworks Association.
Baumann, P., Doudoroff, M. and Stanier, R.Y. (1968). A study of the Moraxella group. II.

Oxidase-negative species (genus Acinetobacter)./ Bacteriol, 95: 1520-1541.
Brisou, J. and Prevot, A.R. (1954). Etudes de systematique bacterienne. X. Revision des

especes reunies dans le genre Achromobacter. Ann Inst Pasteur (Paris), 86: 722-728.
Debast, S.B., Meis, J.F., Melchers, W.J. et al. (1996). Use of interrepeat PCR fingerprinting

to investigate an Acinetobacter baumannii outbreak in an intensive care unit. Scand

J Infect Dis, 28: 577-5 81.
Henriksen, S.D. (1973). Moraxella, Acinetobacter, and the Mimeae. Bacteriol Rev, 37:

522-561.
Jawad, A., Hawkey, P.M. et al. (1994). Description of Leeds Acinetobacter Medium, a new

selective and differential medium for isolation of clinically important Acinetobacter

spp., and comparison with Herellea agar and Holton's agar. / Clin Microbiol, 32:

2353-2358.
Juni, E. (1978). Genetics and physiology of Acinetobacter. Annu Rev Microbiol, 32: 349-371.

Bacteriology

Juni, E. (1984). Genus III. Acinetobacter Brisou et Prevot 1954. In Bergey's Manual of
Systematic Bacteriology, vol. 1, 9th edn, Krieg, N.R. and Holt, J.G. (eds). Baltimore:
Williams and Wilkins, pp. 303-307.

Kappstein, I., Grundmann, H., Hauer, T. et aL (2000). Aerators as a reservoir of Acineto-
bacter junii: an outbreak of bacteraemia in paediatric oncology patients. / Hosp Infect,
44: 27-30.

Lessel, E.F. (1971). Minutes of the Subcommittee on the Taxonomy of Moraxella and
Allied Bacteria. Int J Syst Bacteriol, 21: 213-21 4.

McDonald, L.C., Walker, M., Carson, L. et al. (1998). Outbreak of Acinetobacter spp.
bloodstream infections in a nursery associated with contaminated aerosols and air con-
ditioners. Pediatr Infect Dis J, 17: 716-722.

Mandel, A.D., Wright, K. and McKinnon, J.M. (1964). Selective medium for isolation of
Mima and Herellea organisms./ Bacteriol, 88: 1524-1525.

Pina, P., Guezenec, P., Grosbuis, S. et al. (1998). An Acinetobactr baumanii outbreak at the
Versailles Hospital Center. Pathol Biol (Paris), 46: 385-394.

Ritter, E., Thurm, V., Becker-Boost, E. et al. (1993). Epidemic occurrences of multiresistant
Acinetobacter baumannii strains in a neonatal intensive care unit. Xentralbl Hyg
Umweltmed, 193: 461-470.

Rossau, R., van den Bussche, G. et al. (1989). Ribosomal ribonucleic acid cistron similar-
ities and deoxyribonucleic acid homologies of Neisseria, Kingella, Eikenella, Alysiella,
and Centers for Disease Control Groups EF-4 and M-5 in the amended family
Neisseriaceae. Int J Syst B act eriol, 39: 185-198.

Rossau, R., van Landschoot, A. et al. (1991). Taxonomy of Moraxellaceae famnov., a new
bacterial family to accommodate the genera Moraxella, Acinetobacter and Psychrobac-
ter and related organisms. Int J Syst Bacteriol, 41: 310-319.

Smith, A.W., Freeman, S. et al. (1990). Characterization of a siderophore from Acineto-
bacter calcoaceticus. FEMS Microbiol Lett, 70: 29-32.

Thurm, V. and Ritter, E. (1993). Genetic diversity and clonal relationships of Acinetobac-
ter baumannii strains isolated in a neonatal ward: epidemiological investigations by
allozyme, whole-cell protein and antibiotic resistance analysis. Epidemiol Infect, 111:
491-498.

Towner, K.J., Bergogne-Berezin, E. and Fewson, C.A. (1991). The Biology of Acinetobac-
ter: Taxonomy, Clinical Importance, Molecular Biology, Physiology, Industrial Relevance.
New York: Plenum Press.

A eromonas

Basic microbiology

Aeromonas spp. are rod-shaped (0.3-1.0 [Jim by 1.0-3.5 |xm), Gram-negative,
non-spore forming, oxidase-positive, and facultatively anaerobic bacteria.
While commonly found in freshwater reservoirs, soil and agricultural pro-
duce, Aeromonas have also been isolated from the gastrointestinal contents
of fish, reptiles, amphibia and higher vertebrates. Occasionally Aeromonas,
which are pathogenic to humans, are found in the marine environment.

The genus is divided into two groups, namely the non-motile psychrophilic
aeromonads, which are pathogenic to fish, and the motile mesophiles, which
grow at a temperature range of 15-3 8°C. It is the mesophilic aeromonads,
namely Aeromonas hydrophila, Aeromonas caviae and Aeromonas sobria
which are of human significance and thus important to the water industry and
public health. Despite the mesophilc Aeromonas spp. being highlighted as pri-
mary agents involved in gastroenteritis, controversy still presides over their
pathogenicity and epidemiology. However, in light of the available literature,
it seems more likely that a number of the species within the genus, specifically
Aeromonas hydrophila, deserve recognition as a candidate involved in gastro-
enteritis and related diarroheal infections.

Bacteriology

Origin and taxonomy

The earliest documented evidence of the existence of the bacterium Aeromonas
occurred in 1891. Initially it was classified as Bacillus hydrophilus fuscus follow-
ing its isolation from the blood and lymph of an infected frog (Sanarelli, 1891).
Numerous reports of the presence of Bacillus hydrophilus fuscus occurred in
1897 (Chester, 1897) until its reclassification to Bacterium hydrophila in 1901
(Chester, 1901). As its name implies this was a bacterium that loved water.
During the decades that passed following its first acknowledgment Bacterium
hydrophila was isolated from a wide array of different animals ranging from
birds to reptiles. However, due to the lack of consensus on the name, Bacterium
hydrophila, many researchers misclassified it into different genera. This is appar-
ent when you consider the collection of groups to which Bacterium hydrophila
has been assigned during the many years of reclassification. These groups have
included Aerobacter, Proteus, Pseudomonas, Escherichia, Achromobacter,
Flavobacterium and Vibrio (Table 3.1).

It was not until 1936 that Kluyver and van Niel (1936) proposed the
genus Aeromonas, which literally meant c gas-producing unit'. Aeromonas
was endorsed in the seventh edition of Bergey's Manual of Determinative
Bacteriology. Incorporation of Aeromonas into the family Vibrionaceae in the
eighth edition of the manual occurred later (Schubert, 1974).

The acceptance of the genus Aeromonas, during its early developments as a
human pathogen, was generally slow despite small pockets of evidence emer-
ging in the 1930s suggesting a plausible link to disease. In spite of this, it took
over 20 years to conclude that some infections in humans were due to the
colonization and subsequent pathology associated with infection by Aeromonas
species. It was not until 1954 that a relationship between Aeromonas and
human disease was established. This evolved following a report compiled by

Table 3.1 Historical names given to the genus Aeromonas (adapted from Carnahan
and Altwegg, 1996)

Name Year of first description

Bacillus punctatus
Bacillus ranicida

Bacillus hydrophilus fuscus 1891

Bacterium punctatum 1 89 1

Aerobacter liquefaciens 1 900

Ba cillus hydrophilus Sana re Hi 1901
Bacillus (Proteus, Pseudomonas, Escherichia) ichthyosmius 1917

Achromobacter punctatum 1923

Pseudomonas (Flavobacterium) fermentans 1 930

Pseudomonas punctata 1 930

Proteus melanovogenes 1 936

Pseudomonas caviae 1 936

Pseudomonas formicans 1 954

Vib rio jamaicensis 1955

Aeromonas

Hill et al. (1954) suggesting that Aeromonas, isolated from a biopsy of a
women who had died, had possibly caused fulminant septicaemia with
metastatic myositis. Greater evidence of Aeromonas causing disease did not
transpire until about 1968 when infections caused by Aeromonas hydrophila
were apparent. This occurred because of an outbreak of infections in a hos-
pital in New Haven, Connecticut, USA where 27 cases of infection due to
Aeromonas hydrophila were documented. This paper appeared in the New
England journal of Medicine and suggested the greatest evidence to date of
the positive relationship between aeromonads and diarrhoea (von Graevenitz
and Mensch, 1968).

Within the three main groups of mesophilic Aeromonas, namely the species
caviae, hydrophila and salmonicida, multiple hybridization groups (HGs) are
known to exist, indicating the existence of phenospecies. Studies by Popoff
et al. (1981) and the Centers for Disease Control have shown 12 HGs are evi-
dent in the group Aeromonas. The type strains of named species are assigned
to specific HGs, as in the case of A. hydrophila (HG 1), A. salmonicida (HG 3),
A. caviae (HG 4), A. media (HG 5) and A. sobria (HG 7). Investigations using
DNA-DNA reassociation have identified a number of new species. These
included A. eucrenophila and HG 6, A. jandaei and HG 9, A. veronii and HG
10 and A. schubertii and HG 12. HGs 2 and 11 are currently unnamed with
HG 8 now suggested as a biotype of A. veronii. Two new Aeromonas species,
which do not correspond to any of the original 12 HGs, have also been proposed.
These are A. trota (Carnahan et al., 1991) and A. allosaccharophila.

To date there are 14 species in the genus Aeromonas with only nine of these
implicated in human disease (Janda, 1991) with 17 DNA hybridization groups
(Table 3.2). A further new species, A. popoffi (unassigned DNA hybridization
group), has also been proposed.

Metabolism and physiology

Aeromonads are chemo-organotrophs and have the ability to utilize a wide col-
lection of different sugars and carbon as a source of energy. Glucose is metab-
olized both aerobically and fermentatively with or without the production of
gas (CO2, H 2 ). Salt at high concentrations (6-7%) has been found to have an
inhibitory effect on the growth of Aeromonas, a useful characteristic involved
in bacterial isolation. Despite this, it is well documented that Aeromonas spp.
are able to tolerate pH values in the range 5-10 suggesting these organisms are
very resilient to the demands inflicted on them in unfavourable environments,
particularly those found in fresh waters. The mesophilic aeromonads, in par-
ticular Aeromonas hydrophila, are capable of utilizing a vast array of com-
pounds including amino acids, carbohydrates and carboxylic acids, peptides
and long-chain fatty acids. As a result of this ability Aeromonas are ideally
suited for growth in biofilms in distribution systems.

Bacteriology

Table 3.2 Genospecies and phenospecies of the genus Aeromonas (from WHO)

DNA

Reference strain

hybridization

(T = type strain)

Genospecies

Phenospecies

ATCC 7966T

A. hydrophila

ATCC51108T

A. bestiarum

A. hydrophila

ATCC 33658T

A. salmonicida

CDC 0434-84

A. salmonicida

A. hydrophila

ATCC 15468T

A. caviae

CDC 0862-83

A. media

A. caviae

CDC 0435-84

A. media

ATCC 23309T

A. eucrenophila

CIP7433T

A. sob ha

ATCC 9071

A. veronii

A. veronii biovar sobria

ATCC 49568T

A. jandaei

ATCC 35624T

A. veronii

ATCC 35941

Un-named

Aeromonas spp.
(ornithine-positive)

ATCC 43700T

A. schubertii

ATCC 43946

Un-named

Aeromonas Group 501

ATCC 49657T

A. trota

CECT4199T

A. allosaccharophila a

CECT 4342T

A. encheleia a

'The taxonomic status of A. allosaccharophila and A. encheleia remains to be confirmed.

Clinical features

Aeromonads have been documented as being involved in both intestinal and
extraintestinal human infections. The principal species of Aeromonas associated
with gastroenteritis are A. caviae, A. hydrophila and A. veronii biovar sobria
(Joseph, 1996), with the incidence of A. caviae more specifically evident in
young children under the age of three. Clinical data have suggested that strains
of A. hydrophila and A. sobria are inherently more pathogenic than A. caviae
(Janda etaL, 1994).

While certain strains of Aeromonas have been proposed as aetiological
agents of diarrhoeal disease, their role in specific disease causation still
remains inconclusive. Gastrointestinal conditions associated with Aeromonas
are usually self-limiting, but cases have been documented which show that
Aeromonas can produce a severe life-threatening cholera-like disease (Joseph,
1996). Despite a possible link between Aeromonas and gastrointestinal infec-
tions a large number of investigations have suggested that mesophilic
aeromonads are not primary enteropathogens.

Aeromonas spp. have also been associated with human wounds that are
often localized and mild, and known to cause septicaemia, respiratory tract
infections and a large array of systemic infections (Janda and Duffey, 1988;
Janda and Abbott, 1996; Nichols et al. 9 1996). The range of infections asso-
ciated with mesophilic Aeromonas can be seen in Table 3.3.

Aeromonas

Table 3.3 Occurrence of human infections associated with mesophilic Aeromonas a
(from WHO)

Type of
infection

Characteristics

Relative
frequency 13

Diarrhoea

Secretory
Dysenteric
Chronic
Choleraic

Systemic

Cellulitis
Myonecrosis
Erythema
gangrenosum
Septicaemia
Peritonitis
Pneumonia
Osteomyelitis
Cholecystitis
Eye infections

Acute watery diarrhoea, vomiting
Acute diarrhoea with blood and mucus
Diarrhoea lasting more than 10 days
'Rice water' stools

Inflammation of connective tissue

Haemorrhage, necrosis with/without gas gangrene

Skin lesions with necrotic centre, sepsis

Fever, chills, hypotension, high mortality

Inflammation of peritoneum

Pneumonia with septicaemia, sometimes necrosis

Bone infection following soft-tissue infection

Acute infection of gallbladder

Conjunctivitis, corneal ulcer, endophthalmitis

Very common
Common
Common
Rare

Common

Rare

Uncommon

Fairly common

Uncommon

Rare

a Modified from Janda and Duffey, 1988, and Nichols etal., 1996.

b Frequency of occurrence relative to all cases of Aeromonas infection.

It is highly probable that people are generally unaffected by enteric
Aeromonas and that aeromonads form a natural part of the gut microbiota.
However, we cannot rule out the fact that certain conditions including age,
immunocompetence, infection dose, and sufficient virulence factors of
Aeromonas spp. may aid in its ability to cause disease (Nichols et al., 1996).

Pathogenicity and virulence

While a number of virulence mechanisms for Aeromonas have been docu-
mented the pathogenesis of Aeromonas is still not fully understood (Gosling,
1996). This is probably due to the virulence mechanisms being multifactorial.
Many extracellular enzymes are produced by Aeromonas that may be import-
ant pathogenicity mechanisms. These include nucleases, cytolytic toxins,
stapholysin, lipases, sulphatases, lecithinase and amylase (Gosling, 1996;
Howard et aL, 1996).

In gastroenteritis, the chief Aeromonas virulence factor in the disease
process seems to be the enterotoxin (Janda, 1991). Cytolytic beta-haemolysin,
or aerolysin, produced by several different Aeromonas species, including
A. hydrophila and A. veronii, seems to be the most studied enterotoxin.
Aerolysins are part of a family of Aeromonas toxins that have similar structural
properties with dissimilar genetic, immunological and biological characteristics
(Kozaki et aL 9 1989). One enterotoxin, proaerolysin, has been shown to have

Bacteriology

a major impact on cell integrity by producing a transmembrane channel that
ultimately destroys the host cell (Tucker et al. 9 1990; Parker et al. 9 1994).
A second non-haemolytic toxin, identified in Aeromonas, produces a cytotonic
response in adrenal cells. The weak haemolysin produced by A. hydrophila
and A. salmonicida is called glycerophospholipid cholesterol acyltransferase
(GCAT) (Howard et al. 9 1996). One cytotonic enterotoxin with similar activ-
ity to cholera toxin has also been demonstrated (Ljungh et al. 9 1982; Gosling
et al. 9 1992; Gosling, 1996). There is also documented evidence for plasmid-
encoded expression by A. hydrophila and A. caviae of another cytotoxin. This
seems to be very similar to the Shiga-like toxin 1.

Aeromonas also possess an array of other virulence-mechanisms possibly
playing a role in gastrointestinal disease. These include, but are no means
exhaustive, adhesins, mucinases and mechanisms aiding in their ability to pene-
trate specific eukaryotic cells (Janda, 1991). Also documented are the S layers
(providing protection in the harsh environment and aiding possible resistance
to phagocytosis), beta-haemolysin activity (Chakraborty et al. 9 1987), resist-
ance to complement-mediated lysis (Janda et al. 9 1994), lipopolysaccharides
and proteases (Gosling, 1996).

All the virulence mechanisms mentioned above have been demonstrated in
A. hydrophila and A.veronii but not at present in A. caviae.

Treatment

Aeromonads are known to be ampicillin-resistant but susceptible to the
third generation cephalosporins, aminoglycosides, tetracycline, trimethoprim-
sulphamethoxazole, chloramphenicol and quinolones, carbapenems and mono-
bactams (Burgos etal. 9 1990). Resistance to beta-lactam antibiotics in Aeromonas
species is often documented.

Presence in the environment

Aeromonads can be isolated from virtually any freshwater source and poi-
kilothermic animals during the cold months of the year, when numbers are
relatively low. It is during the warmer months of the year that Aeromonas
numbers are often high and can be isolated from potable water sources. Levels
of Aeromonas are also found at very high densities, all year round, within
both domestic and industrial wastewater. This is due predominantly as a
result of faecal contamination and also high temperatures evident in these
situations.

The density of Aeromonas numbers varies substantially within different
environmental situations. In certain environments it has been suggested that
Aeromonas can exist at levels of >10 8 cfu/ml in sewage sludge, l-10 2 cfu/ml

Aeromonas

in lakes and reservoirs, 10 2 -10 7 cfu/ml in wastewater, 10 4 cfu/ml in rivers,
l-10 2 cfu/ml in drinking water and l-10cfu/ml in groundwater (Carnahan
and Joseph, 1991).

In raw sewage and many effluents the mesophilic Aeromonas species that
shows dominance is A. caviae (Ramteke et al., 1993). Evidence of A. caviae in
sewage treatment and ultimately the receiving waters has often led to proposals
for its use as an indicator of faecal pollutions. However, it has been noted that a
higher percentage of A. hydrophila and A. sobria are toxigenic compared with
A. caviae. Aeromonads in sewage effluents are of concern where such effluents
are used for irrigation of crops or are discharged into recreational waters.

Aeromonads are frequently isolated from surface waters with the most pre-
dominant species being Aeromonas hydrophila closely followed by Aeromonas
caviae. Low numbers of Aeromonas have been documented as occurring in
groundwater with maximum counts of 35cfu/100ml being reported in deep
aerobic and anaerobic groundwaters, even in the absence of coliforms
(Havelaar et al., 1990). Other evidence of the existence of Aeromonas in
groundwater has been documented (Huys et al., 1995). In this study two strains
of aeromonads isolated from two groundwaters in Belgium consisted solely of
strains belonging to hybridization groups 2 and 3 of the A. hydrophila complex.

Despite water treatment plants reporting mean reductions of 99.7% in
aeromonad numbers following flocculation-decantation and chlorination (Huys
et al., 1995; Kersters et al., 1995), low numbers of mesophilic Aeromonas are
often found in drinking waters (Havelaar et al., 1990). Surveys in London and
Essex, UK have shown isolation rates for A. hydrophila from chlorinated drink-
ing water of 25% in summer and 7% in the winter, compared with rates of
82% in the summer and 75% in the winter for untreated water. Researchers
have found that 90% of domestic water supplies in areas of Cairo contain
Aeromonas (Ghanem etal., 1993), while from a survey of three distribution sys-
tems in Sweden (Krovacek et al., 1992), it has been reported that 85% of sam-
ples were positive for Aeromonas (860cfu/100ml). A number of studies have
also been carried out looking at the occurrence of Aeromonas species in the met-
ropolitan water supplies of Perth, Western Australia (Burke et al., 1984). From
these studies a relationship between Aeromonas in chlorinated water with water
temperature and residual chlorine was demonstrated. Despite Aeromonas, like
the majority of organisms, entering the water-distribution system in very low
numbers, development in biofilms results in bacterial aftergrowth and possible
public health concerns (van der Kooij et al., 1995).

When it comes to controlling the numbers of Aeromonas spp. it has been
found, when using clinical and environmental strains of A. hydrophila,
A. sobria, A. caviae and A. veronii ( Knochel, 1991), the mesophilic aeromon-
ads are generally more susceptible to chlorine and monochloramine than the
Enterobacteriaceae (Medema et al., 1991). However, A. hydrophila, within
a mixed heterotrophic bacterial biofllm, are unaffected by the addition of
0.3mg/l monochloramine (Mackerness et al., 1991). From this study it was
also evident that the biofllm-associated A. hydrophila was able to also survive
0.6mg/l monochloramine.

Bacteriology

Methods of isolation and detection

At present there are no regulatory standards for the acceptable number of
aeromonads in drinking water both in Europe and the USA. However, it must
be assumed that if a water treatment process is functioning properly the
heterotrophic plate counts (HPCs) should not increase substantially. On the
whole heterotrophic plate counts should remain at levels below lOcfu/lOOml
in drinking water. Within Europe, guidelines as to the numbers of Aeromonas
in drinking water have been established. Here it is generally accepted that
levels of Aeromonas in drinking water should be no higher than 20 cfu/100 ml
when leaving the water treatment plant and no more than 200 cfu/100 ml in
distribution water.

Membrane filtration is commonly used for the enumeration of Aeromonas
from treated water. The most widely used medium for the recovery of
Aeromonas from drinking water is ampicillin-dextrin agar (ADA) (Havelaar
et al., 1987; Havelaar and Vonk, 1988). Ryan's Aeromonas medium (Holmes
and Sartory, 1993) has also been employed. As with many media employed
for the recovery of drinking-water bacteria these media contain selective
agents that are nutrient-rich, which may result in the recovery of low numbers
of some aeromonads (Gavriel and Lamb, 1995; Holmes et al., 1996).
As Aeromonas species are sensitive to the presence of copper at concentrations
as low as 10mg/l, a complexing agent (50mg/l sodium ethylenediamine tetra-
acetate, Na2EDTA) should therefore be added to samples (Versteegh et al.,
1989; Schets and Medema, 1993). This is particularly relevant during the
sampling of domestic households and municipal buildings that use copper
pipes to transport water. There is also evidence that pre-enrichment with
alkaline peptone water before subculturing to selective media has proved suc-
cessful for recovery of Aeromonas from water (e.g. well water) (Moyer
etal.,1992).

There have been a large number of different media used for the recovery of
Aeromonas from environmental waters. These have included m-aeromonas
agar (Rippey and Cabelli, 1979), ADA, starch-ampicillin agar (Palumbo et al.,
1985), pril-xylose-ampicillin agar (Rogol et al. 9 1979) and SGAP-10C agar
(Huguet and Ribas, 1991).

Because growth requirements are simple, aeromonads can be routinely
cultured on many non-selective, differential and selective agars. Since most
strains ferment sucrose, lactose or both sucrose and lactose, they can be
easily missed on media like MacConkey, Hektoen enteric and xylose-lysine-
desoxycholate (XLD) agars where they resemble non-pathogenic coli-
forms. The best differential/selective media for the isolation of pathogenic
aeromonads is blood agar and cefsulodin-Irgasan-novobiocin (CIN) agar
plates. Blood agar can be used as a screening medium for Aeromonas. Many
Aeromonas strains also produce beta-haemolysis on sheep blood agar, a
phenotypic characteristic useful in separating aeromonads from other enteric
bacilli.

Aeromonas

On non-selective agar media Aeromonas isolates typically appear as buff-
coloured, smooth, convex colonies after 24 hours incubation at 37°C. It has
been documented that most aeromonads are able to grow in media containing
up to 4% NaCl and over a pH range of 5-9 (Austin et al. 9 1989).

Primary identification of the genus Aeromonas is straightforward. Identifi-
cation to phenospecies or genospecies level is, however, more complex
(Millership, 1996). Molecular based technology using polymerase chain reac-
tion (PCR) (Khan and Cerniglia, 1997) has enabled the detection of A. caviae
and A. trota in water and environmental samples (Dorsch et ah, 1994).

Epidemiology

The health significance of detecting mesophilic aeromonads in public water sup-
plies is poorly understood. Early studies have correlated a possible link between
Aeromonas in drinking water and gastroenteritis (Burke et aL 9 1984; Picard and
Goullet, 1987). However, these studies did not adequately type strains. A num-
ber of studies have now reported subtyping of strains of Aeromonas isolated
from water and human cases (Havelaar et al. 9 1992; Hanninen and Siitonen,
1995; Borchardt et al. 9 2003). All the typing studies reported found little simi-
larity between human and environmental strains suggesting that Aeromonas in
drinking water does not constitute a risk to human health. This problem is com-
plex when we consider that Aeromonas may in fact be part of the normal micro-
biota of the human host. It is possible that Aeromonas may not even be a true
enteric pathogen.

Some epidemiological studies have been carried out that suggest people
have become colonized by Aeromonas from drinking untreated water.
However, the possible role of potable water implicated in the transmission of
infections from Aeromonas is still under discussion.

To date even though aeromonads are frequently isolated from drinking
water systems, and some strains may exhibit enterotoxic properties, there is
a need for further epidemiological studies. This will help to ascertain a rela-
tionship between cases of Aeromon as-associated diarrhoea and presence of
these organisms in drinking water.

Risk assessment

Health effects: occurrence of illness, degree of morbidity and mortality, prob-
ability of illness based on infection:

• Aeromon as-related gastroenteritis is generally self-limited, acute, watery
diarrhoea of a few days' to a few weeks' duration. Chronic diarrhoea of

Bacteriology

more than a few weeks and more serious sequelae, such as sepsis, usually
occur only in immunocompromised people. Disseminated, systemic disease
in the immunocompromised (especially those with underlying liver disease
or cancer) causes a high fatality rate. Soft-tissue and skin infection is related
to fresh water exposure.

• The probability of illness given infection is unknown for Aeromonas-related
gastroenteritis. Symptomatic and asymptomatic people shed Aeromonas in
their faeces.

Exposure assessment: routes of exposure and transmission, occurrence in
source water, environmental fate:

• Routes of exposure may include ingestion of contaminated water and food
and dermal contact with water or soil. Most infections are community-
acquired; some are nosocomial (especially in the immunocompromised).
The role of drinking water in transmission is under debate.

• Aeromonas spp. are found at high densities in all types of water: lakes,
rivers, marine environments, wastewater and chlorinated drinking water.
Because Aeromonas is also found frequently in all types of food, animals
and is ubiquitous in the environment, the role of water ingestion in the
development of gastroenteritis is unknown. Exposure to fresh water is the
primary risk factor associated with wound infection.

• The aeromonads occur at generally high concentrations in all types of
source water. River and marine water concentrations have been reported up
to 10 4 cfu/ml; lake concentrations have been reported up to l-10 2 cfu/ml;
and >10 8 cfu/ml in sewage sludge. Concentrations of l-10 2 cfu/ml have been
found in chlorinated drinking water.

• Aeromonas spp. are able to tolerate pH values in the range of 5-10, which
suggests that they are hardy in the environment, and particularly, in natural
waters.

• Survival and amplification in drinking water distribution is significant.
Aeromonas can colonize wells and distribution systems for months and
years. Substantial regrowth can occur after disinfection, and the aero-
monads colonize biofilm, which makes them resistant to disinfectant
residuals.

Risk mitigation: drinking-water treatment, medical treatment:

• Water treatment with filtration and chlorine disinfectant decreases concen-
trations significantly, but not necessarily completely. Water temperature,
contact time, and level of organic material in the source water are all related
to disinfection efficacy. Biofilm-associated A. hydrophila was shown to be
resistant to 0.6mg/l of monochloramine.

• Aeromonas is resistant to many antibiotics, including ampicillins; however,
many other antibiotics, such as third generation cephalosporins and tetra-
cycline, are efficacious in treating all kinds of infections, including chronic
gastroenteritis.

Aeromonas

References

Austin, D.A., Mcintosh, D. and Austin, B. (1989). Taxonomy of fish associated

Aeromonas spp., with the description of Aeromonas salmonicida subsp. smithia subsp.

nov. Syst Appl Microbiol, 11: 277-290.
Burke, V., Robinson, R., Gracey, M. et al. (1984). Isolation of Aeromonas bydropbila from

a metropolitan water supply: seasonal correlation with clinical isolation. Appl Environ

Microbiol, 48: 361-366.
Burgos, A., Quindos, G. et al. (1990). In vitro susceptibility of Aeromonas caviae,

Aeromonas bydropbila and Aeromonas sobria to fifteen antibacterial agents. Eur J Clin

Microbiol Infect Dis, 9: 413-417.
Carnahan, A.M. and Altwegg, M. (1996). Taxonomy. In Tbe Genus Aeromonas, Austin, B.

et al. (eds). London: Wiley, pp. 1-38.
Carnahan, A.M. and Joseph, S.W. (1991). Aeromonas update: new species and global

distribution. 3rd International Workshop on Aeromonas and Plesiomonas. Experienta,

47: 402-403.
Chakraborty, T., Huhle, B. et al. (1987). Marker exchange mutagenesis of the aerolysin

determinant in Aeromonas hydrophila. Infect Immun, 55: 2274-2280.
Chester, F.D. (1897). A preliminary arrangement of the species of the genus Bacterium

Contribution to determinative bacteriology. Part 1. 9th Annu Rep Delaware College

Agric Exp Sta, 92.
Chester, F.D. (1901). A Manual of Determinative Bacteriology. New York: Macmillan.
Dorsch, M. et al. (1994). Rapid identification of Aeromonas species using 16S rDNA

targeted oligonucleotide primers: a molecular approach based on screening of environ-
mental isolates. / Appl Bacteriol, 77: 722-726.
Gavriel, A. and Lamb, A.J. (1995). Assessment of media used for selective isolation of

Aeromonas spp. Lett Appl Microbiol, 21: 313-315.
Ghanem, E.H., Mussa, M.E. and Eraki, H.M. (1993). Aeromonas-associated gastro-enteri-

tis in Egypt. Zentralbl Mikrobiol, 148: 441-447.
Gosling, P.J. (1996). Pathogenic mechanisms. In Tbe Genus Aeromonas, Austin, B. et al.

(eds). London: Wiley, pp. 245-265.
Gosling, P.J. et al. (1992). Isolation of Aeromonas sobria cytotonic enterotoxin and beta-

haemolysin./ Med Microbiol, 38: 227-234.
Hanninen, M.L., Saimi, S. and Siitonen, A. (1995). Maximum growth temperature ranges

of Aeromonas spp. isolated from clinical or environmental sources. Microb Ecol,

29: 259-267.
Havelaar, A.H. and Vonk, M. (1988). The preparation of ampicillin dextrin agar for the

enumeration of Aeromonas in water. Lett Appl Microbiol, 7: 169-171.
Havelaar, A.H., During, M. and Versteegh, J.F.M. (1987). Ampicillin-dextrin agar medium

for the enumeration of Aeromonas species in water by membrane filtration. / Appl

Bacteriol, 62: 279-287.
Havelaar, A.H., Versteegh, J.F.M. and During, M. (1990). The presence of Aeromonas in

drinking water supplies in the Netherlands. Zentralbl Hyg, 190: 236-256.
Havelaar, A.H., Schets, F.M., van Silfhout, A. et al. (1992). Typing of Aeromonas

strains from patients with diarrhoea and from drinking water. / Appl Bacteriol, 72:

435-444.
Hill, K.R., Caselitz, F.H. and Moody, L.M. (1954). A case of acute metastatic myosistis

caused by a new organism of the family Pseudomonodaceae: a preliminary report.

Wind Med, 3:9-11.
Holmes, P. and Sartory, D.P. (1993). An evaluation of media for the membrane filtration

enumeration of Aeromonas from drinking-water. Lett Appl Microbiol, 17: 58-60.
Holmes, P., Niccolls, L.M. and Sartory, D.P. (1996). The ecology of mesophilic Aeromonas

in the aquatic environment. In The Genus Aeromonas, Austin, B. et al. (eds). London:

Wiley, pp. 127-150.
Howard, S.P., Maclntyre, S. and Buckley, J.T. (1996). Toxins. In The Genus Aeromonas,

Austin, B. et al. (eds). London: Wiley, pp. 267-286.

Bacteriology

Huguet, J.M. and Ribas, F. (1991). SGAP-10C agar for the isolation and quantification of
Aeromonas from water. / Appl Bacteriol, 70: 81-88.

Huys, G., Gersters, 1., Vancanneyt, M. et al. (1995). Chemotaxonomic analysis and
genomic fingerprinting of Aeromonas sp. isolated from Flemish drinking water produc-
tion plants. Abstracts of the 5th International Aeromonas-Plesiomonas Symposium,
Edinburgh.

Huys, G., Gersters, L., Vancanneyt, M. et al. (1995). Diversity of Aeromonas sp. in Flemish
drinking water production plants as determined by gas-liquid chromatographic analy-
sis of cellular fatty acid methyl esters (FAMEs). / Appl Bacteriol, 78: 445-455.

Janda, J.M. (1991). Recent advances in the study of the taxonomy, pathogenicity, and infec-
tious syndromes associated with the genus Aeromonas. Clin Microbial, 4: 397-410.

Janda, J.M. and Abbott, S.L. (1996). Human pathogens. In The Genus Aeromonas,
Austin, B. et al. (eds). London: Wiley, pp. 151-173.

Janda, J.M. and Duffey, P.S. (1988). Mesophilic aeromonads in human disease: current
taxonomy, laboratory identification, and infectious disease spectrum. Rev Infect Dis,
10: 980-987.

Janda, J.M., Kokka, R.P. and Guthertz, L.S. (1994). The susceptibility of S-layer-positive
and S-layer-negative Aeromonas strains to complement-mediated lysis. Microbiology,
140: 2899-2905.

Joseph, S.W. (1996). Aeromonas gastrointestinal disease: a case study in causation? In The
Genus Aeromonas, Austin, B. et al. (eds). London: Wiley, pp. 311-335.

Kersters, I., Huys, G., Janssen, P. et al. (1995). Influence of temperature and process technol-
ogy on the occurrence of Aeromonas sp. and hygienic indicator organisms in drinking
water production plants. Presented at the Fifth International Aeromonas-Plesiomonas
Symposium, Edinburgh.

Khan, A. A. and Cerniglia, C.E. (1997). Rapid and sensitive method for the detection of
Aeromonas caviae and Aeromonas trota by polymerase chain reaction. Lett Appl
Microbiol, 24: 233-239.

Kluyer, A.J. and Niel, C.B. (1936). Prospects for a natural system of classification of bac-
teria. Zentralbl Bakteriol Parasitenk Intekionskr Hyg Abt II, 94: 369-403.

Knochel, S. (1991). Chlorine resistance of motile Aeromonas spp. Water Sci Technol, 24:
327-330.

Kozaki, S., Asao, T. et al. (1989). Characterization of Aeromonas sobria hemolysin by use
of monoclonal antibodies against Aeromonas hydrophila hemolysins. / Clin Microbiol,
27: 1782-1786.

Krovacek, K., Faris, A., Baloda, S.B. et al. (1992). Isolation virulence profiles of
Aeromonas spp. from different municipal drinking water supplies in Sweden. Food
Microbiol, 9: 215-222.

Ljungh, A., Eneroth, P. and Wadstrsm, T. (1982). Cytotonic enterotoxin from Aeromonas
hydrophila. Toxicon, 20: 787-794.

Mackerness, C.W., Colbourne, J.S. and Keevil, C.W. (1991). Growth of Aeromonas
hydrophila and Escherichia coli in a distribution system biofilm model. Proc UK Symp
Health-Related Water Microbiology. London: IAWPRC, pp. 131-138.

Medema, G.J., Wondergem, E., van Dijk-Looyaard, A.M. et al. (1991). Effectivity of chlor-
ine dioxide to control Aeromonas in drinking water distribution systems. Water Sci
Technol, 24: 325-326.

Millership, S.E. (1996). Identification. In The Genus Aeromonas, Austin, B. et al. (eds).
London: Wiley, pp. 85-107.

Millership, S.E., Barer, M.R. and Tabaqchali, S. (1986). Toxin production by Aeromonas
spp. from different sources. Med Microbiol, 22: 311-314.

Moyer, N.P. et al. (1992). Application of ribotyping for differentiating aeromonads
isolated from clinical and environmental sources. Appl Environ Microbiol, 58:
1940-1944.

Nichols, G.L. et al. (1996). Health significance of bacteria in distribution systems - review
of Aeromonas. London: UK Water Industry Research Ltd (Report DW-02/A).

Palumbo, S.A. et al. (1985). Starch-ampicillin agar for the quantitative detection of
Aeromonas hydrophila. Appl Environ Microbiol, 50: 1027-1030.

Aeromonas

Parker, M.W., Buckley, J.T. et al. (1994). Structure of the Aeromonas toxin proaerolysin in
its water-soluble and membrane-channel states. Nature, 367: 292-295.

Picard, B. and Goullet, P. (1987). Epidemiological complexity of hospital Aeromonas infec-
tions revealed by electrophoretic typing of esterases. Epidemiol Infect, 98: 5-14.

Popoff, M.Y., Coynault, C, Kiredjian, M. et al. (1981). Polynucleotide sequence related-
ness among motile Aeromonas species. Curr Microbiol, 5: 109-114.

Ramteke, P.W., Pathak, S.P., Gautam, A.R. et al. (1993). Association of Aeromonas caviae
with sewage pollution./ Environ Sci Health, A28: 859-870.

Rippey, S.R. and Cabelli, V.J. (1979). Membrane filter procedure for enumeration of
Aeromonas hydrophila in fresh waters. Appl Environ Microbiol, 38: 108-113.

Rogol, M. et al. (1979). Pril-xylose-ampicillin agar, a new selective medium for the isol-
ation of Aeromonas hydrophila. J Med Microbiol, 12: 229-231.

Sanarelli, G. (1891). Ober einen neuen Mikroorganismus des Wassers, welcher fur Thiere
mit veriinderlicher und konstanter Temperatur pathogen ist. Zentralbl Bakt Parasitenk,
9:222-228.

Schets, F.M. and Medema, G.J. (1993). Prevention of toxicity of metal ions to Aeromonas
and other bacteria in drinking-water samples using nitrilotriaceticacid (NTA) instead of
ethylenediaminetetraaceticacid (EDTA). Lett Appl Microbiol, 16: 75-76.

Schubert, R.H.W. (1974). Genus 11. Aeromonas Kluyver and Van Niel, 1936. In Bergeys
Manual of Determinative Bacteriology, 8th edn, Buchanan, R.E. and Gibbons, N.E.
(eds). Baltimore: Williams and Wilkins, pp. 345-348.

Tucker, A.D., Parker, M.W. et al. (1990). Crystallization of a proform of aerolysin, a hole-
forming toxin from Aeromonas hydrophila. J Mol Biol, 212: 561-562.

van der Kooij, D., Veenendaal, H.R., Baars-Lorist, C. et al. (1995). Biofilm formation on
surfaces of glass and Teflon exposed to treated water. Wat Res, 29: 1655-1662.

van der Goot, F.G., Pattus, F. et al. (1993). Oligomerization of the channel-forming toxin
aerolysin precedes insertion into lipid bilayers. Biochemistry, 32: 2636-2642.

Versteegh, J. F.M. et al. (1989). Complexing of copper in drinking-water samples to
enhance recovery of Aeromonas and other bacteria./ Appl Bacteriol, 67: 561-566.

Von Graevenitz, A. and Mensch, A.H. (1968). The genus Aeromonas in human bacteriology.
N Engl J Med, 278: 245-249.

Arcobacter

Basic microbiology

Arcobacter are Gram-negative, non-spore forming rods, generally 0.2-0.9 fjim
wide by 1-3 fjim long. One example of this is Arcobacter butzleri which are
curved to S-shaped rods, 0.2-0.4 jxm wide by 1-3 fjim long, catalase-positive
or weakly positive. Under some conditions the cells of Arcobacter become
very long. These cells are usually motile with a darting, corkscrew-like move-
ment aided by an unsheathed single polar flagellum. Arcobacter are able to
grow over a diverse temperature range with evidence of growth documented
at 15°C, with viability also evident at 42°C. They are micro-aerophilic but do
not require hydrogen for growth and are aerotolerant at 30°C.

Origin of the organism

The genus Arcobacter was first described by Vandamme and others in 1991
(Vandamme et aL, 1991). It was a name given to a bacterium which was

Bacteriology

aerotolerant and able to grow at low temperatures. Originally called Campylo-
bacter cryaerophila (Neill et aL, 1985), it has since been renamed Arcobacter
cryaerophilus. It is an organism primarily associated with the abortion of pigs
and cattle but strains have now been associated with diarrhoea, mainly in chil-
dren in developing countries (Tee et aL, 1988; Taylor et aL, 1991; Kiehlbauch
et aL, 1991). Arcobacter cryaerophilus at present contains two subgroups. Some-
time after the first isolation of Arcobacter cryaerophilus, Arcobacter butzleri
(16 biotypes and 65 serotypes) was proposed as a species by Kiehlbauch
and coworkers, originally it was only a strain of Arcobacter cryaerophilus
(Kiehlbauch et aL, 1991). Arcobacter nitrofigilis was isolated later from the
roots of plants, it was found not to be associated with any human disease or
infection.

Initially all three species of Arcobacter were classified as Campylobacter spp.
To date Arcobacter are integrated into the family Campylobacteraceae and there
are currently four species, the final one being A. skirrowii.

Metabolism and biochemistry

Arcobacters are metabolically inactive with d-glucose and other carbohydrates
neither fermented nor oxidized. They are, however, able to hydrolyse indoxyl
acetate but not hippurate and able to reduce nitrite and produce indole and
hydrogen sulphide. All Arcobacter strains are oxidase-positive, catalase-positive
and urease-negative (except A. cryaerophila).

There are slight differences in the biochemistry of Arcobacter that aid in
their identification. In the case of A. butzleri, these have weak catalase activity,
are able to grow on MacConkey agar and in 8% glucose and reduce nitrate.
A. skirrowii, on the other hand, differs from A. butzleri, as it is unable to grow
on MacConkey agar and generally does not grow in the presence of 1.5% NaCl.
A. cryaerophila are not able to reduce nitrate. A. nitrofigilis can be differenti-
ated from other arcobacters by its nitrogenase activity and from A. cryaerophilus
and A. skirrowii in particular by its ability to grow in the presence of
1.5% NaCl.

Clinical features

A. butzleri is associated with diarrhoea in humans (Lerner et aL, 1994) and
animals, occasional systemic infection in humans, and abortion in cattle and
pigs. The prevalence of such organisms in abortion and enteritis in other
species of livestock is unknown and their pathogenic role has yet to be defined.
A. cryaerophilus is isolated from normal and aborted ovine and equine fetuses
and has been the cause of an outbreak of bovine mastitis. A. skirrow has been

Arcobacter

isolated from the preputial fluid of bulls, aborted bovine, porcine and ovine
fetuses, and the faeces of animals with diarrhoea.

Occasionally Arcobacter species can be isolated from blood and peritoneal
fluid of patients following acute appendicitis.

The role of A. butzleri as an enteric pathogen is not yet clearly established.

Treatment

At present all species of Arcobacter are susceptible to nalidixic acid, with vari-
able susceptibility to cephalothin.

Environment

A. cryaerophilus has been found in the faeces of domestic animals and occasion-
ally from humans, although some strains originally identified as this species
have since been identified as A. butzleri. Arcobacter butzleri is found in water,
sewage, poultry and other meats. Recent findings have shown that several bio-
types and serogroups of A. butzleri have been associated with human disease
as well as fresh poultry carcasses. Food may therefore be a possible source of
exposure to Arcobacter. A. nitrofigilis is a nitrogen -fixing bacterium found on
the roots of a small marsh plant but is not found in animals or humans.

There is no available evidence on the removal of Arcobacters by water-
treatment processes, though they are likely to be removed to a similar extent
as other bacteria. However, during a two-investigation study, while six strains
of C. jejuni and C. colt were isolated 100 strains of Arcobacter butzleri were
isolated from drinking water treatment plants in Germany (Jacob et al.,
1998).

If we consider the similarity of the genus Arcobacter with that of Campylo-
bacter then we could conclude that Arcobacter should be susceptible to
inactivation by disinfectants such as chlorine and ozone. However, this may
turn out not to be the case.

Isolation and detection

The primary isolation of Arcobacter involves a two-step motility enrichment in
a semisolid medium at 30°C, followed by culture on blood agar containing
a selective agent such as carbenicillin. Direct filtration onto enrichment media as
well as non-selective media has also been used.

Bacteriology

A primary isolation procedure based on swarming in a semisolid blood-free
selective medium at 24°C has been developed for the examination of meat
products. The selective agents used were piperacillin, cefoperazone, trimetho-
prim and cycloheximide, and these were incorporated in a basal medium
comprising Mueller Hinton broth and 0.25% agarose (Arcobacter Selective
Medium). Pure cultures of Arcobacter are then isolated from the swarming
zones which may extend 30-40 mm. Optimal conditions for the isolation of
Arcobacter spp. from clinical specimens have not been determined. Arcobacter
spp. are aerotolerant and have been recovered on certain selective media used
for Campylobacter, such as Campy-CVA.

Arcobacter strains grown on blood agar for 24 hours under microaerobic
conditions at 37°C form small, flat, watery, translucent beige to yellow, irregu-
lar colonies with a campylobacter-like appearance. A. nitrofigilis has a slightly
different appearance with typical colonies that are white and round. Growth
occurs at 15°C and 30°C but only slightly at 42°C. A. cryaerophilus colonies
on blood agar are domed, entire and yellowish in the case of subgroup 2 strains.
A. skirrowii colonies are flat, greyish and alpha-haemolytic on blood agar. It
has been observed that most aerotolerant strains, except A. butzleri, grow
weakly on the common blood agar bases, although recently brain heart infu-
sion agar containing 0.6% yeast extract and 10% blood agar has been used
successfully for routine culturing.

PCR based identification assays (targetted at 16S ribosomal RNA sequences)
have been developed to differentiate Arcobacter species from other closely
related Campylobacters, including Campylobacter and Helicobacter. Biotyping
of A. butzleri can be performed using four biochemical tests and serotyping by
slide agglutination based on heat labile antigenic factors of live bacteria using
absorbed and unabsorbed antisera.

Epidemiology

The exact route as to how humans become infected with Arcobacter is at pre-
sent unknown. It possibly could be due to the consumption of or contact with
contaminated water. However, there is documented evidence which suggests
that Arcobacter may be transmitted by food. One example of this occurred
when A. butzleri-like organisms had been isolated from retail meat products,
in particular poultry meat.

The epidemiology of Arcobacter infections is not well understood. However,
A. butzleri has been isolated from urban drinking water, river water, faeces and
sewage. Isolation from a drinking water reservoir in Germany has also been docu-
mented. It is likely that consumption of contaminated water is a source of expos-
ure to arcobacters, but there is no direct evidence of this to date. This seems
probable if we consider its pathogenic potential (Vandamme etaL 9 1991, 1992).

Arcobacter

Risk assessment

Health effects: occurrence of illness, degree of morbidity and mortality, prob-
ability of illness based on infection:

• Three Arcobacter species have been recovered from humans: A. butzleri,
A. cryaerophilus, and A. skirrowii.

• A. butzleri has most often been associated with human illness - diarrhoea
and occasional systemic infection. A. butzleri has been isolated from
patients with chronic diarrhoea and stomach cramps (Lerner et al., 1994).

• Occasionally Arcobacter species can be isolated from blood and peritoneal
fluid of patients following acute appendicitis.

• The role of A. butzleri as an enteric pathogen is not yet clearly established.

Exposure assessment: routes of exposure and transmission, occurrence in
source water, environmental fate:

• The routes of exposure and transmission in humans are unknown. Because
it has been found in sewage, surface water, drinking water treatment plants
and groundwater, contaminated water is a possible route; however, the preva-
lence of Arcobacters in meat, especially poultry, makes food transmission
probable. There is no direct evidence of either route of exposure.

• A. butzleri was found to attach easily to water-distribution pipe surfaces made
of stainless steel, copper and plastic, making regrowth in the distribution
system a possibility (Assanta et al., 2002).

• A. butzleri is found in water, sewage, poultry and other meats. Recent findings
have shown that several biotypes and serogroups of A. butzleri have been
associated with human disease. Arcobacter are able to grow over a diverse
temperature range with evidence of growth documented at 15°C, with via-
bility also evident at 42°C.

• A. cryaerophilus has been found in the faeces of domestic animals and
occasionally from humans.

• A. skirrowii is slow growing, so may be missed in sample analysis and
therefore overlooked as a human pathogen.

Risk mitigation: drinking-water treatment, medical treatment:

• We know little about the effects of drinking-water treatment on Arcobacter.
One study found A. butzleri to be sensitive to chlorine disinfection (Rice
etaU 1999).

• At present all species of Arcobacter are susceptible to nalidixic acid, with
variable susceptibility to cephalothin. A majority of A. butzleri isolates
were found resistant to antibiotics commonly used for the treatment of
infectious bacterial diseases in humans (Atabay and Aydin, 2002).

Bacteriology

References

Assanta, M.A., Roy, D., Lemay, M.J. et al. (2002). Attachment of Arcobacter butzleri, a

new waterborne pathogen, to water distribution pipe surfaces. / Food Prot, 65(8):

1240-1247.
Atabay, H.I. and Aydin, F. (2002). Susceptibility of Arcobacter butzleri isolates to 23

antimicrobial agents. Lett Appl Microbiol, 33(6): 430-433.
Jacob, J., Woodward, D., Feuerpfeil, I. et al. (1998). Isolation of Arcobacter butzleri in raw

water and drinking water treatment plants in Germany. Zentralbl Hyg Umweltmed,

201: 189-198.
Kiehlbauch, K.A., Brenner, D.J., Nicholson, M.A. et al. (1991). Campylobacter butzleri sp.

nov. isolated from humans and animals with diarrhoeal illness./ Clin Microbiol, 29: 376.
Lerner, J., Brumberger, V. and Preac-Mursic, V. (1994). Severe diarrhea associated with

Arcobacter butzleri. Eur J Clin Microbiol Infect Dis, 3(8): 660-662.
Neill, S.D., Campbell, J.N., O'Brien, J.J. et al. (1985). Taxonomic position of Campylobacter

cryaeropbila sp. nov. Inst J Syst Bact, 35: 342.
Rice, E.W., Rodgers, M.R., Wesley, I.V. et al. (1999). Isolation of Arcobacter butzleri from

ground water. Lett Appl Microbiol, 28(1): 31-35.
Taylor, D.N., Kiehlbauch, J. A., Tee, W. et al. (1991). Isolation of Group 2 aerotolerant

Campylobacter species from Thai children with diarrhoea./ InfDis, 163: 1062.
Tee, W., Baird, R., Dyall-Smith, M. et al. (1988). Campylobacter cryaerophila isolated

from a human./ Clin Microbiol, 26: 2469.
Vandamme, P., Falsen, E., Rossau, R. et al. (1991). Revision of Campylobacter,

Helicobacter and Wolinella taxonomy: emendation of generic descriptions and pro-
posal of Arcobacter gen. Nov. Inst J Syst Bacteriol, 41: 88-103.
Vandamme, P., Pugina, P., Benzi, G. et al. (1992). Outbreak of recurrent abdominal cramps

associated with Arcobacter butzleri in an Italian school./ Clin Microbiol, 30: 2335-2337.

Campylobacter

Basic microbiology

Campylobacter are Gram-negative, curved or spiral rods 0.2-0.4 |xm wide by
0.5-5 fjim long and non-spore forming. All Campylobacters are oxidase-positive,
catalase-positive, urease-negative and motile, by means of a single polar
unsheathed flagellum at one or both ends of the cell, except Campylobacter
gracilis (oxidase-negative and aflagellate). Campylobacters are 'microaerophilic'.
Generally Campylobacter require a 3-5% concentration of carbon dioxide,
a 3-15% concentration of oxygen and a temperature of 42°C for optimum
growth. The ability of Campylobacters to tolerate oxygen is thought to result
from the vulnerability of their strongly electronegative dehydrogenases to super-
oxides and free radicals, especially when they are in a resting state. The coccoid
form of Campylobacter, often formed in old cultures or cultures that have been
exposed to air, has been suggested as the Viable but non-culturable state'
(VBNC) of the organism (Cappelier and Federighi, 1998; Holler et al. 9 1998).
However, a large amount of controversy still presides over the VBNC state of
Campylobacter, particularly when we consider its colonization, pathogenicity
and transmission. The significance of this VBNC state for infection of animals,
and as the cause of disease for humans, is still very tentative. In a review by
Thomas et al. (1999) the statement was made, The virulence of VBNC forms

Bacteriology

should be considered to be equivalent to that of the culturable forms, with the
added risk that they are not detectable by conventional culturable methods.'
This statement has been fully backed up by research conducted in 1999 sug-
gesting that the VBNC state of Campylobacter has an important role to play in
the transmission of infection for a number of strains (Tholozan et al. 9 1999;
Lazaro et al., 1999; Cappelier et al. 9 1999).

Campylobacter are commensals in most animals and are only a major con-
cern in humans, predominantly C. jejuni and C. coli, which are known to
cause both food-borne diarrhoea and enterocolitis (Blaser, 2000).

Campylobacters that are of relevance to the water industry are ascribed to
the class of Campylobacters known as the 'thermophilic' group. This group is
generally composed of C. jejuni, C. coli and C. upsaliensis.

Origin and taxonomy

Over 100 years ago Theodor Escherich showed evidence of Campylobacter
enteritis (Escherich, 1886). He visualized 'Campylobacter' cultures in smears
made from the colonic contents of babies who had died of 'cholera infantum'.
In 1906, Campylobacter was first isolated from the uterine exudate of aborting
sheep (McFadyean and Stockman, 1913). In 1919, Smith isolated a similar
organism from fetuses of aborting cows. The isolated organism was named
Vibrio fetus (Smith and Taylor, 1919). In 1959, Florent showed that a form
of the infection known as bovine infectious infertility was due to a variety of
V. fetus transmitted from carrier bulls to cows during coitus. He named the
organism Vibrio foetus var. venerialis (now Campylobacter fetus subsp. venerealis).

In 1927, another microaerophilic vibrio was isolated from the jejunum of
calves with diarrhoea and later named Vibrio jejuni (Jones et al., 1931). In
1944, Doyle isolated a similar vibrio from pigs suffering from swine dysen-
tery; he named it Vibrio coli (Doyle, 1948). These became C. jejuni and
C. coli, respectively, with the formation of the genus Campylobacter proposed
by Sebald and Veron in 1963, although no examples of the original V. jejuni
or V. coli strains had at that time survived.

The true first isolations of the organism from the faecal samples of patients
with diarrhoea occurred in Australia in 1971. Other evidence emerged in
1972 (Dekeyser et al., 1972). In Europe, however, the extent and prevalence
of Campylobacter and its relevance to gastroenteritis did not occur until
1977. This was primarily due to the problems of isolating this organism.

Metabolism and physiology

Most species of Campylobacter produce catalase and, apart from C. jejuni
subsp. doylei, all reduce nitrate to nitrite, with C. jejuni the only species able to

Campylobacter

hydrolyse sodium hippurate. Campylobacters obtain their energy from amino
acids or tricarboxylic acid cycle intermediaries and are known not to utilize
sugars or produce indole, except C. gracilis. The possession of a complete cit-
ric acid cycle and a complex highly branched respiratory chain together with a
large number of regulatory functions enables Campylobacter to survive in a
large number of different niches. These will be discussed in more detail below.
Some species of Campylobacter are able to grow anaerobically in the presence
of fumarate, aspartate or nitrate, which act as electron acceptors (Veron et al.,
1981).

Clinical features

The principal symptom of infection by Campylobacter in humans is acute diar-
rhoea. The incubation period, following infection, ranges from 1 to 8 days with
2-3 days the more common time period. The infectious dose has been shown to
vary considerably, although infection has been caused by ingestion of a few hun-
dred organisms. Following infection, the onset of diarrhoea is usually sudden,
often preceded by a prodromal flu-like illness, acute abdominal pain, or both.
These symptoms often mimic the symptoms of appendicitis, resulting in incor-
rect diagnosis. The diarrhoea, often just self-limiting to the patient, may be pro-
fuse and watery, probably due to the production of a cholera-like enterotoxin,
or may be dysenteric and contain blood and mucus. In young children, mild
symptoms of watery, non-inflammatory diarrhoea are often seen (Ketley, 1997).
With cases of Campylobacter diarrhoea stools are often culture negative after
3 weeks. Asymptomatic carriage of Campylobacter is, however, a common fea-
ture following infection and a patient may disseminate Campylobacter, in the
faeces, for over 4 months.

Campylobacter enteritis, specifically C. jejuni, can be associated with com-
plications, although these are relatively rare. Such complications are reactive
arthritis and Guillain-Barre syndrome. These are known to occur in about
1-2% of individuals who have been infected with Campylobacter. Guillian-
Barre syndrome is an autoimmune disorder of the peripheral nervous system
leading to acute flaccid paralysis. The characteristics of this syndrome are rap-
idly progressing weakness of the limbs and respiratory muscles.

Pathogenicity and virulence

C. jejuni and C. coli are by far the most important members of the Campylobac-
teraceae of human significance, expressing a number of different virulence
factors (van Vliet and Ketley, 2001). However, studying Campylobacter spp. at
the molecular level has posed many problems to investigators, resulting in limited

Bacteriology

success in the identification of virulence factors. To date, the only identified viru-
lence factors that have been studied in any great detail are the flagellar genes and
antibiotic resistance genes (Taylor, 1992).

For Campylobacter to induce gastrointestinal problems, like most enteric
organisms, it must be able to colonize the intestines. For it to do this it must be
able to penetrate the mucus layer of the gut. Motility, aided by the flagella, com-
bined with its spiral shape allows Campylobacter to penetrate effectively the
mucus layer of the intestinal wall (Guerry et al. 9 1992). Motility and flagella are
very important determinants for attachment and invasion of Campylobacter
(Wassenaar et aL, 1991). The specific components of intestinal mucin known to
be chemotactic to Campylobacter include L-fucose and L-serine. Movement
towards these components may be important in Campylobacters' pathogenesis
of infection (Hugdahl et al. 9 1988; Takata et aL 9 1992). Once Campylobacter
has crossed the mucus layer it has to adhere to the host's epithelial cells and
then invade them. The adhesion of Campylobacter to these cells is mediated by
fimbriae and other adhesions, such as the proteins PEB1 and CadF (fibronectin
binding protein). Once Campylobacter is adhered to the host cell it then has
to internalize itself. The mechanism for this is not known at present. However,
the uptake of Campylobacter into the host cell appears to be dependent
upon bacterial protein synthesis. It is probable that once invasion of the
host cell is successful an inflammatory response is initiated (Konkel et al.,
1992). C. jejuni appear to produce proteins which may be important in inter-
nalization of the organism (Konkel et al., 1993). C. jejuni has also been
reported to produce cytotoxins and a cholera-like enterotoxin (Ruiz-Palacios
et al., 1992; Gillespie et aL, 1993).

Several investigators have examined the ability of C. jejuni to acquire iron,
that is essential for bacterial pathogenesis, from exogenous sources. As levels
of iron in host tissues are low, an organism that has the ability to complex iron
is at a selective advantage. C. jejuni does not appear to produce its own sidero-
phores to utilize iron. However, it is able to utilize exogenous sidero-
phores (ferrichrome and enterochelin) from other bacteria as iron carriers
(Baig etal., 1986).

Another important virulence factor, particularly in C. fetus, is its microcap-
sule, or S layer, which protects the bacteria from serum killing and phago-
cytosis (Blaser and Pei, 1993).

During the past decade, evidence has appeared showing an association of
Campylobacter infection with Guillain-Barre syndrome (GBS) (Mishu and
Blaser, 1993). Several studies have shown that the disease is particularly asso-
ciated with serogroup 0:19 (Penner) strains. Surface polysaccharides, as well
as flagella, following sialylation may be responsible for GBS as a result of
molecular mimicry.

Campylobacter, specifically C. jejuni and C. coli, also have oxidative stress
defence systems. The main superoxide stress defence in Campylobacter is
superoxide dismutase (sodB) and is involved in converting superoxides into
hydrogen peroxide. It is probable that this plays some role in the intracellular
survival of Campylobacter in the host (Pesci et al. 9 1994).

Campylobacter

Heat shock proteins are also evident as a virulence factor in Campylo-
bacter. These enable Campylobacter to survive the extremes of temperature it
may be associated with.

Campylobacter are also associated with the oral cavity and may play a role
in the pathogenesis of periodontal disease (Ogura et al. 9 1995).

Treatment

Campylobacter enteritis is generally self-limiting and requires no more than
fluid and electrolyte replacement. All species of Campylobacter show sensitiv-
ity to erythromycin, ciprofloxacin, metronidazole and fluoroquinones at pre-
sent but resistance to these agents is developing rapidly. To date C. jejuni and
C. coli are showing good resistance to penicillins and cephalosporins.

Survival in the environment and water

Campylobacters are widespread in the environment. They have been isolated
and identified in fresh and marine waters. Campylobacters have also been found
in high numbers in domestic sewage and undisinfected treated sewage effluents.
Numbers of Campylobacters in surface waters are usually low as opposed to the
high numbers detected in sewage effluent. In groundwater (4°C) Campylobacters
are able to survive for several weeks (Gondrosen, 1986). In aquatic envir-
onments C. jejuni appears to be the predominant species, more so than C. coli
or C. lari. In developed countries domestic animals, birds (caged), pigs, sheep
and cows have been suggested as sources of Campylobacter infections. The risk
to human health of the presence of Campylobacter in these environments is
unknown and remains so to date. This seems to be due to the fact that there is
still no clear route for the transfer of Campylobacter from the environment to
the consumer, apart from food of course. Little is known about the survival of
Campylobacter in the environment apart from a number of laboratory-based
studies used to simulate the environment (Koenraad et al., 1997; Beutling,
1998; Buswell et ai, 1998; Holler et aL, 1998). These studies have shown that
Campylobacter is only able to survive a few hours in adverse conditions indica-
tive of the environment with temperature being a major factor in Campylo-
bacter's survival. However, from these studies it was found that the survival of
Campylobacter was greater with decreasing temperature, reaching several days
at 4°C. In fact the survival of Campylobacter was substantially enhanced when
it was present with other organisms within a biofilm (Buswell et al., 1998).
Research has shown that Campylobacters can survive in water for many weeks,
even months, at temperatures below 15°C.

On the available evidence so far it is consensually accepted that fully treated
water, which is subjected to correct disinfection procedures, is regarded as

Bacteriology

being free from Campylobacter. If, however, Campylobacters are found in
chlorinated drinking water, they are present usually as a result of post-treatment
contamination. Saying this, however, much more research is needed on the
survival potential and the role of the VBNC state in Campylobacter in the
environment and in drinking water, particularly biofilms, that provide appro-
priate microniches for their proliferation.

In the environment, specifically water, Campylobacters are only found in
the presence of faecal streptococci and faecal coliforms (Carter et al., 1987;
Arvanitidou et al., 1995). When compared to the coliforms, it is generally
accepted that methodologies used to inactivate these coliforms, specifically in
drinking water, are fully effective against Campylobacters. This is further
accepted when we consider the numerous studies that have shown Campylo-
bacter to be very vulnerable to chlorine.

Methods of detection

A wide spectrum of media has been used in the isolation of Campylobacter,
although most have not been designed for water. In the 1980s most work on
the detection of Campylobacters in water and the environment used Preston
broth (a nutrient broth base containing lysed blood, trimethoprim, rifampicin,
polymyxin B and amphotericin B) together with a supplement known as FBP
(containing ferrous sulphate, sodium metabisulphite, and sodium pyruvate)
followed by plating on Preston agar.

As a general procedure, C. jejuni and C. colt detection in water samples
involves concentration using a 0.22-|xm filter. Following filtration the filter is
placed in non-selective broth containing FBP, supplemented with numerous
antibiotics, and incubated at 42°C for 4 hours. Some studies have shown that
the period of incubation should then be continued for another 24 hours.
Following this the broth is then plated onto Preston agar, in a microaerophilic
environment and incubated for 48 hours. Following the incubation stage
appropriate tests can then be performed.

In clinical terms, faecal specimens from patients who present with a gastro-
intestinal infection are analysed for the presence of Campylobacter. Faecal
samples are usually transported in Cary-Blair media. Maximum recovery of
Campylobacter spp. is achieved in a microaerobic atmosphere containing
approximately 5% 2 , 10% C0 2 and 85% N 2 . Campylobacter spp. may then
be detected by direct Gram-stain examination of faecal samples. In areas
where species other than C. jejuni and C. coli are common, a filtration method
using non-selective medium should also be used (Mishu Alios et al. 9 1995).
In the clinical laboratory most Campylobacters produce visible growth after
24 hours at 37°C. However, a further incubation period of 24 hours enables
the development of appropriate-sized colonies which are generally circular
and convex.

Campylobacter

Epidemiology and waterborne outbreaks

Faecal material, transported from farms, is the probable cause of ground-
water contaminated with Campylobacter (Pearson et aL, 1993; Stanley et aL,
1998; Jones, 2001). It is possible that the conditions found within subsurface
aquifers favour the survival of Campylobacter in these environments aiding in
their survival (Jones, 2001).

Waterborne outbreaks due to Campylobacter have arisen when individuals
have consumed untreated or contaminated water (Taylor et aL, 1983). In fact
Campylobacters have been the main cause of private water outbreaks in the
UK in water that had not been adequately disinfected, a prerequiste for
human consumption (Duke et aL, 1996; Furtado et aL, 1998). Campylobacter
outbreaks from private water supplies have been principally due to Campylo-
bacter jejuni HS50 PT35 (Anon, 2000), which is an organism rarely cultured.

Outbreaks due to Campylobacters have been associated with drinking water
(Sobsey, 1989; Skirrow and Blaser, 1992; Jones and Roworth, 1996; Pebody
et aL, 1997). In Canada, Campylobacter has been indirectly incriminated as a
cause of a waterborne outbreak of enteritis (Borczyk et aL, 1987) and is con-
sidered the most important bacterial agent in waterborne diseases in many
European countries (Strenstrom et aL, 1994; Furtado et aL, 1998). Campylo-
bacter has caused a large number of outbreaks in Sweden involving over 6000
individuals (Furtado et aL, 1998).

Campylobacter enteritis incidences on average are 60-100 cases per 100000
population per year. However, laboratory diagnosed cases of Campylobacter do
not represent a true picture of the true incidence of Campylobacter infections
(Kendall and Tanner, 1982). In the UK over 550 000 and in the USA 2.4 million
Campylobacter infections are reported annually (Sibbald and Sharp, 1985).

A large number of waterborne outbreaks of Campylobacter have been
reported in the literature, often affecting hundreds or even thousands of indi-
viduals (Pebody et aL, 1997) (Table 5.1). Despite a large number of outbreaks

Table 5.1 Campylobacter water (and unknown) outbreaks in
England and Wales (1992-1994) (Pebody etal,, 1997; Frost, 2001)

Vehicle

Setting

Evidence

Water

College

Microbiological

Water

College

Microbiological

Water

Community

Microbiological

Water

Adventure camp

Cohort

Water

Function

Descriptive

Water

College

Descriptive

Unknown

Home of elderly

Descriptive

Unknown

Residential school

Descriptive

Unknown

Dining centre

Descriptive

Unknown

Home for elderly

Descriptive

Unknown

Function

Descriptive

Bacteriology

due to Campylobacter the organism causing the outbreaks has seldom been
isolated. This lack of isolation is possibly due to sporadic occurrence or the
presence of viable but non-culturable Campylobacter. Based on these findings,
it seems logical to suggest that while drinking water may be a major risk factor
of Campylobacter enteritis, there is no evidence to date to indicate that Campylo-
bacters can survive in water-distribution systems.

Many typing systems including phage typing, biotyping (Bolton et al., 1984),
bacteriocin sensitivity, detection of preformed enzymes, auxotyping, lectin
binding, serotyping, multilocus enzyme electrophoresis, and genotypic methods
such as restriction endonuclease analysis, ribotyping and restriction analysis
of polymerase chain reaction (PCR) products have been devised to study the
epidemiology of Campylobacter infections (Patton and Wachsmuth, 1992). How-
ever, for routine epidemiological investigations combinations of phenotypic and
genotypic tests are being used to investigate the presence of Campylobacter
infections in the population (Salama et al., 1990; Khakhria and Lior, 1992;
Nachamkin et al. 9 1993, 1996).

Risk assessment

Health effects: occurrence of illness, degree of morbidity and mortality, prob-
ability of illness based on infection:

• In most industrialized countries, Campylobacter enteritis is the most fre-
quent form of acute infective diarrhoea. Laboratory reports give incidences
in the order of 50-100 cases per 100000 population per year.

• The principal symptom of infection by Campylobacter is acute diarrhoea.
The diarrhoea may be profuse and watery or may be dysenteric and contain
blood and mucus. The diarrhoea produced is usually self-limiting.

• Campylobacter enteritis can be associated with complications, although
these are relatively rare. One such complication is that of reactive arthritis
and Guillain-Barre syndrome which occurs in about 1-2% of individuals
who have been infected.

• Most patients who become infected with C. jejuni were previously healthy
and recover rapidly from infection. However, patients with C. fetus infec-
tions are usually immunocompromised with conditions such as chronic
alcoholism, liver disease, old age, diabetes mellitus and malignancies.

• C. fetus infections may cause intermittent diarrhoea or non-specific abdom-
inal pain without localizing signs.

• Not all Campylobacter infections produce illness. Two of the most import-
ant factors related to infection appear to be the dose of organisms reach-
ing the small intestine and the specific immunity of the host to the
pathogen.

• Volunteers rechallenged with the C. jejuni organism developed infection,
but were protected from illness. Also, in developing countries, where

Campylobacter

C. jejuni infection is hyperendemic, the decreasing case-to-infection ratio
with age suggests the acquisition of immunity.

Exposure assessment: routes of exposure and transmission, occurrence in
source water, environmental fate:

• Campylobacters are transmitted by the faecal-oral route; person-to-person
transmission is relatively uncommon; direct transmission from animals to
humans is relatively common; indirect transmission, through consumption of
contaminated food or water, is by far the most common route of infection.

• Many waterborne outbreaks of Campylobacter have been reported. Despite
these large outbreaks, the organism has seldom been isolated from the drink-
ing water supply. Sources have included surface water, unchlorinated water
storage tanks contaminated with bird faeces, groundwater contaminated by
surface run-off, and water mains contaminated by cross-connection.

• The infectious dose has been shown to vary considerably, although infec-
tion has been caused by ingestion of a few hundred organisms.

• Though they cannot grow in water, Campylobacters have been isolated and
identified in fresh and marine waters and are also found in high numbers in
domestic sewage and undisinfected treated sewage effluents.

• Campylobacters have been shown to survive in water at 4°C for many
weeks, but for only a few days at temperatures above 15°C.

• In water, Campylobacters are generally sensitive to adverse conditions, such
as heat, disinfectants and gamma-irradiation.

• While drinking water may be a risk factor, there is no evidence that Campylo-
bacters can colonize or survive in water-distribution systems, thus, con-
sumption of properly treated water is unlikely to result in infection.

Risk mitigation: drinking-water treatment, medical treatment:

• Treatment methodologies used to inactivate coliforms are fully effective
against Campylobacters. Campylobacter has been shown to be very vulner-
able to chlorine, more so than E. coli.

• Most infections are self-limited. However, antibiotic treatment is advised in
patients with high fever, bloody diarrhoea, or more than eight stools per day;
whose symptoms have not lessened or are worsening at the time the diagnosis
is made; or whose symptoms have lasted more than one week (Blaser, 2000).

References

Anon. (2000). Surveillance of waterborne disease and water quality, July to December

1999. Commun Dis Rep Wkly, 10: 65-67.
Arvanitidou, M., Stathopoulos, G.A., Constantinidis, T.C. et al. (1995). The occurrence of

Salmonella, Campylobacter and Yersinia spp in river and lake waters. Microbiol Res,

150: 153-158.
Baig, B.H., Wachsmuth, I.K. and Morris, G.K. (1986). Utilization of exogenous siderophores

by Campylobacter species./ Clin Microbiol, 23: 431-433.

Bacteriology

Beutling, D. (1998). Incidence and survival of Campylobacter in foods. Arch Lebensm, Hyg,
39: 53-62.

Blaser, M.J. (2000). Campylobacter jejuni and related species. In Mandell, Douglas, and
Bennett's Principles and Practice of Infectious Diseases, vol. 1, 5th edn. Philadelphia:
Churchill Livingstone, pp. 2276-2285.

Blaser, M.J. and Pei, Z. (1993). Pathogenesis of Campylobacter fetus infections: critical
role of high-molecular weight S-layer proteins in virulence. / Infect Dis, 167: 372-377 '.

Bolton, F.J., Holt, A.V. and Hutchinson, D.N. (1984). Campylobacter biotyping scheme of
epidemiological value./ Clin Pathol, 37: 677-681.

Borczyk, A., Thompson, S. et al. (1987). Water-borne outbreak of Campylobacter laridis-
associated gastroenteritis. Lancet, 1: 164-165.

Buswell, CM., Herlihy, Y.M., Lawrence, L.M. et al. (1998). Extended survival and persist-
ence of Campylobacter spp in water and aquatic biofilms and their detection by immuno-
fluorescent-antibody and rRNA staining. Appl Environ Microbiol, 64: 733-741.

Cappelier, J.M. and Federighi, M. (1998). Demonstration of viable but non culturable state
Campylobacter jejuni. Rev Med Vet, 149: 319-326.

Cappelier, J.M., Minet, J., Magras, C. et al. (1999). Recovery in embryonated eggs of
viable but nonculturable C. jejuni cells and maintenance of ability to adhere to HeLa
cells after resuscitation. Appl Environ Microbiol, 65: 5154-5157.

Carter, A.M., Pacha, R.E., Clarke, G.W. et al. (1987). Seasonal occurrence of Campylobac-
ter spp in surface waters and their correlation with standard indicator bacteria. Appl
Environ Microbiol, 53: 523-526.

Dekeyser, P., Gossuin-Detrain, M. et al. (1972). Acute enteritis due to related vibrio: first
positive stool cultures. / Infect Dis, 125: 390-392.

Doyle, L.P. (1948). The etiology of swine dysentery. Am J Vet Res, 9: 50-51.

Duke, L.A., Breathnach, A.S., Jenkins, D.R. et al. (1996). A mixed outbreak of Crypto-
sporidium and Campylobacter infection associated with a private water supply.
Epidemiol Infect, 116: 303-308.

Escherich, T. (1886). Beitrage zur Kenntniss der Darmbacterien. III. Ueber das Vorkommen
von Vibrionen im Darmcanal und den Stuhlgangen der sauglinge. (Articles adding to the
knowledge of intestinal bacteria. III. On the existence of vibrios in the intestines and
faeces of babies.) Munch Med Wochenschr, 33: 815-817.

Florent A. (1959). Les deaux vibrioses genitales de la bete bovine: la vibriose venerienne,
due a V. foetus venerialis, et la vibriose d'origine intestinale due a V. foetus intestinalis,
Proceedings of the 16th International Veterinary Congress, Madrid, 2: 953-957.

Frost, J.A. (2001). Current epidemiological issues in human campylobacteriosis. / Appl
Microbiol, 90: 85S-95S.

Furtado, C, Adak, G.K., Sturt, M. et al. (1998). Outbreaks of waterborne infections intes-
tinal disease in England and Wales, 1992-1995. Epidemiol Infect, 121: 109-119.

Gillespie, M.J., Haraszthy, G.G. and Zambon, J.J. (1993). Isolation and partial character-
ization of the Campylobacter rectus cytotoxin. Microb Pathog, 14: 203-215.

Gondrosen, B. (1986). Survival of thermotolerant Campylobacters in water. Acta Vet
Scand, 79: 1-47.

Guerry, P., Aim, R.A. et al. (1992). Molecular and structural analysis of Campylobacter
flagella. In Campylobacter jejuni: Current Status and Future Trends, Nachamkin, I.,
Blaser, M.J. and Tompkins, L.S. (eds). Washington, DC: American Society for Micro-
biology, pp. 267-281.

Holler, C, Witthuhn, D. and Janzen-Blunck, B. (1998). Effect of low temperatures on
growth, structure and metabolism of Campylobacter coli SP10. Appl Environ Microbiol,
64:581-587.

Hugdahl, M.B., Beery, J.T. and Doyle, M.P. (1988). Chemotactic behavior of Campylo-
bacter jejuni. Infect Immun, 56: 1560-1566.

Jones, F.S., Orcutt, M. and Little, R.B. (1931). Vibrios (Vibrio jejuni) associated with intes-
tinal disorders of cows and calves./ Exp Med, 53: 853-864.

Jones, I.G. and Roworth, M. (1996). An outbreak of Escherichia coli 0157 and campy-
lobacterosis associated with contamination of a drinking water supply. Public Hlth,
110:277-282.

Campylobacter

Jones, K. (2001). Campylobacters in water, sewage and the environment./ Appl Microbiol,
Symposium Supplement, 90: 68S-79S.

Kendall, E.J.C. and Tanner, E.I. (1982). Campylobacter enteritis in general practice./ Hyg,
88: 155-163.

Ketley, J.M. (1997). Pathogenesis of enteric infection by Campylobacter. Microbiology,
143:5-21.

Khakhria, R. and Lior, H. (1992). Extended phage-typing scheme for Campylobacter jejuni
and Campylobacter coli. Epidemiol Infect, 108: 403-414.

Koenraad, P.M.F.J., Rombouts, F.M. and Notermans, S.H.W. (1997). Epidemiological
aspects of thermophilic Campylobacter in water-related environments: a review. Water
Environ Res, 69: 52-63.

Konkel, M.E., Mead, D.J. and Cieplak, W. (1993). Kinetic and antigenic characterization
of altered protein synthesis by Campylobacter jejuni during cultivation with human
epithelial cells. / Infect Dis, 168: 948-954.

Konkel, M.E., Hayes, S.F. et al. (1992). Characteristics of the internalization and intracel-
lular survival of Campylobacter jejuni in human epithelial cell cultures. Microb Patbog,
13: 357-370.

Lazaro, B., Carcamo, J., Audicana, A. et al. (1999). Viability and DNA maintenance in
nonculturable spiral C. jejuni cells after long-term exposure to low temperatures. Appl
Environ Microbiol, 65: 4677-4681.

McFadyean, J. and Stockman, S. (1913). Report of the Departmental Committee
appointed by the Board of Agriculture and Fisheries to inquire into Epizootic Abortion.
Part III. Abortion in Sheep. London: HMSO.

Mishu Alios, B., Blaser, M.J. and Lastovica, A.J. (1995). Atypical Campylobacters and
related microorganisms. In Infections of the Gastrointestinal Tract, Blase, M.J., Smith,
P.D. etal. (eds). New York: Raven Press, pp. 849-865.

Mishu, B. and Blaser, M.J. (1993). Role of infection due to Campylobacter jejuni in the ini-
tiation of Guillain-Barre syndrome. Clin Infect Dis, 17: 104-108.

Nachamkin, I., Bohachick, K. and Patton, CM. (1993). Flagellin gene typing of Campylo-
bacter jejuni by restriction fragment length polymorphism analysis. / Clin Microbiol,
31: 1531-1536.

Nachamkin, I., Ung, H. and Patton, CM. (1996). Analysis of O and HL serotypes of
Campylobacter by the flagellin gene typing system./ Clin Microbiol, 34: 277-281.

Ogura, N., Shibata, Y. et al. (1995). Effect of Campylobacter rectus LPS on plasminogen
activator-plasmin system in human gingival fibroblast cells. / Per iodont Res, 30: 132-140.

Patton, CM. and Wachsmuth, I.K. (1992). Typing schemes: are current methods useful?
In Campylobacter jejuni: Current Status and Future Trends, Nachamkin, I., Blaser, M.J. and
Tompkins, L.S. (eds). Washington, DC: American Society for Microbiology, pp. 110-128.

Pearson, A.D., Greenwood, M. et al. (1993). Colonization of broiler chickens by water-
borne Campylobacter jejuni. Appl Environ Microbiol, 59: 987-996.

Pebody, R.G., Ryn, M.J. and Wall, P.G. (1997). Outbreaks of Campylobacter infection: rare
events for a common pathogen. Communicable disease report. CDR Rev, 7: 33-37.

Pesci, E.C, Cottle, D.L. and Pickett, CL. (1994). Genetic, enzymatic and pathogenic studies
of the iron superoxide dismutase of Campylobacter jejuni. Infect Immun, 62: 2687-2694.

Ruiz-Palacios, G.M., Cervantes, L.E. et al. (1992). In vitro models for studying Campylo-
bacter infections. In Campylobacter jejuni: Current Status and Future Trends,
Nachamkin, I., Blaser, M.J. and Tompkins, L.S. (eds). Washington, DC: American Society
for Microbiology, pp. 176-183.

Salama, S.M., Bolton, F.J. and Hutchinson, D.N. (1990). Application of a new phagetyping
scheme to Campylobacters isolated during outbreaks. Epidemiol Infect, 104: 405-411.

Sebald, M. and Veron, M. (1963). Teneur en bases de l'ADN et classification des vibrions.
Ann Inst Pasteur (Paris), 105: 897-910.

Sibbald, C.J. and Sharp, J. CM. (1985). Campylobacter infections in urban and rural popu-
lations in Scotland./ Hyg, 95: 87-93.

Skirrow, M.B. and Blaser, M.J. (1992). Clinical and epidemiological considerations.
In Campylobacter jejuni: Current Status and Future Trends, Nachamkin, I., Blaser, M.J.
and Tompkins, L.S. (eds). Washington, DC: American Society for Microbiology, pp. 3-9.

Bacteriology

Smith, T. and Taylor, M. (1919). Some morphological and biological characters of the
spirilla (Vibrio fetus, n. sp.) associated with disease of the fetal membranes in cattle.
/ Exp Med, 30: 299-311.

Sobsey, M.D. (1989). Inactivation of health related microorganisms in water by disinfec-
tion processes. Water Sci Technol, 21: 179-196.

Stanley, K.N., Cunningham, R. and Jones, K. (1998). Thermophilic Campylobacters in
groundwater. / Appl Microbiol, 85: 187-191.

Stenstrom, T.A., Boisen, R, Georgsen, F. et al. (1994). Vattenburna Infektioner I Norden
(Waterborne Outbreaks in Northern Europe). Copenhagen, Den: TemaNord, Nordisk
Minist.

Takata, T., Fujimoto, S. and Amako, K. (1992). Isolation of nonchemotactic mutants of
Campylobacter jejuni and their colonization of the mouse intestinal tract. Infect
Immun, 60: 3596-3600.

Taylor, D.N. (1992). Campylobacter infections in developing countries. In Campylobacter
jejuni: Current Status and Future Trends, Nachamkin, I., Blaser, M.J. and Tompkins, L.S.
(eds). Washington, DC: American Society for Microbiology.

Taylor, D.N., McDermott, K.T. et al. (1983). Campylobacter enteritis from untreated
water in the Rocky Mountains. Ann Intern Med, 99: 38-40.

Tholozan, J.L., Cappelier, J.M., Tissier, J.P. et al. (1999). Physiological characterization of
viable but nonculturable C. jejuni cells. Appl Environ Microbiol, 65: 1110-1116.

Thomas, C, Gibson, H., Hill, D.J. et al. (1999). Campylobacter epidemiology: an aquatic
perspective. / Appl Microbiol, Symposium Supplement, 85: 168S-177S.

van Vliet, A.H.M. and Ketley, J.M. (2001). Pathogenesis of enteric Campylobacter infec-
tion. / Appl Microbiol, 90: 45S-56S.

Veron, M., Lenvoise-Furet, A. and Beaune, P. (1981). Anaerobic respiration of fumarate
as a differential test between Campylobacter fetus and Campylobacter jejuni. Curr
Microbiol, 6: 349-354.

Wassenaar, T.M., Bleumink-Pluym, N.M. and van der Zeijst, B.A. (1991). Inactivation of
Campylobacter jejuni flagellin genes by homologous recombination demonstrates that
flaA but not flaB is required for invasion. EMBO J, 10: 2055-2061.

Cyanobacteria

Basic microbiology

Cyanobacteria are Gram-negative bacteria, formerly known as blue-green
algae, that are able to fix nitrogen in the dark. They range in size from 1 |xm
for unicellular cyanobacteria to over 30 fjim for multicellular species. They
perform oxygenic photosynthesis, deriving electrons from water to reduce
carbon dioxide to cellular material. During the warmer months of the year
cyanobacteria produce algal blooms, specifically in freshwater lakes and
reservoirs. The abundance and concentration of these blooms have been
increasing over the last decade, more specifically in the summer months. These
toxic blooms have been reported in many parts of Europe, the USA, Australia,
Africa, Asia and New Zealand. A survey in the UK has found that 75% of
cyanobacterial blooms contain toxins (Baxter, 1991; Carmichael, 1992).

Origin and taxonomy

In the USA, five genera of cyanobacteria have been identified as toxin pro-
ducers, including two strains of Anabaena flosaquae, Aphanizomenon

Bacteriology

flosaquae, Microcystis aeruginosa and Nodularia spp. World-wide, six genera
and at least 13 species, have been identified. Cyanobacteria were originally
classified into groups on the basis of morphology, namely the unicellular
cyanobacteria and the filamentous forms. This taxonomic principle was not
upheld by rDNA studies. One classically defined group that has withstood
rDNA analysis was the filamentous cyanobacteria that contained heterocysts,
specialized cells that fix nitrogen, referred to as Anabaena cylindrica. Heterocysts
protect the oxygen-labile enzyme nitrogenase, since, unlike vegetative cells, hete-
rocysts do not carry out oxygen-evolving photosynthesis and further protect the
nitrogenase by having an oxygen-impermeable glycolipid layer and a suite of res-
piratory enzymes.

There are about 25 species of cyanobacteria that have been associated with
adverse health effects (Gold et al., 1989).

Clinical features

While high densities of cyanobacteria may appear in the faeces of infected
animals, cyanobacterial illness is not infectious, but due to contact with or
consumption of various toxins (Carmichael, 1994). The exact symptoms experi-
enced following cyanobacterial poisoning depend on the route of exposure and
the nature and concentration of the particular toxin (Hunter, 1998). Follow-
ing recreational contact the most common reported features are allergic in
nature, acute dermatitis or a hay-fever-like syndrome characterized by rhinitis,
conjunctivitis and asthma. Gastroenteritis following recreation exposure is
generally mild and short lived. Atypical pneumonia has been described in one
outbreak (Turner et al., 1990).

Acute illness following consumption of drinking water contaminated by
cyanobacteria is more commonly gastroenteritis (Hunter, 1998). A subclinical
hepatitis has been reported in one study (Falconer et al., 1983) and there has
been one outbreak of severe systemic illness including hepatitis and nephro-
pathy that has been linked to cyanobacterial contamination of drinking water
(Byth, 1980; Bourke et al. 9 1983). An outbreak of hepatitis linked to cyano-
bacterial poisoning occurred in a dialysis unit; patients presented with visual
disturbances, nausea and vomiting and almost 50% of those affected died
(Jochimsen et al., 1998).

Endotoxic reaction in dialysis patients, characterized by chills, fever,
myalgia, nausea and vomiting, and hypotension, have also been described
(Hindman et al. 9 1975).

Probably the most concerning outcome is primary liver cell (PLC) carcinoma.
Most of the epidemiological evidence of an association comes from the south-
east coastal area of China where the mortality due to PLC is particularly high
(>30/100000 person years) (Yu, 1989, 1995; Yeh, 1989; Zhu et al. 9 1989). The
most significant risk factor for PLC is infection with either hepatitis B or C virus.

Cyanobacteria

However, in people who are infected with these viruses, drinking water from
ponds and ditches that are at increased risk of cyanobacterial contamination sub-
stantially increases the risk. Furthermore, the incidence of PLC in a community
declined when the source of water was changed from ponds to other wells.

Pathogenicity and virulence

The cyanobacteria do not invade the animal/human body but, as mentioned
above, some produce potent toxins. Cyanobacterial toxins are of three main
types: hepatotoxins, neurotoxins and lipopolysaccharide (LPS) endotoxins.

Several hepatotoxins have been described, though the most important are
microcystin and nodularin, low molecular weight cyclic peptide toxins.
Microcystin is a seven amino acid ring, two amino acids of which are unique
to microcystin, N-methyl-dehydroalanine (Mdha) and 3-amino-9-methoxy-
2,6,8-trimethyl-10-phenyldeca-4,6-dienoic acid (ADDA). Nodularin is a five
amino acid ring hepatotoxin. Following absorption across the ileum, hepato-
toxins are transported to the liver and then taken up by hepatocytes. These
toxins are highly potent inhibitors of protein phosphatases types 1 and
2A, being active at nanomolar concentrations (MacKintosh et al. 9 1990;
Matsushima etaL 9 1990; Yoshizawa etal., 1990; Runnegar etal., 1995). Protein
phosphatases play an essential role in various cellular processes and their
inhibition can lead to cell death and to mutagenic change, including cell growth
and tumour suppression. Following acute ingestion, experimental animals
develop weakness, anorexia, pallor of the mucous membranes, vomiting,
cold extremities and diarrhoea. When death occurs it may be within a few
hours due to intrahepatic haemorrhage and shock or a after a few days due to
hepatic insufficiency (Hunter et al., 1999). The LD 50 by intraperitoneal injec-
tion in mice for most hepatotoxins is in the range 60-70 |xg/kg body weight.
Microcystin has also been shown to be an extremely active tumour promoter
for primary liver cell cancer in both in vivo and in vitro models (Zhou and Yu,
1990; Nishiwaki-Matsushima et aL, 1992; Yu, 1995). In a two-stage hepatocar-
cinogenesis model in rats, the number of gamma-glutamyl-transferase (GGT)
altered foci after partial hepatectomy was increased in rats given pond/ditch
water to drink. This is strong evidence that a component of the pond water
was a cancer promoter.

The two main neurotoxins are the anatoxins and neosaxitoxin. Anatoxin-a
(antx-a), is a secondary amine, 2-acetyl-9-azabicyclo[4.2.1]non-2-ene, an
alkaloid neurotoxin and is reported to be the most potent nicotinic agonist
so far described. Antx-a will cause a depolarizing neuromuscular blockade
(Carmichael, 1992). In experimental animals, acute poisoning causes muscle
fasciculations, decreased movement, collapse, exaggerated abdominal breath-
ing, cyanosis, convulsions and death, due to respiratory failure (Hunter et al.,
1999). Death will occur within minutes of administration. The LD 50 (intra-
peritoneal in mice) is about 200 |xg/kg body weight.

Bacteriology

Neosaxitoxin is related to saxitoxin, the causative toxin in paralytic shell-
fish poisoning. These two toxins act by inhibiting nerve conduction by block-
ing sodium, but not potassium, transport across the axon membrane. Features
of acute administration include loss of coordination, twitching, irregular
breathing and death by respiratory failure (Hunter et al., 1999). The LD 50
(intraperitoneal in mice) is about 10 |xg/kg.

Cyanobacterial lipopolysaccharide (LPS) endotoxin differs from that found
in other Gram-negative bacteria in that it lacks phosphate in the lipid A core
(Keleti and Sykora, 1982). Cyanobacterial LPS is also somewhat less toxic
than enterobacteriaceal LPS, at least in animal studies. Although there has
been relatively little research into the pathological effects of cyanobacterial
LPS, it is possible that it may play a role in the allergic reactions and the gastro-
enteritis seen in people after contact with cyanobacterial material.

It is still not known to any large extent what factors are responsible for
toxin production by cyanobacteria. It is known that toxin production varies
substantially between strains and in the same strain over time (Hunter et al.,
1999). There remains a considerable amount of uncertainty about the factors
promoting the synthesis and release of cyanobacterial toxins. A number of
environmental factors have been proposed as affecting toxin production such
as temperature, light intensity and phosphate levels, though different studies
have often yielded conflicting results.

Survival in the environment and water

Cyanobacteria are not dependent on a fixed source of carbon and, as such, are
widely distributed throughout aquatic environments. These include fresh-
water and marine environments and in some soils. In fact, cyanobacteria are
found in the early stages of soil formation being associated with converting
bare rock or decomposing debris. Stagnant water, sediments and soil appear to
be the significant reservoirs for these organisms. In the environment, cyanobac-
teria are at a selective advantage over eukaryotes under adverse conditions,
as they can fix nitrogen.

No species in the cyanobacteria group is classed as a true pathogen, but cer-
tain freshwater strains, such as Anabaena and Microcystis, produce saxitoxin-
like neurotoxins in the waters in which they grow. This is usually not a problem
for humans, but farm animals can be poisoned by drinking water from ponds
containing dense cyanobacterial blooms. It has been hypothesized that these
toxins are produced by cyanobacteria to kill fish, thereby releasing nutrients.

Cyanobacteria naturally occur in stream sediments, slow-moving streams,
receiving waters for a variety of waste discharges and treatment effluents,
rural storm runoff, drainage canals, and marine waters. Massive growth of
these bacteria often occurs during the summer in surface waters. Densities of
500 cells or more per millilitre have been recorded at this time of the year.

Cyanobacteria

Polluted surface waters that become stagnant due to slow flows under sum-
mer drought conditions often support persisting populations of cyanobac-
teria. Growth of cyanobacteria is stimulated by high water temperatures and
high concentrations of inorganic nitrogen (N) and phosphorus (P). Their
capability to be a source of biological nitrogen fixation in soils and water is
also a significant contributor to long-term survival and an important role in
the ecological succession of microorganisms in the environment.

Control of cyanobacteria is a problem as research has shown that the tox-
ins can remain potent for days even after the organisms have been destroyed
by chlorination and copper sulphate (El Saadi et aL 9 1995). While there are
toxicity data obtained from mouse models, further research is needed on the
acute and chronic toxicity of cyanobacterial toxins and suitable methods need
to be developed for monitoring the types and concentrations of cyanobacter-
ial toxins in natural as well as treated drinking water.

Methods of detection

Direct microscopic examination of bloom material will allow identification of
the cyanobacterial species present. In many cases this will enable distinction
between those species that are potentially toxic and those that are not (Chorus
and Bartram, 1999). A semiquantitative estimate of cyanobacterial biomass
can also be made from a simple microscopic analysis. A more accurate quan-
titative method has been recommended. To culture cyanobacteria, water sam-
ples are first blended with glass beads or treated by ultrasound to break
filamentous forms prior to streaking on agar plates (AWWA, 1999). There are
a variety of selected mineral media available (D-medium, ASM-1, BG-11, and
WC) with incubation at 25°C under cool white fluorescent light. Some recal-
citrant cyanobacteria may not be freed easily of contaminants, thus, physical
and chemical separation schemes may be necessary.

There are also several methods for detecting cyanobacterial toxins in water
(Chorus and Bartram, 1999). Such methods include bioassays in mice, or the brine
shrimp (Artemia salina). There are also a range of analytical methods including
high pressure liquid chromatography (HPLC), gas chromatography (GC) and
liquid chromatography/mass spectroscopy (Chorus and Bartram, 1999).

Epidemiology of waterborne outbreaks

There have been numerous reports of poisonings of livestock, pets and
wildlife with waters containing cyanobacteria blooms (Hunter et al., 1999).
Compared to animal poisoning, reports of outbreaks of human illness have
been less common. Most reports of human illness have been associated with

Bacteriology

recreational contact (Hunter, 1998). Nevertheless, there have still been several
reports of illness associated with consumption of drinking water.

Most of the reported outbreaks associated with drinking water have been of
gastroenteritis. Zilberg (1966) reported a sharp annual increase in admissions
during winter from an area supplied by a reservoir that regularly suffered
from algal blooms in Salisbury, Rhodesia (now Harare, Zimbabwe) but not
in neighbouring populations. The peak incidence in cases corresponded with
the death of the algae in the affected reservoir. Lippy and Erb (1976) reported
an outbreak of gastroenteritis that affected an estimated 5000 people (62%
of the population) supplied by a single reservoir in Sewickley, Pennsylvania,
USA. On examination of the water reservoir, the remains of a recently dead
bloom of Schizothrix calicola were found. These two outbreaks suggest that
the main risk of gastroenteritis is associated with death of a cyanobacterial
bloom. Cell death is probably associated with the release of various toxins
into the water.

An outbreak of gastroenteritis affecting about 2000 people, of whom 88 died,
in Brazil was found to be associated with drinking water taken from a new dam
(Teixeira et al. 9 1993). The only abnormal results were high levels of Anabaena
and Microcystis in untreated dam water and the outbreak declined rapidly after
copper sulphate treatment of the dam water to reduce cyanobacterial counts.

In a prospective case-control study of people taking their drinking water
from the Murray River, cases of gastroenteritis were more likely to have
drunk chlorinated river water rather than rain or spring water compared to
controls (El Saadi et aL 9 1995). There was also a correlation between the
weekly mean log cyanobacterial cell counts in the river and the number of
patients presenting to medical practitioners with gastroenteritis.

There is also evidence that drinking water can be associated with hepatitis. An
outbreak of hepatitis illness affected 139 children and 10 adults of aboriginal
descent in the Palm Island Community, Queensland, Australia (Byth, 1980;
Bourke et aL, 1983). Illness was strongly correlated with consumption of the main
supply, which came from a dam that had suffered from a heavy algal bloom. The
outbreaks started 5 days after the dam had been treated with copper sulphate to
kill the algae. Evidence of subclinical hepatitis was found to be related to an algal
bloom on a drinking water reservoir when increased liver enzymes were detected
in blood samples taken for other purposes (Falconer et al. 9 1983).

There have been at least two outbreaks linked to contamination of water
used for renal dialysis. The first outbreak was of pyogenic reactions in a dialy-
sis centre in Washington DC, USA (Hindman et al. 9 1975). This was found
to be due to high levels of endotoxin in the potable water supply to the clinic,
which correlated in turn with a period of high blue-green algal bloom counts
in the supply reservoir. The second was an outbreak of hepatitis in a dialysis
centre in Brazil during February 1996 (Jochimsen et al. 9 1998). Of 130
patients attending the centre, 116 (89%) became ill and 50 patients died with
liver failure. Microcystin was detected in the water used for dialysis and in
samples from affected patients. The water had been taken from a local reser-
voir and not treated before delivery to the hospital.

Cyanobacteria

For utilities using surface water supplies, cyanobacteria are well known for
their association with taste-and-odour problems, often regarded as a matter
of aesthetics. In light of recent information on cyanobacteria, granular acti-
vated carbon (GAC) may be very important to toxin removal. Furthermore,
for those water systems using disinfection as the only surface water treatment,
there is always the threat of a seasonal passage of cyanobacteria and depos-
ition of their dead cells in the distribution pipe network. Such an occurrence
provides a source of assimilable organic carbon (AOC), which is a potential
nutrient for bacterial regrowth.

The US Environmental Protection Agency (USEPA) and European govern-
ments regulate neither cyanobacteria nor their metabolites, except under guide-
lines stating that drinking water must be potable. The World Health
Organization has established guideline values for the tolerable concentration of
microcystin in drinking water - a value of 1 jxg/1 was obtained (Fawell, 1993;
Falconer, 1994). Contrary to this a value of <0.5 |xg/l was determined by work
in Canada (Kuiper-Goodman et al., 1994). The engineering and water supply
department of South Australia has also developed interim guidelines for accept-
able numbers of cyanobcteria in water supplies (El Saadi et al. 9 1995).

Risk assessment

Health effects: occurrence of illness, degree of morbidity and mortality, prob-
ability of illness based on infection:

• There are about 25 species of cyanobacteria that have been associated with
adverse health effects.

• Cyanobacteria act primarily as hepatotoxins and neurotoxins, but can also
cause skin irritation. They are extremely potent toxins and, therefore, have
the potential to be fatal to humans. However, acute oral or dermal expos-
ures have not resulted in any known human deaths. Reported illnesses in
humans exposed to cyanobacterial toxins range from dermatitis and gastro-
enteritis to hepatitis and allergic reactions. Illness is self-limited.

• Outcomes from cyanobacterial toxins are mostly acute; however, the
microcystins and other toxins have been shown to be tumour promoters in
animal studies, and epidemiological evidence in humans suggests that
chronic exposure to microcystins in drinking water is associated with an
increase in hepatocellular cancer.

Exposure assessment: routes of exposure and transmission, occurrence in
source water, environmental fate:

• World-wide, the number of humans acutely affected by cyanobacterial
toxins is low compared with other waterborne contaminants. However,
because of decreasing water quality, the potential for an increase in inci-
dents is high.

Bacteriology

• Routes of exposure are primarily ingestion through drinking water or recre-
ational water contact, also, dermal exposure and possibly aerosolization.

• Most acute exposures result from recreational water use; low levels in
drinking water are associated with an increase in hepatocellular cancer in
certain exposed populations.

• Cyanobacteria are found in all types of water: lakes, rivers, marine environ-
ments, and drinking-water reservoirs. Surface waters that receive waste
effluents are at special risk for contamination. High water temperatures and
high concentrations of inorganic nitrogen and phosphorus stimulate growth.

• Surface water systems that use only disinfectant may get deposition of dead
cells in the distribution system with potential for regrowth.

• Cyanobacterial toxins are ubiquitous, though their occurrence is dependent
on the conditions that contribute to algal bloom formation. Concentrations
vary widely depending on the species of bloom and the stage of the bloom's
formation and deterioration. Toxin concentrations have been reported as
ranging from 0.2 /xg/1 to 8.5 mg/1.

• The toxic dose is unknown for humans. The no observed adverse effect level
(NOAEL) for mice dosed orally with microcystin-LR has been reported to
be 40 (xg/kg/day for 13 weeks. The NOAEL reported for mice dosed orally
with anatoxin-a has been reported to be 0.1 mg/kg/day. Intraperitoneal and
intranasal exposure is more potent than oral ingestion for both toxins.

Risk mitigation: drinking-water treatment, medical treatment:

• There is some question as to the efficacy of standard drinking water treat-
ment (e.g. coagulation, sedimentation, disinfection and filtration) for
removing all but large concentrations of cyanobacterial toxins, though cur-
rent methods are effective enough to prevent any acute effects.

• Evidence on the efficacy of chlorine on the microcystins is equivocal; chlor-
ine is ineffective on anatoxin-a. Activated carbon treatment appears to be
the best removal method for treated water.

• Preventing the formation of blooms in the source water is the best way to
assure cyanobacteria-free drinking water.

• Membrane filtration technology has the potential to remove virtually any
cyanobacteria or their toxins from drinking water.

• Efficacious medical treatment is unknown in an acute exposure; however,
antihistamines and steroids may be helpful for allergic reactions. If given in
a timely manner, activated charcoal or an emetic could have a positive effect
on the toxic response.

References

AWWA. (1999). Manual of Water Supply Practices: Waterborne Pathogens. Washington,

DC: American Water Works Association.
Baxter, P.J. (1991). Toxic marine and freshwater algae: an occuptational hazard? Br J Ind

Med, 49: 505-506.

Cyanobacteria

Bourke, A.T.C., Hawes, R.B., Neilson, A. et al. (1983). An outbreak of hepato-enteritis

(the Palm Island mystery disease) possibly caused by algal intoxication. Toxicon,

3(SuppL): 45-48.
Byth, S. (1980). Palm Island mystery disease. Med J Aust, 2: 40-42.
Carmichael, W.W. (1992). Cyanobacteria secondary metabolites - the cyanotoxins./ Appl

Bacteriol, 72: 445-459.
Carmicheal, W.W. (1994). The toxins of cyanobacteria. Sci Am, 270: 78-86.
Chorus, I. and Bartram, J. (1999). Toxic Cyanobacteria in Water. London: E & FN Spon.
El Saadi, O., Easterman, A.J., Camerson, S. et al. (1995). Murray River water, raised

cyanobacterial cell counts and gastrointestinal and dermatological symptoms. Med J

Aust, 162: 122-125.
Fawell, J.K. (1993). Toxins from blue-green algae: Toxicological Assessment of

Micro cystin-LR. Vol. 4. Microcystin-LR:13 week oral (gavage) toxicity study in the

mouse. Final Report. Medmenham, UK: Water Res. Cent., pp. 1-259.
Falconer, I.R. (1994). Health problems from exposure to cyanobacteria and proposed

safety guidelines for drinking water and recreational water. In Detection Methods

for Cyanobacterial Toxins, Codd, G.A., Jefferies, T.M., Keevil, C.W. et al. (eds).

Cambridge, UK: R. Soc. Chem., pp. 3-10.
Falconer, I.R., Beresford, A.M. and Runnegar, M.T.C. (1983). Evidence of liver damage by

toxin from a bloom of the blue-green alga, Microcystis aeruginosa. Med J Aust,

1:511-514.
Gold, G.A., Bell, S.G. and Brooks, W.P. (1989). Cyanobacterial toxins in water. Water Sci

Tecbnol, 21: 1-13.
Hindman, S.H., Favero, M.S., Carson, L.A. et al. (1975). Pyogenic reactions during

haemodialysis caused by extramural endotoxin. Lancet, ii: 732-734.
Hunter, P.R. (1998). Cyanobacterial toxins and human health. / Appl Bacteriol,

84(Suppl.): 35S-40S.
Hunter, P.R., Petersen, A., Merrett, H. et al. (1999). Investigations of toxins produced by

cyanobacteria: Literature review of the hazards and risks posed by freshwater

cyanobacteria to human and animal health. R&D Report Series No. 4. Johnstown

Castle Estate, County Wexford, Ireland: Environmental Protection Agency.
Jochimsen, E.M., Carmichael, W.W., An, J.S. et al. (1998). Liver failure and death after expos-
ure to microcystins at a haemodyalisis center in Brazil. New Engl J Med, 338: 873-878.
Keleti, G. and Sykora, J.L. (1982). Production and properties of cyanobacterial endo-
toxins. Appl Environ Microbiol, 43: 104-109.
Kuiper-Goodman, T., Gupta, S., Combley, H. et al. (1994). Microcystins in drinking water:

risk assessment and derivation of a possible guidance value for drinking water. In

Toxic Cyanobacteria: A Global Perspective, Steffensen, D.A. and Nicholson, B.C. (eds).

Salisbury, S. Australia: Aust. Cent. Water Qual. Res., pp. 17-23.
Lippy, E.C. and Erb, J. (1976). Gastrointestinal illness at Sewickley, Pa. JAWWA, 76:

60-70.
MacKintosh, C, Beattie, K.A., Klumpp, S. et al. (1990). Cyanobacterial microcystin-LR is

a potent and specific inhibitor of protein phosphatases 1 and 2A from both mammals

and higher plants. FEBS Lett, 264: 187-192.
Matsushima, R., Yoshizawa, S., Watanabe, M.F. et al. (1990). In vitro and in vivo effects

of protein phosphatase inhibitors, microcystin and nodularin, on mouse skin and

fibroblasts. Biochem Biophys Res Comm, 171: 867-874.
Nishiwaki-Matsushima, R., Ohta, T., Nishiwaki, S. et al. (1992). Liver cancer promoted by

the cyanobacterial cyclic peptide toxin microcystin LR. / Cancer Res Clin Oncol, 118:

420-424.
Runnegar, M., Berndt, N. and Kaplowitz, N. (1995). Microcystin uptake and inhibition of

protein phosphatases: effects of chemoprotectants and self-inhibition in relation to

known hepatic transporters. Toxicol Appl Pharmacol, 134: 264-272.
Teixeira, M.G.L.C., Costa, M.C.N., de Carvalho, V.L.P. et al. (1993). Gastroenteritis

epidemic in the area of the Itaparica dam, Bahia, Brazil. Bull PAHO, 27: 244-253.
Turner, P.C., Gammie, A.J., Hollinrake, K. et al. (1990). Pneumonia associated with

contact with cyanobacteria. Br Med J, 300: 1440-1441.

Bacteriology

Yeh, F.-S. (1989). Primary liver cancer in Guangxi. In Primary Liver Cancer ', Tang, Z.-Y.,

Wu, M.-C. and Xia, S.-S. (eds). Beijing: China Academic Publishers, pp. 223-236.
Yoshizawa, S., Matsushi, R., Watanabe, M.F. et al. (1990). Inhibition of protein

phosphatases by microcystis and nodularin associated with hepatotoxicity. / Cancer

Res, 116: 609-614.
Yu, S.-Z. (1989). Drinking water and primary liver cancer. In Primary Liver Cancer,

Tang, Z.-Y., Wu, M.-C. and Xia, S.-S. (eds). Beijing: China Academic Publishers,

pp. 30-37.
Yu, S.-Z. (1995). Primary prevention of hepatocellular carcinoma. / Gastroenterol

Hepatol, 10: 674-682.
Zhou, T.L. and Yu, S.Z. (1990). Laboratory study on the relationship between drinking

water and hepatoma: Quantitative evaluation using GGT method. Chin J Prevent Med,

24: 203-205.
Zhu, Y.-R., Chen, J.-G. and Huang, X.-Y. (1989). Hepatocellular carcinoma in Qidong

County. In Primary Liver Cancer, Tang, Z.-Y., Wu, M.-C. and Xia, S.-S. (eds). Beijing:

China Academic Publishers, pp. 204-222.
Zilberg, B. (1966). Gastroenteritis in Salisbury European children - a five year study. Cent

Afr J Med, 12:164-168.

Escherichia coli

Basic microbiology

Escherichia coli are non-spore-forming, Gram-negative bacteria, usually motile
by peritrichous flagella. They are facultative anaerobes with gas usually
produced from fermentable carbohydrates. E. coli form rod-shaped cells
2.0-6.0 fjim in length and 1.1-1.5 (xm in width, however, cells may vary from
coccal to long filamentous rods. Capsules or microcapsules, made of acidic
polysaccharides, are common in E. coli. Mucoid strains sometimes produce
extracellular polymers, generally referred to as K antigens and acid poly-
saccharides, composed of colanic acid, known as M antigens. E. coli produce
different kinds of fimbriae which are important during the adhesion of host
cells. Fimbriae vary both structurally and antigenically in different strains of
E. coli.

Some less commonly encountered E. coli strains found both within the envir-
onment and potable water systems are capable of giving rise to diseases usu-
ally in the form of diarrhoea. The process by which E. coli causes diarrhoea
varies between strains. These strains can be grouped depending upon which
mechanism is used by a particular strain.

Bacteriology

Origins of the organism

E. coli was first identified by the German paediatrician Theodor Escherich dur-
ing his studies of the intestinal flora of infants. He described the organism in
1885 as Bacterium coli commune and established its pathogenic properties
in extraintestinal infections. The name Bacterium coli was widely used until
1919. The genus Escherichia and the type species E. coli was used thereafter.
It was not until 1964 that the genus Escherichia was defined in Wilson
and Miles, Topley and Wilson's Principles of Bacteriology and Immunity as
a motile or non-motile organism that produced characteristics conforming to
the definition Enterobacteriaceae. Work conducted in the 1920s concluded
that Bacterium coli was a very antigenically heterogeneous species. However,
it was not until the 1940s that a classification scheme dividing E. coli into more
than 70 different serogroups, primarily based on the O (somatic) antigens
(Kauffmann, 1947) was developed. Today over 50 H (flagella) antigens and
over 100 K (capsular antigens) are now recognized which enable E. coli to be
further subdivided into serotypes.

In 1987, the type genus Escherichia contained four new species in addition
to E. coli: E. blattae (isolated from the hind gut of the cockroach and has not
been reported in clinical material) was first described in 1973 (Burgess et al.,
1973), E. fergusoni (isolated from clinical material) first described by Farmer
and coworkers in 1985, £. hermannii (isolated from wounds) first described
by Brenner and others in 1982 (Brenner etal.^ 1982a) and E. vulneris (isolated
from human wounds) first described by Brenner and coworkers also in 1982
(Brenner et al., 1982b).

Metabolism and physiology

The biochemical characteristics of the genus Escherichia are shown in Table 7.1
and the characteristics of E. coli are shown in Table 7.2. Most strains of
£. coli ferment lactose. They produce indole, fail to hydrolyse urea and to grow
in Moller's KCN broth. H 2 S production is not detectable on triple sugar iron
(TSI) agar, phenylalanine is not deaminated, and gelatine is not liquefied.
Most strains decarboxylate lysine and utilize sodium acetate, but they do not
grow on Simmons' citrate agar.

Clinical features

In 1921, Muir and Ritchie first described the pathogenic properties of B. coli
being associated with infections of the intestine and urinary tract, some cases

Escherichia coli

Table 7.1 Biochemical characteristics of Escherichia (adapted from Wilson and
Miles, 1964; Barrow and Feltham, 1995)

Characteristics

Reaction

Motility

MacConkey growth

Mannitol fermentation

Lactose, 37°C

Lactose, 44°C

Adonitol

Inositol

Indole at 37°C

Indole at 44°C

Methyl red reaction

Voges-Proskauer reaction

Urea

Phenylalanine deamination

Kligler's H 2 S (hydrogen sulphide) medium

Moller's KCN (potassium cyanide) medium

Gluconate oxidation

Gelatine liquefaction

Glutamine acid decarboxylase

Lysine decarboxylase

+ , usually gas

Acid +, gas +

Seldom fermented

Usually produced

No hydrolysis

No blackening
No growth

Table 7.2 Characteristics of Escherichia coli (adapted from Wilson and Miles, 1964;
Barrow and Feltham, 1995)

Characteristics

Reaction

Gram stain

Negative

Morphology

Straight rods

Motility

+ (peritrichous) some non-motile

Aerobic and anaerobic growth

Oxidase

—

Catalase

MacConkey growth

D-mannitol fermentation

Lactose, 37°C

Lactose, 44°C

D-adonitol

Inositol

D-glucose

Indole at 37°C

Indole at 44°C

Methyl red reaction

Voges-Proskauer reaction

Urea

Phenylalanine deamination

H 2 S (triple sugar iron) medium

KCN (potassium cyanide) medium

Gelatine liquefaction

Glutamine acid decarboxylase

Lysine decarboxylase

+ , usually gas (over 90% of strains)

Acid + , gas + (over 90% of strains)

Acid +, gas + (over 90% of strains)

Seldom fermented (over 90% of strains)

Seldom fermented

Acid

Usually produced

+ (over 90% of strains)

- (over 90% of strains)
No hydrolysis

- (over 90% of strains)

No blackening (over 90% of strains)
No growth

- (over 90% of strains)
+

+ (75-89% of strains)

Bacteriology

of summer diarrhoea (cholera nostras) and some cases of infantile diarrhoea
and food poisoning. From Topley and Wilson's first addition (1929) the path-
ogenicity of Bact. coli was summarized as: c Bact. Coli is a normal inhabitant
of the intestine of man and other animals. In certain circumstances it acquires
pathogenicity, and may cause local or general infection. It is a frequent cause
of acute and chronic infection of the urinary tract, and may give rise to an
acute or chronic cholecystitis.'

To date, £. coli is classed as a harmless member of the normal microbiota
of the human located at the distal end of the intestinal tract. The organism is
generally acquired at birth or by the faecal-oral route from the mother and also
from the environment. Most strains of E. coli are not pathogenic. A list of
some of the strains of £. coli that can cause a number of illnesses is given in
Table 7.3. The illnesses related to E. coli are shown in Table 7.4.

E. coli is the most common cause of acute urinary tract infections as well as
urinary tract sepsis. It has also been known to cause neonatal meningitis and
sepsis and also abscesses in a number of organ systems. E. coli may also cause
acute enteritis in humans as well as animals and is a general cause of 'traveller's
diarrhoea', a dysentery-like disease affecting humans, and haemorrhagic col-
itis often referred to as 'bloody diarrhoea'. Many oral challenge studies have
been done with a number of £. coli types to determine necessary doses required
to cause infection. The results of these studies suggest that levels of 10 5 -10 10
enteropathogenic (EPEC) organisms, 10 8 -10 10 enterotoxigenic (ETEC) and 10 8
cells of enteroinvasive (EIEC) need to be ingested to produce diarrhoea and
infection. These numbers will of course vary depending on age, sex and acid-
ity of the stomach. In the case of Vero cytotoxigenic E. coli (VTEC) the infective
dose that is capable of causing infection is <100 cells (Advisory committee on
the Microbiology Safety of Food, 1995; Bolton et al. 9 1996).

VTEC is the major E. coli of concern. It is known to cause haemolytic uraemic
syndrome (HUS) that is characterized by acute renal failure, haemolytic
anaemia and thrombocytopaenia that usually occurs in children under the age
of 5. On average 10% of patients infected with VTEC 0157 develop HUS and
some go onto develop thrombotic thrombocytopaenic purpura (TTP). Approx-
imately 5% of cases of VTEC develop haemorrhagic colitis which then develop
into HUS, in which case fatality can be as high as 10%. Diarrhoea caused
by E. coli 0157 is sometimes self-limiting. Strains of E. coli fall into at least
five groups with different pathogenic mechanisms. These will be discussed
in turn.

Enterotoxigenic E. coli (ETEC)

Enterotoxigenic E. coli (ETEC) cause gastroenteritis with profuse watery diar-
rhoea with abdominal cramps, vomiting and fever evident in a small percent-
age of patients. The severity of illness as a result of infection with ETEC varies
from relatively mild and short-lived to a severe life-threatening illness. They
are one of the major causes of death in children below the age of 5 years in

Escherichia coli

Table 7.3 Serogroups and disease associations of E. coli (Rowe, 1983; Salyers and Whitt, 1994;
Beutin etal., 1997; Sussman, 1997; Willshaw etal., 1997; Bell and Kyriakides, 1998).

Virulence type

Serogroup

Disease

Summary of host cell interaction

Enteropathogenic
(EPEC)

Enterotoxigenic
(ETEC)

Vero cytotoxigenic
(VTEC) (including
enterohaemor-
rhagic, EHEC)

Enteroinvasive
(EIEC)

Enteroaggregative
(EAECf

Diffusely adherent
(DAEC)

055 H6, NM
086 H34, NM
0111 H2, H12, NM
0119 H6, NM
0125ac H21
0126 H27, NM
0128 H2, H12
0142 H6
06 H16
08 H9
011 H27
015 H11
020 NM

025 H42, NM
027 H7

078 H11, H12
0128 H7

0148 H28

0149 H10
0159 H20
0173 NM

026 H11, H32, NM
055 H7

0111ab H8, NM
0113 H21
0117H14
0157 H7

028ab NM
029 NM
0112ac NM
0124 H30, NM
0136 NM

0143 NM

0144 NM
0152 NM
0159 H2, NM
0164 NM

0167 H4, H5, NM

03 H2

015 H18

044 H18

086 NM

077 H18

0111 H21

0127 H2

0? H10

Not yet

established

Enteritis in
infants
Traveller's
diarrhoea

Diarrhoea,
vomiting
and fever
Traveller's
diarrhoea

Shigella-like

dysentery

(stools

contain

blood and

mucus)

Haemolytic

uraemic

syndrome

Shigella-like

dysentery

Persistent
diarrhoea in
children

Childhood
diarrhoea

EPEC attach to intestinal
mucosal cells causing cell
structure alterations (attaching
and effacing), EPEC cells
invade the mucosal cells

ETEC adhere to the small
intestinal mucosa and produce
toxins that act on the
mucosal cells

EHEC attach to and efface
mucosal cells and produce toxin

EIEC invade cells in the colon
and spread laterally, cell to eel

EAEC bind in clumps
(aggregate to cells of the
small intestine) and produce
toxins

Fimbrial and non-fimbrial
adhesions identified

Bacteriology

Table 7.4 E. coli related illnesses (adapted from Bell and Kyriakides, 1998)

Time to onset of
Pathogenic E. coli illness

Duration
of illness

Symptoms

EPEC

1 7-72 h
(average 36 h)

6 h to 3 days
(average 24 h)

ETEC

8-44 h
(average 26 h)

3-1 9 days

VTEC

3-9 days
(average 4 days)

2-9 days
(average 4 days)

EIEC

8-24 h
(average 11 h)

Days to weeks

Severe diarrhoea in
infants which may persist
for more than 14 days.
Also fever, vomiting and
abdominal pain. In adults,
severe watery diarrhoea,
with prominent amounts
of mucus without blood,
and nausea, vomiting,
abdominal cramps,
headache, fever and chills.
Watery diarrhoea, low-
grade fever, abdominal
cramps, malaise, nausea.
When severe, causes
cholera-like extreme diar-
rhoea with rice water-like
stools, leading to
dehydration.
Haemorrhagic colitis
(HC): sudden onset of
severe crampy abdominal
pain, grossly bloody diar-
rhoea, vomiting, no fever.
Haemolytic uraemic
syndrome (HUS):
prodrome of bloody
diarrhoea, acute renal
failure in children,
thrombocytopaenia,
acute nephropathy,
seizures, coma, death.
Thrombotic thrombocy-
topaenic purpura (TTP):
similar to HUS but also
fever, central nervous
system disorders,
abdominal pain,
gastrointestinal
haemorrhage, blood clots
in the brain, death.
Profuse diarrhoea or
dysentery, chills, fever,
headache, muscular pain,
abdominal cramps.

developing countries. Adults and older children in tropical countries can also
become infected by ETEC but, generally, these become asymptomatic carriers
as a result of mucosal immunity. Individuals who do not acquire this sort of
immunity develop a condition known as traveller's diarrhoea. Traveller's

Escherichia coli

diarrhoea is usually brief episodes of loose stools sometimes accompanied with
nausea, vomiting and abdominal pains.

ETEC produce two enterotoxins, a heat stable toxin and a heat labile
oligopeptide toxin. Both toxins enter the cell and increase the concentration
of cyclic guanine monophosphate (cGMP) or cyclic adenine monophosphate
(cAMP) known to affect electrolyte transport causing excessive fluid loss. The
incubation period associated with infection of this group of £. coli is 12-72
hours, with duration of illness often 3-5 days.

Approximately 2-8% of the E. coli found in water are enteropathogenic
E. coli which cause traveller's diarrhoea. While water and also foods are
instrumental in the transmission and spread of E. coli the required dose
for this pathogen to cause infection is high, typically in the range of 10 6 -10 9
organisms.

Enteropathogenic E. coli (EPEC)

EPEC is a major cause of infant (less than 2 years of age) watery diarrhoea
infections. It has, however, been documented as a major cause of watery diar-
rhoea in children less than 6 months of age specifically in developing countries.
Mortality can be as high as 30%. The infectious dose of EPEC is very high,
about 10 8 -10 10 organisms. The transmission of infection is directly from person
to person with no evidence to date that it is transmitted in water.

Enterohaemorrhagic E. coli (EHEC) or Vero cytotoxigenic E. coli
(VTEC)

Enterohaemorrhagic E. coli (EHEC) produces a shiga-like toxin that is cytotoxic
to Vero cells. The commonest EHEC is 0157. The incubation period following
ingestion is 3-8 days and duration of infection is 1-12 days. Following inges-
tion of the required dose of less than 100 organisms symptoms develop which
include watery and bloody diarrhoea associated with vomiting. EHEC can
also cause two distinct conditions, haemorrhagic colitis and haemolytic uraemic
syndrome (HUS). HUS is characterized by thrombocytopaenia, microangio-
pathic, haemolytic anaemia and renal failure.

Symptoms resulting from EHEC include a colicky abdominal pain,
followed by diarrhoea, vomiting in some cases, that becomes bloody. This loss
of blood is often severe after only a few days following infection. EHEC has
also been associated with sporadic outbreaks of haemorrhagic colitis which is
a grossly bloody diarrhoea. Fatality rates are very high in the elderly and the
very young with 10% of children under 10 years developing a combination of
haemolytic anaemia, thrombocytopaenic purpura and acute renal failure. In
the UK, HUS is now the commonest cause of acute renal failure in children.

Bacteriology

Enteroinvasive E. colt (EIEC)

Patients who become infected with EIEC present with watery diarrhoea, with
a small proportion developing bloody diarrhoea. The infectious dose is high,
between 10 6 and 10 10 organisms. EIEC are closely linked to shigellae in that
they both cause the bacillary dysentery using similar processes. The incubation
period following ingestion is usually 1-3 days and the duration of infection is
1-2 weeks.

En tero aggregative E. colt (EAEC)

EAEC causes a watery mucoid diarrhoea, which usually contains no blood
and there is no fever. The infectious dose is generally high and it is primarily a
disease of developing countries.

Virulence and pathogenicity

ETEC

In order for ETEC to initiate effects on the host it must be able to adhere to
the mucosal surface of the epithelial cells in the small intestine. Many col-
onization factors have been discovered in ETEC, suggesting the process of
adhesion is more complex than originally thought and it still remains an area
of great debate. However, the process of adhesion seems to be achieved by
fimbriae, known to have specific binding receptors. Once ETEC is adhered the
role of toxins in the infection process comes into play. ETEC strains have been
found to produce two toxins, a short polypetide chained toxin called the heat-
stable (ST) toxin, and/or a heat-labile (LT), large oligometric enterotoxin.
Both the ST and LT toxins have two antigenic types, namely STa and STb and
LT-I and LT-II respectively. LT-I is closely related to the toxin produced by
Vibrio cholerae. It is an oligomeric toxin composed of one polypeptide A sub-
unit and five identical polypeptide B subunits. The molecular weights of sub-
units A and B are 25 000 and 115 000 respectively. The five B subunits form a
very stable doughnut shape with a central aqueous channel and are important
in binding the toxin to the epithelial cells. The A subunit sits above and is
partly inserted into this central channel.

After the toxin molecule has bound to the enterocyte cell surface through its
five B subunits the toxin molecule is taken into the cell by the process of endo-
cytosis. Within the cell the A subunit is split into A 1 and A 2 fragments. The
activity lies in the A 1 fragment which is an ADP-ribosyltransferase. It is sub-
unit A which catalases the nicotinamide adenine dinucleotide (NAD) dependent
activation of adenylate cyclase, causing an increase in the concentration of

Escherichia coli

intracellular cyclic adenosine 5 '-monophosphate (cAMP). There is a marked
net excretion of chloride from the enterocyte into the gut lumen. This is
because cAMP inhibits the absorption of sodium in the intestinal villus cells
and therefore chloride ions and water. The net result of these reactions is that
the intraluminal osmolality increases and water is drawn into the gut. In the
crypt cells cAMP causes an increase in sodium secretion, with a loss of chlor-
ide ions and water. This then leads to the development of a watery diarrhoea.
Other more complex mechanisms have also been suggested as alternative to
this possible mechanism (Sears and Kaper, 1996).

The A subunit in LT-II causes a similar effect to that of LT-I. LT-II also causes
a net excretion of chloride through an increase in intracellular cAMP, though
the initial intracellular binding site is different.

The heat stable enterotoxins (STa) of E. coli are low molecular weight
toxins, generally poorly immunogenic. As mentioned there are two major
classes, STa and STb. The STa (18 amino acids) toxin activates guanylate
cyclase C (GC-C), an enzyme present in the luminal membrane of enterocytes.
The binding of STa to GC-C causes an increase in intracellular cGMP. The
activity of STa is quick and the end result is increased chloride excretion in
the same way as for LT. LT, as opposed to STa, does have a lag period
before it acts. STb differs from STa in being larger, 48 amino acids, and not
causing an increase in cAMP or cGMP. Also chloride transport is not directly
affected, instead there is a net increase in bicarbonate excretion. Furthermore,
STb is associated with histological evidence of cellular damage and STa is
plasmid encoded and can be distinguished from STb as it is soluble in
methanol.

EPEC

EPEC produce characteristic histological features in the intestinal tract of
patients. These are known as attaching and effacing lesions which show the
adherence of the bacteria to the epithelial cell membrane with effacement of
the microvilli. The lesions develop in three stages. At stage one E. coli attaches,
aided by fimbriae, to the intestinal enterocyte cells. Following attachment to
the enterocyte the secretion of various extracellular proteins causes dissol-
ution of the microvilli (Bain et aL 9 1998). Following on from this the bac-
terium binds closely to the enterocyte membrane. There has been much recent
interest in EPEC (DeVinney et aL, 1999; Vallance and Finlay, 2000). Recent
work has shown that EPEC inserts its own receptor for intimate adherence,
Tir (translocated intimin receptor) into the host cell membrane. Once
attached, translocated EPEC proteins activate signalling pathways within the
underlying cell. These cause the reorganization of the host actin cytoskeleton.
EPEC once attached become partially surrounded by cup-like projections
(pedestal-like structures) beneath the adherent bacteria.

The most prevalent serogroups of EPEC are 06, 08, 025, Olll, 0119,
0125-0128.

Bacteriology

EHEC

EHEC has the same virulence mechanisms as EPEC but produces potent
toxins known as Shiga (Stx) toxins. Stxl has been shown to be identical to the
Shiga toxin of S. dysenteriae type 1 with Stx 2 only some 55-57% homolo-
gous. Both Stxl and 2 are made up of an A subunit and five B subunits. The
A subunit possesses the biological activity while the B subunits are thought to
mediate binding to the cell surface and thus aid uptake of the toxin.

Shiga toxins bind to a glycolipid receptor, globotriaosylceramide. The toxin
can then be taken into the cell by a process of endocytosis and transported to
the endoplasmic reticulum. Once the toxin is attached subunit A is released
into the cytoplasm where it disrupts protein synthesis, leading to cell death.
The toxin removes a single adenine residue from the 28S rRNA of eukaryotic
ribosomes which brings about this effect.

Associated with EHEC is the production of diarrhoea possibly caused by
the death of the intestinal absorptive cells, leaving the secretory cells intact.
Haemolytic uraemic syndrome (HUS) is thought occur because the toxin is
transported via the blood to the kidneys where it can cause kidney failure.
However, some research has shown that while strains of E. coli 0157:H7 had
been isolated from cases of HUS they in fact lacked the ability to produce the
Shiga toxin (Schmidt et aL, 1999).

EIEC

Enteroinvasive E. coli bind to enterocysts of the large intestine where they
destroy cells resulting in tissue damage which results in an inflammatory
response. The invasive property of EIEC is controlled by a 140 MD plasmid.
As the name suggests there is invasion of the epithelial cell itself with subse-
quent intracellular multiplication and lateral movement through the cell to
allow subsequent penetration of adjacent cells. Although not yet fully under-
stood, there also seems to be production of an enterotoxin which is thought to
be responsible for the initial watery diarrhoea.

EIEC causes dysentery-like symptoms by invading colonic epithelium (cf.
Shigella). These strains belong to serogroups 028, 052, 0112, 0115, 0124,
0136, 0143, 0145, 0147. These strains account for <0.5% infection in the
West but about 0.5% of dysentery in Asia.

EAEC

EAEC are named after their characteristic aggregative adherence to Hep-2 cells
in tissue culture. The mechanism of the pathogenicity of EAEC is poorly under-
stood, however, a model of pathogenicity has been suggested by Nataro and
Kaper (1998). This model is composed of three stages with the first two stages
involving the adherence of EAEC to the mucosal membrane and the stimulation

Escherichia coli

of mucus secretion leading to the formation of a thick mucus biofllm. In the
biofllm the bacteria tend to clump forming an enteroaggregative. The final
stage of the pathogenicity model involves the secretion of a cytotoxin responsible
for causing damage to the mucosal lining.

DAEC

Infection with DAEC causes a watery diarrhoea, mostly in older children. DAEC
like EAEC adhere to Hep-2 cells. The pattern by which DAEC attaches is,
however, more diffuse than EAEC, a distinguishing feature. However, the
pathogenic mechanisms for DAEC are to date still inconclusive.

Treatment

With £. coli diarrhoeal infections fluid and electrolyte correction is obligatory.
Extraintestinal £. coli infections, however, are treated with antibiotics. If this
method of treatment is to be employed it must be borne in mind that £. coli
does possess intrinsic resistance to benzylpenicillin. As a general rule
£. coli are still sensitive to the antibiotics ampicillin, tetracycline, aminogly-
cosides, trimethoprim and the cephalosporins. However, because of the wide-
spread evidence of antibiotic resistance being acquired by plasmid transfer,
mounting numbers of £. coli are becoming resistant to streptomycin and tetra-
cycline. For this reason antibiograms should be performed, especially for
epidemiological purposes. The use of antibiotics in treating £. coli 0157 is
questionable due to the problem with the growing numbers of £. coli 0157
developing antibiotic resistance.

Survival in the environment

£. coli has a reservoir in the intestines of man and other warm-blooded
animals, and is excreted in the faeces. It is known to survive in the environ-
ment but not reproduce (Feachem et aL 9 1983), however, in the tropical envi-
ronment there is evidence that £. coli can survive and multiply (Rivera et al.,
1988). Subsequently, if £. coli is detected in the environment it is indicative of
faecal pollution but in the tropics this may not be the case which warrants fur-
ther investigation. For this reason this research suggests that in fact £. coli
may not be a good indicator of faecal pollution. Escherichia coli was introduced
into water bacteriology because it was a useful marker of faecal pollution and
thus became an important marker in food and water hygiene. The theory was

Bacteriology

that if £. coli was present then so could be pathogenic enteric bacteria such as
Shigella and Salmonella spp. (Gleeson and Gray, 1997).

The natural reservoirs of enteropathogenic strains are the intestines of humans
(EPEC, ETEC, EIEC, EAEC) or domestic animals (ETEC, EHEC). The organ-
isms are transmitted by direct contact or via contaminated food and water.

Water

All enterovirulent £. coli are acquired directly or indirectly from a human or
animal carrier. Risk from drinking water, therefore, only follows from faecal
contamination of the supply. Given the sensitivity of £. coli to chlorine and
other disinfectants, even if the organisms did contaminate the supply adequate
chlorination would effectively remove any risk (Hunter, 2003).

However, standards do not exist for £. coli 0157 (EHEC) within the 1980
European Drinking Water Inspectorate Directive. In 1997 the Environment
Group of the former Scottish Office of Agriculture, Environment and Fisheries
Department (SOAEFD) commissioned the Water Research Council (WRC) to
carry out a study to examine the existing evidence for waterborne trans-
mission of £. coli 0157. From this study they found no evidence to indicate
that £. coli 0157 was more persistent in the environment or more resistant
to water- treatment processes than the non-pathogenic £. coli found in the
human gastrointestinal tract. Currently the Scottish Agricultural College and
the University of Aberdeen are doing research into The survival and dispersal
of £. coli 0157 in Scottish soils and potential for contamination of private
water supplies'. It is expected results from this study will be available in 2004.
Findings to date suggest that £. coli 0157 can survive for up to 21 days in
water but that £. coli 0157 is as susceptible to chlorination as any other
£. coli strain.

Concern exists, however, about the potential role of biofllm in protecting
enterovirulent £. coli. The very high infectious doses required for all
enterovirulent £. coli, other than EHEC, suggests that this possible route of
transmission is unlikely as a risk. The lower infectious dose of EHEC does
potentially increase the risk of infection from biofllms in water but there have
been no outbreaks or studies of sporadic cases of EHEC implicating inad-
equately disinfected water supply.

Detection

Escherichia spp. may be detected as part of the normal microbiota (gastro-
intestinal tract) of both man and animals. £. coli can also be recovered from
patients suffering from urinary tract and wound infections, meningitis and
septicaemia. Samples, which have been recovered from sterile sites on the

Escherichia coli

human body, can be plated out on non-selective media such as blood, chocolate
or nutrient agar. If contamination of samples is probable then MacConkey or
eosin-methylene blue (EMB) agar should also be incorporated into any form
of microbiological analysis.

For isolating enterovirulent E. coli from fresh stool samples MacConkey or
EMB agar is used. For convalescents or patients with previous antibiotic treat-
ment a number of samples should be taken for analysis. As £. coli is part of
the normal intestinal flora, at least 10, but preferably more, colonies should
be picked from the isolation media and tested for the presence of virulence
markers.

Polymerase chain reaction (PCR) targeting virulence-associated gene sequences
can be performed on colony sweeps from MacConkey. Escherichia are identi-
fied by biochemical reactions and slide agglutination. For the identification of
different enterovirulent E. coli biological, immunological and molecular methods
can be employed.

EPEC

For the detection of EPEC, stool specimens are plated out on MacConkey
agar. Lactose-fermenting colonies of E. coli groups can be tested by use of
agglutinating antisera for the EPEC serogroups. For EPEC outbreaks ten or
more lactose-positive colonies are subcultured on blood or nutrient agar and
agglutinated with a set of polyvalent and monovalent OK antisera. After sub-
culture a clear agglutination with a monovalent OK antiserum is heated at
100°C for 30-60 minutes and titrated in parallel with a reference culture of
the same O group. After overnight incubation at 50°C the agglutination is read.
Presence of the particular O group is confirmed only if both the test and the
reference O antigens are agglutinated to nearly the same titre. EPEC strains
can be identified by fluorescence actin staining method and cell culture tests
with HEp-2 or HeLa cells to demonstrate localized or diffuse adherence.
Molecular diagnosis of EPEC using PCR or colony hybridization of genes
encoding for intimin (eaeA), bundle-forming pili (bfpA) or the EPEC adherence
factor (EAF) plasmid have also been used for the diagnosis of EPEC.

EAEC

EAEC have the ability to clump and are characterized by a unique, 'stacked
brick-like' adherence which can be demonstrated in HEp-2 cells. Based on these
characteristics and a particular heat-stable enterotoxin, EAEC can be confirmed.

ETEC

Cell culture techniques (heat-labile enterotoxin, LT) or animal models (the
suckling mouse assay for the heat-stable enterotoxin, ST, and the rabbit ileal

Bacteriology

loop for LT and ST) are used for the detection of LT. Exposure of cells, e.g.
Vero monkey kidney cells to culture supernatant containing LT leads to mor-
phological changes in cell structure which can be observed using microscopic
techniques.

Many immunological tests are available for the detection of LT. These are
available commercially and include enzyme-linked immunoassay (ELISA) and
solid phase radio-immunoassay (RIA).

PCR or DNA probes can detect genes encoding for LT and ST production.
Many of these probes have also been developed for use in food and water.

EIEC

EIEC do not decarboxylate lysine and are mostly non-motile and usually non-
or late lactose fermenters. After an appropriate infection period in Hep-2 and
HeLa cells the presence of EIEC cells can be observed by microscopy. EIEC
strains are confirmed by proof of invasivenes (Sereny or cell culture tests) or
by demonstration of the ipaH gene present on the chromosome and on the
virulence plasmid. When invasion by EIEC has been demonstrated biochemical
tests will enable identification.

VTEC

Serovars 0157:H7 and 0157:H— , plus a few others, are the most important
VTEC strains of £. coll. One of the most important problem £. coli is the
VT-producing strains of O group 157. While a very high percentage of £. colt
ferment sorbitol, the strain 0157 does not ferment sorbitol in 24 hours.
This characteristic is then used as a basic test used to detect them by using
MacConkey (SMAC) agar. However, it is now documented that some strains
of £. colt 0157:H - isolated in Europe - are able to ferment sorbitol rapidly.
Therefore SMAC alone should not be used for their detection. By supple-
menting SMAC with cefixime or cefixime-tellurite the selectivity for sorbitol-
negative EHEC 0157 can be improved.

Agglutinating sera and a latex coagglutination tests are available for VTEC
detection.

Most strains of £. colt 0157:H7 produce a plasmid-encoded haemolysin.
This can now be detected by PCR.

VT production in £. coli can be detected by the cytotoxin effects observed
in the Vero cell culture test. Commercially available enzyme immunoassay
(Meridian Diagnostics, Cincinnati, USA and Milan, Italy) can be used for
cytotoxin detection in both stool specimens and £. coli isolates.

Because of the low number of excreted VTEC evident in stool specimens,
especially in patients with the haemolytic uraemic syndrome (HUS), testing
of 20-100 isolated colonies of £. coli is necessary to achieve a sufficient level
of diagnostic sensitivity. PCR can be performed on colony sweeps from

Escherichia coli

MacConkey plates giving an excellent sensitivity for the detection of VTEC
strains.

Following the identification of a VT-producing colony biochemical deter-
mination of the species should follow. To prevent the loss of encoding genes,
E. coli cultures should be suspended in Luria-Bertani or tryptic soy broth sup-
plemented with 20-50% glycerol and stored at -70°C.

Epidemiology

The epidemiology of waterborne E. coli infections are reviewed elsewhere
(Hunter, 2003).

EPEC

Few epidemics of EPEC have been documented in Europe and the USA. How-
ever, the incidence of sporadic cases of infantile enteritis peaks in summer
months. Cases seem to be more common in areas that have very poor hygiene
(Regua et al., 1990). While water may be implicated in its transmission, the
evidence to date suggests that EPEC is not transmitted in water.

ETEC

ETEC have occasionally been reported in regions of the world that have good
hygiene but outbreaks of diarrhoea, due to ETEC, are one of the major causes
of deaths in children under 5 in developing countries. ETEC are probably the
commenest cause of traveller's diarrhoea. The most probable cause of ETEC
transmission seems to be via a water source with person-to-person transmission
uncommon.

Waterborne outbreaks due to ETEC have been documented. A large out-
break, affecting more than 2000 staff and visitors to an American National
Park in Oregon, occurred in the summer of 1975 (Rosenberg et aL, 1977).
From this study enterotoxigenic E. coli were isolated from 20 (16.7%) of 120
rectal swabs that were examined. A strong correlation between illness and
drinking park water in park staff and visitors was also found (P < 0.00001).
However, no association with drinking water occurred in visitors on 7-9 July
when chlorination of the water supply was being more closely monitored.
Another outbreak affecting 251 passengers and 51 crew on a Mediterranean
cruise (O'Mahony et al., 1986) has also suggested a route of transmission of
ETEC. From this study faecal coliforms were isolated from tap water suggesting
the only plausible cause of this outbreak of ETEC. From an investigation of
this outbreak faulty chlorination and faulty covers, possibly allowing bilge

Bacteriology

water into the drinking water tanks, was a probable cause of the outbreak.
Other studies of ETEC transmission in water have been documented by Huerta
et al. (2000). In a further outbreak, 175 Israeli military personnel and at least
54 civilians in the Golan Heights were infected by ETEC. Samples of water
from several points along the distribution system which supplied the commu-
nity showed inadequate chlorination and high concentrations of E. colt.

Outbreaks on ships are quite common which have been due to consuming
drinks with ice cubes and drinking unbottled water (Daniels et al., 2000).

Children aged 7-10 months in Ecuador have been shown to have anti-
bodies to ETEC which has been linked to the consumption of low quality
drinking water (Brussow et al., 1992).

EHEC

EHEC was recognized in the early 1980s as a severe disease of the very young
and the elderly (Riley et al., 1983). As with most E. coli EHEC is found in the
intestines of several animal species and as such a faecal-oral spread from
infected animals or other humans is evident. Also faecal contamination of
food or water provides a means of EHEC transmission. Serotype 0157:H7 is
linked to outbreaks in drinking water in Europe and North America. Other
non-0157 EHEC are now becoming recognized as causes of foodborne out-
breaks, principally beef products, and person-to-person transmission.

Evidence of E. coli 0157:H7 was found to be strongly linked to the con-
sumption of drinking water in Missouri in 1990 (Swerdlow et al., 1992).
From this study it was found that, of a population of 3126, 243 people devel-
oped illness of whom 86 developed bloody diarrhoea, 36 were hospitalized
and four died. Based on a case-control study of 53 cases, the only significant
factor was that those infected drank more cups of municipal water per day
(7.9) than did controls (6.1) (P = 0.04). Following an investigation of the
water supply to the city it was noted that two mains water breaks had
occurred after the start of the outbreak but before its main peak. The town's
drinking water came from two deep-ground water sources and it was found
that the sewage system was inadequate and sewage overflow crossing drink-
ing water mains, provided the means for contamination of water supplies.

Outbreaks of E. coli 0157:H7 have occurred in Scotland. In this study, dur-
ing the hot summer of 1990, four people developed haemorrhagic colitis (Dev
et al., 1991). Because of the hot weather water levels in the water supply
extraction points were low. As a result of this water from two subsidiary reser-
voirs was used. However, one of the reservoirs was fed from a source which
may have been contaminated by cattle slurry.

In South Africa and Swaziland, in 1992, an outbreak of bloody diarrhoea
affecting thousands of individuals was documented (Isaacson et al., 1993).
There were fatalities and cases of renal failure. Most cases were from men
who drank surface water in the fields and women and children who drank
borehole water. From the microbiological analysis it was found that E. coli

Escherichia coli

0157:H7 was isolated from 14.3% of 42 samples of cattle dung and 18.4%
of 76 randomly collected water samples. The conclusion drawn from this
study was that cattle carcasses and dung washed into rivers and dams by
heavy rains after a period of drought contaminating the water.

In Europe a large outbreak in Fuerteventura, Canary Islands occurred in
March 1997 (Pebody et aL 9 1999). From this study 14 confirmed and one
probable case were identified. The cases occurred in four different hotels.
It was established following investigations that three of the four hotels were
supplied with water from a private well.

The largest and most disreputable outbreak of EHEC associated with drinking
water occurred during May and June 2000 among residents of Walkerton,
Ontario (Anon, 2000). It was found in this study that 1346 cases of illness
were identified. However, many people who presented with symptoms and
were then analysed were infected with Campylobacter instead of EHEC.
However, in addition to the 1346 cases a further 65 people were admitted to
the local hospital. Twenty-seven of these patients developed haemolytic
uraemic syndrome with six fatalities. An association between drinking water
consumption was identified. The drinking water came from a number of wells
and there was strong evidence suggesting that one of the wells had become
contaminated with cattle faeces following heavy rains and flooding.

EIEC

EIEC transmission is common from person to person, although it is princi-
pally thought to be acquired by a food or waterborne route. A water route is
uncommon with evidence of only one outbreak of EIEC due to water,
reported in 1959 (Lanyi et aL, 1959).

EAEC

One outbreak of EAEC linking it to drinking water has been documented.
This occurred in a small Indian village where people who drank water from a
borehole were found to be much less likely to have diarrhoea than people
drinking from shallow wells. From this study it was found that E. coli were
isolated from the shallow wells but not the borehole (Pai et al. 9 1997).

Risk assessment

Health effects: occurrence of illness, degree of morbidity and mortality:

• The pathogenic £. coli serotypes are grouped based on their mechanism of
causing symptoms. The six are enteropathogenic, enterotoxigenic, verocyto-
genic (which included enterohaemorrhagic), enteroinvasive, enteroaggregative

Bacteriology

and diffusely adherent. All pathogenic £. coli cause diarrhoea to various
degrees of severity.

- Enteropathogenic: traveller's diarrhoea, enteritis in infants

- Enterotoxigenic: traveller's diarrhoea, diarrhoea, vomiting and fever

- Verocytotoxigenic: Shigella -like dysentery (stools with blood and mucus),
haemolytic uraemic syndrome (especially in children)

- Enteroaggregative: chronic diarrhoea in children

- Diffusely adherent: diarrhoea in children

• £. coli also causes urinary tract infections and can cause sepsis and meningitis
in neonates.

• Pathogenic £. coli strains cause the majority of childhood diarrhoea in the
world.

• The degree of morbidity and mortality varies according to the strain and
the host's characteristics. In developing countries, diarrhoeal disease is
much more likely to result in serious illness and death. In developed coun-
tries, though childhood diarrhoea is less of a problem, infection with vero-
cytotoxigenic £. coli can result in haemolytic uraemic syndrome -
especially in children under five - and thrombotic thrombocytopaenia pur-
pura. These conditions can cause acute kidney failure and death.

Exposure assessment-, infectious dose, routes of exposure and transmission,
occurrence in source water, environmental fate:

• The infectious doses of most pathogenic £. coli are high: ranging from 10 5
to 10 10 organisms. This varies based on the host characteristics, such as age
and stomach acidity. The exception is the group of verocytotoxigenic
£. coli serogroups, which includes £. coli 0157:H7. The infectious dose for
this group appears to be less than 100 organisms.

• The routes of exposure and transmission in humans are faecal-oral; there-
fore, food, water, and person to person. Water contaminated with sewage
has caused water-related outbreaks. The verocytotoxigenic types are most
associated with waterborne outbreaks, probably because of their low infec-
tious doses.

• The major source of pathogenic strains is human sewage, or in the case of
verocytotoxigenic strains, faecal contamination from cattle.

• Humans and domestic animals are the reservoirs for enteropathogenic
strains of £. coli; the connection between undercooked beef and £. coli
0157:H7 infection is well known. Though £. coli strains can survive in the
environment, they do not reproduce. Survival times in water vary based on
the water characteristics (e.g. source, temperature, etc.).

Risk mitigation: drinking-water treatment, medical treatment:

• £. coli is very sensitive to chlorine and other disinfectants. Adequate residual
disinfection should take care of any contamination in the distribution sys-
tem. Waterborne outbreaks have resulted from treatment failures or from
untreated water sources contaminated with faecal matter.

Escherichia coli

All diarrhoeas should be treated with fluids and electrolytes, if appropriate.
Though most E. coli strains have been sensitive to many antibiotics, includ-
ing ampicillin, cephalosporins, and tetracycline, strains are becoming more
and more antibiotic-resistant. Antibiotic treatment is generally not recom-
mended for infection with E. coli 0157:H7.

References

Advisory Committee on the Microbiology Safety of Food. (1995). Report on Vero

Cytotoxin-Producing Escherichia coli. London: HMSO.
Anon. (2000). Waterborne outbreak of gastroenteritis associated with a contaminated

municipal water supply, Walkerton, Ontario, May-June 2000. Canada Commun Dis

Rep, 26: 170-173.
Bain, C, Keller, R., Collington, G.K. et al. (1998). Increased levels of intracellular calcium

are not required for the formation of attaching and effacing lesions by enteropathogenic

and enterohemorrhagic Escherichia coli. Infect Immun, 66: 3900-3908.
Barrow, G.I. and Feltham, R.K.A. (eds) (1995). Cowan and Steel's Manual for the

Identification of Medical Bacteria, 3rd edn. Cambridge University Press.
Bell, C. and Kyriakides, A. (1998). E. coli: A Practical Approach to the Organism and Its

Control in Foods. London: Blackie Academic and Professional.
Beutin, L., Gleier, K., Kontny, I. et al. (1997). Origin and characteristics of enteroinvasive

strains of Escherichia coli (EIEC) isolated in Germany. Epidemiol Infect, 118: 199-205.
Bolton, F.J., Crozier, L. and Williamson, J.K. (1996). Isolation of Escherichia coli 0157

from raw meat products. Lett Appl Microbiol, 23: 317-312.
Brenner, D.J., Davis, B.R., Steigerwalt, A.G. et al. (1982a). Atypical biogroups of

Escherichia coli found in clinical specimens and description of Eschericha hermannii

sp. nov. / Clin Microbiol, 15: 703.
Brenner, D.J., McWhorter, A.C., Knutson, J.K.L. et al. (1982b). Escherichia vulneris: a new

species of Enterobacteriaceae associated with human wounds./ Clin Microbiol, 15: 1133.
Brussow, H., Rahim, H. and Freire, W. (1992). Epidemiological analysis of serologically

determined rotavirus and enterotoxigenic Escherichia coli infections in Ecuadorian

children. / Clin Microbiol, 30: 1585-1587.
Burgess, N.R.H., McDermott, S.N. and Whiting, J. (1973). Laboratory transmission of

Enterobacteriaceae by the oriental cockroach, Blatta orientalis. J Hyg Camb, 71: 9.
Daniels, N.A., Neimann, J., Karpati, A. et al. (2000). Traveler's diarrhea at sea: three

outbreaks of waterborne enterotoxigenic Escherichia coli on cruise ships. / Infect Dis,

181: 1491-1495.
DeVinney, R., Gauthier, A., Abe, A. et al. (1999). Enteropathogenic Escherichia coli:

a pathogen that inserts its own receptor into host cells. Cell Mol Life Sci, 55: 961-976.
Dev, V.J., Main, M. and Gould, I. (1991). Waterborne outbreak of Escherichia coli 0157.

Lancet, 337: 1412.
Farmer, J.J. Ill, Fanning, G.R., Davis, B.R. et al. (1985). Escherichia fergusonii and

Enterobacter taylorae, two new species of Enterobacteriaceae isolated from clinical

specimens./ Clin Microbiol, 21: 77.
Feachem, R.G., Bradley, D.J., Garelick, H. et al. (1983). Sanitation and Disease: Health

Aspects of Excreta and Wastewater Management. Chichester: John Wiley &C Sons.
Gleeson, C. and Gray, N. (1997). The Coliforms Index and Waterborne Disease. London:

E and FN Spon.
Huerta, M., Grotto, L, Gdalevich, M. et al. (2000). A waterborne outbreak of gastro-
enteritis in the Golan Heights due to enterotoxigenic Escherichia coli. Infection, 28:

267-271.
Hunter, P.R. (1997). Waterborne Disease: Epidemiology and Ecology. Chichester: Wiley.

Bacteriology

Hunter, P.R. (2003). Drinking water and diarrhoeal disease due to Escherichia coli.J Water

Hlth, 1: 65-72.
Isaacson, M., Canter, P.H., Effler, P. et al. (1993). Haemorrhagic colitis epidemic in Africa.

Lancet, 341: 961.
Kauffmann, F. (1947). Review, the serology of the coli group./ Immunol, 57: 71-100.
Lanyi, B., Szita, J., Ringelhann, A. et al. (1959). A waterborne outbreak of enteritis associ-
ated with Escherichia coli serotype 124:72:32. Acta Microbiol Hung, 6: 77-78.
Muir, R. and Ritchie, J. (1921). Manual of Bacteriology. London: Oxford University Press,

pp. 353-359.
Nataro, J.P. and Kaper, J.B. (1998). Diarrheagenic E. coli. Clin Rev Microbiol, 11:

142-201.
O'Mahony, M.C., Noah, N.D., Evans, B. et al. (1986). An outbreak of gastroenteritis on a

passenger cruise ship./ Hyg, 97: 229-236.
Pai, M., Kang, G., Ramakrishna, B.S. etal. (1997). An epidemic of diarrhoea in south India

caused by enteroaggregative Escherichia coli. Ind J Med Res, 106: 7-12.
Pebody, R.G., Furtado, C, Rojas, A. et al. (1999). An international outbreak of Vero cyto-

toxin-producing Escherichia coli 0157 infection amongst tourists; a challenge for the

European infectious disease surveillance network. Epidemiol Infect, 123: 217-223.
Regua, A.H., Bravo, V.L.R., Leal, M.C. et al. (1990). Epidemiological survey of the

enteropathogenic Escherichia coli isolated from children with diarrhoea. / Trop

Paediatr, 36: 176-179.
Riley, L.W., Remis, R.S., Helgerson, S.D. et al. (1983). Hemorrhagic colitis associated with

a rare Escherichia coli serotype. New Engl J Med, 308: 681-685.
Rivera, S.C., Hazen, T.C. and Toranzos, G.A. (1988). Isolation of fecal coliforms from

pristine sites in a tropical rain forest. Appl Environ Microbiol, 54: 513-517.
Rosenberg, M.L., Koplan, J.P., Wachsmuth, I.K. et al. (1977). Epidemic diarrhea at Crater

Lake from Enterotoxigenic Escherichia coli. A large waterborne outbreak. Ann Intern

Med, 86: 7 14-7 '18.
Rowe, B. (1983). Escherichia coli diarrhoea. Culture, 4(1): 1-3.
Salyers, A. A. and Whitt, D.D. (1994). Bacterial Pathogenesis - A Molecular Approach.

Washington, DC: ASM Press, pp. 190-204.
Schmidt, H., Scheef, J., Huppertz, H.I. et al. (1999). Escherichia coli 0157:H7 and

0157:H( — ) strains that do not produce Shiga toxin: phenotypic and genetic characteri-
zation of isolates associated with diarrhea and hemolytic-uremic syndrome. / Clin

Microbiol, 37: 3491-3496.
Sears, C.L. and Kaper, J.B. (1996). Enteric bacterial toxins: mechanisms of action and link-
age to intestinal secretion. Microbiol Rev, 60: 167-215.
Sussman, M. (1997). Escherichia coli and human disease. In Escherichia coli; Mechanisms

of Virulence, Sussman, M. (ed.). Cambridge: Cambridge University Press, pp. 3-48.
Swerdlow, D.L., Woodruff, B.A., Brady, R.C. et al. (1992). A waterborne outbreak in

Missouri of Escherichia coli 0157:H7 associated with bloody diarrhoea and death.

Ann Intern Med, 117: 812-819.
Vallance, B.A. and Finlay, B.B. (2000). Exploitation of host cells by enteropathogenic

Escherichia coli. Proc Natl Acad Sci USA, 97: 8799-8806.
Willshaw, F.A., Scotland, S.M. and Rowe, B. (1997). Vero cytotoxin-producing

Escherichia coli. In Escherichia coli: Mechanisms of Virulence, Sussman, M. (ed.).

Cambridge: Cambridge University Press, pp. 421-448.
Wilson, G.S. and Miles, A.A. (1964). Topley and Wilson's Principles of Bacteriology and

Immunity, vol. 1, 5th edn. London: Edward Arnold, pp. 806-826.

Helicobacter pylori

Basic microbiology

Helicobacter pylori is an important pathogen which colonizes the mucus layer
and epithelial mucus of the stomach in approximately 50% of humans world-
wide. H. pylori is a curved (0.6 [Jim width, 2-5 fjim length), Gram-negative bac-
terium which is typically motile with five to six sheathed unipolar flagella.
Morphologies include spiral, curved, rod (bacillary), gull-winged, U-shaped
and circular (coccoid) forms. H. pylori is urease- and oxidase-positive and
negative for indoxyl acetate and hippurate hydrolysis (Foliguet et al. 9 1989;
Popovic-Uroic et al. 9 1990; On and Holmes, 1992). Genomes of two strains of
H. pylori have been completely sequenced (Tomb et al. 9 1997; Aim et al. 9
1999). The genomes contain nearly 1.7 million base pairs encoding approxi-
mately 1600 genes, 17% of which are unique to H. pylori (Doig et al. 9 1999;
Pawlowski et al. 9 1999).

Under conditions of stress, H. pylori can transform from a spiral, bacillary
morphology to a condensed, coccoid morphology. While it has been speculated
that it may be a dormant form and play a significant role in transmission, con-
siderable controversy surrounds the function and viability of H. pylori coccoid
forms (Cellini et al. 9 1994; Kusters et al. 9 1997; Mizoguchi et al 9 1998; Enroth
et al. 9 1999; Kurokawa etal 9 1999; Ren etal. 9 1999). Certain H. pylori coccoid
forms may represent a viable but non-culturable (VBNC) state similar to those

Bacteriology

of Vibrio and Campylobacter spp. in the natural environment (Roszak and
Colwell, 1987; Cappelier et al. 9 1999). The putative VBNC state has been con-
sidered to represent a specific form of dormancy occurring in non-sporulating
organisms (Oliver, 1995). Cells in this state cannot be grown by the culture
method normally used for the organism concerned but are believed to have the
capacity to return to the vegetative state when exposed to appropriate stimuli
(Gribbon and Barer, 1995). Therefore, by definition, VBNC bacteria are unable
to form colonies on traditional media, while potentially maintaining their patho-
genic capabilities (McFeters, 1990). It is probable that VBNC indicator bacteria
can pass undetected from a water-treatment system into the distribution net-
work and result in the underestimation of indicator and pathogenic bacterial
populations (Bucklin et al. 9 1991). A distinction between VBNC H. pylori and
coccoid H. pylori should be emphasized, as the VBNC state of H. pylori is
morphology-independent. H. pylori enter a VBNC state without change from
vibriod morphology (Cellini et al. 9 1994). Coccoid H. pylori exhibit diversity
in ultra structure following exposure to different stresses and it is likely that
coccoids exhibit variation in viability (Mizoguchi et al. 9 1998). Some coccoid
forms are electron-dense with intact cellular membranes and flagella, indicating
that they are likely to be viable (Zheng et al. 9 1999). These forms remain capa-
ble of reducing tetrazolium (INT) for extended periods in water (Gribbon and
Barer, 1995). SDS-PAGE (sodium dodecyl sulphate polyacrylamide gel elec-
trophoresis) demonstrated that most protein bands appeared to be similar in
both the spiral and coccoid forms, thus supporting the notion that some of the
coccoid forms of H. pylori are likely to be viable (Zheng et al. 9 1999). H. pylori
undergoes substantial modification of its unique muropeptide composition dur-
ing morphological transition to coccoid forms. The accumulation of dipeptide
monomers and a concomitant reduction in tri- and tetrapeptide monomers con-
stitutes the most dramatic modification (Costa et al. 9 1999). This suggests that
activation of a 7-glutamyl-diaminopimelate endopeptidase leads to massive
conversion of tri- and tetrapeptide monomers into dipeptide monomers, as pre-
viously observed in sporulating Bacillus sphaericus (Vacheron et aL 9 1979). This
accumulation of disaccharide-dipeptides appears to be a result of convergent
evolution between the distantly related bacteria H. pylori and B. sphaericus in
the genesis of resistant forms, i.e. coccoid cells and endospores, respectively.
Muropeptide composition remains essentially stable from the time coccoid cells
become predominant and remains so for at least 2 weeks (Costa et al. 9 1999).

The role of the VBNC H. pylori and coccoid forms in infection and trans-
mission remains unclear. Attempts at successful infection of animals with coc-
coid H. pylori are conflicting (Cellini et al. 9 1994; Eaton et al. 9 1995; Wang
et al. 9 1997). In vitro reversion of the coccoid morphology to the bacillary form
has been reported (Anderson etal. 9 1997; Kurokawa et al. 9 1999). Deprivation
of nutrients and a non-permissive temperature act as a powerful trigger for
H. pylori programmed cell death, apparently as a means of species preservation
(Cellini et al. 9 2001). Pre-incubation in non-nutrient solution and high density
of bacterial concentration appear to be important for recovery of H. pylori cul-
tured for a prolonged time under anaerobic conditions (Yamaguchi et al. 9 1999).

Helicobacter pylori

Origin and taxonomy

The discovery of H. pylori by Warren and Marshall in Western Australia in
1983 (Warren and Marshall, 1983) not only introduced a whole new group of
bacteria to science, but also revolutionized our concept of gastroduodenal
pathology, in particular peptic ulcer disease. Although the presence of spiral
bacteria in the human stomach had been reported in the literature (Freedberg
and Barron, 1940; Steer and Colin-Jones, 1975) H. pylori was not success-
fully cultured until the 1980s (Marshall and Warren, 1984). H. pylori was
subsequently found to cause peptic ulcer disease and gastric cancer (Marshall
and Warren, 1984; Blaser, 1987). This spiral organism isolated from the stom-
ach was originally called Campylobacter pyloridis, but this was later changed
to the more grammatically correct C. pylori. Further studies showed that the
organism differed sufficiently from true Campylobacters justifying the need of
a new genus, Helicobacter. It soon became apparent that similar organisms
colonized the stomach of a wide variety of animals other than humans, and
those spiral bacteria colonizing the intestines of rodents and other animals
also belonged to Helicobacter.

Helicobacter are phylogenetically closely related to Campylobacter,
Wolinella and Arcobacter, thus belonging to the delta-epsilon group of
Proteobacteria (Wesley et al., 1995; Bunn et al., 1997). Initially classified as
a species in the genus Campylobacter, its unique 16S rRNA sequence data,
presence of sheathed flagella, and a distinct fatty acid and outer membrane
protein profile led to the establishment of the genus Helicobacter (Goodwin
et al., 1989). Approximately 20 Helicobacter species have been described in
the literature, and some are now regarded as human pathogens. Figure 8.1
summarizes the taxonomy of the Helicobacter genus as determined by
sequencing of the 16S ribosomal RNA.

Metabolism and physiology

H. pylori is a fascinating organism. It has a mixture of aerobic and anaerobic
physiologies that combine to produce a microaerophilic physiology, the
molecular basis for which is not fully understood. Part of the microaerophilic
nature is probably due to oxygen-sensitive enzymes within its central meta-
bolic pathways. The biochemical basis for the requirement for CO2 has not
been completely explained and there is a surprising lack of anaplerotic carbox-
ylation (C0 2 fixation) enzymes (Kelly, 1998).

The organism has solute transport systems and an incomplete tricarboxylic
acid (TCA) cycle, all contributing to complex nutritional requirements.
H. pylori has an absolute growth requirement for arginine, histidine, leucine,
isoleucine, valine, methionine and phenylalanine (Doig et al. 9 1999). The
respiratory chain of H. pylori is remarkably simple, apparently with a single

Bacteriology

Wolinella succinogenes

Campylobacter jejuni

Helicobacter pullorum

Helicobacter trogontum

Helicobacter fennelliae

Helicobacter sp. Mainz

Helicobacter hepaticus

Helicobacter muridarum

Helicobacter cinaedi

Flexispira rappini

Helicobacter bilis

Helicobacter canis

Helicobacter pametensis

Helicobacter mustelae

Helicobacter acinonyx

Helicobacter pylori

Helicobacter nemestrinae

Helicobacter heilmannii

Helicobacter felis

0.1

Figure 8.1 Phylogenetic tree of Helicobacter spp. and related organisms (adapted
from Cantet etal., 1999).

terminal oxidase and with fumarate reductase as the only reductase for anaer-
obic respiration. NADPH appears to be the preferred electron donor in vivo,
rather than NADH as in most other bacteria (Kelly, 1998). The non -cyclic
TCA cycle, characteristic of anaerobic metabolism, is directed towards the
production of succinate in the reductive dicarboxylic acid branch and alpha-
keto-glutarate in the oxidative tricarboxylic acid branch. Both branches are
metabolically linked by the presence of alpha-ketoglutarate oxidase activity.
H. pylori does not possess a gamma-aminobutyrate shunt, owing to the absence
of both gamma-aminobutyrate transaminase and succinic semialdehyde dehydro-
genase activities (Pitson et ah, 1999). Genomic analyses suggest that glucose or

Helicobacter pylori

malate is not the primary source for production of pyruvate in H. pylori but,
rather, that lactate, L-alanine, L-serine, and D-amino acids are the primary
sources. The Entner-Doudoroff and pentose phosphate pathways have been
shown to be active rather than glycolytic pathways (Doig et aL 9 1999).

While the urease of most bacterial species has been found to possess three
distinct subunits, urease-producing Helicobacter produce a unique two-
subunit enzyme which exhibits remarkable submillimolar K values (Mobley
et al.> 1995). The small subunit (urea) of H. pylori probably represents a
structural gene fusion. A complex urease-independent chemotaxis signal
transduction system suggests that chemotaxis (toward urea, Na + , and bicar-
bonate) is an important feature of H. pylori physiology (Yoshiyama et al.,
1998). Since urea and sodium bicarbonate are secreted through the gastric
epithelial surface and hydrolysis of urea by urease on the bacterial surface is
essential for colonization, the chemotactic response of H. pylori may be cru-
cial for its colonization and persistence in the stomach (Mizote et aL 9 1997).
Urease enzyme activity is the best-characterized colonization factor, protect-
ing the bacteria from gastric acid exposure (Dunn, 1993). Urease catalyses the
hydrolysis of urea to yield carbonic acid and two molecules of ammonia,
which in solution are in equilibrium with their protonated and deprotonated
forms, respectively. The net effect of this reaction is an increase in pH. In add-
ition to buffering acidic environments, Helicobacter urease activity plays
a critical role in nitrogen assimilation (Chin et al., 2001). Ammonia can be
assimilated in H. pylori by conversion of glutamate to glutamine. Lack of a
glutamate synthase suggests that a-ketoglutarate is transformed into glutamate
by glutamate dehydrogenase and that H. pylori is adapted to an ammonia rich
environment (Marais et al., 1999).

Clinical features

Once colonized by Helicobacter pylori, the human host can remain infected
for life unless intensive antimicrobial therapy is administered. It is the princi-
pal cause of chronic active gastritis, stomach and peptic ulceration, and clas-
sified as a Class I carcinogen for gastric cancer and gastric mucosa-associated
lymphoid tissue (MALT) lymphoma (Tytgat and Rauws, 1990; Aruin, 1997;
Nedrud and Czinn, 1999). Acquisition of H. pylori infection early in life
appears to be associated with early-onset gastric corpus atrophy and meta-
plasia, and a higher risk of cancer. Peptic ulcer disease (PUD) occurs at the
rate of 1% per annum in infected individuals (Figure 8.2).

Gastric carcinogenesis is a multifactorial, multistep process, in which
chronic inflammation plays a major role. H. pylori infection induces physio-
logical changes and DNA adduct formation in the gastric microenvironment
(Farinati et al., 1988). Reduction in antioxidant (ascorbate) levels increases
the risk of carcinogenesis and damage to DNA from intragastric release of

Bacteriology

gastric lymphoma

duodenal ulcer

gastric cancer

Figure 8.2 Relation of H. pylori infection to clinical pathology (source: The
Helicobacter Foundation, www.pylori.com).

Normal gastric mucosa

H. pylori

Genetic makeup

Dietary factors
smoking

Chronic active gastritis

Antisecretory Rx
gastric pH \

pHf

Multifocal atrophic gastritis

H. pylori

Ascorbic acid J

Intestinal metaplasia

Mutagens

Dysplasia

Cancer

Figure 8.3 Proposed role of H. pylori in gastric cancer.

free radicals. Ingestion of dietary carcinogens, deficiencies in dietary antioxi-
dants, smoking, and anti-secretory (proton pump inhibitor) medications are
thought to be important cofactors in the genesis of Helicobacter-* elated
cancer (Figure 8.3).

In persons who do not develop peptic ulcer, long-term H. pylori gastritis
can lead to intestinal metaplasia, associated with gastric adenocarcinoma.
Chronic gastritis is also associated with the presence of lymphoid follicles
in the lamina propria. This mucosa-associated lymphoid tissue (MALT) can
develop into lymphoma. In cases of gastric MALT lymphoma, eradication of

Helicobacter pylori

H. pylori infection has been shown to cure over 50% of patients
(Wotherspoon et al. 9 1993).

According to observations from human and animal Helicobacter infections,
the genus has the capacity to colonize and cause inflammation in the stomach
(H. pylori, H. heilmannii, H. mustelae), colon (H. fennelliae) and liver (H.
bilis). Many people who are infected with H. pylori are asymptomatic. For
those who have gastritis or ulcers, the most common symptoms are nausea,
abdominal pain, heartburn, or bleeding. Gastritis is an irritation in the lining
of the stomach; an ulcer is a sore in the lining of the stomach or duodenum.
If left untreated, ulcers can become life threatening. While some medications or
too much stomach acid can also cause gastritis and ulcers, the most common
cause is H. pylori infection. Acute H. pylori infection manifests as nausea and
abdominal pain, lasting between 3 and 14 days. During this time hypochlor-
hydria develops and may persist for up to a year. Following infection of the
gastric epithelium, a marked inflammatory response in the mucosa occurs
within days. This is induced initially by secretion of interleukin-8 from surface
epithelial cells, accompanied by release of the intercellular adhesion molecule
ICAM-1. The initial superficial gastritis can progress to one of three forms.
The first is chronic active non-atrophic gastritis, predominantly antral,
referred to as diffuse antral gastritis. This form increases the risk of duodenal
ulcer. The second is chronic active gastritis with some atrophy, which places
the person at risk from gastric ulcers. The third is chronic atrophic pangastritis
with significant atrophy, which increases the risk of gastric cancer (Blaser,
1987). There are more than 2 million physician visits per year for duodenal
ulcers, 90% of which are attributable to H. pylori (other causes include non-
steroidal anti-inflammatory drug (NSAID) usage and Zollinger-Ellison syn-
drome), and more than 3 million physician visits per year for gastric ulcers,
in which 80% of patients with non-NSAID-induced gastric ulcers are infected
(Blaser, 1997). In a prospective study, the risk of developing duodenal ulcer
disease in H. pylori-'miected patients followed for 10 years exceeded 10%; in
contrast, it was less than 1% in uninfected patients (Monath et al., 1998).
Development of atrophy and metaplasia of the gastric mucosa is strongly
associated with H. pylori infection. Approximately 40-50% of infected sub-
jects develop these conditions, which are rare in non-infected subjects
(Kuipers, 1999). The risk of developing gastric cancer is estimated to be three-
to sixfold higher in infected than in uninfected individuals (Antonioli, 1994).
H. pylori is found in close contact with the mucosa in the antrum of the stom-
ach, often between cells under a stable layer of mucus that shields the organ-
isms from gastric acid. Evidence suggests that H. pylori infection of humans is
ancient, and often the interaction is not overtly destructive. H. pylori is almost
always present when inflammation is present, and is rare when inflammation
is absent, indicating H. pylori are not simple commensals. An equilibrium
between virulence of H. pylori and the mucosal defence mechanism has been
proposed (Loffeld and Arends, 1993). In this case, the bacterium is neither
commensal nor pathogen, being able to change its role depending on the local
microenvironment.

Bacteriology

Virulence

H. pylori is a remarkably well-adapted organism which can persist indefin-
itely in the hostile environment of the stomach, including the vigorous
humoral and cellular immune response mounted against it. Despite the vigour
of this response, the infection in many cases is not eradicated. H. pylori must
possess several physiological adaptations to allow it to persist in the hostile
environment of the human stomach. The mechanisms by which H. pylori can
colonize and persist at 10 4 -10 7 H. pylori per gram of gastric mucus are prob-
ably complex (Nowak et al., 1997). Motility, urease activity and association
with gastric mucosal cells are important colonization factors (Nakamura
et al., 1998; Beier et aL 9 1997). H. pylori, with optimal growth in the neutral
pH range, is not an acidophile and requires mechanisms to survive stomach
acid. By breaking down urea present in the gastric juice and extracellular
fluid, the organism is able to generate bicarbonate and ammonia in its peri-
cellular environment, so that hydrogen ions are effectively neutralized before
damaging the cell. H. pylori is thereby able to survive in gastric acid long
enough for it to colonize the gastric mucosa.

Once attached to the gastric mucosa, H. pylori can damage host tissue by
causing vacuolation in the epithelial cells. This vacuolation is caused by the pro-
duction of a cytotoxin called vacuolating cytotoxin A (VacA), a protein which
is endocytosed by epithelial cells where it causes endosome-lysosome fusion
(Ge and Taylor, 1999). Variations of cytotoxin occur such that more aggressive
forms are likely to be associated with peptic ulcer, whereas more benign forms
of cytotoxin may be associated with gastritis without ulcers (Atherton et al.,
1995). Although individuals infected with H. pylori are often asymptomatic,
they always have histological changes of chronic gastritis. Colonization of gas-
tric mucosa by H. pylori induces gastritis with resultant neutrophil infiltration
(Fujioka, 1995). Chronic gastritis is characterized by the accumulation of
oxidative DNA damage with mutagenic and carcinogenic potential (Farinati
et al. 9 1988). H. pylori seems to be particularly resistant to the oxidative
inflammatory response of neutrophils, which can in turn damage the host gas-
tric mucosa (Basso et al. 9 1999). The nature of this resistance is not well under-
stood (Odenbreit et aL 9 1996). Most H. pylori bacteria found in the stomach
are in the layer of mucus overlying the epithelium. Some penetrate the mucous
layer and adhere to the gastric epithelial cells with formation of a pedestal. A
few H. pylori are seen between adjacent cells in proximity to tight junctions.

Several H. pylori adhesins have been described that bind to Lewis b anti-
gens with terminal fucose residues (found in humans with blood group O), to
sialic acid-lactose residues and to phosphatidylethanolamine, a glycolipid
receptor on the gastric antral mucosa. The adhesin binding to blood group
O-specific antigens may help explain the predispostion of people with blood
group O to peptic ulcer disease and gastric adenocarcinoma.

Apart from urease, which may cause damage to host cells, a haemolysin and
vacA have been described. CagA, a high molecular weight protein product of

Helicobacter pylori

a cytotoxin associated gene (cagA), is produced by about 60% of H. pylori
strains and is associated with the expression of vacA. CagA-positive strains
are also associated with the presence of duodenal ulceration, although it is
unclear whether CagA has an independent pathogenic role. CagA is an island
of approximately 30 genes which have apparently been acquired by H. pylori
from another organism since the guanine and cytosine content of this island dif-
fers from that seen in the rest of the H. pylori genome. The CagA pathogen-
icity island contains genes which function as a Type IV secretory system.
Subsequent induction of interleukin-8 production attracts neutrophils through
the lamina propria, and emerges between the epithelial cells. Rapid recruitment
of inflammatory cells into the mucosa results, and these cells then express a
repertoire of cytokines. The intense release of inflammatory mediators, along
with substances that H. pylori itself elaborates, results in damage to the gastric
epithelial barrier. In addition, H. pylori lipopolysaccharide (LPS) is suspected
of driving production of auto-antibodies through molecular mimicry of human
Lewis blood group antigens present on parietal cell H + /K + -ATPase (Namavar
et al, 1998).

The lipopolysaccharide of H. pylori appears to have an unusual structure
related to the composition of the fatty acids that form the hydrophobic region
of lipid A with the presence of 3-hydroxyoctadecanoic acid. The lipid A portion
of Helicobacter LPS appears to have lower biological activity than LPS from
other enteric bacteria. Some strains of H. pylori possess LPS with a ladder-like
side chain like those of 'smooth' strains of Enterobacteriaceae; others give a
'rough'-type profile. Strains may change from smooth to rough LPS when
grown on conventional solid media and may be reversible when grown in
liquid medium. Although LPS core antigens are shared, side chain antigens are
strain-specific. The distribution of specific LPS antigens among strains is differ-
ent from that of protein antigens. Antigenic differences in LPS from different
strains has been detected by immunoblotting and haemagglutination assays.

Treatment

Helicobacter pylori is sensitive to penicillins (including benzylpenicillin),
cephalosporins, tetracycline, erythromycin, rifampicin, aminoglycosides and
nitrofurans, but resistant to nalidixic acid, though sensitive to the more active
quinolones such as ciprofloxacin. It is highly resistant to trimethoprim and
moderately resistant to polymyxins. H. pylori is usually susceptible to metron-
idazole but resistance rates are variable and may reach 50% in some areas.
H. pylori is sensitive to colloidal bismuth compounds commonly prescribed for
gastric disease in concentrations easily attainable in the stomach. The proton
pump inhibitor omeprazole has mild in vitro activity against H. pylori, 1% bile
salts are also inhibitory. Eradication of H. pylori has been shown to be a defini-
tive cure for duodenal ulcer and most gastric ulcers. In the 1980s, treatment for

Bacteriology

H. pylori was difficult since combinations of bismuth, tetracycline and metron-
idazole were required for adequate eradication of the organism (Borody et al. 9
1989). The action of amoxicillin was found to be greatly enhanced when gas-
tric acid was suppressed with a proton pump inhibitor, notably omeprazole
(Unge et al. 9 1989). As a result, H. pylori is treated with a 7-day therapy of
omeprazole (to render the gastric pH neutral) in combination with two anti-
biotics, usually amoxicillin and clarithromycin. Omeprazole, clarithromycin,
and metronidazole combinations have achieved similar high cure rates. For dif-
ficult to eradicate infections, bismuth, tetracycline, metronidazole and omepra-
zole are usually successful (Kung et al. 9 1997).

No current treatment regimen for H. pylori is universally effective, even
with triple and quadruple therapies (Vyas and Sihorkar, 1999). Three of the
six most widely prescribed medications in the USA are for the treatment of
ulcers. PUD treatment costs exceed $4 billion each year, not including indirect
costs due to work and productivity loss (Vakil, 1997). Recent efforts toward
therapeutic immunization by oral recombinant H. heilmannii or H. pylori
urease given with cholera toxin or E. coli heat-labile enterotoxin, respectively,
induced gastric corpus atrophy in mice and failed to eradicate H. pylori in
infected individuals (Dieterich et al. 9 1999; Michetti et al. 9 1999). Expression
of stable immunogenic antigens in an attenuated Salmonella typhi vector,
an effective vaccine approach for typhoid fever, failed to induce detectable
mucosal or systemic antibody responses for H. pylori in human volunteers
(DiPetrillo e* a/., 1999).

Once the exact mode of transmission is understood in different commu-
nities then effective public health measures can be started. In areas where
H. pylori exists in the environment, humans do not seem to be able to mount
a protective immune response following natural infection. Therefore, treat-
ment is useless, and the only effective way of eliminating H. pylori from the
population would be via public health measures, i.e. improved sanitation and
standard of living, or vaccination.

Survival in the environment

Helicobacters can be broadly divided into those that colonize the stomach
mucosa and those that colonize the intestines of humans and animals,
although some intestinal species are capable of colonizing the stomach when
acid secretion is defective. Some gastric species only colonize the non-acid
secreting mucosa, whereas others, such as Helicobacter felis, can colonize the
canaliculi of acid-secreting oxyntic cells.

The surface of the human stomach mucosa is the major habitat of H. pylori.
Almost all isolations are from gastric biopsy specimens, but the organism has
occasionally been detected in gastric juices, saliva, dental plaque, bile and fae-
ces. In developed countries prevalence rates increase with age, from about 20%
in young adults to about 50% in people over 50 years old, but in developing

100

Helicobacter pylori

countries rates are much higher and children are commonly infected.
H. pylori has been isolated from domestic cats which raises the question
about the role of this domestic animal in transmission of disease to humans.
There is some suggestion that water may be a source for Helicobacter infec-
tion, but whether the organisms merely exist or have some ecological niche in
natural waters is unknown.

The precise mechanism of transmission of H. pylori is unknown, but any
mode that introduces the organism into the stomach of a susceptible person
may lead to infection. Several routes of transmission have been proposed:
gastric-oral, oral-oral, faecal-oral, zoonotic and water/food-borne. Data sup-
porting all routes have been published and it is likely that infection occurs
through multiple transmission pathways (Goodman et al. 9 1996; Velazquez
and Feirtag, 1999). Recent observations in persons infected with H. pylori
caused to vomit or have diarrhoea showed that an actively unwell person with
these symptoms could spread H. pylori in the immediate vicinity by aerosol,
splashing of vomitus, infected vomitus and infected diarrhoea (Parsonnet
et aL 9 1999). In developed countries, H. pylori is more difficult to acquire and
usually is transmitted from one family member to another, possibly by the fae-
cal-oral route, or by the oral-oral route.

Faecal-oral transmission of H. pylori has been suggested by several
researchers being implicated with crowding, socioecomonic status and con-
sumption of raw, sewage-contaminated vegetables as risk factors for infection.
Studies in Peru have identified the type of water supply as a risk factor for
infection and have found that water source appeared to be more important as
a risk factor than socioecomonic factors. The number of people in the house-
hold, sharing a bed, and a lack of proper sanitation and a permanent hot water
supply all increase the risk (Mendall et al. 9 1992; Whitaker et al. 9 1993; Malaty
and Graham, 1994). The gastric-oral mode of transmission of H. pylori would
favour spread in small units of interacting people, such as in families.
Interfamilial transmission is likely to occur, however, diverse H. pylori species
have been detected in infected couples, suggesting other possible reservoirs
(Kuo et al. 9 1999). In developing countries, many children are already infected
by the age of 10 years, and relapse can be a serious problem (Dooley et al. 9
1989; Ramirez-Ramos et al. 9 1997; Gurel et al. 9 1999). Data regarding
acquisition and loss of H. pylori infection are critical to understanding the
epidemiology of the infection and to developing treatment and vaccination
strategies. In a Canadian prospective 3 -year cohort study of 316 randomly
selected subjects, a continuous risk of acquisition of 1% per year rather than a
cohort effect was concluded.

The prevalence of H. pylori infection varies according to age and country
of origin. In developed countries prevalence of infection with H. pylori is
clearly age related, but varies according to ethnic group and socioeconomic
status with only about 0.5% of adults developing new infections each
year (Graham et al. 9 1991; Vaira et al. 9 1998). H. pylori seropositivity was
found among 1161 (45%) of 2598 healthy Italian adults in which age
(67% seropositivity in subjects aged 50 or older) and sociocultural class were

101

Bacteriology

identified as persisting determinant features of H. pylori infection (Russo
et al. 9 1999). In countries whose economy has gone from developing to affluent
over the past 50 years, marked decline in H. pylori prevalence has been noted.

Although the natural niche for H. pylori is the human stomach, for wide-
spread infection the organism may need to survive in the external environ-
ment (Brown, 2000). PCR detection of H. pylori in oral and fingernail
samples in a rural population in Guatemala suggests that oral carriage of H.
pylori may play a role in the transmission of infection and that the hand may
be instrumental in transmission (Dowsett et al. 9 1999). However, the low fre-
quency (<1%) of H. pylori in dental plaque suggests that oral transmission is
not a significant reservoir in adult populations (Luman et al. 9 1996; Cave,
1997; Kamat et al. 9 1998; Oshowo et al. 9 1998a, b). A recent study found that
20 of 21 Japanese couples (both members were positive for H. pylori) showed
restriction enzyme patterns that differed between spouses, indicating that in
Japan, interspousal transmission of H. pylori occurs rarely (Suzuki et al. 9
1999).

Evidence seems to favour a faecal-oral route by which the bacterium,
excreted with faeces, might colonize water sources and subsequently be trans-
mitted to humans (Xia and Talley, 1997). H. pylori can be cultured from the
faeces of infected individuals (Thomas et al. 9 1992; Kelly et al. 9 1994) and
specific DNA sequences have been amplified from raw sewage (Forrest et al. 9
1998). Experimental murine models support a faecal-oral, rather than oral-
oral route as the mode of transmission of H. pylori infection (Cellini et al. 9
1999; Yoshimatsu et al. 9 2000).

H. pylori infection in Italian shepherds approaches 100%, who are about
80 times as likely to be infected with H. pylori as are their siblings who are not
shepherds (Dore et al. 9 1999a). Raw milk samples from 60% of sheep on Italian
farms contained traces of H. pylori DNA, and two H. pylori isolates from
sheep milk and tissue were isolated, indicating that H. pylori infection
includes phases in the environment (Dore et al. 9 1999b). The authors suggest
that sheep may be the source of infection in humans, however, they did not
rule out the possibility that humans might contaminate the sheep samples.
Domestic cats and old-world macaques have been found to be colonized with
H. pylori 9 however, it is doubtful whether this species provides an important
reservoir for human infection (Bode et al. 9 1997; Osata et al. 9 1997). A poten-
tial role of flies in vectoral spread of H. pylori from human faeces to food was
supported by the amplification of H. pylori DNA from flies in developing
countries (Grubel et al. 9 1998). H. pylori was rarely detected by PCR on chop-
sticks following eating and hence, the risk of infection via the use of chop-
sticks was concluded to be low (Leung et al. 9 1999).

While the principal route of transmission remains inconclusive, geographic
prevalence of H. pylori infection (Figure 8.4) shows a strong correlation with
access to clean water (Figure 8.5).

Children in Peru whose homes were supplied with municipal water were 12
times more likely to be infected than those whose water supply came from com-
munity wells (Klein etal. 9 1991). Similar seroprevalence patterns in a community

102

Helicobacter pylori

7g/6* t , "

Figure 8.4 Geographic prevalence of H. pylori infection (data: Helicobacter
Foundation, www.pylori.com).

PERCENT OF POPULATION IN URBAN AREAS HAVING ACCESS TO CLEAN WATER

| LESS THAN 75

75 TO 94.9

95 OR MORE □ NO DATA

Figure 8.5 Geographic access to clean water (1980-90) (source: The Water
Resources Institute, www.sciam.com/1197issue/1197scicit5.html).

study between H. pylori and hepatitis A virus (HAV) indicate H. pylori may
have spread in a manner similar to organisms transmitted by a faecal-oral route
(Pretolani et aL, 1997). A survey of 3289 residents (416 families) in northern
Italy for prevalence of H. pylori IgG antibodies suggested a common source of

103

Bacteriology

exposure for infection (Dominici etal. 9 1999). In the Colombian Andes, swim-
ming in streams, using streams as a drinking water source, and frequent con-
sumption of raw vegetables increased the odds of infection (Goodman et al. 9
1996). Other epidemiological reports do support an association between
H. pylori infection and consumption of untreated well or spring water (Carballo
et al 9 1997; Bunn et al. 9 1997). A longitudinal seroconversion study of children
from Bolivia found a striking increase in the incidence of infection, which rose
from less than 10% in 2 year olds or younger to over 35% in children over
2 years (Friedman, 1998). Interestingly, this study was set up on the back of
a social programme to improve the hygiene by giving families plastic canisters
of chlorinated drinking water. Factors associated with seroconversion included
H. pylori infection in another household member as a positive risk factor and
the use of chlorinated drinking water as a strong protective factor.

Water and water biofilm reservoirs of infection could account for many of
the published conclusions, which have been attributed to other factors: e.g.
socioeconomic factors such as overcrowding, which will also bring associated
sharing of common water supplies. This idea is supported by a study of
H. pylori infection rates in French submarine crewmembers with a common
tanked water supply (Hammermeister et al. 9 1992). The argument for a water-
borne route of H. pylori transmission is supported by maintenance of viability
in spiked natural water (West et al. 9 1992; Shahamat et al. 9 1993; Hunter,
1997; Fan et al. 9 1998; Jiang and Doyle, 1998; Sato et al. 9 1999).

H. pylori conserve the capability to produce acid-inducible proteins for at
least 100 days when stored at 4°C in either phosphate-buffered saline (PBS) or
distilled water (Mizoguchi et al. 9 1999). While attempts to culture H. pylori
from environmental water samples have been unsuccessful, closely related
microaerophilic organisms, Campylobacter jejuni and Arcobacter butzleri 9
have been cultured from ground and surface waters (Arvanitidou et al. 9 1994;
Stanley et al. 9 1998; Rice et al. 9 1999) and associated with waterborne out-
breaks (MMWR, 1999).

Initial molecular probe evidence of H. pylori from environmental water sam-
ples came from PCR amplification of samples from Columbia, where infection
rates are over 90% (Schauer et al. 9 1995; Handwerker et al. 9 1995). A two-
stage in vitro method for detection of H. pylori in spiked water and faecal sam-
ples using an initial concentration by immunomagnetic separation followed by
PCR detection (IMS/PCR) has been described (Enroth and Engstand, 1995).
IMS/PCR was used to detect H. pylori from water in Peru and Sweden (Hulten
et al. 9 1996, 1998). Water samples from a water delivery truck and two lakes
in the Canadian Arctic were PCR positive for H. pylori (McKeown et al. 9
1999). In a region of Japan with a high infection rate, H. pylori-specific DNA
was detected in water, field soil, flies and cow faeces by nested PCR, and the
aligned urease partial sequences of the PCR products were highly homologous
(96-100%) with the H. pylori sequence in the GenBank database (Sasaki et al. 9
1999). PCR analysis of river water in Japan detected H. pylori-specific DNA
in 8 of 62 (13%) of samples (Fukuyama et al. 9 1999). Actively respiring
H. pylori from surface and well water in the USA were detected using fluorescent

104

Helicobacter pylori

antibody-tetrazolium reduction (FACTC) microscopy (Hegarty and Baker,
1999) and confirmed using species -specific PCR (Baker and Hegarty, 2001). All
of these findings suggest the presence of H. pylori in the natural environment
and a possible waterborne route of transmission.

The challenge remains to determine the importance of waterborne, as
opposed to other modes, of transmission. It is possible that a specifically
adapted form of H. pylori, or association with a biofilm community, may be
required for environmental persistence and transmission (Vincent, 1995).
Waterborne bacteria can attach to surfaces by aggregating in a hydrated exo-
polymer glycocalyx matrix of their own synthesis to form biofilms (Costerton
et aL 9 1999). Association with biofilm communities within a water distribu-
tion system can offer bacteria protection from disinfection and protozoan preda-
tion (Sibille £tftf/., 1998).

Water

The geographic prevalence of H. pylori infection shows a strong correlation
with access to clean water. In 1998, the US Environmental Protection Agency
(US EPA) Office of Ground Water and Drinking Water included H. pylori on
its Contaminant Candidate List (CCL) (62 Federal Register 52193). This list
designates contaminants which are not regulated under current EPA drink-
ing water regulations, but which are anticipated to occur in water systems.
Inadequate, interrupted or intermittent drinking water treatment has repeat-
edly been associated with waterborne outbreaks (Moore etal. 9 1994). Among
waterborne disease outbreaks between 1971 and 1994, the distribution sys-
tem has been considered the cause of 30% of the 272 outbreaks in community
systems. At least two of these outbreaks resulted in hospitalizations and six
deaths (Shaw and Regli, 1999). Regulations to reduce microbial risk have
focused primarily on reducing risk from contaminants in the source water by
providing some means of control at a treatment plant. Risks from growth of
bacteria in the distribution system or from intrusion of pathogens have not
been comprehensively addressed.

Distribution-system failures have been attributed to intrusion of pathogens
through leaks, surges, backflow, cross-connections or as the result of unsani-
tary maintenance practices. Transient pressure gradients routinely occur in
distribution pipes and even well-run systems can experience leakage of
10-20% of the produced water. Therefore ample opportunities exist for
intrusion of contaminated water. Sources of pathogenic microorganisms
come primarily from leaking sewer lines located within 18 inches of the
drinking water pipeline. Bacteria can enter water via either point or non-
point sources of contamination. Point sources are those that are readily
identifiable and typically discharge water through a system of pipes. Sewered
communities may not have enough capacity to treat the extremely large
volume of water sometimes experienced after heavy rainfalls, and treatment

105

Bacteriology

facilities may need to bypass some of the wastewater. During bypass or
other overflow events, bacteria-laden water is discharged directly into the
surface water.

The existing literature on oxidative disinfection of H. pylori is not well dev-
eloped. H. pylori in brain-heart infusion with 10% horse serum survived up to
24 hours following addition of 0.1 mM H2O2, a potent source of reactive oxygen
metabolites (ROM), whereas E. colt had limited (<5 hours) survival under iden-
tical conditions (Barton et al. 9 1997). Chlorine and ozonation studies indicate
that H. pylori may be capable of surviving disinfection practices adequate to
remove £. coli 9 thereby allowing entry into public and private drinking water
distribution systems (Johnson et al. 9 1997; Hegarty et al. 9 1999). The mechan-
ism by which H. pylori effectively persists within an environment of chronic
oxidative stress (Hazen et al. 9 1996; McGowan et al. 9 1997; Suzuki et al. 9 1997;
Henderson et ai 9 1999; Santra et ai 9 2000) is intriguing in regard to transmission
implications.

The dissemination of viable H. pylori cells through faecal material (Thomas
et al. 9 1992; Mapstone et al. 9 1993; Kelly et al. 9 1994) may provide a route for
the contamination of drinking water and H. pylori may survive for prolonged
periods in water over a range of physical variables (West et al. 9 1992).
H. pylori strains were found to survive for longer periods in physiological
saline concentrations, low temperatures and a pH ranging from 5.8 to 6.9.
The addition of urea was found to result in a reduction in survival times,
whereas the addition of bovine serum albumin led to variable survival times.
Using autoradiography, temperature was found to be an important factor in
survival of the bacterium in water. Indeed H. pylori was found to remain
viable for periods ranging from 48 hours to between 20 and 30 days depend-
ing on the conditions under which the organism was studied (Shahamat et al. 9
1993) indicating strong evidence for support of a waterborne for route for
H. pylori. However, H. pylori cells are readily inactivated by chlorine sug-
gesting that the organism would be controlled by disinfection regimes nor-
mally employed in the treatment of drinking water (Johnson et al. 9 1997).

The absence of a proven analytical method or a standard protocol for
assessing cell viability in water has resulted in the publication of a variety of
methods and survival times for H. pylori in the aquatic environment.
Although evidence has been published that viable but non-culturable
H. pylori cells are still alive (Barer et al. 9 1993; Oliver, 1993; Shahamat et al. 9
1993), no evidence has been published on resuscitation of these cells or on
their ability to cause infection.

As mentioned previously, in drinking-water systems microorganisms are
predominantly associated with surfaces as biofllms rather than in the water
itself (Costerton et al. 9 1999). Waterborne enteric pathogens such as E. colt
(Mackerness et al. 9 1993) and Campylobacter spp. make use of biofllms as a
vector and reservoir, but do not survive for prolonged periods in drinking
water itself. MacKay's work suggests that biofllms in water distribution sys-
tems may also harbour H. pylori (MacKay et al. 9 1999; Park et al. 9 2001).
However, viability and culturability was not measured in these studies.

106

Helicobacter pylori

Disinfection study evidence for water transmission of Helicobacter pylori
There is currently a paucity of data concerning the effectiveness of standard
drinking water disinfection processes on Helicobacter pylori. In a paper by
Baker et al. (2002) Helicobacter pylori was found to be more resistant to low
doses of free chlorine than either Campylobacter jejuni or E. coli. From this
study it was found that exposure to 0.2mg/l free chlorine for 1 minute
resulted in a 4 log reduction in the number of E. coli and C. jejuni cells but less
than a two log reduction of Helicobacter pylori cells. Calculated free chlorine
CT99 values, in this study, for C. jejuni, E. coli and H. pylori were 0.03, 0.13
and 0.25 mg/l-min respectively. A similar result was observed in the ozonation
studies carried out by the same group.

Based on the study by Baker et al. (2002) it seems feasible to suggest that
under conditions of inadequate disinfection, Helicobacter pylori may persist
in certain aquatic environments in the absence of E. coli. It may therefore be
reasonable to suggest that H. pylori may be present in water when enumer-
ation of coliforms on standard selective media indicates that the water is
potable. Therefore current monitoring methods for drinking water may indi-
cate that water is safe while Helicobcter pylori may actually be present. In
addition to this, it may be conceivable to suggest that H. pylori may be able
to persist for extended periods in drinking-water distribution systems con-
taining low levels of residual chlorine and/or chloramines. This is reinforced
when we consider the CT 99 values (at pH 6.0) of 0.045 and 0.12 mg/l-min for
E. coli and Helicobacter pylori, respectively (Johnson et al., 1997) which
suggest that H. pylori may be more resistant to free chlorine than E. coli. The
data generated by Baker (2002) do not support the speculation by Johnson
et al. (1997) that the difference in CT 99 values between E. coli and H. pylori
in their disinfection experiments can be solely attributed to the presence of
agar debris or H. pylori cell aggregates. Differences between the Baker et al.
(2002) and Johnson et al. (1997) studies might have accounted for differ-
ences in observed CT 99 values. These differences include the presence and
absence of inorganic salts, absence and presence of phosphate buffer and
temperature (20°C and 5°C) respectively. In addition, while disinfection
CT 99 values for E. coli were determined at pH 7.0 in Baker's study, Johnson
et al. referred to pH 6.0 data determined by an unrelated study. The some-
what higher H. pylor'v.E. coli CT 99 ratio (0.12:0.045) of Johnson et al. com-
pared with Baker's study (0.25:0.13) can be attributed to the reported
presence of H. pylori cell aggregates and debris in the former study.
Considering the tendency of H. pylori to form aggregates and biofllms, the
actual CT 99 of H. pylori in a natural aquatic environment may be consider-
ably larger.

Chlorine residuals in distribution water systems often range from 0.1 to
0.3 mg/1 (Geldreich, 1990). Inadequate, interrupted, or intermittent treatment
has repeatedly been associated with waterborne disease outbreaks. Reduced
chlorine residuals may not provide adequate inactivation of H. pylori to
prevent entry into and persistence within the water distribution systems.
Assuming the mean chlorine residual (free and combined chlorine) of treated

107

Bacteriology

water is l.lmg/1 (WQDC, 1992), then approximately half of all treatment
systems have a residual less than that, which may leave the distribution sys-
tems exposed to surface infiltration, susceptible to H. pylori contamination.
This would be a particular concern if infiltration occurs prior to point of use,
resulting in limited disinfection residual contact time. Short-term survival in
the presence of low residual disinfectants would present a risk of infection of
individuals consuming marginally treated drinking water.

Association with biofllm communities (i.e. water-distribution pipe) can
offer bacteria a source of nutrients and protection from disinfection. H. pylori
is known to produce a water-soluble biofllm when grown under high
carbon initrogen ratio conditions (Stark et al. 9 1999). H. pylori incorporated
into a laboratory-scale biofllm, created by continuous 28-day flow of natural
water through a modified Robbins device (MacKay et al. 9 1999) persisted >196
hours (as determined by PCR).

Recovery of water adapted Helicobacter pylori

The crux of developing a culturable method for Helicobacter in water is
dependent upon effective enumeration of all viable H. pylori cells including
'stressed' cells which, while not-immediately culturable, can maintain their
infectivity. This is illustrated by the rapid decline of colonies observed on stan-
dard plate media, irrespective of changes in morphology (i.e. formation of coc-
coid cells), following introduction of H. pylori into an oligotrophic water
microcosm. Understanding the mechanisms affecting persistence and reversible
loss of colony formation is critical to the development of an effective culture
method for H. pylori isolated from environmental water. Low nutrient and
hyperosmotic conditions can induce a rapid VBNC state when H. pylori are
exposed to defined natural water systems, similar to that observed with
Campylobacter jejuni (Tholozan et al., 1999). H. pylori conserves the capabil-
ity to produce acid -inducible proteins for at least 100 days when stored at 4°C
in either phosphate-buffered saline or distilled water (Mizoguchi et al., 1999).
This induction of protein expression may determine the potential for morpho-
logical reversion during incubation.

Spent culture supernatant from early stationary phase M. tuberculosis cul-
tures increased the viability of bacilli from aged cultures and allowed small
inocula to initiate growth in liquid culture. The resuscitation factor was asso-
ciated with an acid-labile, heat stable component. Density dependent growth
of H. pylori has been observed in a number of laboratories and the use of
H. pylori spent culture could be investigated for critical recovery from envir-
onmental sources.

Human gastrin was recently reported to be a specific dose-dependent
growth factor for H. pylori. Human gastrin shortened the lag time, increased
growth rate in the log phase and increased final bacterial concentration at the
stationary phase. Controls consisting of cholecystokinin, pentagastrin,
somatostatin and epidermal growth factor did not stimulate growth (Chowers
et al., 1999). A structurally restricted receptor-mediated gastrin-specific effect,

108

Helicobacter pylori

which may have a role in the adaptation of H. pylori to its unique gastric
habitat, was suggested by the authors.

The use of water-adapted bacteria in a defined natural water provides con-
ditions more representative of real-world water disinfection compared to
common disinfection models using log phase cultures in water. Water adapta-
tion (hyperosmotic shock, nutrient limitation, temperature downshift) of bac-
terial cultures grown in rich media can cause bacteria to assume significantly
altered physiologies (Kolter et al. 9 1993). Nutrient limitation can induce
polyphosphate (poly P) accumulation (Zago et al. 9 1999) and temperature
downshift has a significant effect on the activity of metabolic enzymes and can
induce cold-shock proteins in some bacteria.

Possible mechanisms of persistence and cultur ability of Helicobacter
pylori in water

Physiological adaptations (including economy of global response regulators)
intrinsic to H. pylori may allow this organism better to survive oxidative
stress. Compared to E. coli and Campylobacter jejuni , Helicobacter pylori has
far fewer regulatory and DNA binding proteins that coordinate gene expres-
sion as a bacterium enters a new environment (Tomb et al. 9 1997). Notably
lacking are two component 'sensor-regulator' systems. A two component reg-
ulatory system designated the RacR-RacS (reduced ability to colonize) system
that is involved in a temperature-dependent signalling pathway has been
described in Campylobacter jejuni (Bras et al. 9 1999). Flagellar biosynthesis is
not as highly regulated in H. pylori as in other bacteria and appears to be
linked with urease activity (Doig et al. 9 1999; McGee et al. 9 1999). Unpublished
observations indicate that the H. pylori urease operon is regulated by decay of
mRNA (Akada et al. 9 1999). This indicates a minimal degree of environmen-
tally responsive gene expression suggesting that in the human stomach the
enzymatic pathway of H. pylori (at least those required for gastric survival)
are continually switched on. A noteworthy feature of H. pylori is the absence
of a global regulator such as OxyR, which controls the expression of impor-
tant oxidative stress-related genes in enteric bacteria. This characteristic is
shared by Mycobacterium tuberculosis, an organism with high resistance to
killing by hydrogen peroxide and organic peroxides, compared to related
Mycobacterium spp. with functional OxyR genes. Resistance may be medi-
ated by unregulated peroxidase or alkyl hydroperoxidase reductase. It is also
interesting to note that OxyR regulatory protein is not induced during expo-
sure of E. coli to free chlorine.

Although iron repression in Gram-negative bacteria is usually carried out
by the Fur protein, the iron repressed ahpC gene of C. jejuni is Fur independ-
ent. Recent work (van Vliet et al. 9 1999) has demonstrated that a C. jejuni fur
homologue PerR acts as a 'peroxide stress regulator', whereby PerR insertion
mutants showed depressed expression of both AhpC and KatA, but not other
iron regulated genes. Interestingly, the PerR mutation made by C. jejuni was
hyper-resistant to oxidative stress caused by hydrogen peroxide and cumene

109

Bacteriology

hydroperoxide, a finding consistent with the high levels of KatA and AhpC
expression. C. jejuni is the first Gram-negative bacterium where non-OxyR
regulation of peroxide stress genes has been described. The differences in the
regulation of AhpC gene expression in the closely related organism, C. jejuni
(regulated, iron repressed) and H. pylori (unregulated, constitutive) may
cause differing sensitivities to oxidative stresses. To support this idea, depres-
sion of OxyR control of AhpC, KatG (catalase) and Dps expression in
Salmonella typhimurium has been shown to allow survival of oxidative stress
within macrophages (Farr et aL, 1991). In addition, a mechanism of increased
oxidative resistance (100 mM H2O2) caused by upregulated, constitutive
expression of AhpC, KatA and Dps homologues in a Bacillus fragilis mutant
has been demonstrated (Rocha and Smith, 1998).

H. pylori may be protected from chlorination by the production of a consti-
tutive gene product(s) that may rapidly neutralize this oxidative disinfectant
effect. H. pylori has a lack of transcriptional and post-transcriptional regula-
tion (i.e. multiple sigma factors, positive activators and post-translational
adenylation) of metabolic gene expression (i.e. urease, glutamate synthetase,
AhpC, NapA (DPs), gamma glutamyltranspeptidase) that results in constitu-
tive gene expression under all physiological conditions (Garner et al., 1998;
Chevalier et al., 1999). Purified disinfection by-products have recently been
shown to form extremely stable crystals with DNA almost instantaneously
within which DNA is sequestered, providing effective protection against varied
assaults (Wolf et al., 1999). Constitutive enzyme expression can provide rapid
response to environmental stress, while storage of large quantities of pre-
formed products (i.e. urease, polyposhpate), characteristic for H. pylori
(McNulty and Dent, 1987; Bode et al. 9 1993; Xia et al. 9 1994; Bauerfeind et aL,
1997), could assist response to sudden oxidative stress by bypassing the inher-
ent lag of regulated gene expression.

In adaptation to stress, cells must coordinate major changes in the rates of
transcription, translation and replication as well as make choices in the genes
expressed (Kolter et al., 1993). An E. colt mutant lacking the enzyme polyphos-
phate kinase (ppK) that makes long chains of inorganic polyphospahte (polyp)
was deficient in functions expressed in the stationary phase of growth. After
2 days of growth in a medium-limited carbon source, only 7% of the mutant
survived compared with nearly 100% of the wild type; loss in viability of the
mutant was even more pronounced in a rich medium (Rao and Kornberg,
1999). Likewise a ppk insertion mutant of H. pylori exhibited a dramatic
decrease in survival during stationary phase.

Polyphospahte (polyP) could provide activated phosphates or coordinate
an adaptive response by binding metals and/or specific proteins (Rao and
Kornberg, 1999). Many distinctive functions appear likely for polyP depend-
ing on its abundance, chain length, biological source and subcellular location:
an energy supply and ATP substitute, a reservoir for inorganic phosphate, a
chelator of metals, a buffer against alkali, a channel for DNA entry, a cell cap-
sule and/or of major interest a regulator of responses to stresses and adjust-
ments for survival in the stationary phase of culture growth (Kornberg, 1999).

110

Helicobacter pylori

Due to the lack of a stringent response in H. pylori (Scoarughi et al. 9 1999)
low exogenous phosphate levels could promote a decline of intracellular
polyphosphate levels and may result in rapid loss of colony formation
(Crooke et al. 9 1994). Hypo-osmotic shock causes rapid reversible hydrolysis
up to 95% of intracellular polyphosphate with concomitant increase in cyto-
plasmic pH in Neurospora crassa and exogenous pyrophosphate (10m) has
been shown significantly to increase growth yield of E. coli cells in glucose
minimal media and enhanced stationary phase survival (Biville et al. 9 1996).
Pre-incubation in a non-nutrient phosphate buffer, at a high bacterial density
(possible quorum effect) might be important for recovery of H. pylori
(Yamaguchi et al. 9 1999). Bacteria within biofllms encounter higher osmolarity
conditions, oxygen limitations, and higher cell density than in the liquid phase.
It is possible that persistence of H. pylori in aquatic environments may be
associated with biofllms.

Bacteria exposed to hyperosmotic stress conditions efflux large amounts of
intracellular potassium. Potassium has a critical role in the maintenance of cell
turgor pressure, maintenance of intracellular energy, internal pH homeostasis
and enzyme activation (Bakker, 1993), all of which could contribute to rapid
loss (reversible or irreversible) of colony formation. Potassium/putrescine
(diamine) has been readily absorbed by E. coli cells and has been accompanied
with a proportional loss of intracellular putrescine with enhanced DNA super-
coiling. Interestingly, heat shock results in a decrease in the extent of DNA
supercoiling due to the dissociation of the putrescine-DNA complexes. The
global level of DNA supercoiling influences the topology of oriC (origin of
replication) and thereby the sequence of events leading to initiation of DNA
replication in E. coli (von Freiesleben and Rasmussen, 1992). Changes in DNA
topology may therefore serve as important global regulatory and replication
(colony formation) factors.

Methods of detection

H. pylori is most commonly detected in infected patients by its potent urease
enzyme. Biopsies of gastric mucosa are placed in a gel containing urea and the
subsequent ammonia production causes a pH change which is a highly accur-
ate indicator of the infection. Similarly, patients can swallow urea labelled
with an isotope, either 14 C or 13 C, and isotope-labelled C0 2 in the breath is
indicative of urease activity. Infected persons carry antibodies, usually IgG, dir-
ected against H. pylori. Serological tests exist for detecting antibodies against
the organism.

While H. pylori accounts for the vast majority of Helicobacter infections of
humans, evidence for the presence of H. heilmannii 9 H. felis, H. rappini 9
H. cinaedi, H. fennelliae and H. pullorum has been associated with gastroenteri-
tis in humans (Kusters and Kuipers, 1998; Gibson et al. 9 1999). H. heilmannii
(initially named G astro spirillum hominis 9 but reclassified as a Helicobacter

111

Bacteriology

(based on 16S rRNA analysis) is the most commonly described non-pylori
Helicobacter in humans, with colonization usually associated with mild gas-
tritis (Hoick et al. 9 1997). Differentiation between the closely-related H. heil-
mannii and H. pylori requires molecular characterization as they can assume
identical morphologies and H. heilmannii is rarely culturable in vitro (Fawcett
et al. 9 1999). PCR amplification and sequencing of more than 80% of the 16S
rDNA is believed to be the only method to reach a precise identification (Chen
et al. 9 1997; Cantet et al. 9 1999).

Gastric biopsy specimens are the only ones likely to be used for the primary
isolation of H. pylori. They should be transported in a moist state and cultured
within 2 hours of collection. Storage beyond this time should be at 4°C, or
at — 20°C if the period is more than 2 days. Various transport media have been
described for transporting biopsy samples, including cysteine brucella broth,
normal saline, glucose, milk, Stuart's medium, semi-solid agar, brain-heart infu-
sion broth, Cary-Blair medium and, more recently, cysteine-Albimi medium
containing 20% glycerol. Various non-selective and selective culture media have
been used for the isolation of H. pylori and other Helicobacter spp. Chocolate
agar, or more complex media such as brain-heart infusion or brucella agar sup-
plemented with 5-7% horse blood, are suitable non-selective media. Examples
of selective media are Dent's, Glupczynski's Brussels charcoal medium and
Skirrow's Campylobacter medium. Samples should be prepared for direct exam-
ination and plating on non-selective and selective media and plates should be
incubated in a microaerobic environment at 35-3 7°C. Colonies may be visible
after 3-5 days of incubation but may take longer to appear on primary isol-
ation. Colonies are small, domed, translucent and sometimes weakly haemolytic.

Because of the unique characteristics of H. pylori 9 only a few simple tests are
needed for identification, including Gram-stain appearance, catalase, oxidase
and urease tests. Intestinal helicobacters, such as H. cinaedi and H. fennelliae 9
can be isolated on some campylobacter-selective agars, or by filtration on non-
selective agar. Like Campylobacters, H. pylori is strictly microaerophilic and
CO2 (5-20%) and high humidity are required for growth. H. pylori requires
media containing supplements similar to those used for Campylobacters: blood,
haemin, serum, starch or charcoal. However, H. pylori is inhibited by the
bisulphite in the FBP Campylobacter 'aerotolerance' supplement. Growth is
best on media such as moist freshly prepared heated (chocolated) blood agar,
or brain-heart infusion agar with 5% horse blood and 1% IsoVitaleX. Strains
grow in various liquid media supplemented with fetal calf or horse serum.
Some strains grow in serum-free media, notably bisulphite-free brucella broth.
Sheep blood, either whole or laked, when used as a broth supplement, has been
shown to inhibit the growth of H. pylori. All strains grow at 37°C, some grow
poorly at 30° and 42°C but none grows at 25°C. Colonies from primary cul-
tures at 37°C usually take 3-5 days to appear and are circular, convex
and translucent like those of C. fetus. They seldom grow bigger than 2 mm
in diameter even if incubation is extended beyond 1 week. They are weakly
haemolytic on 5% horse blood agar. Motility is weak or absent when grown
on agar.

112

Helicobacter pylori

Like Campylobacters, H. pylori is inactive in most conventional biochemical
tests. Notable exceptions are the strong production of urease, catalase and alka-
line phosphatase. All strains produce DNAase, leucine aminopeptidase, and
gamma-glutamylaminopeptidase. The urease is a nickel containing high molecu-
lar weight protein (c. 600 kDa) with maximum activity at 45°C and pH 8.2.

Epidemiology of waterborne outbreaks

There are no recorded waterborne outbreaks associated with this organism.
This may be due to a latent period of many years prior to phenotypic pathology
and difficulty in culturing the organism from water.

Risk assessment

While epidemiological evidence suggests transmission by multiple pathways,
H. pylori infection has been associated with consumption of contaminated
drinking water (Klein et al., 1991; Baker and Hegarty, 2001). To assess accur-
ately risks from waterborne disease, it is necessary to understand pathogen dis-
tribution and survival within water-distribution systems and to apply
methodologies that can detect not only the presence, but also the viability and
infectivity of the pathogen. H. pylori (10 4 -10 6 cells) have been demonstrated
to colonize the stomach mucosa of mice. Interestingly, coccoid forms of
H. pylori induced colonization with only 10 2 cells (Aleljung et al. 9 1996).
Compared with illness to infection ratios of other infectious diseases, that of
H. /ry/on-associated peptic ulcer, 1:5, is high (Cullen et aL, 1993). In compari-
son, such ratios are about 1:25 for hepatitis B-associated chronic liver disease
and about 1:10 for Mycobacterium tuberculosis. Although the ratio for
H. pylori-associated gastric adenocarcinoma is lower (1:200), the morbidity
and mortality associated with this disease are substantial (Monath etaL 9 1998).
Although H. pylori colonizes half of the world's human population, no large
reservoir outside the human stomach has been identified. Current understand-
ing of the microbial ecology of this organism outside of its gastric niche is
extremely limited. The 13 000 research papers on H. pylori in the past decade
have focused primarily upon clinical treatment issues. There are few data con-
cerning the occurrence, persistence or ecology of H. pylori organisms in envir-
onmental waters or the efficacy of water-treatment and disinfection practices for
controlling this organism. The potential presence of a microbial Class I carcino-
gen in source water necessitates the study of procedures to prevent human infec-
tion. Elucidation of the primary transmission routes of H. pylori may allow for
preventative control of this human pathogen. Experimental and epidemiologi-
cal evidence suggests that H. pylori transmission may involve, under certain

113

Bacteriology

circumstances, the consumption of contaminated drinking water. However, no
evidence yet exists for waterborne transmission outside the developing world.
H. pylori survives poorly in drinking water compared to other pathogens such
as Escherichia coli 9 but survival for up to 12 hours under environmental condi-
tions would, in some parts of the developing world, be long enough for trans-
mission to occur. Overall, the transmission of H. pylori within water is of
particular concern within developing countries where contaminated drinking
water and poor sanitation and hygiene conditions enable the organism to be
transmitted by the faecal-oral routes. Studies using mature heterotrophic mixed-
species biofilms, using continuous chemostat systems, have shown that when
challenged with H. pylori, the bacteria can be associated with the biofllm 8 days
post-challenge. This suggests that biofilms in potable water systems could be a
possible and unrecognized reservoir of H. pylori. Water and water biofllm reser-
voirs of infection could account for many of the published conclusions which
have been attributed to others factors: e.g. socioeconomic factors such as over-
crowding, which will also bring associated sharing of common water supplies.
This idea is supported by a study of H. pylori infection rates in French sub-
marine crew members with a common tanked water supply (Hammermeister
et al. 9 1992). Overall, the argument for a waterborne route of H. pylori trans-
mission is supported primarily by maintenance of viability in spiked natural
water (West et al. 9 1992; Shahamat et al. 9 1993; Hunter, 1997; Fan et al. 9 1998;
Jiang and Doyle, 1998; Sato et al. 9 1999).

Overall risk assessment

Health effects: occurrence of illness, degree of morbidity and mortality:

• Approximately 50% of the world's population is infected with H. pylori
(Parsonnet, 1998). H. pylori infection is more prevalent in developing
countries; the rate of infection has been decreasing steadily in developed
countries over the last few decades.

• The majority of people infected with H. pylori are asymptomatic, though
they may live their whole lives with the organism. Some will develop peptic
ulcer disease, and a very small fraction will develop gastric cancer. Though
only a small percentage of H. py/on-infected people develop gastric cancer,
it is the second leading cause of cancer deaths in the world - and is directly
related to the prevalence level of H. pylori in the population.

• Compared with illness-to-infection ratios of other infectious diseases, 1:5,
the ratio of H. pylori-associated peptic ulcer is high (Cullen et al. 9 1993). In
comparison, other such ratios are about 1:25 for hepatitis B-associated
chronic liver disease and about 1:10 for Mycobacterium tuberculosis.
Although the ratio for H. pylori-associated gastric adenocarcinoma is lower
(1:200), the morbidity and mortality associated with this disease are sub-
stantial (Monath et al. 9 1998).

114

Helicobacter pylori

Exposure assessment: routes of exposure and transmission, occurrence in source
water, environmental fate:

• Because H. pylori is such a fastidious organism, culturing in environmental
and clinical samples can be challenging. PCR has been used more success-
fully to detect H. pylori DNA in environmental samples, however, little is
known about the occurrence of the organism in water.

• Under conditions of stress, H. pylori can transform to a durable, coccoid
morphology. While this form could be significant in transmission, contro-
versy surrounds its function and viability (Cellini et al. 9 1994; Kusters et al.,
1997; Ren et ai, 1999).

• Few data on environmental occurrence are available; however, of 42 sur-
face water and 20 well water samples in the USA, 40% and 65% respec-
tively, tested positive using fluorescent antibody and PCR methods (Hegarty
and Baker, 1999).

• Limited studies are available indicating that Helicobacter spp. could be pre-
sent in water distribution system biofilm (Stark et ai, 1999; Park et ai, 2001).

• While epidemiological evidence suggests transmission by multiple path-
ways, H. pylorfs exact route of transmission is unknown. H. pylori infec-
tion has been associated with consumption of contaminated drinking water
(Klein et al. 9 1991; Baker and Hegarty, 2001). Evidence is limited, but the
waterborne route of exposure is probably important in some populations.

• Occurrence data indicate that infection clusters in families as well as group-
living situations (i.e. orphanages, mental institutions). Secondary spread,
particularly among children, is likely.

Risk mitigation: drinking-water treatment, medical treatment:

• Multiple medical therapies using a variety of antibiotics and bismuth
preparations are generally 70-95% effective at eradicating H. pylori in-
fection. Re-infection in adults seems to be a rare occurrence, though the
re-infection rate for children is unknown. Infection eradication's ability to
improve gastric carcinoma precursors is controversial; however, antibiotic
treatment can successfully cause tumour regression in gastric lymphoma
cases.

• The ability of conventional drinking water treatment to remove H. pylori
from drinking water is unknown. The results of studies on the efficacy of
chlorine as a disinfectant of H. pylori are equivocal (Johnson et al., 1997;
Hegarty et al. 9 1999).

• Until the exact mode of transmission is understood, it will be difficult to
launch public health intervention programmes. Meanwhile, improved sani-
tation and living conditions have resulted in decreased incidence.

• Though it is unlikely that drinking water is a significant source of infection
in countries with adequate treatment, more research into the role of water
in the spread of H. pylori is necessary.

115

Bacteriology

References

Akada, J.K., Shirai, M., Takeuchi, H. et al. (1999). The urease operon in Helicobacter
pylori is regulated by decay of mRNA. Yamaguchi University School of Medicine,
Department of Microbiology. Unpublished communication , T. Nakazawa, Oct 1999.

Aleljung, P., Nilsson, H.O., Wang, X. et al. (1996). Gastrointestinal colonisation of
BALB/cA mice by Helicobacter pylori monitored by heparin immunomagnetic separ-
ation. FEMS Immun Med Microbiol, 13: 303-309.

Aim, R.A., Ling, L.S., Moir, D.T. et al. (1999). Genomic-sequence comparison of two
unrelated isolates of the human gastric pathogen Helicobacter pylori. Nature, 14:
176-180.

Anderson, A.P., Elliott, D.A., Lawson, M. et al. (1997). Growth and morphological trans-
formations of Helicobacter pylori in broth media. / Clin Microbiol, 35: 2918-2922.

Antonioli, D.A. (1994). Precursors of gastric carcinoma: A critical review with a brief
description of early (curable) gastric cancer. Hum Pathol, 25: 994-1005.

Aruin, L.I. (1997). Helicobacter pylori infection is carcinogenic for humans. Arkh Patol,
59(3): 74-78.

Arvanitidou, M., Stathopoulos, G.A. and Katsouyannopoulos, V.C. (1994). Isolation of
Campylobacter and Yersinia spp. from drinking waters./ Travel Med, 1: 156-159.

Atherton, J.C., Cao, P., Peek, R.M. Jr et al. (1995). Mosaicism in vacuolating cytotoxin
alleles of Helicobacter pylori: association of specific vacA types with cytotoxin produc-
tion and peptic ulceration./ Biol Chem, 270: 17771-17777.

Baker, K.H. and Hegarty, J.P. (2001). Presence of Helicobacter pylori in drinking water is
associated with clinical infection. Scan J Infect Dis, 33(10): 744-746.

Baker, K.H., Hegarty, J.P., Redmond, B. et al. (2002). Effect of oxidizing disinfectants
(chlorine, monochloramine, and ozone) on Helicobacter pylori. Appl Environ Microbiol,
68: 981-984.

Bakker, E.P. (ed.) (1993). Cell K~ and K + transport systems in procaryotes. In Alkali Cation
Transport Systems in Prokayotes. Boca Raton: CRC Press, pp. 205-224.

Barer, M.R., Gribbon, L.T., Harwood, C.R. et al. (1993). The viable but non-culturable
hypothesis and medical bacteriology. R Med Microbiol, 4: 183-191.

Barton, S.G.R.G., Young, K.A., Hardie, J.M. et al. (1997). CagA status does not influence sur-
vival of H. pylori after exposure to hydrogen peroxide. European Helicobacter pylori Study
Group, Sept 12-14, Lisbon, Portugal. 10th International Workshop Abstracts, p. All.

Basso, D., Stefani, A., Gallo, N. et al. (1999). Polymorphonuclear oxidative burst after
Helicobacter pylori water extract stimulation is not influenced by the cytotoxic geno-
type but indicates infection and gastritis grade. Clin Chem Lab Med, 37: 223-229.

Bauerfeind, P., Garner, R., Dunn, B.E. et al. (1997). Synthesis and activity of Helicobacter
pylori urease and catalase at low pH. Gut, 40: 25-30.

Beier, D., Spohn, G., Rappuoli, R. et al. (1997). Identification and characterization of an
operon of Helicobacter pylori that is involved in motility and stress adaptation.
/ Bacteriol, 179(15): 4676-4683.

Biville, F., Laurent-Winter, C. and Danchin, A. (1996). In vivo positive effects of exogen-
ous pyrophosphate on Escherichia coli cell growth and stationary phase survival. Res
Microbiol, 147(8): 597-608.

Blaser, M.J. (1987). Gastric Campy lobacter-like organisms, gastritis, and peptic ulcer dis-
ease. Gastroenterology, 93(2): 371-383.

Blaser, M.J. (1997). Helicobacter pylori eradication and its implications for the future.
Aliment Pharmacol Ther, 1: 103-107.

Bode, G., Mauch, F., Ditschuneit, H. et al. (1993). Identification of structures containing
polyphosphate in Helicobacter pylori. J Gen Microbiol, 139: 3029-3033.

Bode, G., Rothenbacher, D., Brenner, H. et al. (1997). Pets are no risk for Helicobacter
pylori infection in young children. Results of a population based study in southern
Germany. European Helicobacter pylori Study Group, Sept 12-14, Lisbon, Portugal.
10th International Workshop Abstracts, p. A37.

116

Helicobacter pylori

Borody, T.J., Cole, P., Noonan, S. et al. (1989). Recurrence of duodenal ulcer and
Campylobacter pylori infection after eradication. Med J Aust, 151(8): 431-435.

Bras, A.M., Chatterjee, S., Wren, B.W. et al. (1999). A novel Campylobacter jejuni two
component regulatory system important for temperature-dependent growth and colon-
ization. / Bacteriol, 181(10): 3298-3302.

Brown, L.M. (2000). Helicobacter pylori: epidemiology and routes of transmission.
Epidemiol Rev, 22(2): 283-297.

Bucklin, K.E., McFeters, G.A. and Amirtharaja, A. (1991). Penetration of coliforms
through municipal drinking water filters. Water Res, 25: 1013-1017.

Bunn, J. E.G., Thomas, J.E., Harding, M. et al. (1997). Supplemental water in early
infancy: A risk factor for H. pylori} European Helicobacter pylori Study Group, Sept
12-14, Lisbon, Portugal. 10th International Workshop Abstracts, p. A37.

Cantet, F., Magras, C., Marais, A. et al. (1999). Helicobacter species colonizing pig stom-
ach: Molecular characterization and determination of prevalence. Appl Environ
Microbiol, 65(10): 4672-4676.

Cappelier, J.M., Minet, J., Magras, C. et al. (1999). Recovery in embryonated eggs of
viable but nonculturable Campylobacter jejuni cells and maintenance of ability to
adhere to HeLa cells after resuscitation. Appl Environ Microbiol, 65(11): 5154-5157.

Carballo, F., Caballero, P., Parra, T. et al. (1997). Untreated drinking water is a source of
H. pylori infection. European Helicobacter pylori Study Group, Sept 12-14, Lisbon,
Portugal. 10th International Workshop, p. A3 9.

Cave, D.R. (1997). How is Helicobacter pylori transmitted? Gastroenterology, 113: S14.

Cellini, L., Allocati, N., Angelucci, D. et al. (1994). Coccoid Helicobacter pylori not cul-
turable in vitro reverts in mice. Microbiol Immunol, 38(11): 843-850.

Cellini, L., Dainelli, B., Angelucci, D. et al. (1999). Evidence for an oral-faecal transmission
of Helicobacter pylori infection in an experimental murine model. APMIS, 107(5):
477-484.

Cellini, L., Robuffo, L, Maraldi, N.M. et al. (2001). Searching the point of no return in
Helicobacter pylori life: necrosis and/or programmed death? / Appl Microbiol, 90(5):
727-732.

Chen, Y., Wang, J.D. and Xu, Z.M. (1997). Using polymerase chain reaction to detect
Helicobacter heilmannii in gastric biopsy materials. Chung Hua Liu Hsing Ring Hsueh
TsaCbib, 18(4): 241-243.

Chevalier, C, Thiberge, J.M., Ferrero, R.L. et al. (1999). Essential role of Helicobacter
pylori gammaglutamyltranspeptidase for the colonization of the gastric mucosa of mice.
Mol Microbiol, 31(5): 1359-1372.

Chin, E.Y., Whary, M.T., Fox, J.G. et al. (2001). Urease is required for intestinal colon-
ization of the mouse by the pathogen Helicobacter bepaticus. (Personal communication
from David B. Schauer, MIT.)

Chowers, M.Y., Keller, N., Tal, R. et al. (1999). Human gastrin: A Helicobacter pylori-
specific growth factor. Gastroenterology, 17(5): 1113-1118.

Costa, K., Bacher, G., Allmaier, G. et al. (1999). The morphological transition of
Helicobacter pylori cells from spiral to coccoid is preceded by a substantial modification
of the cell wall. / Bacteriol, 181(12): 3710-3715.

Costerton, J.W., Stewart, P.S. and Greenberg, E.P. (1999). Bacterial biofilms: A common
cause of persistent infections. Science, 284: 1318-1322.

Crooke E., Akiyama M., Rao N.N. et al. (1994). Genetically altered levels of inorganic
polyphosphate in Escherichia coli. J Biol Chem, 269(9): 6290-6295.

Cullen, D.J., Collins, B.J. and Christiansen, K.J. (1993). When is Helicobacter pylori infec-
tion acquired? Gut, 34: 1681-1682.

Dieterich, C, Bouzourene, H., Blum, A.L. et al. (1999). Urease-based mucosal immuniza-
tion against Helicobacter heilmannii infection induces corpus atrophy in mice. Infect
Immun, 67(11): 6206-6209.

DiPetrillo, M.D., Tibbetts, T., Kleanthous, H. et al. (1999). Safety and immunogenicity
of phoP/phoQ-deleted Salmonella typhi expressing Helicobacter pylori urease in adult
volunteers. Vaccine, 18: 449-459.

117

Bacteriology

Doig, P., de Jonge, B.L., Aim, R.A. et al. (1999). Helicobacter pylori physiology predicted

from genomic comparison of two strains. Microbiol Mol Biol Rev, 63(3): 675-707.
Dominici, P., Bellentani, S., Di Biase, A.R. et al. (1999). Familial clustering of Helicobacter

pylori infection: Population based study. Br Med J, 319: 537-541.
Dooley, C.P., Cohen, H. and Fitzgibbons, P.L. (1989). Prevalence of Helicobacter pylori

infection and histologic gastritis in asymptomatic persons. New Engl J Med, 321:

1562-1566.
Dore, M.P., Bilotta, M., Vaira, D. et al. (1999a). High prevalence of Helicobacter pylori

infection in shepherds. Dig Dis Sci, 44(6): 1161-1164.
Dore, M.P., Sepulveda, A.R., Osato, M.S. et al. (1999b). Helicobacter pylori in sheep milk.

Lancet, 354: 132.
Dowsett, S.A., Archila, L., Segreto, V.A. et al. (1999). Helicobacter pylori infection in

indigenous families of Central America: Serostatus and oral and fingernail carriage.

/ Clin Microbiol, 37(8): 2456-2460.
Dunn, B.E. (1993). Pathogenic mechanisms of Helicobacter pylori. Gastroenterol Clin

North Am, 22(1): 43-57.
Eaton, K.A., Catrenich, C.E., Makin, K.M. et al. (1995). Virulence of coccoid and bacillary

forms of Helicobacter pylori in gnotobiotic piglets. / Infect Dis, 171(2): 459-462.
Enroth, H. and Engstrand, L. (1995). Immunomagnetic separation and PCR for detection

of Helicobacter pylori in water and stool specimens. / Clin Microbiol, 33(8):

2162-2165.
Enroth, H., Wreiber, K., Rigo, R. et al. (1999). In vitro aging of Helicobacter pylori:

Changes in morphology, intracellular composition and surface properties. Helicobacter,

4(1): 7-16.
Fan, X.G., Chua, A., Li, T.G. et al. (1998). Survival of Helicobacter pylori in milk and tap

water./ Gastroenterol Hepatol, 13(11): 1096-1098.
Farinati, F., Cardin, R., Degan, P. et al. (1988). Oxidative DNA damage accumulation in

gastric carcinogenesis. Gut, 42(3): 351-356.
Farr, S.B. and Kogoma, T. (1991). Oxidative stress responses in Escherichia coli and

Salmonella typhimurium. Microbiol Rev, 55(4): 561-585.
Fawcett, P.T., Gibney, K.M. and Vinette, K.M.B. (1999). Helicobacter pylori can be

induced to assume the morphology of Helicobacter beilmannii.J Clin Microbiol, 37(4):

1045-1048.
Foliguet, B., Vicari, F., Guedenet, J.C. et al. (1989). Scanning electron microscopic study

of Campylobacter pylori and associated gastroduodenal lesions. Gastroenterol Clin

Biol, 13: 65B-70B.
Forrest, K., Stinson, M. and Wright, S.M. (1998). The presence of Helicobacter pylori in

sewage. American Society for Microbiology, Abstracts to the 98th General Meeting,

Atlanta GA, p. 445.
Freedberg, A.S. and Barron, L.E. (1940). The presence of spirochetes in human gastric

mucosa. Am J Dig Dis, 7: 443-445.
Friedman, B. (1998). European Helicobacter Pylori Study Group. Xlth International

Workshop on Gastroduodenal Pathology and Helicobacter pylori. Gut, 43: 2.
Fujioka, T. (1995). Virulence and pathogenesis of Helicobacter pylori infection. Rinsho

Byori, 43(6): 557-561.
Fukuyama, M., Arimatu, M., Sakamato, K. et al. (1999). PCR detection of Helicobacter

pylori in water samples collected from rivers in Japan. In Abstracts of 10th

International CHRO Workshop, Baltimore, MD, Sept 14, p. 117.
Garner, R.M., Fulkerson, J. Jr and Mobley, H.L. (1998). Helicobacter pylori glutamine

synthetase lacks features associated with transcriptional and posttranstational regula-
tion. Infect Immun, 66(5): 1839-1847.
Ge, Z. and Taylor, D.E. (1999). Contributions of genome sequencing to understanding the

biology of Helicobacter pylori (Review). Ann Rev Micro, 53: 353-387.
Geldreich, E.E. (1990). Microbiological quality of source waters for water supply. In

Drinking Water Microbiology, McFeters, G.A. (ed.). New York: Springer, pp. 3-31.
Gibson, J.R., Ferrus, M.A., Woodward, D. et al. (1999). Genetic diversity in Helicobacter

pullorum from human and poultry sources identified by an amplified fragment length

118

Helicobacter pylori

polymorphism technique and pulsed-field gel electrophoresis./ Appl Microbiol, 87(4):

602-610.
Goodman, K.J., Correa, P., Tengana Aux, H.J. et al. (1996). Helicobacter pylori infection

in the Colombian Andes: a population-based study of transmission pathways. Am J

Epidemiol, 144(3): 290-299.
Goodwin, C.S., Armstrong, J. A., Chlivers, T. et al. (1989). Transfer of Campylobacter

pylori and C. mustelae to Helicobacter gen. nov. as H. pylori comb. nov. and H. muste-

lae comb nov., respectively. Int J Syst Bacteriol, 39: 397-405.
Graham, D.Y., Malaty, H.M., Evans, D.G. et al. (1991). Epidemiology of Helicobacter

pylori in asymptomatic population in the U.S. Gastroenterology, 100: 1495-1501.
Gribbon, L.T. and Barer, M.R. (1995). Oxidative metabolism in nonculturable

Helicobacter pylori and Vibrio vulnificus cells studied by substrate-enhanced tetra-

zolium reduction and digital image processing. Appl Environ Microbiol, 61(9):

3379-3384.
Grubel, P., Huang, L., Masubuchi, N. et al. (1998). Detection of Helicobacter pylori DNA

in houseflies (Musca domestica) on three continents. Lancet, 352: 788-789.
Gurel, S., Besisk, F., Demir, K. et al. (1999). After the eradication of Helicobacter pylori

infection, relapse is a serious problem in Turkey./ Clin Gastroenterol, 28(3): 241-244.
Hammermeister, I., Janus, G., Schamorowski, F. et al. (1992). Elevated risk of Helicobacter

pylori infection in submarine crews. / Clin Microbiol Infect Dis, 11: 9-14.
Handwerker, J., Fox, J.G. and Schauer, D.B. (1995). Detection of Helicobacter pylori

in drinking water using polymerase chain reaction amplification. American Society for

Microbiology, In Abstracts to the 95th General Meeting, Washington DC, p. 435.
Hazen, S.L., Hsu, F.F., Mueller, D.M. et al. (1996). Human neutrophils employ chlorine

gas as an oxidant during phagocytosis./ Clin Invest, 98(6): 1283-1289.
Hegarty, J.P. and Baker, K.H. (1999). Occurrence of Helicobacter pylori in surface water

in the United States. / Appl Microbiol, 87(5): 697-701.
Hegarty, J.P., Herson, D.S., Redmond, B.W. et al. (1999). Comparative efficacy of oxidative

disinfectants on Helicobacter pylori. American Society for Microbiology, In Abstracts to

the 99th General Meeting, Chicago, Illinois, June 1. Session 129. Paper Q-132.
Henderson, J.P., Byun, J. and Heinecke, J.W. (1999). Molecular chlorine generated by the

myeloperoxidase-hydrogen peroxide-chloride system of phagocytes produces 5-chloro-

cytosine in bacterial RNA./ Biol Chem, 274(47): 33440-33448.
Hoick, S., Ingeholm, P., Blom, J. et al. (1997). The histopathology of human gastric

mucosa inhabited by Helicobacter keilmannii-iike (Gastrospirillum hominis) organ-
isms, including the first culturable case. APMIS, 105(10): 746-756.
Hulten, K., Han, S.W., Enroth, H. et al. (1996). Helicobacter pylori in the drinking water

in Peru. Gastroenterology, 110(4): 1031-1035.
Hulten, K., Enroth, H., Nystrom, T. et al. (1998). Presence of Helicobacter species DNA in

Swedish water. / Appl Microbiol, 85(2): 282-286.
Hunter, P.R. (1997). Waterborne Diseases. New York: John Wiley & Sons, pp. 202-205.
Jiang, X. and Doyle, M.P. (1998). Effect of environmental and substrate factors on survival

and growth of Helicobacter pylori. J Food Prot, 61(8): 929-933.
Johnson, C.H., Rice, E.W. and Reasoner, D.J. (1997). Inactivation of Helicobacter pylori

by chlorination. Appl Environ Microbiol, 63(12): 4969-4970.
Kamat, A.H., Mehta, P.R., Natu, A.A. et al. (1998). Dental plaque: An unlikely reservoir

of Helicobacter pylori. Indian] Gastroenterol, 17(4): 138-140.
Kelly, D.J. (1998). The physiology and metabolism of the human gastric pathogen

Helicobacter pylori. Adv Microb Physiol, 40: 137-189.
Kelly, S.M., Pitcher, M.C., Farmery, S.M. et al. (1994). Isolation of Helicobacter pylori

from feces of patients with dyspepsia in the United Kingdom. Gastroenterology, 107(6):

1671-1674.
Klein, P.D., Graham, D.Y., Gaillour, A. et al. (1991). Water source as risk factor for

Helicobacter pylori infection in Peruvian children. Gastrointestinal Physiology Working

Group. Lancet, 337(8756): 1503-1506.
Kolter, R., Siegele, D.A. and Tortno, A. (1993). The stationary phase of the bacterial life

cycle. Annu Rev Microbiol, 47: 855-874.

119

Bacteriology

Kornberg, A. (1999). Inorganic polyphosphate: A molecule of many functions. Prog Mol

Sub cell Biol, 23: 1-18.
Kuipers, E.J. (1999). Exploring the link between Helicobacter pylori and gastric cancer.

Aliment Pharmacol Ther, 13(Suppl.): 3—11.
Kung, N.N., Sung, J.J., Yuen, N.W. et al. (1997). Anti-Helico barter pylori treatment in

bleeding ulcers: randomized controlled trial comparing 2-day versus 7-day bismuth

quadruple therapy. Am J Gastroenterol, 92: 438-441.
Kuo, C.H., Poon, S.K., Su, Y.C. et al. (1999). Heterogeneous Helicobacter pylori isolates

from H. py/oW-infected couples in Taiwan./ Infect Dis, 180(6): 2064-2068.
Kurokawa, M., Nukina, M., Nakanishi, H. et al. (1999). Resuscitation from the viable but

nonculturable state of Helicobacter pylori. Kansensbogaku Zasshi, 73(1): 15-19.
Kusters, J.G., Gerrits, M.M., Van Strijp, J.A. et al. (1997). Coccoid forms of Helicobacter

pylori are the morphologic manifestation of cell death. Infect Immun, 65(9): 3672-3679.
Kusters, J.G. and Kuipers, E.J. (1998). Non-pylori Helicobacter infections in humans. Eur

J Gastroenterol Hepatol, 10(3): 239-241.
Leung, W.K., Sung, J.J., Ling, T.K. et al. (1999). Use of chopsticks for eating and

Helicobacter pylori infection. Dig Dis Sci, 44(6): 1173-1176.
Loffeld, R.J. and Arends, J.W. (1993). The role of Helicobacter pylori in non-ulcer dys-
pepsia and gastritis. Neth J Med, 42(1-2): 73-79.
Luman, W., Alkout, A.M., Blackwell, C.C. et al. (1996). Helicobacter pylori in the mouth:

Negative isolation from dental plaque and saliva. Eur J Gastroenterol Hepatol, 8(1):

11-14.
Mackay, W.G., Gribbon, L.T., Barer, M.R. et al. (1999). Are drinking water biofilms a

source of Helicobacter pylori} J Appl Microbiol, 85(Suppl.): 52S-59S.
Mackerness, C.W., Colbourne, J.S., Dennis, P.J.L. et al. (1993). Formation and control of

coliform biofilms in drinking water distribution systems. In Society for Applied

Bacteriology Technical Series, vol. 30, Denyer, S.P. (ed.). London: Blackwell Scientific,

pp. 217-227.
Malaty, H.M. and Graham, D.Y. (1994). Effect of childhood socio-economic status on the

current prevalence of Helicobacter pylori infection. Gut, 35: 742-745.
Mapstone, N.P., Lewis, F.A., Tompkins, D.S. et al. (1993). PCR identification of

Helicobacter pylori from gastritis patients. Lancet, 341: 447.
Marais, A., Mendz, G.L., Hazell, S.L. et al. (1999). Metabolism and genetics of

Helicobacter pylori: the genome era. Microbiol Mol Biol Rev, 63(3): 642-674.
Marshall, B.J. and Warren, J.R. (1984). Unidentified curved bacilli in the stomach of

patients with gastritis and peptic ulceration. Lancet, i: 1311-1315.
McFeters, G.A. (1990). Enumeration, occurrence and significance of injured bacteria in

drinking water. In Drinking Water Microbiology: Progress and Recent Developments,

McFeters, G.A. (ed.). New York: Springer, pp. 478-492.
McGee, D.J., May, C.A., Garner, R.M. et al. (1999). Isolation of Helicobacter pylori genes

that modulate urease activity./ Bacteriol, 81(8): 2477-2484.
McGowan, C.C., Necheva, A.S. and Cover, T.L. (1997). Acid-induced expression of oxida-
tive stress protein homologs in H. pylori. European Helicobacter pylori Study Group,

10th International Workshop Abstracts, Sept 12-14.
McKeown, I., Orr, P., Macdonald, S. et al. (1999). Helicobacter pylori in the Canadian arc-
tic: Seroprevalence and detection in community water samples. Am J Gastroenterol,

94(7): 1823-1829.
McNulty, C.A. and Dent, J.C. (1987). Rapid identification of Campylobacter pylori

(C pyloridis) by preformed enzymes./ Clin Microbiol, 25(9): 1683-1686.
Mendall, M.A., Goggin, P.M. and Molineaux, N. (1992). Childhood living conditions and

Helicobacter pylori seropositivity in adult life. Lancet, 339: 896-897.
Michetti, P., Kreiss, C, Kotloff, K.L. et al. (1999). Oral immunization with urease and

Escherichia coli heat-labile enterotoxin is safe and immunogenic in Helicobacter pylori-

infected adults. Gastroenterology, 116(4): 804-812.
Mizoguchi, H., Fujioka, T., Kishi, K. et al. (1998). Diversity in protein synthesis and via-
bility of Helicobacter pylori coccoid forms in response to various stimuli. Infect Immun,

66(11): 5555-5560.

120

Helicobacter pylori

Mizoguchi, H., Fujioka, T. and Nasu, M. (1999). Evidence for viability of coccoid forms of

Helicobacter pylori. J Gastroenterol, 34(Suppl. 11): 32-36.
Mizote, T., Yoshiyama, H. and Nakazawa, T. (1997). Urease-independent chemotactic

responses of Helicobacter pylori to urea, urease inhibitors, and sodium bicarbonate.

Infect Immun, 65(4): 1519-1521.
MMWR. (1999). Outbreak of Escherichia coli 0157:H7 and Campylobacter among attend-
ees of the Washington County Fair - New York. MMWR, 48(36): 803-805.
Mobley, H.L.T., Island, M.D. and Hausinger, R.P. (1995). Molecular biology of ureases.

Microbiol Rev, 59: 451-480.
Monath, T.P., Lee, C.K., Ermak, T.H. et al. (1998). The search for vaccines against

Helicobacter pylori. Infect Med, 15(8): 534-546.
Moore, A.C., Herwaldt B.L., Craun, G.F. et al. (1994). Waterborne disease in the United

States, 1991 and 1992. JAWWA, 86: 87-99.
Nakamura, H., Yoshiyama, H., Takeuchi, H. et al. (1998). Urease plays an important role

in the chemotactic motility of Helicobacter pylori in a viscous environment. Infect

Immun, 66(10): 4832-4837.
Namavar, F., Sparrius, M., Veerman, E.C. et al. (1998). Neutrophil-activating protein

mediates adhesion of Helicobacter pylori to sulfated carbohydrates on high-molecular-
weight salivary mucin. Infect Immun, 66(2): 444-447.
Nedrud, J.G. and Czinn, S.J. (1999). Host, heredity and Helicobacter. Gut, 45: 323-324.
Nowak, J. A., Forouzandeh, B. and Nowak, J. A. (1997). Estimates of Helicobacter pylori

densities in the gastric mucus layer by PCR, histologic examination, and CLOtest. Am

J Clin Pathol, 108(3): 284-288.
Odenbreit, S., Wieland, B. and Haas, R. (1996). Cloning and genetic characterization of

Helicobacter pylori catalase and construction of a catalase-deflcient mutant strain.

JBacteriol, 178(23): 6960-6967.
Oliver, J.D. (1993). Formation of viable but non-culturable cells. In Starvation in Bacteria,

Kjelleberg, S. (ed.). New York: Plenum, pp. 239-272.
Oliver, J.D. (1995). The viable but non-culturable state in the human pathogen Vibrio

vulnificus. FEMS Microbiol Lett, 133: 203-208.
On, S.L. and Holmes, B. (1992). Assessment of enzyme detection tests useful in identifica-
tion of Campylobacteria./ Clin Microbiol, 30(3): 746-749.
Osata, M.S., Le, H.H., Ayoub, K. et al. (1997). Houseflies are an unlikely reservoir for

Helicobacter pylori. European Helicobacter pylori Study Group, 10th International

Workshop Abstracts, Sept 12-14.
Oshowo, A., Gillam, D., Botha, A. et al. (1998a). Helicobacter pylori: The mouth, stom-
ach, and gut axis. Ann Periodontol, 3(1): 276-280.
Oshowo, A., Tunio, M., Gillam, D. et al. (1998b). Oral colonization is unlikely to play an

important role in Helicobacter pylori infection. Br J Surg, 85(6): 850-852.
Park, S.R., Mackay, W.G. and Reid, D.C. (2001). Helicobacter sp. recovered from drinking

water biofilm sampled from a water distribution system. Water Res, 35(6): 1624-1626.
Parsonnet, J. (1998). Helicobacter pylori. Infect Dis Clin North Am, 12: 185-197.
Parsonnet, J., Shmuely, H. and Haggerty, T. (1999). Fecal and oral shedding of

Helicobacter pylori from healthy infected adults. JAMA, 282(23): 2240-2245.
Pawlowski, K., Zhang, B., Rychlewski, L. et al. (1999). The Helicobacter pylori genome:

From sequence analysis to structural and functional predictions. Proteins, 36(1): 20-30.
Pitson, S.M., Mendz, G.L., Srinivasan, S. et al. (1999). The tricarboxylic acid cycle of

Helicobacter pylori. Eur J Biochem, 260(1): 258-267.
Popovic-Uroic, T., Patton, CM., Nicholson, M.A. et al. (1990). Evaluation of the indoxyl

acetate hydrolysis test for rapid differentiation of Campylobacter, Helicobacter, and

Wolinella species./ Clin Microbiol, 28(10): 2335-2339.
Pretolani, S., Stroffolini, T., Rapicetta, M. et al. (1997). Seroprevalence of hepatitis A virus

and Helicobacter pylori infections in the general population of a developed European

country (the San Marino study): Evidence for similar pattern of spread. Eur J

Gastroenterol Hepatol, 9(11): 1081-1084.
Ramirez-Ramos, A., Gilman, R.H., Leon-Barua, R. et al. (1997). Rapid recurrence of

Helicobacter pylori infection in Peruvian patients after successful eradication.

121

Bacteriology

Gastrointestinal Physiology Working Group of the Universidad Peruana Cayetano

Heredia and The Johns Hopkins University. Clin Infect Dis, 25(5): 1027-1031.
Rao, N.N. and Kornberg, A. (1999). Inorganic polyphosphate regulates responses of

Escherichia coli to nutritional stringencies, environmental stresses and survival in the

stationary phase. Frog Mol Subcell Biol, 23: 183-195.
Ren, Z., Pang, G., Musicka, M. et al. (1999). Coccoid forms of Helicobacter pylori can be

viable. Microbios, 97(388): 153-163.
Rice, E.W., Rodgers, M.R., Wesley, I.V. et al. (1999). Isolation of Arcobacter butzleri from

ground water. Lett Appl Microbiol, 28(1): 31-35.
Rocha, E.R. and Smith, C.J. (1998). Characterization of a peroxide-resistant mutant of the

anaerobic bacterium Bacteroides fragilis. J Bacteriol, 180(22): 5906-5912.
Roszak, D.B. and Colwell, R.R. (1987). Survival strategies of bacteria in the natural envir-
onment. Microbiol Rev, 51: 365-379.
Russo, A., Eboli, M., Pizzetti, P. et al. (1999). Determinants of Helicobacter pylori

seroprevalence among Italian blood donors. Eur J Gastroenterol Hepatol, 11(8):

867-873.
Santra, A., Chowdhury, A., Chaudhuri, S. et al. (2000). Oxidative stress in gastric mucosa

in Helicobacter pylori infection. Indian J Gastroenterol, 19(1): 21-23.
Sasaki, K., Tajiri, Y., Sata, M. et al. (1999). Helicobacter pylori in the natural environment.

Scand J Infect Dis, 31(3): 275-279.
Sato, E, Saito, N., Shouji, E. et al. (1999). The maintenance of viability and spiral morph-
ology of Helicobacter pylori in mineral water. / Med Microbiol, 48(10): 971.
Schauer, D.B., Handwerker, J., Correa, P. et al. (1995). Detection of H. pylori in drinking

water using PCR amplification. European Helicobacter pylori Study Group 8th

International Conference.
Scoarughi, G.L., Cimmino, C. and Donini, P. (1999). Helicobacter pylori: a eubacterium

lacking the stringent response./ Bacteriol, 181(2): 552-555.
Shahamat, M., Mai, U., Paszko-Kolva, C. et al. (1993). Use of autoradiography to assess

viability of Helicobacter pylori in water. Appl Environ Microbiol, 59(4): 1231-1235.
Shaw, S.E. and Regli, S. (1999). U.S. regulations on residual disinfection. JAW WA, 91(1):

75-80.
Sibille, I., Sime-Ngando, T., Mathieu, L. et al. (1998). Protozoan bacterivory and

Escherichia coli survival in drinking water distribution systems. Appl Environ

Microbiol, 64(1): 197-202.
Stanley, K., Cunningham, R. and Jones, K. (1998). Isolation of Campylobacter jejuni from

groundwater. / Appl Microbiol, 85(1): 187-191.
Stark, R.M., Gerwig, G.J., Pitman, R.S. et al. (1999). Biofilm formation by Helicobacter

pylori. Lett Appl Microbiol, 28(2): 121-126.
Steer, H.W. and Colin-Jones, D.G. (1975). Mucosal changes in gastric ulceration and their

response to carbenoxolone sodium. Gut, 16: 590-597.
Suzuki, H., Mori, M., Suzuki, M. et al. (1997). Extensive DNA damage induced by mono-

chloramine in gastric cells. Cancer Lett, 115(2): 243-248.
Suzuki, J., Muraoka, H., Kobayasi, I. et al. (1999). Rare incidence of interspousal trans-
mission of Helicobacter pylori in asymptomatic individuals in Japan. / Clin Microbiol,

37(12): 4174-4176.
Tholozan, J.L., Cappelier, J.M., Tissier, J.P. et al. (1999). Physiological characterization

of viable-but-nonculturable Campylobacter jejuni cells. Appl Environ Microbiol, 65:

1110-1116.
Thomas, J.E., Gibson, G.R., Darboe, M.K. et al. (1992). Isolation of Helicobacter pylori

from human feces. Lancet, 340(8829): 1194-1195.
Tomb, J. -F., White, O., Kerlavage, A.R. etal. (1997). The complete genome sequence of the

gastric pathogen Helicobacter pylori. Nature, 388(6642): 539-547.
Tytgat, G.N.J, and Rauws, E.A.J. (1990). Campylobacter pylori and its role in peptic ulcer

disease. Gastroenterol Clin North Am, 19: 183-196.
Unge, P., Gad, A., Gnarpe, H. et al. (1989). Does omeprazole improve antimicrobial ther-
apy directed towards gastric Campylobacter pylori in patients with antral gastritis?

A pilot study. Scand J Gastroenterol Suppl, 167: 49-54.

122

Helicobacter pylori

Vacheron, M.J., Guinand, M., Francon, A. et al. (1979). Characterisation of a new

endopeptidase from sporulating Bacillus spbaericus which is specific for the gamma-D-

glutamyl-L-lysine and gamma-D-glutamyl-(L)meso-diaminopimelate linkages of pepti-

doglycan substrates. Eur J Biochem, 100: 189-196.
Vakil, N.B. (1997). Managing patients with peptic ulcer disease: Improving the value of

care. Drug Benefit Trends, 9(4): 30-32.
Vaira, D., Holton, J., Menegatti, M. et al. (1998). Routes of transmission of Helicobacter

pylori infection. Ital J Gastroenterol Hepatol, 30: S279-S285.
van Vliet, A.H., Baillon, M.L., Penn, C.W. et al. (1999). Campylobacter jejuni contains

two Fur homologs: Characterization of iron-responsive regulation of peroxide stress

defense genes by the PerR repressor. / Bacteriol, 181(20): 6371-6376.
Velazquez, M. and Feirtag, J.M. (1999). Helicobacter pylori: characteristics, pathogenicity,

detection methods and mode of transmission implicating foods and water. Int J Food

Microbiol, 53(2-3): 95-104.
Vincent, P. (1995). Transmission and acquisition of Helicobacter pylori: Evidences and

hypothesis. Biomed Pbarmacotber, 49: 11-18.
von Freiesleben, U. and Rasmussen, K.V. (1992). The level of supercoiling affects the regu-
lation of DNA replication in Escherichia coli. Res Microbiol, 143(7): 655-663.
Vyas, S.P. and Sihorkar, V. (1999). Exploring novel vaccines against Helicobacter pylori:

protective and therapeutic immunization./ Clin Pbarm Ther, 24(4): 259-272.
Wang, X., Sturegard, E., Rupar, R. et al. (1997). Infection of BALB/c A mice by spiral and

coccoid forms of Helicobacter pylori. J Med Microbiol, 46(8): 657-663.
Warren, J.R. and Marshall, B.J. (1983). Unidentified curved bacilli on gastric epithelium in

active chronic gastritis. Lancet, 1: 1273-1275.
Wesley, I.V., Schroeder- Tucker, L., Baetz, A.L. et al. (1995). Arcobacter-speci&c and

Arcobacter butlzeri-specific 16S rRNA-based DNA probes. / Clin Microbiol, 33:

1691-1698.
West, A.P., Millar, M.R. and Tompkins, D.S. (1992). Effect of physical environment on sur-
vival of Helicobacter pylori. J Clin Pathol, 45(3): 228-231.
Whitaker, C.J., Dubiel, A.J. and Galpin, O.P. (1993). Social and geographical risk factors

in Helicobacter pylori infection. Epidemiol Infect, 111: 63-70.
Wolf, S.G., Frenkiel, D., Arad, T. et al. (1999). DNA protection by stress-induced biocrys-

tallization. Nature, 400: 83-85.
Wotherspoon, A.C. (1998). Helicobacter pylori infection and gastric lymphoma. Br Med

Bull, 54(1): 79-85.
WQDC, Water Quality Disinfection Committee (1992). Survey of water utility disinfction

practices. JAWWA, 82: 121-128.
Xia, H.H. and Talley, N.J. (1997). Natural acquisition and spontaneous elimination of

Helicobacter pylori infection: Clinical implications. Am J Gastroenterol, 92(10):

1780-1787.
Xia, H.X., Keane, C.T. and O'Morain, C.A. (1994). Pre-formed urease activity of

Helicobacter pylori as determined by a viable cell count technique: Clinical implica-
tions. / Med Microbiol, 40(6): 435-439.
Yamaguchi, H., Osaki T., Takahashi M. et al. (1999). Colony formation by Helicobacter

pylori after long-term incubation under anaerobic conditions. FEMS Microbiol Lett,

175(1): 107-111.
Yoshimatsu, T., Shirai, M., Nagata, K. et al. (2000). Transmission of Helicobacter pylori

from challanged to nonchallanged nude mice kept in a single cage. Dig Dis Sci, 45(9):

1747-1753.
Yoshiyama, H., Mizote, T., Nakamura, H. et al. (1998). Chemotaxis of Helicobacter

pylori: A urease-independent response./ Gastroenterol, 33: 1-5.
Zago, A., Chugani, S. and Chakrabarty, A.M. (1999). Cloning and characterization of

polyphosphate kinase and exopolyphosphatase genes from Pseudomonas aeruginosa

8830. Appl Environ Microbiol, 65(5): 2065-2071.
Zheng, P.Y., Hua, J., Ng, H.C. et al. (1999). Unchanged characteristics of Helicobacter

pylori during its morphological conversion. Microbios, 98: 51-64.

123

Other heterotrophic plate
count bacteria

(Flavobacterium, Klebsiella, Pseudomonas,
Serratia, Staphylococcus)

Basic microbiology

Flavobacterium

The genus Flavobacterium are aerobic, Gram-negative, asporogenous bacilli
that are non-motile, oxidase-positive, non-fermentative- and non-glucose
oxidizing. They produce pigmented colonies (when grown at incubation
temperature below 35°C) that may be yellow, orange, and red to brown.
Flavobacterium meningosepticum is clinically the most significant of all
Flavobacterium. It lives in moist environments, particularly the soil, and is
frequently isolated from nebulizers. Flavobacterium meningosepticum is a
common nosocomial infection in children and has been isolated from a num-
ber of meningitis outbreaks.

Bacteriology

Klebsiella

Klebsiella are lactose-fermenting, Gram-negative bacilli. Klebsiella are straight
rods 1-2 (xm long and 0.5-0.8 (xm wide. They are non-motile and many strains
are fimbriated. They grow best at temperatures between 12 and 43 °C and are
killed by moist heat at 55°C for 30 minutes. Under certain conditions they form
a gelatinous encapsulation. They are facultatively anaerobic with poor growth in
anaerobic conditions. This is particularly evident in the strain Klebsiella pneumo-
niae. Five species of Klebsiella including K. pneumoniae, K. oxytoca, K. rhinoscle-
romatis, K. planticola and K. ozaenae are known to be clinically significant.

Most Klebsiella strains are of environmental origin without significance to
human health.

Pseudomonas

Pseudomonads are aerobic, Gram-negative bacilli and non-spore forming.
They are oxidase- and catalase-positive and motile by use of polar flagella.
Of the 200 species that were originally present in the genus Pseudomonas,
the most important medical strain is that of Pseudomonas aeruginosa.
Pseudomonas aeruginosa is motile by means of one or two polar flagella and
is frequently isolated from soil and water. P. aeruginosa produces two main
types of soluble diffusible pigments, pyocyanin (blue phenazine) and, in low
iron media, the fluorescent pigment pyoverdin (yellow-green). The latter is
produced abundantly in media with a low-iron content functioning as a
siderophore. Two additional pigments are also produced by Pseudomonas
aeruginosa, pyorubrin (red) and melanin (brown).

Pseudomonas aeruginosa is renowned for its resistance to antibiotics. Its nat-
ural resistance to many antibiotics is principally due to the permeability barrier
afforded by its outer membrane (lipopolysaccharides). As its natural habitat is the
soil, living in association with the bacilli, actinomycetes and moulds, it has devel-
oped resistance to a variety of their naturally-occurring antibiotics. Pseudomonas
also has antibiotic resistance plasmids and it also is able to transfer these genes
to other bacteria. Aside from Pseudomonas aeruginosa, several other species
of Pseudomonas have been associated with clinical specimens, including
Pseudomonas putida, Pseudomonas fluorescens and Pseudomonas stutzeri.

To date many of the species that originally resided in the genus Pseudomonas
now belong to new genera, namely Burkholderia, Comamonas, Stenotro-
phomonas and Brevundimonas. Of these new genera Burkholderia pseudo-
mallei and Burkholderia cepacia have emerged as very important pathogens,
particularly in immunocompromised patients.

Serratia

The Serratia genus is composed of small, Gram-negative, motile, coccobacilli
that characteristically give positive reactions for citrate, Voges-Proskauer and

126

Other heterotrophic plate count bacteria

ONPG. They are aerobic and facultatively anaerobic, catalase-positive but
oxidase-negative. Serratia do not normally develop a capsule but capsular
material has been shown to be formed when Serratia are grown in well-
aerated medium with low levels of nitrogen and phosphate. Serratia ferment
sugars including mannitol and trehalose often with the production of gas.

Serratia marcescens is the type species and produces a non-diffusible red
pigment that is more pronounced on certain media at temperatures between
25 and 30°C. In the environment and clinical laboratories non-pigmented
strains of Serratia are very common. As well as Serratia marcescens other
Serratia have been isolated from the clinical environment and include S. lique-
faciens and S. odorifera.

Staphylococcus

Staphylococci are Gram-positive cocci. Under the microscope the cocci occur
as single cells, in pairs, packet clusters, or as short chains of several individual
cells. Species in this genus are usually non-motile, catalase-positive and fer-
ment glucose. By far the most significant and pathogenic strain in the genus is
Staphylococcus aureus. Staphylococcus aureus strains are coagulase-positive
and often form a yellow to orange pigmentation on some media after 2-3
days of incubation. Some strains are encapsulated or form a slime layer aiding
in their resistance to antimicrobial agents. It is well documented that while
most species are facultative anaerobes, some strains of Staphylococcus aureus
grow more favourably in aerobic environments.

In drinking water, Staphylococcus aureus are the potential pathogens of the
genus. They are able to grow in the presence of 10% sodium chloride and can
withstand high temperature changes. Of the coagulase-negative species of the
genus Staphylococcus, S. epidermidis and S. saprophyticus have been associ-
ated with human infections.

Origin of the organism

Flavobacterium

The exact taxonomy of some of these organisms is still not resolved, principally
because of atypical reactions in most biochemical identification media. Estimates
of the number of Flavobacterium species presently known to date vary from 40
to 70. The genus comprises seven well-defined taxa that fall into three natural
groups. The first group comprises three saccharolytic, indole-positive species; the
second a single non-saccharolytic indole-negative species; and the third group
three saccharolytic, indole-negative species. Flavobacterium meningosepticum is
the species of major clinical significance.

127

Bacteriology

Klebsiella

The genus Klebsiella includes at least seven currently recognized species and
72 serotypes. Originally called Friedlander's bacillus (later called K. pneumo-
niae) the problem with Klebsiella was that it had to be distinguished from
Aerobacter aerogenes, later called K. aerogenes. Historically many workers
called Klebsiella aerogenes K. pneumoniae. A number of studies have concluded,
from DNA-DNA hybridization studies, that the original Klebsiella species
aerogenes, ozaenae, pneumoniae and rhinoscleromatis were in fact only bio-
types of K. pneumoniae. Therefore it was acknowledged and in the interest
of uniformity, that all subspecies of Klebsiella should be grouped into one
species, namely K. pneumoniae.

Pseudomonas

Until relatively recently Pseudomonas aeruginosa and Pseudomonas fluo-
rescens were virtually the only species of Pseudomonas studied. Pseudomonas
aeruginosa is of most significance and belongs to the bacterial family
Pseudomonadaceae.

Serratia

Serratia marcescens was the first Serratia to be identified and later, in 1948,
Serratia rubidaea was described. This was found to have very similar charac-
teristics to that of Serratia marcescens. Since that time other species of Serratia
have been identified and often associated with clinical conditions, namely
S. ficaria and S. plymuthica.

Staphylococcus

Pasteur observed small spherical bacteria in the pus of furuncles and
osteomyelitis and thought that these bacteria might be pathogenic.

The criteria used in the classification of Staphylococcus were based on both
cell morphology and type of cell aggregation. Colony colour was the criterion
for species classification. Staphylococci were suggested as being a group of
saprophytic, tetrad-forming micrococci.

There are 32 species currently acknowledged in the genus Staphylococcus.
Staphylococcus spp. are classified first of all on the basis of DNA-DNA
hybridization as determined by the relative binding of DNAs in reassociation
reactions conducted at non-restrictive (optimal), restrictive (stringent) condi-
tions, or a combination of both (Kloos, 1980; Schleifer, 1986).

128

Other heterotrophic plate count bacteria

Metabolism and physiology

Flavobacterium

With Flavobacterium all carbohydrates are attacked slowly by oxidation or
not at all. The type species is F. aquatile.

Klebsiella

Klebsiella typically use citrate and give a negative methyl red and positive
Voges-Proskauer reaction, although some strains have atypical biochemical
reactions, such as fermenting glucose at 5°C or lactose fermentation at 44.5°C
(faecal Klebsiella). An estimated 60-85% of all Klebsiella isolated from faeces
and clinical specimens are positive in the faecal coliform test and identify as
K. pneumoniae. As a consequence, classification of these bacteria into a clear-
cut scheme has been difficult.

Pseudomonas

Although the bacterium is respiratory and never fermentative, it will grow in
the absence of O2 if NO3 is available as a respiratory electron acceptor.
P. aeruginosa possesses the metabolic versatility for which pseudomonads are so
renowned. Organic growth factors are not required, and it can use more than
30 organic compounds for growth. Pseudomonas aeruginosa is often observed
growing in 'distilled water' and shampoos, evidence of its minimal nutritional
requirements. While its optimum temperature for growth is 37°C, it is able to
grow at temperatures as high as 42°C. It is able to tolerate a wide variety of phys-
ical conditions. Pseudomonas aeruginosa obtains energy from carbohydrates by
an oxidative rather than a fermentative metabolism but, generally, it is only able
to utilize glucose.

Serratia

Serratia attack sugars fermentatively often with gas production. They are
gluconate-positive and some strains produce orthithine decarboxylase.

Staphylococcus

Staphylococci are capable of using a variety of carbohydrates as carbon and
energy sources. These carbohydrates can be taken up, unmodified, and accumu-
lated inside the cell or it can be modified during uptake. In most cases the carbo-
hydrate is phosphorylated during the transport process mediated by the

129

Bacteriology

phosphoenolpyruvate (PEP)-carbohydrate phosphotransferase system (PTS).
Glucose, mannose, glucosamine, fructose, lactose, galactose, mannitol, N-acetyl-
glucosamine and beta-glucosides are taken up by the PTS (Reizer et aL 9 1988).

The Embden-Meyerhof-Parnas (EMP, glycolytic) pathway and the oxida-
tive hexose monophosphate pathway (HMP) are the two central routes used
by staphylococci for glucose metabolism (Blumenthal, 1972). S. aureus and
S. epidermidis metabolize glucose mainly by glycolysis. The major end prod-
uct of anaerobic glucose metabolism in S. aureus is lactate (73-94%); smaller
quantities of acetate (4-7%) and traces of pyruvate are also formed. Under
aerobic conditions, only 5-10% of the glucose carbon appears as lactate and
most of it appears as acetate and CO2.

In S. epidermidis, lactate is the major end product of anaerobic glucose
metabolism. The other end products are acetate, formate and CO2 and these
are formed in only trace amounts. S. sacch arolyticus ferments glucose mainly to
ethanol, acetic acid and CO2 together with small amounts of lactic and formic
acid (Kilpper-Balz and Schleifer, 1981). In members of the S. saprophyticus
species group and S. intermedius, S. capitis, S. haemolyticus and S. warneri, lac-
tose and galactose are metabolized via the Leloir pathway, whereby galactose-
1-phosphate is epimerized to glucose-1 -phosphate (Schleifer, 1986). S. aureus,
S. epidermidis, S. hominis, S. chromogenes, S. sciuri and S. lentus metabolize
lactose and galactose via the tagatose-6-phosphate pathway, whereby galactose-
6-phosphate is isomerized to tagatose-6-phosphate, which is further converted to
tagatose-l,6-diphosphate and then cleaved to triosephosphates (Schleifer, 1986).

Most staphylococcal species are capable of synthesizing a large proportion
of the different amino acids needed for growth. The amino acid requirements
of approximately half the recognized species of staphylococci have been deter-
mined in vitro with the use of chemically defined media (Hussain et ah, 1991).
Members of the S. epidermidis species group have numerous (c. 5-13) amino
acid requirements. Most of the strains of species in this group also require
isoleucine-valine and proline and, with the exception of S. epidermidis, most
require histidine. S. aureus requires arginine and has either an absolute or par-
tial requirement for proline. Most strains of this species also require isoleucine-
valine and have either an absolute or partial requirement for cysteine and
leucine. Different ecovars of S. aureus demonstrate some differences in their
amino acid requirements (Tschape, 1973).

Members of the S. saprophyticus species require fewer amino acids than the
other species and some do not even require an amino acid or an organic nitro-
gen source. Some strains of S. saprophyticus have an absolute or partial
requirement for proline and isoleucine-valine.

Clinical features

Most species in the genera found in heterotrophic plate count (HPC) bacteria
have only very rarely caused disease in humans. Pseudomonas aeruginosa,

130

Other heterotrophic plate count bacteria

however, has been frequently associated with diseases such as urinary tract
infections, respiratory disease, and ear and eye infections. These have also been
associated with bacteraemia, osteomyelitis and meningitis (Pollack, 2000).

Acinetobacter spp., an opportunist pathogen in hospital patients, is known
to cause bacteraemia and respiratory infections (Allen and Hartman, 2000).

Chryseobacterium (Flavobacterium) meningosepticum, Stenotrophomonas
maltophilia and Pseudomonas paucimobilis have also been associated with
diseases in the hospital setting.

Flavobacterium

Clinically, the most important species of genus Flavobacterium are F.
meningosepticum, F. breve and F. odoratum. F. meningosepticum is the species
most frequently involved as an opportunistic pathogen in nosocomial infec-
tions, including meningitis (particularly in infants), pneumonia, endocarditis
and septicaemia.

Klebsiella

Klebsiella are a common cause of urinary tract infections, sometimes giving
rise to bronchopneumonia, sometimes with chronic destructive lesions and
multiple abscesses. They are occasionally associated with bacteraemia often
with a high mortality rate and are a major cause of nosocomial infections.

Klebsiella strains that are responsible for most infections are generally
K. pneumoniae being associated with clinical sepsis, particularly in surgical
wounds and urinary tract infections. Colonization in the respiratory tract is
a very common concern, particularly in those who are receiving antibiotics.
K. pneumoniae subspecies, ozaenae and rhino scleromatis, cause upper respi-
ratory tract disease (rhino scleromatis) . This seems to be particularly prevalent
in Eastern Europe. Infections caused by Klebsiella suggest it to be the primary
aetiological agent but more often it is found in mixed infection or as an
opportunistic invader.

Pseudomonas

Pseudomonas aeruginosa became apparent after the 1950s as a very import-
ant organism associated with a wide range of infections. Pseudomonas aerug-
inosa is the major cause of hospital acquired (nosocomial) infections.
It infects mainly immunocompromised individuals, burn victims, and individ-
uals on respirators or with indwelling catheters. Pseudomonas aeruginosa
also colonizes the lungs of cystic fibrosis patients, increasing the mortality rate
of individuals with the disease. Infections can lead to sepsis, pneumonia,
pharyngitis, and many other problems. Rarely will Pseudomonas be a cause
of infection in healthy individuals.

131

Bacteriology

Serratia

S. marcescens, S. odorifera, S. rubidaea, and the subgroup S. liquefaciens are
recognized as potential opportunistic pathogens that may spread in epidemic
proportions causing nosocomial infections in hospital patients. S. marcescens
is a frequent cause of infections ranging from cystitis to life-threatening blood-
stream and central nervous system infections. Serratia marcescens may cause
occasional infections, particularly in children. S. ficaria, normally associated
with figs, has been isolated from a respiratory specimen and from a leg ulcer
and S. plymuthica has been isolated from a burn site.

Staphylococcus

S. aureus, S. epidermidis and S. saprophyticus are the opportunistic pathogens
associated with infections of the skin (cellulitis, pustules, boils, carbuncles and
impetigo), bacteraemia, peritonitis associated with dialysis, genitourinary
infections, and postoperative wound infections. S. aureus is also a cause of
meningitis, osteomyelitis, and violent diarrhoea and vomiting from ingestion
of the enterotoxin caused by the organism growing in food. Infections caused
by this species are often acute and pyogenic and, if untreated, may spread to
surrounding tissue or to metastatic sites involving other organs.

S. aureus is a common aetiological agent of postoperative wound infections,
bacteraemia, pneumonia, osteomyelitis, acute endocarditis, mastitis, toxic
shock syndrome and abscesses of the muscle, urogenital tract, central nervous
system and various intra-abdominal organs. Food poisoning is frequently
attributed to staphylococcal enterotoxin.

S. epidermidis and other members of the S. epidermidis species group are
commonly associated with hospital-acquired bacteraemia, especially in
patients in intensive therapy units. In most cases, the focus of infection is an
intravascular catheter.

The Staphylococcus species of relevance to water is S. aureus. Concentrations
in drinking water can be a health concern for individuals in contact with
water for extended periods, such as from dishwashing, whirlpool therapy, and
dental hygiene. Densities of 200-400 cocci/ml have been shown to set up a
carrier state in the nose of 50% of newborn infants, and an S. aureus density
of a few hundred cells per ml in a water contact may induce infection in trau-
matized skin.

Pathogenicity and virulence

Klebsiella

132

By far the most significant Klebsiella is the capsulated strain Klebsiella
pnemoniae. Virulence, however, does not appear to depend on capsule formation.

Other heterotrophic plate count bacteria

In a study of 94 hospitals, the infection rate (16.7 infections per 100 patients)
for pathogenic K. pneumoniae was the cause of 1.1% of all nosocomical
deaths. Infections of the urinary system, lower respiratory tract, and surgical
wounds were the most frequent cause of Klebsiella-associated illness or deaths.

It is documented that between 30 and 40% of all warm-blooded animals
have Klebsiella in their intestinal tract. K. planticola and K. terrigena have
their origins in the environment, being found on fruit and vegetables, dairy
products, seed embryos, internal and external tree tissues, hay and cotton.

Klebsiella strains possess a high affinity iron uptake system, one employing
aerobactin the other, enterochelin. Klebsiella also show resistance to comple-
ment-mediated serum killing and phagocytosis if both the K and O antigens
are present. The virulence of K. pneumoniae in animal models varies consid-
erably and does not appear to depend on capsule formation.

Pseudomonas

Pseudomonas aeruginosa is able to infect both internal and external sites. It
has been associated with many superficial and mild infections such as otitis
externa. Most strains of Pseudomonas aeruginosa produce two exotoxins,
exotoxin A and exoenzyme S, as well as a variety of cytotoxins including pro-
teases, phospholipases, rhamnolipids and also a pycocyanin. All these viru-
lence factors depend on the site and the nature of infection. For example, the
proteases are particularly important in corneal infections; exotoxin and pro-
teases are important in burn infections; phospholipases, proteases and aligi-
nate are important in chronic pulmonary colonization. Production of
fluorescein is important as it allows Pseudomonas aeruginosa to compete with
human iron-binding proteins.

Serratia

Pigmented forms of S. marcescens as well as non-pigmented forms are found
occasionally in the human respiratory tract and faeces. Most infections occur
in hospitals causing urinary tract infections, respiratory tract infections,
meningitis, wound infections, septicaemia and endocarditis. Some strains are
endemic, particularly in hospitals, as strains are often able to multiply at room
temperature.

Staphylococcus

Staphylococus aureus possess a large collection of virulence mechanisms
which are important to overcome the body's defences and also invade, survive
and adhere to tissues. On average, over 60% of all Staphylococcus aureus
strains produce at least five enterotoxins (A-E). These toxins are often produced
alone or in combination and are found to be very resistant to heat. They

133

Bacteriology

are responsible for staphylococcal food intoxication, inducing symptoms of
nausea, vomiting and diarrhoea within a few hours. Staphylococcal toxic
shock syndrome toxin (TSST-1) is also produced by Staphylococcus aureus
strains, which are often referred to as super antigens in that they are potent
activators of T lymphocytes inducing an immune response with the release of
large amounts of cytokines and tumour necrosis factor. Certain of the cell sur-
face adhesins, exocellular materials and extracellular proteins produced by S.
aureus are believed to play important roles in the pathogenicity of this organ-
ism. Epidermolytic toxins (A and B) are associated with intraepidermal blis-
tering. This is particularly prevalent in children's nurseries and nursing homes.
One very important effect of these toxins is a condition known as scalded
baby syndrome where extensive blistering lesions occur resulting in a look of
scalding on babies.

Treatment

Klebsiella

Clinical isolates of Klebsiella are naturally resistant to ampicllin, amoxycillin and
most other penicillins. They are generally found to be sensitive to cephalosporins,
gentamicin and other aminoglycosides. With infections of the urinary tract
caused by Klebsiella, trimethoprim, nitrofurantoin, co-amoxiclav or oral
cephalosporins are administered. In cases of Klebisella pneumoniae treatment
with an aminoglycoside or a cephalosporin, such as cefotaxime, is required.

Pseudomonas

Only a few antibiotics are effective against Pseudomonas. These have
included fluoroquinolone, gentamicin and imipenem, but even these antibi-
otics are not effective against all strains. Pseudomonas infections associated
with cystic fibrosis patients become very difficult to treat because of the high
prevalence of resistant strains. Until about 1960 all pseudomonas infections
were treated with polymyxins. Now treatment with gentamicin and tobramycin
combined with an aminoglycoside and a beta-lactam is commonly adopted. In
general ciprofloxacin shows good activity against Pseudomonas aeruginosa.

Serratia

Serratia are commonly resistant to cephalosporins with variable resistance to
ampicillin and gentamicin documented. Generally, an aminoglycoside such as
gentamicin is usually the most reliable first-line defence for treating infections
caused by Serratia. In recalcitrant cases fluoroquinolones or carbapenems are
often used.

134

Other heterotrophic plate count bacteria

Staphylococcus

Staphylococcus aureus are inherently sensitive to many antibiotics. However,
in the hospital setting 90% of Staphylococcus aureus are resistant to ben-
zylpenicillin. This is brought about by the production of the enzyme pencilli-
nase which inactivates penicillin. It is methicillin-resistant Staphylococcus
aureus (MRSA) that has become a major problem, particularly in the hospital
setting. Vancomycin is being used to treat infections caused by MRSA but
vancomycin-resistant Staphylococcus aureus (VRSA) are now being docu-
mented, with the first one isolated in the USA in June 2002. The choice of
antibiotic for treating Staphylococcus aureus infections should principally be
based on the results of sensitivity tests. Treatment initially should, in cases of
severe infection, begin with flucloxacillin, unless MRSA is highly endemic, in
which case, a glycopeptide should be used. In general terms, if the staphylo-
coccus causing the infection is sensitive to penicillin, benzylpenicillin should
be used but with patients who are hypersensitive to this then erythromycin,
clindamycin or vancomycin should be used.

Survival in the environment and water

Flavobacterium

In the aquatic environment, flavobacteria are ubiquitous, being found in soil,
water, sewage, vegetation and dairy products. Soil and water appear to be sig-
nificant reservoirs for Flavobacterium with evidence that some are chlorine-
resistant. Stagnation of building plumbing systems can provide opportunities
for Flavobacterium to adhere to water pipes, particularly in hospital water
systems. Flavobacterium has been introduced into drinking water through
microbial growth on devices connected to the water supply. On the availabil-
ity of current evidence it is feasible to suggest that the route of exposure of
Flavobacterium may be by ingestion, body contact with water supply, or by
person-to-person contact in hospitals.

Current evidence suggests that conditions contributing to Flavobacterium
regrowth in drinking water include absence of free chlorine residual, water
temperatures above 15°C, accumulations of bacterial nutrients in pipe sedi-
ments, and static water conditions.

Klebsiella

Surface water and unprotected groundwater receive Klebsiella from both
environmental and faecal sources. Klebsiella has been shown to survive for 20
days in laboratory 'pure water'. Environmental strains are introduced to
source water from urban and rural runoff and by discharges from industrial

135

Bacteriology

waters. Faecal Klebsiella enter the water cycle from municipal sewage and
meat processing works and source discharge from farm animal waste runoff.

Klebsiella organisms are included in the national and international guide-
lines for total coliform occurrences. Generally, these standards limit total coli-
forms to less than one organism or their absence in 100 ml of treated drinking
water. World Health Organization guidelines have established a limit of less
than 10 total coliforms per 100 ml of untreated groundwater, provided no fae-
cal coliforms are detected in the sample.

Klebsiella is often transmitted via body contact with water supply during
bathing, ingestion or by person-to-person contact through poor hand-washing
habits in hospitals but, more specifically, in senior citizen care institutions.
Inhalation of moisture from vaporizers using drinking water contaminated
with Klebsiella should also be considered a risk to some individuals. No com-
munity waterborne outbreaks have been reported to be caused by Klebsiella
in public water supplies. Most of the Klebsiella waterborne occurrences do
not involve faecal strains. In those infrequent situations in which the laboratory
analyses reveal faecal Klebsiella in the distribution system, the colonization sites
must be destroyed to avoid more frequent releases of this opportunistic
pathogen at higher densities into the water supply. Infective dose (ID50) values
for environmental and clinical isolates of Klebsiella have been reported to be
between 3.5 X 10 and 7.9 X 10 5 cells/ml. Therefore, ingestion of 100 ml of drin-
king water (approximately one glass of water) containing 3.5 X 10 Klebsiella
per ml could present a risk to susceptible individuals.

Klebsiellae can be controlled effectively by adequate disinfection in a clean
pipe environment - these organisms can be protected by particulate material,
porous pipe sediments, biological debris, macroinvertebrates and disinfection
demand products. Furthermore, Klebsiella can encapsulate, which provides
some protection from disinfectants.

Pseudomonas

Surveys undertaken within potable water supplies have shown Pseudomonas to
constitute around 2-3% of the total heterotrophic plate count and is responsible
for some 10% of nosocomial infections. A large number of Pseudomonas species
are often isolated from potable water. These have included Pseudomonas
aeruginosa, fluorescens, alcaligenes, mendocina, putida, cepacia, allei, mal-
tophila, testosteroni, vescularis, flava, pseudoflava, palleroni, rhodos, eckinoides,
radiora and mesophilica. Recreational and occupational infections are associated
with pseudomonas, which include Jacuzzi or whirlpool rashes.

Molecular subtyping of isolates of P. aeruginosa from patients and water
have proven that water is the source of many Pseudomonas infections.
Pseudomonas aeruginosa is usually found in water that has been contam-
inated with faecal material such as surface waters. It is excreted in the faeces
of many healthy adults. The occasional occurrence of Pseudomonas aerugi-
nosa in drinking water often indicates deterioration of the quality of water.

136

Other heterotrophic plate count bacteria

It is a typical biofilm bacterium being isolated from many materials that are in
contact with water.

Serratia

Serratia are considered to be ubiquitous in the environment. They can be found
in surface and groundwater, soil, decaying vegetation, insects, decaying meat and
spoiled milk. Serratia strains are spread by person-to-person contact and by con-
taminated water from sumps in hospital equipment. S. marcescens infections
have also been transmitted via medical solutions and peritoneal-dialysis effluents.
Serratia appear to occur seasonally in high-quality waters (such as private
wells, distribution systems, finished reservoir supplies, and bottled water).
Colonization may occur in a variety of attachment devices, including drinking
water fountains, ice machines, point-of-use treated water, laboratory high
quality water systems, humidifying units and haemodialysis equipment.
Serratia have also been found in the heterotrophic bacterial population on
media substrate in granular activated carbon (GAC) filters. Densities in water
are variable, most often being less than 100 organisms/ml unless predominate
colonization occurs in a biofilm. The persistence of Serratia in tap water is
about 100 days, and much longer in contaminated well water. In distilled
water, Serratia may survive for 48 days at room temperature.

Staphylococcus

Staphylococci have been detected in the pharynx, conjunctiva, mouth, blood,
mammary glands, faeces, bodily discharges, excretions and intestinal, geni-
tourinary and respiratory tracts of their hosts. Small numbers of staphylococci
have been found in air, dust, soil and water, and on inanimate surfaces or
fomites, molluscs, insects and plants in areas frequented by mammals and
birds. Is it any wonder the major reservoirs of staphylococci include warm-
blooded animals (in the skin, nose, ear, and mucous membranes), sewage and
stormwater runoff?

The major concern with S. aureus in water transmission is contact with cuts
and scratches on the skin, infecting the ears or the eyes during bathing, or in
water used to prepare uncooked foods. Ingestion is the pathway for gastro-
intestinal infections from contaminated foods, or for individuals on intensive
antibiotic therapy. The density of coagulase-positive Staphylococcus (i.e. S.
aureus) ranges from 10 2 to 10 4 /g in the normal human faecal flora, but occur-
rence is quite variable (10-93%). Coagulase-negative strains of Staphylo-
coccus conversely may be found in 31-59% of faeces from healthy people. It
is not surprising then that Staphylococcus are the most numerous bacteria
shed by swimmers in natural bathing waters and chlorinated swimming pools.
In one study, approximately two-thirds of the staphylococci in bathing waters

137

Bacteriology

were S. aureus. Densities of S. aureus in both studies ranged from a few to sev-
eral hundred cells/100 ml. S. aureus was found to be one of the two most con-
centrated opportunistic pathogens in urban stormwater, ranging from 10 to
1000 organisms/ml. In private water supplies S. aureus is present in quite high
concentrations often as high as 400 organisms/ml. Private water supplies con-
stitute a very important concern in water-related diseases.

S. aureus in the hospital environment and water supplies are generally classed
as opportunistic pathogens, however, they may play a subordinate role in disease.
No waterborne outbreaks have been caused by Staph ylococccus aureus.
However, the risk of staphylococci infection acquired from poor quality small
water systems and private water has been documented. In drinking water,
Staphylococcus may persist at 20°C for 20-30 days, provided trace amounts
of organic nutrients are available. Growth in water is slow at temperatures
below 20°C, and merely at subsistence rate below 10°C.

Methods of detection

Flavobacterium

Flavobacterium can be recovered easily from water with R2A spread plates
incubated between 22 and 28°C for at least 7 days. Following growth all pig-
mented colonies should be tested using conventional biochemical tests.

Klebsiella

Biochemical characteristics are used in Klebsiella speciation when purified on
M-Endo agar from water. Other differential media can be used for the isolation
of Klebsiella, such as M-Kleb agar, often in connection with the membrane
filtration. On this agar Klebsiella form dark blue to dark grey colonies.

Pseudomonas

Pseudomonas aeruginosa grow on most common culture media. For culture
from soil or water a selective medium containing acetamide as the sole carbon
source and nitrogen source should be used. Formal identification can be done
using biochemical tests. P. aeruginosa isolates may produce a number of
different colony types depending on the environmental selective pressure.
Isolates of Pseudomonas identified from soil or water classically produce a
small, rough colony. As a comparison, isolates from clinical samples yield one
or another of two smooth colony types. One type has a fried-egg appearance,
which is large, smooth, with flat edges, and an elevated appearance. Another
type, frequently obtained from respiratory and urinary tract secretions, has a
mucoid appearance, which is attributed to the production of alginate slime.

138

Other heterotrophic plate count bacteria

It is documented that the smooth and mucoid colonies play a role in coloniza-
tion and virulence.

Serratia

Serratia in water samples can be isolated on R2A agar when incubated at tem-
peratures between 20 and 30°C using either the membrane filtration (MF)
procedure or a spread plate technique. Colonies of S. marcescens are generally
homogeneous for the first day or two and then often become convex, pig-
mented with a relatively opaque centre and an effuse, colourless, almost trans-
parent periphery with an irregular crenated edge. Pigment is only formed in
the presence of oxygen and at an appropriate temperature. The red pigment
(prodigiosin) is soluble in alcohol (absolute) but insoluble in water.

Staphylococcus

Pigmented Staphylococcus colonies can be seen on heterotrophic plate count
(HPC) cultures performed on samples taken from the water distribution sys-
tem. No specific media are available for the detection of Staphylococcus from
water. Some success has been reported for the use of M-staphylococcus broth
in a modified multiple-tube procedure or Baird-Parker agar in the membrane
filter (MF) procedure. Presumptive results are verified by placing pure cultures
from the MF into a commercial multi-test system for biochemical reactions.
Any turbid tubes in the multiple-tube test are confirmed in Lipoviettin-salt
mannitol agar streak plates.

Staphylococci produce distinctive colonies on a variety of commercial, select-
ive and non-selective agar media. The commonly used selective media in the med-
ical setting include mannitol-salt agar, lipase-salt-mannitol agar, phenylethyl
alcohol agar, Columbia colistin -nalidixic acid (CNA) agar, and Baird-Parker agar
base supplemented with egg yolk tellurite enrichment. These media inhibit the
growth of Gram-negative bacteria, but allow the growth of staphylococci and
certain other Gram-positive bacteria. Baird-Parker agar when supplemented
with egg yolk tellurite enrichment is widely recommended for the detection and
enumeration of S. aureus species that share some properties with S. aureus, it is
now recognized that such a distinction. Schleifer-Kramer agar is used by some
laboratories for the selective isolation and enumeration of staphylococci from
foods and other heavily contaminated sources. Incubation of cultures on selective
media should be for at least 48-72 hours at 35-3 7°C for colony development.
Many of the staphylococcal species can produce abundant anaerobic growth in
a semi-solid Brewer's thioglycollate medium within 24-72 hours at 35-3 7°C.

Staphylococci from a variety of clinical specimens are usually isolated in
primary culture on blood agar (e.g. tryptic soy agar supplemented with 5%
sheep blood), following an incubation period of 18-24 hours at 35-3 7°C.
In this short time, most staphylococcal colonies will be 1-3 mm in diameter,
circular, smooth and raised, with a butyrous consistency.

139

Bacteriology

All staphylococcal species grow well on tryptic soy agar. Some strains, how-
ever, should be cultured on tryptic soy agar or tryptic soy agar supplemented
with blood. Brain-heart infusion agar and nutrient agar support good growth
of staphylococci, although not many species comparisons have been made of
colonies growing on these media.

Epidemiology and waterborne outbreaks

Epidemiological studies showing the association between HPC and human
health effects are scarce (Ferley et al. 9 1986; Zmirou et al. 9 1987). A study by
Payment has found an association between illness and total counts of bacteria
using a randomized controlled trial (Payment et al. 9 1991a). From this study
it was found that there was an association between gastrointestinal illnesses
and heterotrophic plate counts at 35°C. An association between HPC bacteria
and human illness was also found in a second trial (Payment et al. 9 1997).
This study demonstrated no association between counts of HPC bacteria and
gastrointestinal illness in humans.

Overall, from the available literature to date there is no evidence that there is
an association between HPC counts in drinking water and disease in humans.
There have, however, been a number of outbreaks of disease linked to water
supplies. These have included an outbreak of multiresistant Chryseobacterium
(F lav o bacterium) meningosepticum, which affected eight neonates on a neo-
natal intensive care unit (Hoque et al. 9 2001). Pseudomonas aeruginosa geno-
types have been detectable in tap water causing infections in hospitals (Bert
et al. 9 1998; Ferroni et al. 9 1998; Trautmann et al. 9 2001). Stenotrophomonas
maltophilia was cultured from endotracheal aspirate samples from five preterm
infants in a neonatal intensive care unit of whom four were colonized and
one died from septicaemia (Verweij etal. 9 1998). Six patients in an intensive care
unit (ICU) were colonized or infected with Pseudomonas paucimobilis (Crane
et al. 9 1981) which were subsequently recovered from the ICU hot water line.

Although many HPC bacteria have been associated infrequently with disease
no epidemiological study has demonstrated an association with drinking water
in the community. Severely ill hospitalized patients, such as those infected with
AIDS, can acquire Mycobacterium avium complex from hospital water systems.

Risk assessment

Health effects: occurrence of illness, degree of morbidity and mortality, prob-
ability of illness based on infection:

• No studies have shown a relationship between gastrointestinal illness and
HPC bacteria in drinking water.

140

Other heterotrophic plate count bacteria

• Certain Flavobacterium species are associated with nosocomial infections,
including meningitis, pneumonia, endocarditis and septicaemia.

• Klebsiella are a common cause of urinary tract infections. They are occa-
sionally associated with bacteraemia often with a high mortality rate and
are a major cause of nosocomial infections.

• Fseudomonas aeruginosa is an important organism associated with a wide
range of infections. Fseudomonas aeruginosa is the major cause of hospital-
acquired infections in the very ill. Infections can lead to sepsis, pneumonia,
pharyngitis, and other problems. Rarely will Fseudomonas cause infection
in healthy individuals.

• Serratia spp. are recognized as potential opportunistic pathogens that
may spread in epidemic proportions causing nosocomial infections in
hospital patients. S. marcescens is a frequent cause of infections ranging
from cystitis to life-threatening bloodstream and central nervous system
infections.

• Staphylococcus aureus, S. epidermidis and S. saprophyticus are opportunis-
tic pathogens associated with infections of the skin, bacteraemia, peritoni-
tis associated with dialysis, genitourinary infections and postoperative
wound infections. S. aureus is also a cause of meningitis, osteomyelitis and
violent diarrhoea and vomiting from ingestion of the enterotoxin caused by
the organism growing in food. The Staphylococcus species of relevance to
water is S. aureus. Concentrations in drinking water can be a health con-
cern for individuals in contact with water for extended periods, such as
from dishwashing, whirlpool therapy, and dental hygiene.

• With the possible exception of P. aeruginosa urinary tract infections, cases
of infection due to HPC bacteria outside of hospital are very rare. Even
within hospitals many HPC bacteria are only very occasionally associated
with disease.

Exposure assessment: routes of exposure and transmission, occurrence in
source water, environmental fate:

• HPC bacteria are generally found plentifully in the environment and fre-
quently colonize the bodies of healthy people, who excrete organisms in
their faeces.

• Exposure through water is typically from contact with cuts and scratches
on the skin rather than ingestion, though transmission may come from
ingestion, contact, or inhalation.

• Some HPC can colonize water pipes and systems like drinking fountains,
humidifying units and whirlpool spas.

• Many HPC species colonize biofllms, where they can live for long periods
in contact with water.

• Soil and water appear to be significant reservoirs for Flavobacterium with
evidence some are chlorine-resistant. Current evidence suggests that condi-
tions contributing to Flavobacterium regrowth in drinking water include
absence of free chlorine residual, water temperatures above 15 °C, accumu-
lations of bacterial nutrients in pipe sediments and static water conditions.

141

Bacteriology

• The persistence of Serratia in tap water is about 100 days and much longer
in contaminated well water. In distilled water, Serratia may survive for
48 days at room temperature.

• In drinking water, Staphylococcus may persist at 20°C for 20-30 days, pro-
vided trace amounts of organic nutrients are available. Growth in water is
slow at temperatures below 20°C, and merely at subsistence rate below
10°C.

Risk mitigation: drinking-water treatment, medical treatment:

• Traditional water treatment with coagulation, sedimentation, and disinfec-
tion should handle HPC bacteria in drinking water sources; however, they
can regrow in the absence of residual disinfection or in treated water that
contains enough nutrients for growth. Bacteria in biofllm are more difficult
to treat, but maintaining adequate residual throughout the distribution sys-
tem, reducing nutrients available for regrowth, and regular flushing of dis-
tribution lines should control HPC.

• HPC bacteria are often resistant to antibiotics:

- Clinical isolates of Klebsiella are generally sensitive to cephalosporins,
gentamicin and other aminoglycosides.

- Only a few antibiotics are effective against Pseudomonas. These have
included fluoroquinolone, gentamicin and imipenem.

- Generally an aminoglycoside, such as gentamicin, is usually reliable in
treating infections caused by Serratia. In recalcitrant cases fluoro-
quinolones or carbapenems are used.

- The choice of antibiotic for treating Staphylococcus aureus infections
should principally be based on the results of sensitivity tests. Benzyl-
penicillin should be used if the isolate is sensitive. Antibiotic-resistant
strains of Staphylococcus have become a major health concern.

References

Allen, D.M. and Hartman, B.J. (2000). Acinetobacter species. In Principles and Practice of
Infectious Diseases, 5th edn, Mandell, G.L., Bennett, J.E. and Dolin, R. (eds). Philadelphia:
Churchill Livingstone, pp. 2339-2342.

Bert, E, Maubec, E., Bruneau, B. et al. (1998). Multi-resistant Pseudomonas aeruginosa
outbreak associated with contaminated tap water in a neurosurgery intensive care unit.
/ Hosp Infect, 39: 53-62.

Blumenthal, H.J. (1972). Glucose catabolism in staphylococci. In The Staphylococci,
Cohen, J.O. (ed.). New York: John Wiley, pp. 111-135.

Crane, L.R., Tagle, L.C. and Palutke, W.A. (1981). Outbreak of Pseudomonas paucimo-
bilis in an intensive care facility. / Am Med Assoc, 246: 985-987.

Ferley, J.P., Zmirou, D., Collin, J.F. et al. (1986). Etude longitudinale des risques lies a la
consommation d'eaux non conformes aux normes bacteriologiques. Rev Epidemiol
Sante Publ, 34: 89-99.

Ferroni, A., Nguyen, L., Pron, B. et al. (1998). Outbreak of nosocomial urinary tract infec-
tions due to Pseudomonas aeruginosa in a paediatric surgical unit associated with tap-
water contamination./ Hosp Infect, 39: 301-307.

142

Other heterotrophic plate count bacteria

Hoque, S.N., Graham, J., Kaufmann, M.E. etal. (2001). Cbryseobacterium (Flavo bacterium)

meningosepticum outbreak associated with colonization of water taps in a neonatal

intensive care unit. / Hosp Infect, 47: 188-192.
Hussain, M., Hastings, J.G.M. and White, P.J. (1991). Isolation and composition of the

extracellular slime made by coagulase-negative staphylococci in a chemically denned

medium./ Infect D is, 163: 534-541.
Kilpper-Balz, R. and Schleifer, K.H. (1981). Transfer of Peptococcus saccbarolyticus

Foubert and Douglas to the genus Staphylococcus: Staphylococcus saccbarolyticus

(Foubert and Douglas) comb. nov. Zentralbl Bakteriol Mikrobiol Hyg Abt 1 Orig C, 2:

324-331.
Kloos, W.E. (1980). Natural populations of the genus Staphylococcus. Annu Rev

Microbiol, 34: 559-592.
Payment, P., Franco, E., Richardson, L. et al. (1991a). Gastrointestinal health effects asso-
ciated with the consumption of drinking water produced by point-of-use domestic

reverse-osmosis filtration units. Appl Environ Microbiol, 57: 945-948.
Payment, P., Siemiatycki, J., Richardson, L. et al. (1997). A prospective epidemiological

study of gastrointestinal health effects due to the consumption of drinking water. Int J

Environ Hltb Res, 7: 5-31.
Pollack, M. (2000). Pseudomonas aeruginosa. In Principles and Practice of Infectious

Diseases, 5th edn, Mandell, G.L., Bennett, J.E. and Dolin, R. (eds). Philadelphia:

Churchill Livingstone, pp. 2310-2335.
Reizer, J., Saier, M.H. Jr et al. (1988). The phosphoenolpyruvate:sugar phosphotransferase

system in Gram-positive bacteria: properties, mechanisms, and regulation. Crit Rev

Microbiol, 15:297-338.
Schleifer, K.H. (1986). Taxonomy of coagulase-negative staphylococci. In Coagulase-
negative Staphylococci, Mardh, P.-A. and Schleifer, K.H. (eds). Stockholm: Almqvist &

Wiksell International, pp. 11-26.
Trautmann, M., Michalsky, T., Wiedeck, H. et al. (2001). Tap water colonization with

Pseudomonas aeruginosa in a surgical intensive care unit (ICU) and relation to

Pseudomonas infections of ICU patients. Infect Control Hosp Epidemiol, 22: 49-52.
Tschape, H. (1973). Genetic studies on nutrient markers and their taxonomic importance.

In Staphylococci and Staphylococcal Infections, Jeljaszewicz, J. (ed.). Warsaw: Polish

Medical Publishers, pp. 57-62.
Verweij, P.E., Meis, J.F., Christmann, V. etal. (1998). Nosocomial outbreak of colonization

and infection with Stenotrophomonas maltophilia in preterm infants associated with

contaminated tap water. Epidemiol Infect, 120: 251-256.
Zmirou, D., Ferley, J.P., Collin, J.F. et al. (1987). A follow-up study of gastro-intestinal

diseases related to bacteriologically substandard drinking water. Am J Public Hlth, 77:

582-584.

143

Legionella

Basic microbiology

Legionellae stain very poorly as Gram-negative bacteria. They are catalase-
positive, and are not able to reduce nitrate. They also do not utilize carbohy-
drates by either oxidation or fermentation. They are short, rod-shaped bacteria,
often coccobacillary and non-spore-forming. The rods of free-living Legionella
are irregular with non-parallel sides approximately 0.3-0.9 |Jim wide and
1-3 fjim long ( Rodger s et al. 9 1978). In vitro Legionella will grow to 2-6 fjim,
but can form even longer filamentous forms in old cultures. All Legionella are
motile except L. oakridgensis, L. londinensis and L. nautarum. This motility is
facilitated by one or more polar or subpolar unsheathed flagella (Chandler
etai, 1980).

Legionella are strict aerobes. For their isolation they require an enriched agar
supplemented with L-cysteine and ferric salts. Optimally they grow at 35 °C.

In water and the environment Legionella require the presence of other bac-
teria or protozoa in order to grow (Wadowsky and Yee, 1985). Protozoa
known to support the growth of legionellae include species of Vahlkampfia,
Hartmanella, Acanth amoeba, Naegleria, Echinamoeba and Tetrahymena
(Fields, 1993). The growth of Legionella in the environment in the absence
of protozoa has not been demonstrated which suggests that protozoa are the

Bacteriology

natural reservoir for Legionella (Fields, 1993). It is possible that the biofilm
may be an area where Legionella may multiply. This, however, needs further
investigation. The growth of Legionella pneumophila also occurs in associa-
tion with cyanobacteria (Tison et al. 9 1980).

Origin and taxonomy of the organism

It was in 1943 that the first strains of Legionella were isolated from guinea-pigs.
In 1954, a bacterium was isolated from free-living amoebae which was not,
however, classified as a species of Legionella until 1996 (Hookey etaL, 1996).
The genus Legionella and the strain pneumophila were discovered following
an outbreak, in 1976, during a convention of the Philadelphia Branch of the
American Legion. In this outbreak 184 attendees fell ill with an apparent
pneumonia. Of these 29 died. Over the next few years, the organism was iden-
tified as the cause of a number of other large outbreaks of similar illnesses.
Subsequent serological studies from previous outbreaks of a similar nature
allowed retrospective diagnosis, confirming that the isolate causing the pneu-
monia was Legionella pneumophila.

Legionella is the sole genus of the family Legionellaceae. It is composed of
many species and serogroups and following 16S rRNA analysis it now belongs
to the gamma-2 subgroup of the class Proteobacteria. To date, the group
Legionella now contains 48 species consisting of 70 serogroups.

Metabolism and physiology

Legionellae have an absolute requirement for iron and utilize amino acids
for energy rather than carbohydrates. The essential amino acids utilized by
Legionella include arginine, cysteine, methionine, serine, threonine and valine.
A number of strains of Legionella also require isoleucine, leucine, phenylala-
nine and tyrosine for growth. Most species of Legionella are beta-lactamase
positive and are able to liquefy gelatine, with the only exception being L.
micdadei. Most species of Legionella do not hydrolyse hippurate, the excep-
tion being L. pneumophila (except serogroups 4 and 15). Legionella are
superoxide dismutase-positive, weakly peroxidase-positive and possess the
cytochromes a-d.

Clinical features

Legionella has been detected in humans, guinea-pigs, rats, mice, marmosets
and monkeys. Infection with Legionella presents as two distinct entities,

146

Legionella

Legionnaires' disease and Pontiac fever. Legionnaires' disease, which has a high
fatality rate, produces pneumonia and also affects the nervous, gastrointestinal
and urinary systems (Mayock et al. 9 1983). Pontiac fever, on the other hand,
is a mild non-pneumonic, self-limited flu-like illness. One to 15% of Legionella
can be community-acquired with up to 50% hospital-acquired (Butler and
Breiman, 1998). The incubation period for Legionnaires' disease is approxi-
mately 2-14 days. The infection has, however, been documented as lasting
several months.

Symptoms of Legionella infection include pneumonia, myalgia, malaise and
headache. Other symptoms include fever and chills, a cough, chest pain and diar-
rhoea. The symptoms of Pontiac fever include chills, fever, myalgia and
headache. The incubation period for this disease is 5-66 hours and can last up to
7 days.

Smoking and alcoholism are commonly acknowledged to be predisposing
factors of acquiring Legionella, with immunocompromised individuals at a
very high risk of developing disease. It is also well documented that infection
is more common in males than females and individuals who are over 40 years
of age (World Health Organization, 1990). Other risk factors have been doc-
umented including people receiving steroid treatments, patients with chronic
lung disease and diabetes.

L. pneumophila serogroup 1 is the most commonly isolated serotype of
L. pneumophila from infected individuals, followed by L. pneumophila
serogroup 6 (Tang and Krishnan, 1993). L. micdadei is the second most fre-
quent cause of Legionnaires' disease in the USA (Goldberg et al., 1989) with
L. longbeachae an important cause of Legionnaires' disease in Australia
(Steele etal, 1990).

While 15 serotypes of L. pneumophila are known to exist, over 70% of all
culture confirmed are caused by L. pneumophila serogroup 1.

Pathogenicity and virulence

Legionellosis occurs following inhalation of aerosolized droplets of Legionella.
Once inhaled, the organism then attaches to alveolar macrophages (Horwitz,
1993). Legionellae are then phagocytosed by these macrophages. The process
involves complement fragment C3 and the monocyte complement receptors
CR1 and CR3. Virulent strains of Legionella are able to multiply inside
macrophages as they are able to inhibit the fusion of phagosomes with lyso-
somes. With the non-virulent strains of Legionella it seems that multiplication
in macrophages is not possible (Horwitz, 1993).

Two products of Legionella are thought to be the main virulence factors for
Legionella (Fields, 1996). These include a macrophage invasion protein (MIP),
which is 24kDa, and an integral protein of the cytoplasmic membrane
(113 kDa), which is the product of the dot A gene (Roy and Isberg, 1997). Other

147

Bacteriology

proteins have been suggested as virulence factors, but their role is unknown to
date (Barker et al. 9 1993; Kwaik and Engleberg, 1994).

Treatment

Legionella are not detectable in body fluids or tissues but antibody levels in
the blood reach high levels during an infection (Fallon et al., 1993).

Antibiotics used in the treatment of Legionnaires' disease include erythromy-
cin. Other antibiotics that have proved clinically effective include rifampicin,
doxycycline, co-trimoxazole and members of the 5-fluoroquinolones. In the
case of Pontiac fever most patients recover without the need for any therapy.

Newer antibiotics are now available to treat Legionnaires' disease. These
include clarithromycin, ciprofloxacin, azithromycin, tetracycline, dexycycline
and now azithromycin and levofloxacin.

Survival in the environment

Legionella have been isolated in small numbers from a wide range of fresh water
habitats including groundwater (Lye et al., 1997), rivers, lakes and natural
thermal pools (Verissimo et al., 1991), specifically man-made water supplies.
The man-made supplies include hot water systems in hospitals, hotels and, in
particular, cooling towers. In fact in drinking water the long-term survival of
Legionella has been reported.

The optimal growth temperature of Legionella pneumophila is 35°C,
however, it is able to multiply at temperatures between 25 and 42°C. The occur-
rence of Legionella organisms in water supplies does not constitute a major
cause for concern as its presence does not necessarily lead to a Legionella
outbreak. Of concern, however, is the aerosolizing of the organism that is a
risk factor for nosocomial infections, particularly in immunocompromised
patients.

As mentioned previously, Legionella multiplies intracellularly in amoe-
bae and ciliates of the genera Hartmanella, Acanthamoeba, Naegleria,
Echinamoeba, Tetrahymena and Cyclidium (Breiman et al., 1990). Under
adverse conditions amoebae form a cyst entrapping Legionella and the cyst is
blown away in the air. The amoebic cyst provides a protective environment for
viable Legionella.

A number of other species of Legionella including L. bozemanii,
L. longbeachae, L. jordanis, L. wadsworthii, L. birminghamensis,
L. cincinnatiensis, L. oakridgeiensis and L. tucsonensis have been isolated from
the environment but not generally implicated in disease.

148

Legionella

Good management is required for the control of Legionella in water.
Legionellae can be killed when temperatures are elevated above 50°C
(Schulze-Rsbbecke et aL, 1987).

Methods of detection

In environmental samples Legionellae can be concentrated by centrifugation
(e.g. 12 000g for 10 minutes at about 20°C) or by membrane filtration. To
reduce the presence of contaminating unwanted bacterial species heat or acid
treatment could be applied. Following decontamination the sample is trans-
ferred onto buffered charcoal-yeast extract (BCYE) agar (Edelstein, 1981)
and then incubated for 5 days at 37°C. On BCYE, Legionella are identified by
their colonial morphology and the requirement of L-cysteine. After 5 days'
growth on BCYE Legionella colonies have a characteristic 'cut glass' appearance:
grey-white, glistening, convex and 3-4 mm in diameter. Isolates that grow on
BCYE react with appropriate antisera allowing for confirmation of legionellae.
Species within Legionellaceae can be easily distinguished using the appropriate
biochemical, serology and nucleic acid analysis. Environmental and clinical isol-
ates of Legionella can be subtyped by molecular techniques such as ribotyping,
macrorestriction analysis by pulsed-field gel electrophoresis, or PCR-based
methods (Van Belkum et aL, 1996).

When exposed to long- wave UV light at 366 nm, the colonies of some
Legionella spp. autofluoresce brilliant blue-white, some species appear red or
yellow-green, whereas L. pneumophila is not autofluorescent.

Epidemiology

Legionnaires' disease accounts for 1-4% of all cases of pneumonia. This, how-
ever, has been documented as being as high as 30% (Macfarlane et aL, 1982;
Fang et aL, 1990). Pontiac fever incidences are unknown.

Aerosols provide a means of transporting Legionella over immense distances,
suggesting a low infective dose for Legionella. The infectivity of Legionella is
enhanced if amoebae are inhaled or aspirated (Brieland et aL, 1996; Berk et aL,
1998). Aspiration following ingestion of contaminated water, ice and food
has also been implicated as the route of infection in some cases (Marrie et aL,
1991; Venezia et aL, 1994; Graman et aL, 1997).

Outbreaks of Legionnaires' disease are not very common. However, a number
of outbreaks, often sporadic, have been traced to air-conditioning systems,
decorative fountains, room humidifiers, cooling towers and ultrasonic nebu-
lizers, hot whirlpool (Castellani Pastoris et aL, 1999) and spa baths (Jernigan
et aL, 1996), hot water from taps and showers, and medical devices containing

149

Bacteriology

water (Butler and Breiman, 1998). Most cases of Legionella in hospitals are
associated with contaminated potable water and hot-water systems (Joseph
etal.,1994).

Risk assessment

Health effects: occurrence of illness, degree of morbidity and mortality, prob-
ability of illness based on infection:

• Legionella causes legionellosis, which has two forms: Pontiac fever and
Legionnaires' disease.

• Pontiac fever is a relatively mild, self-limiting, influenza-like illness. Pontiac
fever presents in otherwise healthy individuals as pleuritic pain in the
absence of pneumonic or multisystem manifestations.

• Pontiac fever has a short incubation period, a high attack rate, but a very
low mortality ratio.

• Legionnaires' disease is a severe respiratory illness. The infection can last
for weeks to months. The symptoms of infection include pneumonia with
anorexia, malaise, myalgia and headache, rapid fever and chills, a cough,
chest pain, abdominal pain and diarrhoea. Acute renal failure, dissemi-
nated intravascular coagulation, shock, respiratory insufficiency, coma and
circulatory collapse are the major factors precipitating death.

• Legionnaires' disease has a low attack rate (1-6%), but a mortality rate of
about 10%.

• Legionnaires' disease is more common in those with underlying illnesses,
smokers, the elderly, or the immunocompromised (for whom the prognosis
is poor).

• Legionnaires' disease accounts for 1-4% of all cases of pneumonia,
although rates as high as 30% have been reported. The incidence of Pontiac
fever is however unknown.

Exposure assessment: routes of exposure and transmission, occurrence in
source water, environmental fate:

• Legionella are ubiquitous in the environment and have been isolated from
a wide range of fresh water habitats including ground water, rivers, lakes
and natural thermal pools. Higher numbers of Legionella have been docu-
mented in man-made water supplies. These have included air-conditioning
condensers, cooling tower effluent, humidifiers, nebulizers, potable and hot
water supplies, domestic and hospital showerheads, whirlpool spas, deco-
rative fountains and vegetable misting machines.

• The association of Legionella with fresh water amoebae in the aquatic envir-
onment is well documented. The organism multiplies intracellularly in
amoebae and ciliates, especially when water temperatures are elevated. This

150

Legionella

relationship protects the bacteria against dry conditions, extremes of tem-
perature, and treatment with biocides.

• Delivery of the agent to the respiratory tract in the form of water droplets
5-15 fjim in diameter serves as the primary vehicle for disease transmission;
aerosols containing the organism constitute a major risk factor for nosoco-
mial infections and for those who are immunocompromised.

• Person-to-person transmission of Legionella is not thought to occur.

• The infectious dose is unknown.

Risk mitigation: drinking-water treatment, medical treatment:

• Legionella is susceptible to both heat and chlorine. Intermittent temperature
increases in the water supply of up to 60°C, as well as the use of chlorination
procedures to give a continuous 1-2 ppm residual are effective. In addition
to disinfection, remedial cleaning and flushing of water systems as well as
the removal of sediment from hot water tanks, coupled with water treat-
ment and constant water monitoring are necessary.

• It is imperative that when designing the plumbing systems of new buildings
and those undergoing modification, dead space volumes must be reduced,
sediment build-up and stagnation prevented as these conditions favour the
growth of legionellae, particularly in biofllms.

• The antibiotic most often given for legionellosis is erythromycin, however
ciprofloxacin or rifampicin may be added. With Pontiac fever most patients
recover without specific therapy.

• In immunocompromised individuals the death rate for Legionnaires' disease
is relatively high, despite appropriate therapy.

References

Barker, J., Lambert, P.A. and Brown, M.R.W. (1993). Influence of the intra-amoebic and
other growth conditions on the surface properties of Legionella pneumophila. Infect
Immun, 61: 3503-3510.

Berk, S.G. et al. (1998). Production of respirable vesicles containing live Legionella
pneumophila cells by two Acanthamoeba spp. Appl Environ Microbiol, 64: 279-286.

Breiman, R.F., Fields, B.S. et al. (1990). Association of shower use with Legionnaires' dis-
ease. Possible role of amoeba. JAMA, 263: 2924-2926.

Brieland, J. et al. (1996). Coinoculation with Hartmannella vermiformis enhances replica-
tive Legionella pneumophila lung infection in a murine model of Legionnaires' disease.
Infect Immun, 64: 2449-2456.

Butler, J.C. and Breiman, R.F. (1998). Legionellosis. In Bacterial infections of humans,
Evans, A.S. and Brachman, P.S. (eds). New York: Kluwer Academic/Plenum, pp. 355-375.

Castellani Pastoris, M. et al. (1999). Legionnaires' disease on a cruise ship linked to the
water supply system: clinical and public health implications. Clin Infect Dis, 28: 33-38.

Chandler, F.W., Roth, K. et al. (1980). Flagella on Legionnaires' disease bacteria: ultra-
structural observations. Ann Intern Med, 93: 711-714.

Edelstein, P.H. (1981). Improved semiselective medium for isolation of Legionella
pneumophila from contaminated clinical and environmental specimens. / Clin
Microbiol, 14:298-303.

151

Bacteriology

Fallon, R.J. et al. (1993). Pontiac fever in children. In Legionella: Current Status and Emer-
ging Perspectives, Barbaree, J.M., Breiman, R.F. and Dufour, A.P. (eds). Washington,
DC: American Society for Microbiology, pp. 50-51.

Fang, G.D., Fine, M. et al. (1990). New and emerging etiologies for community acquired
pneumonia with implications for therapy. A prospective multicenter study of 359 cases.
Medicine (Baltimore), 69: 307-316.

Fields, B.S. (1993). Legionella and protozoa: interaction of a pathogen and its natural host.
In Legionella: Current Status and Emerging Perspectives, Barbaree, J.M., Breiman, R.F.
and Dufour, A.P. (eds). Washington, DC: American Society for Microbiology, pp. 129-136.

Fields, B.S. (1996). The molecular ecology of legionellae. Trends Microbiol, 4: 286-290.

Goldberg, D.J. et al. (1989). Lochgoilhead fever: outbreak of non-pneumonic legionellosis
due to Legionella micdadei. Lancet, i: 316-318.

Graman, P.S., Quinlan, G.A. and Rank, J. A. (1997). Nosocomial legionellosis traced to a
contaminated ice machine. Infect Control Hosp Epidemiol, 18: 637-640.

Hookey, J.V. et al. (1996). Phylogeny of Legionellaceae on small-subunit ribosomal DNA
sequences and proposal of Legionella lytica comb. nov. for Legionella-like amoebal
pathogens. Int J Systematic Bacteriol, 46: 526-531.

Horwitz, M.A. (1993). Toward an understanding of host and bacterial molecules mediating
Legionella pneumophila pathogenesis. In Legionella: Current Status and Emerging Per-
spectives, Barbaree, J.M., Breiman, R.F. and Dufour, A.P. (eds). Washington, DC:
American Society for Microbiology, pp. 55-62.

Jernigan, D.B. et al. (1996). Outbreak of Legionnaires' disease among cruise ship passen-
gers exposed to a contaminated whirlpool spa. Lancet, 347: 494-499.

Joseph, C.A. et al. (1994). Nosocomial Legionnaires' disease in England and Wales,
1980-1992. Epidemiol Infect, 112: 329-345.

Kwaik, Y.A. and Engleberg, N.C. (1994). Cloning and molecular characterization of
Legionella pneumophila upon infection of macrophages. Infect Immun, 61: 1320-1329.

Lye, D. et al. (1997). Survey of ground, surface, and potable waters for the presence of
Legionella species by EnviroAmpO PCR Legionella kit, culture, and immunofluores-
cent staining. Water Res, 31: 287-293.

Macfarlane, J.T., Finch, R.G. et al. (1982). Hospital study of adult community acquired
pneumonia. Lancet, 2: 255-258.

Marrie, T.J. et al. (1991). Control of endemic nosocomial Legionnaires' disease by using
sterile potable water for high risk patients. Epidemiol Infect, 107: 591-605.

Mayock, R., Skale, B. and Kohler, R.B. (1983). Legionella pneumophila pericarditis proved
by culture of pericardial fluid. Am J Med, 75: 534-536.

Pruckler, J.M. et al. (1995). Comparison of Legionella pneumophila isolates by arbitrarily
primed PCR and pulsed-field gel electrophoresis: analysis from seven epidemic investi-
gations./ Clin Microbiol, 33: 2872-2875.

Rodgers, F.G., Macrae, A.D. and Lewis, M.J. (1978). Electron microscopy of the organism
of Legionnaires' disease. Nature, 272: 825-826.

Roy, C.R. and Isberg, P.R. (1997). Topology of Legionella pneumophila dotA: an
inner membrane protein required for replication in macrophages. Infect Immun, 65:
571-578.

Schulze-Rsbbecke, R., Rsdder, M. and Exner, M. (1987). Multiplication and killing tem-
peratures of naturally occurring Legionellae. Zentralbl Bakteriol Mikrobiol Hyg, 184:
495-500.

Steele, T.W., Lanser, J. and Sangster, N. (1990). Isolation of Legionella longbeachae
serogroup 1 from potting mixes. Appl Environ Microbiol, 56: 49-53.

Tang, P. and Krishnan, C. (1993). Legionellosis in Ontario, Canada: Laboratory aspects. In
Legionella: Current Status and Emerging Perspectives, Barbaree, J.M., Breiman, R.F.
and Dufour, A.P. (eds). Washington, DC: American Society for Microbiology, pp. 16-17.

Tison, D.L. et al. (1980). Growth of Legionella pneumophila in association with blue-green
algae (Cyano bacteria) . Appl Environ Microbiol, 39: 456-459.

Van Belkum, A. et al. (1996). Serotyping, ribotyping, PCR-mediated ribosomal 16S-23S
spacer analysis and arbitrarily primed PCR for epidemiological studies on Legionella
pneumophila. Res Microbiol, 147: 405-413.

152

Legionella

Venezia, R.A. et al. (1994). Nosocomial legionellosis associated with aspiration of naso-
gastric feedings diluted in tap water. Infect Control Hosp Epidemiol, 15: 529-533.

Verissimo, A. et al. (1991). Distribution of Legionella spp. in hydrothermal areas in conti-
nental Portugal and the island of Sao Miguel, Azores. Appl Environ Microbiol, 57:
2921-2927.

Wadowsky, R.M. and Yee, R.B. (1985). Effect of non-Legionellaceae bacteria on the multi-
plication of Legionella pneumophila in potable water. Appl Environ Microbiol, 49:
1206-1210.

World Health Organization. (1990). Epidemiology, prevention and control of legionellosis:
memorandum from a WHO meeting. Bull World Hlth Org, 68: 155-164.

153

The Mycobacterium avium
complex

Basic microbiology

Mycobacteria are Gram-positive, slender, non-motile, non-spore forming, rod-
shaped bacilli, which may appear bent or curved. Cells are pleomorphic, ranging
from a coccoid form to long slender rods, 0.2-0.6 fjim wide and 1.0-10 |xm
long. Most mycobacteria are aerobic organisms, but some species may be
microaerobic (Falkinham, 1996). All mycobacteria are catalase-positive. The
high lipid content of the bacterial cell wall provides the bacteria with resist-
ance to acid and alkali. Certain mycobacteria, particularly M. avium paratu-
berculosis (MAP), can shed their cell walls, forming a spheroplast (Thompson,
1994). It is the cell walls of bacteria that pick up stains, therefore the sphero-
plast form of the bacteria cannot be detected using the acid-fast stain test.
Because the mycobacterial cell wall is hydrophobic Mycobacterium tend to
grow in clumps and mycobacterial colonies float on the surface of liquid
media. As a consequence detergents, such as Tween 80, are often used in cul-
ture media to help break down the clumping of organisms.

Many species of mycobacteria exist as saprophytes in soil and water and are
referred to as environmental mycobacteria; they do not generally cause human

Bacteriology

infections. Some species of mycobacteria, however, do cause disease in animals
and humans and are classified according to their host species, i.e. human,
bovine, avian, murine and piscine, with species of mycobacteria categorized by
their potential pathogenicity to humans. Mycobacterial species are very closely
related to Nocardia, Rho do coccus and Corynebacterium spp. However, the
first requirement for an unknown microorganism to be classified with the
mycobacteria, is that it be 'acid fast'. Nocardia and Rhodococcus are referred
to as 'partially acid fast'.

The most important disease causing Mycobacterium is Mycobacterium tuber-
culosis which causes tuberculosis in humans. Tuberculosis is also caused
by Mycobacterium bovis, which causes tuberculosis in cattle and humans and
Mycobacterium africanum, a rare cause of human tuberculosis in central Africa.
Mycobacterium tuberculosis, bovis and africanum all display some phenotypic
differences. They are genetically very closely related and generally classed as the
'Mycobacterium tuberculosis complex'. However, epidemiological and treat-
ment differences exist for diseases caused by Mycobacterium tuberculosis and
Mycobacterium bovis - these two are still usually regarded as separate species.

Other important pathogenic mycobacteria include the 'Mycobacterium
avium intracellular •#' complex and the 'atypical' mycobacteria. These groups
of Mycobacterium are frequent opportunistic pathogens in the immunosup-
pressed, however, they are documented as causing disease in individuals with
normal immune systems.

The genus Mycobacterium can also be divided into two groups on the basis
of growth rate: rapid growers and slow growers. Rapid growers form visible
colonies on solid media within 7 days, while slow growers require longer than
7 days to produce visible colonies. Mycobacterium avium complex (MAC)
species are defined as slow growers. In the Runyon Classification of non-
tuberculous Mycobacteria, MAC organisms are classed in group III, non-pho-
tochromogenic slow growers. There are 28 known serotypes belonging to this
MAC complex (Wolinsky, 1992). They are discussed in more detail below.

Origins of the organism

There are 71 validly named species of Mycobacterium and an additional three
subspecies. The principal pathogens in the genus are M. bovis, M. leprae and
M. tuberculosis but, in all, 32 species are known to be pathogenic to humans
or animals. Species of Mycobacterium other than, M. bovis, M. leprae and
M. tuberculosis are often referred to as 'atypical mycobacteria' (see Table 11.1).
The most commonly encountered pathogens among the atypical mycobacte-
ria are species of the Mycobacterium avium complex. The M. avium complex
(MAC) is considered to contain M. avium, M. avium subspecies paratubercu-
losis, M. avium subspecies silvaticum and M. intracellular. However, poorly
identified strains which show some similarity to M. avium are also frequently,

156

The Mycobacterium avium complex

and incorrectly, allocated to the complex. There are over 20 recognized
serotypes within the M. avium complex.

Mycobacteria were first discovered in 1874 when Armauer Hansen found
acid-fast bacilli in individuals suffering from leprosy (Hansen, 1874).
Mycobacterium tuberculosis was discovered in culturable form in 1882 when
Robert Koch, using coagulated bovine serum, isolated it (Koch, 1882). In
1896 the generic name Mycobacterium was recognized (Goodfellow and
Wayne, 1982).

Johne first discovered Mycobacterium avium paratuberculosis in 1895, an
organism now thought to cause the disease pseudotuberculosis enteritis (Johne
and Frothingham, 1895). Mycobacterium avium paratuberculosis has been isol-
ated from many species of animals. It was not until 1910 that Twort successfully
isolated the first M. avium paratuberculosis (Twort, 1911). At this time the isol-
ate of Mycobacterium was named Mycobacterium enteriditis chronicae pseudo-
tuberculosae bovis Johne (Twort and Ingram, 1912). However, in 1932 the
bacterium was renamed M. paratuberculosis. To date Mycobacterium avium
subspecies paratuberculosis (MAP) is a member of the Mycobacterium avium
complex (MAC) (Thorel et al., 1990). M. paratuberculosis seems to be linked to
Crohn's disease in humans. However, the evidence for this is very limited at pre-
sent. Crohn's disease was first recognized in 1913 by Kennedy Dalziel, a surgeon
from Glasgow who believed the disease must have a mycobacterial cause.
However, he was unable to culture viable organisms from tissue samples (Dalziel,
1913). In 1985, the IS900 insertion sequence, unique to M. paratuberculosis was
discovered. IS900 was the first DNA insertion sequence to be found in mycobac-
teria, it comprises a 1451-1453 base pair repetitive element (Green etal. 9 1989).
The finding of this species-specific marker and the amplification of the poly-
merase chain reaction (PCR) made specific identification of MAP possible.

Table 11.1 Principal types of disease in humans and causative agents

Disease Agent

Lymphadenopathy M. avium complex

M. scrofulaceum
Skin lesions

Post-trauma abscess M. chelonae

M. fortuitum
M. terrae
Swimming pool granuloma M. marinum

Buruli ulcer M. ulcerans

Pulmonary disease M. avium complex

M. kansasii
M. xenopi
M. malmoense
Disseminated disease

AIDS-related M. avium complex

M. genevense
Non-AIDS related M. a vium complex

M. chelonae

157

Bacteriology

In 1959, a classification scheme for Mycobacterium was developed (Runyon).
This scheme enabled mycobacteria to be classified according to rate of growth
(slow or rapid) and production of pigment (yellow or orange). Runyon's scheme

included:

• Group I Photochromogens which produced pigment on exposure to light

• Group II Scotochromogens which produced pigment in the dark

• Group III Non-chromogens which produced no pigment

• Group IV Rapid growers.

In this scheme all strains of Mycobacterium in groups I, II and III grew slowly;
rapid growers (group IV) may be photochromogens, scotochromogens or
non-chromogens. Runyon's classification scheme separates Mycobacteria
tuberculosis from most of the 'atypicals'. The preliminary division of the atyp-
icals in Runyon's classification is still used today for the classification
of mycobacteria species, when combined with other biochemical and growth
characteristics. Identification of species of mycobacteria today is undertaken
by specialist reference laboratories. These are usually based on cultural char-
acteristics (rate and temperature of growth and pigmentation), various bio-
chemical reactions, resistance to antimicrobials and lipid content of the
cell wall.

The Mycobacterium which belong to group I photochromogens are colour-
less cultures in the dark but form bright yellow or orange pigmentation when
they are exposed to light. Three important species make up this group, namely
M. kansasii, M. simiae and M. marinium. M. kansasii grows well at 37°C and
produces a yellow pigment, it rapidly hydrolyses tween in 3 days and has
strong pyrazinamidase activity. It is an organism that is commonly isolated
from chronic pulmonary disease. M. simiae also grows well at 37°C and
hydrolyses tween slowly, taking 10 days or more to do this. Like M. kansasii
it has a high thermostable catalase and is isolated from cases of pulmonary
disease. M. marinum, on the other hand, is the fish tubercule bacillus that
causes warty skin infection and is often referred to as swimming pool granu-
loma or fish tank granuloma. Its optimal growth temperature is 30-32°C with
poor growth at 37°C. It is urease-positive, has no heat stable catalase, but
does produce pyrazinamidase.

Mycobacteria belonging to group II are the scotochromogens and include
the three species Mycobacterium scrofulaceum, M. gordonae and M. szulgai.
M. scrofulaceum is associated with scrofula or cervical lymphadenitis and
also causes pulmonary disease. It is a slow grower (4-6 weeks) and produces
a light yellow to deep orange pigment which is independent of light exposure.
It fails to hydrolyse tween, produces catalase at 68°C but does not reduce
nitrate. M. szulgai as a comparison to M. scrofulaceum is an uncommon cause
of pulmonary disease and bursitis. It has a temperature dependent pigment
production and is a rapid grower (2 weeks at 37°C). It hydrolyses tween
slowly and is intolerant to 5% sodium chloride. M. gordonae is also a rare
cause of lung disease and is frequently found in water. However, it is often a
common contaminant of clinical material. It is able to hydrolyse tween and

158

The Mycobacterium avium complex

produces a stable catalase. It is urease-negative and does not reduce nitrates.
It is resistant to isoniazid and streptomycin but is susceptible to rifampicin
and ethanbutol.

Group III mycobacteria are referred to as the non-chromogens and include
M. avium and M. intracellular e (MAI or MAC) and M. malmoense, M. xenopi,
M. ulcerans and M. terrae. The Mycobacterium avium complex constitutes the
most prevalent and important opportunistic group of human pathogens.
M. avium is the avian tubercule bacillus and M. intracellular e is the battery
bacillus. Most of the MAC clinical isolates are smooth and easily emulsifiable.
There are 28 serotypes delineated by agglutination by specific antisera. MAC is
responsible for lymphadenitis, pulmonary lesions and disseminated disease,
notably in patients with AIDS. MAC does not produce a heat stable catalase
and is generally inert biochemically. M. xenopi was first isolated from the
xenopus toad. It is a thermophile that grows at 45°C and has been isolated
from pulmonary lesions. M. malmoense causes pulmonary disease and lym-
phadenitis and grows very slowly (10 weeks) and is therefore likely to be
missed if cultures are not maintained for up to 12 weeks. M. ulcerans is the
cause of Buruli ulcers. It is a very slow growing bacterium and produces a toxin
which causes tissue necrosis. On inoculation into the foot pads of mice it
causes disease, including swelling, ulceration and often autoamputation.

M. haemophilum is characterized by a growth requirement for haem or
other iron sources. It is a rare cause of granulomatous or ulcerative skin lesion
in xenograft recipients and other immunocompromised patients. It has an
optimum growth at 28-32°C which is stimulated by 10% C0 2 . The only bio-
chemical test that is positive is pyrazinimidase production.

The final group of Mycobacterium are found in Group IV and referred to as
the rapid growers. M. chelonae and M. fortuitum are the two well-recognized
human pathogens found in this group. They are principally responsible for
post-injection abscesses, wound infections and corneal ulcers. They occasionally
cause pulmonary or disseminated disease. Both species are non-chromogenic.

The Mycobacterium species as an entire group have a large number of non-
culturable medical forms. These are referred to as the very poorly growing
species that are mostly isolated from the blood of AIDS patients. The classifi-
cation of this group is based on unique base sequences in their 16S ribosomal
RNA. These c non-culturable' Mycobacterium include: M. genevense, M. con-
fluentis, M. intermedium and M. interjectum.

Clinical features

Mycobacterium, but more specifically the environmental mycobacteria, have
been known to be opportunistic pathogens of many disseminated diseases
with patients dying of AIDS, skin lesions such as post-injection abscesses,
swimming pool granulomas and Buruli ulcers, tuberculosis-like lesions and
also lympadenitis following infection. Thirty-two of the 71 species of

159

Bacteriology

Mycobacterium are known to be pathogenic to humans or animals. The most
common non-tuberculous, or atypical mycobacterial pathogens are species
found within the Mycobacterium avium complex (MAC). MAC contains
M. avium^ M. intracellular '£, M. avium subspecies paratuberculosis (MAP),
and M. avium subspecies silvaticum.

MAC organisms cause disease in pigs and poultry. The organisms are con-
tinually excreted in the faeces of birds and can live in the soil for long periods
of time (Inderlied et al., 1993).

In normal healthy individuals MAC rarely cause disease. However, individ-
uals who have suppressed immune systems, i.e. the elderly and the very young
and AIDS suffers, are susceptible to both pulmonary and non-pulmonary infec-
tions. Individuals who are HIV positive disseminate MAC to other parts of
the body including joints, skin, blood, lungs, liver and brain. MAC entry
into the host is via the respiratory and gastrointestinal tracts (Chin et al.,
1994).

M. paratuberculosis is suspected to cause the chronic, inflammatory, gas-
trointestinal disease of humans known as Crohn's disease (Hermon -Taylor,
1998). It has been found to be present in milk, meat and faeces of infected
cattle. In cattle the disease caused by MAP is called Johne's disease. Johne's
disease has clinical, systemic and pathological characteristics similar to infec-
tion in Crohn's disease in humans (Chiodini, 1989).

Crohn's disease is often referred to as an autoimmune disease where the
body's immune system attacks and inflames its own tissues. The place where
this occurs is the gastrointestinal tract (Chiodini, 1996). As the GI tract becomes
inflamed this leads to a narrowing of the digestive system leading to severe
pain and uncontrollable bowel movements. A number of research papers have
been published that have patients going into remission when treated with
appropriate antibiotics directed at M. paratuberculosis. The first documented
case of human disease caused by M. paratuberculosis was published in 1998
following a 7-year-old boy who developed cervical lymphadenitis. Following
the removal of the boy's lymph nodes the biopsy was found to contain M.
paratuberculosis. Follow up of this patient 5 years after the operation found
that the boy was now suffering from a chronic inflammation of his intestine.
Following treatment with antibiotics active against M. paratuberculosis the
boy's intestinal disease went into remission (Barnes et al., 1998).

MAP is known to be an intracellular pathogen, which is able to colonize
and multiply in macrophages. This may well be a way that MAP is transmit-
ted in milk (cow's milk contains white blood cells) following ingestion of con-
taminated milk from infected cattle. (Hermon-Taylor, 1993). Research has
shown that the normal pasteurization process of 72°C for 15 seconds does not
ensure the complete eradication of MAP (Grant et al. 9 1998). Results from a
study conducted in 1991 to 1993 found MAP organisms in commercially pas-
teurized milk samples throughout central and southern England (Millar,
1996). It is probable that as M. paratuberculosis is known to form aggregates
or possible biofilms this may well aid in its survival during the pasteurization
process.

160

The Mycobacterium avium complex

Treatment

Atypical mycobacteria are often found to be resistant to many anti-TB drugs
in vitro. However, using appropriate patient management and applying a
combination of drugs, non-typical mycobacteria infections can be treated.
Regimens with five or six drugs are used for pulmonary disease due to MAC,
M. kansasii and M. xenopi. Triple therapy using isoniazid, rifampicin and
ethambutol has also been used over an 18-month period with positive results.
There are multiple drug regimens for AIDS-associated MAC disease. These
include clarithromycon or azithromycin with two other drugs selected from
rifabutin, ethambutol, clofazimine, fluoroquinolones and amikacin. Local
skin lesions, which are due to mycobacterium, are usually excised where pos-
sible. With M. chelonae keratitis relapses are very frequent and therefore
surgical intervention is often necessary. The treatment of non-tuberculosis
diseases are usually unsuccessful in a large minority of cases.

Survival in the environment

MAC have been isolated world-wide from all natural water systems, includ-
ing marine water and soil (Grange et al. 9 1990). Man-made habitats, such as
water distribution systems, are colonized by considerably high numbers of
mycobacterial species. Aquatic mycobacteria will colonize biofilms at
solid-water and air-water interfaces, which seem to be an important replica-
tion site in oligotrophic habitats (Schulze-Robbecke, 1993).

Mycobacteria therefore seem to survive well in the environment. They are
also documented as being able to survive in temperatures below 0°C. Iivanainen
et al. (1995) collected samples from shallow brooks to investigate the effect of
prolonged storage on mycobacteria and other heterotrophic bacteria. Water
concentrates were stored in nutrient broth at — 75°C for 15 months. Viable
counts were taken from the fresh samples and again following storage. Their
results showed a threefold increase in the numbers of mycobacteria present
following storage (Iivanainen, 1995). This suggests a very sophisticated survival
mechanism and one which may aid in the transmission of Mycobacterium in the
environment. Food, in this case defrosted water from fish, has been shown to
be a good medium for the transmission of mycobacteria (Mediel et al., 2000).
From a study conducted in California a variety of foods collected from super-
markets and market stalls were shown to contain Mycobacterium spp. From
this study six species of non-tuberculous mycobacteria were isolated from 25
out of the 121 foods tested. M. avium was found to be the most frequently isol-
ated species of Mycobacterium (Argueta et al., 2000).

Mycobacterium avium paratuberculosis can also survive and possibly even
replicate in natural waters and soil. However, to date this has not been deter-
mined. It is probable that MAP is able to replicate in farm animals. These

161

Bacteriology

infected animals then disseminate MAP in their faeces that are then reingested
by other animals (Hermon-Taylor et al. 9 1994).

Water seems to be the main reservoir for environmental mycobacteria
(Wallace, 1987; Iivanainen et al. 9 1999). However, water- treatment processes
are known to help reduce the numbers of MAC organisms present in water
supplies. Replication in biofllms in drinking water pipes may suggest a public
health concern. Mycobacteria are able to survive the normal chlorination
process used in drinking water because of the thick, lipid-rich cell wall. It is
well documented that chlorine levels used in water purification are unlikely to
be effective against mycobacteria, including the MAC organisms (Pelletier,
1988). Evidence for this lies in a study of nosocomial outbreaks, occurring
when water supplies from two hospitals, in Boston and New Hampshire, were
the source of a MAC strain, showing the same genetic pattern as MAC strains
infecting groups of AIDS patients in the hospitals (Von Rehn et al. 9 1994).
Also the current water-treatment processes are unlikely to eradicate MAP
organisms from the water supply (Hermon-Taylor, 1998). Domestic water
systems have been associated with human non-tuberculous mycobacterial
infections even at high temperatures (Schulze-Robbecke and Buchholtz, 1992;
Miyamoto et al. 9 2000).

A number of studies have shown that mycobacteria are present in tap water.
A study by Peters et al. 9 in 1995, showed the presence of Mycobacterium in tap
waters in Berlin. In this study mycobacteria were found in 42.4% of samples;
28% were Mycobacterium gordonae and 1.7% MAC (Peters et al. 9 1995).

Covert et al. (1999) sampled water for the presence of mycobacteria in 21
states in the USA. The water sources included potable water, water in cisterns,
bottled water, drinking-water-treatment samples and ice-machine samples.
The results of the study found that non-tuberculous mycobacteria were detected
in 54% of the ice samples and 35% of the public drinking water samples
analysed. The species isolated in all waters were: M. gordonae; M. peregrinum;
M. mucogenicum; M. intracellular m e; M. scrofulaceum; M. gastri/kansasii;
M. fortuitum; M. avium and M. chelonae. M. mucogenicum was the isolate
most commonly found (Covert et al. 9 1999).

Another study in Los Angeles examined potable water to estimate if
M. avium complex strains isolated from the water were related to strains iso-
lated from AIDS patients in the Los Angeles area. This study concluded that
there was a possible spread of M. avium to immunocompromised patients,
particularly those suffering from AIDS (Aronson, 1999).

When Schulze-Robbecke et al. (1992) analysed 50 biofllm samples obtained
from water-treatment plants, domestic water supplies and aquaria, 90% of
the samples contained mycobacteria.

A study by Falkinham and coworkers (2001) over an 18-month period
found Mycobacterium in water and biofllm samples collected from eight
water-distribution systems. Mycobacteria were isolated in only 15% of sam-
ples tested. Water-treatment processes substantially reduced the numbers of
mycobacteria from the raw water. The highest numbers of mycobacteria were
present in the distribution system samples. M. intracellular e was mostly

162

The Mycobacterium avium complex

recovered from biofilm samples, whereas M. avium was predominant in the
water samples (Falkinham et al., 2001).

Cirillo (1999) investigated the possibility that M. avium could interact with
and replicate within Acanthamoeba castellanii. The results from this study
showed that M. avium was able to replicate within the amoebae (Cirillo et al.,
1997). The ability of M. avium to enter and replicate inside cells may aid
its pathogenicity. From a study by Cirillo et al. (1997) it was found that
M. avium grown within amoebae had increased invasion of cells and intracel-
lular replication when compared to M. avium grown in broth (Cirillo et al.,
1997). In addition, intracellular M. avium within A. castellanii have increased
resistance to disinfectants and also antibiotics (Miltner and Bermudez, 2000).

Detection methods

To isolate and enumerate Mycobacterium from the environment an environ-
mental sample needs first to be decontaminated for the growth of
Mycobacterium. This is because environmental samples contain a lot of fast-
growing heterotrophic bacteria that will overgrow, preventing the isolation of
slow growing Mycobacterium. The most commonly used decontaminating
agent used in the isolation of Mycobacterium was sodium hydroxide. It was
first used as a decontaminant in the culture of M. paratuberculosis from clin-
ical specimens (Petroff, 1915). However, sodium hydroxide does have detri-
mental effects on the growth of mycobacteria. Other decontaminating agents
used in the isolation of Mycobacterium have included sulphuric acid (Kamala
et al. 9 1994), sodium triphosphate, oxalic acid (Portaels et al., 1988) and
cetylpyridinium chloride (CPC) (Neumann et al., 1997).

In order to grow M. avium and M. intracellulare a solid medium supple-
mented with egg or albumin is required. As M. paratuberculosis is the slowest
growing of the culturable mycobacteria Mycobactin J is required as a growth
factor (Portaels et al., 1988). For the M. avium subspecies silvaticum any
media used for its isolation must be supplemented with pyruvate and an acidic
pH. MAC will grow over a temperature range of 25-45°C, however, 30°C
appears to be the optimum temperature for maximum growth (Neumann
et al., 1997). The incubation period for the culture of environmental
mycobacteria can take 6 months or longer (Iivanainen et al., 1993).

One of the most popular media for the growth of Mycobacterium is
Lowenstein-Jensen medium. It is composed of whole egg, citrate, salts, potato
starch, asparagine, glycerol and malachite green. Glycerol is used as a carbon
source for the growing bacteria and malachite green helps to inhibit other
microbial growth and also provides a contrasting colour to enable the small,
pale mycobacterial colonies to be more visible (Jenkins et al., 1982).
Lowenstein-Jensen medium has a pH of 7.0. Many non-tuberculous mycobac-
teria however have a pH optimum of pH 5.4-6.5 (M. avium is pH 5-5.5)
(Portaels and Pattyn, 1982; Falkinham, 1996). Ogawa egg yolk medium has

163

Bacteriology

a lower pH (pH 6), and this medium seems to be more suitable than
Lowenstein-Jensen for the growth of environmental species (Portaels et aL 9
1988). Middlebrook 7H10 agar has also been used for the isolation of
mycobacteria (Iivanainen et al. 9 1999).

The best decontamination methods and culture media will determine the
highest numbers and greatest diversity of species of mycobacteria isolated
from environmental samples. A study by Iivanainen (1995) involving 15 com-
binations of decontamination methods established that the best method for
greatest recovery of Mycobacterium was achieved by using egg media supple-
mented with glycerol at pH 6.5, following decontamination with NaOH,
malachite green and cycloheximide (Iivanainen, 1995).

In Germany, Neumann et al. (1997), using 12 isolation methods, undertook
a study to optimize methods for the isolation of mycobacteria from water sam-
ples. They analysed 109 treated water samples and 26 surface water samples.
For the treated water samples cetylpyridinium chloride (CPC) was used at con-
centrations of 0.005% or 0.05%, with an exposure time of 30 minutes. For the
surface waters, an additional method was also used. For this analysis Tryptic
soy broth (TSB) was added to the sample and incubated at 37°C for 5 hours.
Malachite green oxalate, cycloheximide and NaOH were then added, with an
exposure time of 30 minutes, followed by the addition of hydrochloric acid.
Lowenstein-Jensen media containing glycerol, Ogawa egg yolk medium and
Ogawa medium containing ofloxacin and ethambutol were used for primary
culture. From this study 586 mycobacterial strains were isolated. Strains were
identified using mycolic acid thin-layer chromatography and PCR restriction
analysis.

The best methods for isolation of mycobacterium in treated water involved
decontamination in 0.005% CPC, followed by culture on Lowenstein-Jensen
medium with an incubation temperature of 30°C. For surface water the best
three overall methods were: decontamination in 0.05% CPC, followed by cul-
ture on Lowenstein-Jensen media and incubation at 30°C; decontamination
with the TSB method, culture on Lowenstein-Jensen medium and incubation
at 30°C or decontamination with the TSB method, culture on Ogawa egg yolk
medium and incubation at 30°C (Neumann et al., 1997).

Automated culture systems are available for the culture and isolation
of mycobacterial species. The first to be developed was BACTEC 460. This
system uses Middlebrook 7H9 broth supplemented with antibiotics and
C-palmitic acid (Iivanainen et al., 1999). The ESP Culture System II uses
Middlebrook 7H9 broth supplemented with antibiotics and enrichments. This
system detects mycobacterial respiration as a decrease in gas pressure. MGIT,
BBL and BACTEC 9000 use an oxygen-sensitive fluorescent sensor to detect
a decrease in oxygen levels indicating bacterial respiration. The BacT/Alert
System measures an increase in carbon dioxide levels using a reflectometer. The
BacT/Alert commenced in the USA in June 1990.

The rate of mycobacterial isolation and recovery from 5208 samples was
compared between the MB/BacT culture system and Lowenstein-Jensen

164

The Mycobacterium avium complex

medium (Palacios et al. 9 1999). Accuprobe DNA probes were used to identify
any mycobacteria isolates. From this study mycobacteria were isolated from
only 301 samples. This study established that the detection rate of MB/BacT
was higher than with Lowenstein-Jensen medium, in fact MB/BacT was found
to be more efficient and faster at isolating and detecting mycobacteria than
the Lowenstein-Jensen medium (Palacios et al. 9 1999).

The BACTEC 460 TB system has become the 'gold standard' by which other
culture and detection systems are evaluated (Woods et al. 9 1997). A number of
comparisons between the BACTEC 460 TB and the MB/BacT system have
taken place. A study by Roggenkemp et al. (1999) found the MB/BacT system
had a lower sensitivity than the BACTEC 460. A Swiss study evaluated the
MB/BacT system in comparison to the BACTEC 460 system and solid media.
From this study the BACTEC system had a shorter time to detection with less
contamination of cultures. However, the MB/BacT system is fully automated
and is a closed system, reducing the risk of cross-contamination (Rohner et al. 9
1997). Other comparisons have taken place which have found that MB/BacT
system took a longer time to detect all mycobacterial isolates and had a higher
contamination rate than other systems of detection (Benjamin et al. 9 1998).
Brunello et al. (1999) compared the sensitivity and detection time of the
MB/BacT system, BACTEC 460 TB system and conventional solid media,
Lowenstein-Jensen slopes. From this study there was no significant difference
in the recovery rates of the three systems examined. Another study in Toronto
evaluated the performance of the MB/BacT system and the BACTEC
460 TB system in comparison with the solid egg based medium Lowenstein-
Gruft, in detecting mycobacterial growth. In this study the MB/BacT system
recovered more mycobacteria than other systems of recovery (Laverdiere
et al. 9 2000).

A study in 2000 by Harris and coworkers involved analysing 681 clinical
samples and eight control samples to compare the recovery of mycobacteria
using three isolation systems, the MB/BacT rapid culture system, the
BACTEC 460 radiometric system and conventional egg media. It was found
that none of the systems recovered all of the positive isolates.

Once mycobacteria have been recovered by an appropriate method they
then have to be identified. Mycobacteria can be identified using standard
taxonomy methods. Identification between species can be very difficult and
complex. It is generally achieved using gene probes or DNA analysis. Myco-
bacteria are usually detected using staining and fluorescence microscopy. The
method involves staining Mycobacterium using the Ziehl-Neelsen stain, the
Kinyoun method which uses light microscopy to view bacterial cells stained
red by carbol fuchsin and the use of auramine O.

Mycobacteria can be identified using high-pressure liquid chromatography
or gas-liquid chromatography (Butler et al. 9 1992; Glickman et al. 9 1994). In
these methods mycolic acids, specific to mycobacterial cells, can be extracted,
methylated and then analysed. Molecular techniques have been applied for
the detection of mycobacteria. These have used target or non-targeted areas of

165

Bacteriology

DNA, such as pulse-field gel electrophoresis and restriction fragment length
polymorphism. rRNA gene sequences can be used to identify members of the
MAC complex (Saito et al., 1990) including M. paratuberculosis (Thoresen
and Saxegaard, 1991). However, to distinguish between MAC and MAP a
DNA probe based on IS900 can be used. Commercially available probes,
AccuProbe and Gen-Probe, are used for the identification of mycobacterial
species.

Monoclonal antibodies against a phosphoglycolipid found in the cell wall
of MAC bacteria have been used to produce a latex agglutination test (Olano
et al., 1998). MAC organisms can be identified using DNA sequence analysis
of 16S rRNA (Frothingham and Wilson, 1994). There are three genetic
differences between MAC and MAP. The IS900 DNA insertion element,
specific for M. paratuberculosis, is present in multiple copies, 14 to 18 per
bacterium, allowing PCR methods to be used to identify positively MAP
(Green et al., 1989). M. paratuberculosis has a single copy of a genetic ele-
ment, which has been termed C GS' (Tizard et al., 1998). A gene called hspX
is located within M. paratuberculosis in a third genomic region (Ellingson
etal, 1998).

The differentiation of Mycobacterium species is best achieved by using a
two-step assay of gene amplification and restriction fragment length poly-
morphism (Plikaytis et al, 1992; Roth et al., 2000).

Epidemiology

There have been a number of waterborne outbreaks due to atypical
Mycobacteria but many of these have been referred to as pseudo-outbreaks
(Sniadack et al., 1993; Bennett et al., 1994; Wallace et al., 1998; Lalande
et al., 2001). From these studies the source of contamination by Mycobacterium
was in the laboratory.

The first true outbreak of Mycobacterium associated with water was estab-
lished in a sternal wound infection. It was found that the wound infection was
due to M. fortuitum (Kuritsky et al., 1983). From this study it was found that
the same strain was isolated from both clinical samples and a number of water
samples taken from the hospital environment. A similar route of infection was
established during a study in India. From this study Chadha and colleagues
(1998) reported an outbreak of post-surgical wound infections due to
M. abscessus. The organism responsible for this infection was linked to con-
taminated tap water.

MAC infections are associated predominately with HIV disease. From a
study by Von Reyn et al. (1994) four types of Mycobacterium were identified.
Two types of Mycobacterium were simultaneously isolated from patients
and their respective hospital but not from patients' homes. These findings
provided circumstantial evidence that infection may be related to hospital
water supplies.

166

The Mycobacterium avium complex

Many epidemiological studies have found that potable water is not a risk factor
for Mycobacterium acquisition. However, the strongest evidence of a transmission
route of Mycobacterium via water relates specifically to immunocompromised
individuals. This seems to be particularly relevant in the hospital environment.

Risk assessment

Health effects: occurrence of illness, degree of morbidity and mortality, prob-
ability of illness based on infection:

• MAC generally causes three different types of diseases in humans: (1) pul-
monary disease, (2) cervical lymphadenitis and (3) disseminated MAC,
both AIDS and non-AIDS related. Data suggest that the incidence of the
first two types of MAC disease continue to be increasing, and disseminated
MAC disease is one of the most common opportunistic infections to strike
AIDS patients.

• MAC is generally an opportunistic infection; however, the incidence of pul-
monary MAC in people (especially elderly women) without risk factors and
lymphadenitis in children has been increasing. Data from the late 1970s
and early 1980s showed an annual rate of 1.3 cases/100 000 (O'Brien et al.,
1987). In AIDS patients, disseminated MAC occurs in 20-40% of the pop-
ulation (Horsburgh, 1991), though that number has decreased with new
AIDS drug therapies.

• Pulmonary MAC can cause significant pulmonary and systemic symptoms;
disseminated MAC commonly causes fever, anaemia, night sweats and
weight loss. Both can decrease expected lifespan in patients, but the mor-
tality rate is unknown.

Exposure assessment: routes of exposure and transmission, occurrence in
source water, environmental fate:

• Routes of exposure are ingestion through the gastrointestinal tract and the
inhalation of aerosols. Some believe the latter is responsible for more trans-
mission than the former.

• The infectious dose of MAC organisms in humans is unknown.

• Mycobacteria are common in the environment and have been isolated from
all natural water systems world-wide. They are very hardy in the environ-
ment and have the ability to survive temperatures below 0°C and thrive in
temperatures of 52-5 7°C.

• Many studies have implicated water as the transmission route for MAC
infection. However, MAC is ubiquitous throughout the environment, and
infection undoubtedly occurs through other routes of exposure, such as
food. No evidence supports any person-to-person transmission.

• MAC's colonization in biofilm of water distribution systems is a major
problem and has been associated with human infection. The organism is

167

Bacteriology

thermophilic and MAC has been frequently isolated from showerheads and
recirculating hot water systems in institutions. They are more frequently
found colonizing hot water systems rather than cold.

Risk mitigation: drinking-water treatment, medical treatment:

• Chlorine is an ineffective disinfectant for mycobacteria, especially at residual
levels. Physical treatment, such as sedimentation, coagulation and sand
filtration is effective at removing most, but not all, of the organisms.

• Pulmonary MAC and disseminated MAC are difficult to treat and are resist-
ant to antimicrobials; symptomatic infection can be highly problematic and
result in lengthy and complicated chemotherapy. Relapse is relatively com-
mon. The incidence of disseminated MAC infections has decreased dramat-
ically due to antiretroviral treatment protocols for AIDS patients.

• Untreated, pulmonary MAC can cause numerous respiratory and systemic
symptoms. Available treatment is difficult and sometimes ineffective and
can last for months or years. Permanent lung damage depends on the extent
of infection when the patient presents for treatment. AIDS patients who
develop disseminated MAC are usually in the end-stage of their disease.

References

Aronson, T., Holtzman, A., Glover, N. et al. (1999). Comparison of large restriction frag-
ments of Mycobacterium avium isolates recovered from AIDS and non-AIDS patients
with those of isolates from potable water./ Clin Microbiol, 37(4): 1008-1012.

Argueta, C, Yoder, S., Holtzman, A.E. et al. (2000). Isolation and identification of non-
tuberculous mycobacteria from foods as possible exposure sources./ Food Prot, 63(7):
930-933.

Barnes, N., Clarke, C, Finlayson, C. et al. (1998). Mycobacterium paratuberculosis
cervical lymphadenitis followed five years later by terminal ileitis similar to Crohn's
disease. Br Med], 316: 449-453.

Benjamin, W.H. Jr, Waites, K.B. and Beverley, A. (1998). Comparison of the MB/BacT sys-
tem with a revised antibiotic supplement kit to the BACTEC 460 system for detection
of mycobacteria in clinical specimens./ Clin Microbiol, 36(11): 3234-3238.

Bennett, S.N., Peterson, D.E., Johnson, D.R. et al. (1994). Bronchoscopy-associated
Mycobacterium xenopi pseudoinfections. Am J Resp Crit Care Med, 150: 245-250.

Brunello, E, Favari, E and Fontana, R. (1999). Comparison of the MB/BacT and BACTEC
460 TB systems for recovery of mycobacteria from various clinical specimens. / Clin
Microbiol, 37(4): 1206-1209.

Butler, W.R., Thibert, L. and Kilburn, J.O. (1992). Identification of Mycobacterium avium
complex strains and some similar species by high-performance liquid chromatography.
/ Clin Microbiol, 30: 2698-2704.

Chadha, R., Grover, M., Sharma, A. et al. (1998). An outbreak of post-surgical wound
infections due to Mycobacterium abscessus. Pediatr Surg Internatl, 13: 406-410.

Chin, D.P., Hopewell, P.C., Yajko, D.M. et al. (1994). Mycobacterium avium complex in
the respiratory or gastrointestinal tract and the risk of M. avium bacteremia in patients
with the human immunodeficiency virus. / Infect Dis, 169: 289-295.

Chiodini, R.J. (1989). Crohn's disease and the mycobacterioses: a review and comparison
of two disease entities. Clin Microbiol Rev, 2: 90-117.

Chiodini, R.J. (1996). M. paratuberculosis in Foods and the Public Health Implications. In
Proceedings of the Fifth International Colloquium on Paratuberculosis, Chiodini, R.K.,

168

The Mycobacterium avium complex

Hines, M.E. and Collins, M.T. (eds). Madison, WI: International Association for
Paratuberculosis, pp. 353-365.

Cirillo, J.D. (1999). Exploring a novel perspective on pathogenic relationships. Trend
Microbiol, 7: 96-98.

Cirillo, J.D. , Falkow, S., Tompkins, L.S. etal. (1997). Interaction of Mycobacterium avium
with environmental amoebae enhances virulence. Infect Immun, 65(9): 3759-3767.

Covert, T.C., Rodgers, M.R., Reyes, A.L. et al. (1999). Occurrence of nontuberculous
mycobacteria in environmental samples. Appl Environ Microbiol, 65(6): 2492-2496.

Dalziel, T.K. (1913). Chronic interstitial enteritis. Br Med], 3(2): 1068-1070.

Ellingson, J.L., Bolin, C.A. and Stabel, J.R. (1998). Identification of a gene unique to
Mycobacterium avium subspecies paratuberculosis and application to diagnosis of
paratuberculosis. Mol Cell Probes, 12: 133-142.

Falkinham, J.O. III. (1996). Epidemiology of infection by nontuberculosis mycobacteria.
Clin Microbiol Rev, 9: 177-215.

Falkinham, J.O. Ill, Norton, CD. and Le Chevallier, M.W. (2001). Factors influencing
numbers of Mycobacterium avium, Mycobacterium intracellular e and other myco-
bacteria in drinking water distribution systems. Appl Environ Microbiol, 67(3):
1225-1231.

Frothingham, R. and Wilson, K.H. (1994). Molecular phylogeny of the Mycobacterium
avium complex demonstrates clinically meaningful divisions./ Infect Dis, 169: 305-312.

Glickman, S.E., Kilburn, J.O., Butler, W.R. et al. (1994). Rapid identification of mycolic
acid patterns of mycobacteria by liquid chromatography using pattern recognition soft-
ware and a Mycobacterium library./ Clin Microbiol, 32: 740-745.

Goodfellow, M. and Wayne, L.G. (1982). Taxonomy and nomenclature. In The Biology of
the Mycobacteria, vol. 1, Ratledge, C. and Stanford, J. (eds). London: Academic Press,
pp. 471-521.

Grange, J.M., Yates, M.D. and Boughton, E. (1990). Review: the avian tubercle bacillus
and its relatives. / Appl Bacteriol, 68: 411-431.

Grant, I.R., Ball, H.J. and Rowe, M.T. (1998). Effect of high-temperature, short-time
(HTST) pasteurisation on milk containing low numbers of Mycobacterium paratuber-
culosis. Lett Appl Microbiol, 26: 166-170.

Green, E.P., Tizard, M.L.V., Moss, M.T. et al. (1989). Sequence and characteristics
of IS900, an insertion element identified in a human Crohn's disease isolate of
Mycobacterium paratuberculosis. Nuc Acid Res, 17: 9063-9073.

Hansen, G.A. (1874). Undersogelser angaende spendalskhedens arsager. Norsk Magazin
for Laegevidenskaben, 4: 1-88.

Harris, G., Rayner, A., Blair, J. et al. (2000). Comparison of three isolation systems for the
culture of mycobacteria from respiratory and non-respiratory samples. / Clin Pathol,
53:615-618.

Hermon-Taylor, J. (1993). Causation of Crohn's disease: the impact of clusters.
Gastroenterology, 104: 643-646.

Hermon-Taylor, J. (1998). The causation of Crohn's disease and treatment with antimicro-
bial drugs. Ital J Gastroenterol Hepatol, 30(6): 607-610.

Hermon-Taylor, J., Tizard, J., Sanderson, J. et al. (1994). Mycobacteria and the aetiology
of Crohn's disease. Inflammatory Bowel Dis, http://iol.ie/alank/CROHNS/paratub.htm.

Horsburgh, C.R. Jr. (1991). Mycobacterium avium complex infection in the acquired
immunodeficiency syndrome. New Engl J Med, 324: 1332-1338.

Iivanainen, E. (1995). Isolation of mycobacteria from acidic forest soil samples: compari-
son of culture methods. / Appl Bacteriol, 78: 663-66S.

Iivanainen, E., Martikainen, P.J., Vaananen, P.K. et al. (1993). Environmental factors
affecting the occurrence of mycobacteria in brook waters. Appl Environ Microbiol,
32: 398-404.

Iivanainen, E., Martikainen, P.J. and Katila, M.L. (1995). Effect of freezing of water sam-
ples on viable counts of environmental mycobacteria. Lett Appl Microbiol, 21: 257-260.

Iivanainen, E., Martikainen, P.J., Vaananen, P.K. et al. (1999). Environmental factors
affecting the occurrence of mycobacteria in brook sediments. / Appl Microbiol,
86(4): 673-681.

169

Bacteriology

Inderlied, C.B., Kemper, C.A. and Bermudez, L.E.M. (1993). The Mycobacterium avium

complex. Clin Microbiol Rev ■, 6(3): 266-310.
Jenkins, P. A., Pattyn, S.R. and Portaels, F. (1982). Diagnostic bacteriology. In The Biology

of the Mycobacteria, vol. 1, Ratledge, C. and Stanford, J. (eds). London: Academic

Press, pp. 441-470.
Johne, H.A. and Frothingham, L. (1895). Em eigenthuemlicher Fall von tuberkulose beim

Rind. Dtscb Ztschr Tiermed Path, 21: 438-454.
Kamala, T., Paramasivan, C.N., Herbert, D. et al. (1994). Evaluation of procedures for

isolation of nontuberculous mycobacteria from soil and water. Appl Environ Microbiol,

60(3): 1021-1024.
Koch, R. (1882). Die aetilogie der tuberculose. Berl Klin Wochensch, 19: 221-230.
Kuritsky, J.N., Bullen, M.G., Broome, C.V. et al. (1983). Sternal wound infections and

endocarditis due to organisms of the Mycobacterium fortuitum complex. Ann Intern

Med, 98: 938-939.
Lalande, V., Barbut, F., Varnerot, A. et al. (2001). Pseudo-outbreak of Mycobacterium gor-

donae associated with water from refrigerated fountains./ Hosp Infect, 48: 76-79.
Laverdiere, M., Poirier, L., Weiss, K. et al. (2000). Comparative evaluation of the MB/BacT

and BACTEC 460 TB systems for the detection of mycobacteria from clinical speci-
mens: clinical relevance of higher recovery rates from broth-based detection systems.

Diag Microbiol Infect Dis, 36: 1-5.
Mediel, M.J., Rodriguez, V., Codina, G. et al. (2000). Isolation of Mycobacteria from

frozen fish destined for human consumption. Appl Environ Microbiol, 43: 3637-3638.
Millar, D., Ford, J., Sanderson, J.D. et al. (1996). IS900 PCR to detect Mycobacterium

paratuberculosis in retail supplies of whole pasteurised cows' milk in England and

Wales. Appl Environ Microbiol, 62: 3446-3452.
Miltner, E.C. and Bermudez, L.E. (2000). Mycobacterium avium grown in Acanthamoeba

castellanii is protected from the effects of antimicrobials. Antimicrob Agents

Chemother, 44(7): 1990-1994.
Miyamoto, M., Yamaguchi, Y. and Sasatsu, M. (2000). Disinfectant effects of hot water,

ultraviolet light, silver ions and chlorine on strains of Legionella and nontuberculous

mycobacteria. Microbios, 101(398): 7-13.
Neumann, M., Schulze-Robbecke, R., Hagenau, C. et al. (1997). Comparison of methods

for isolation of mycobacteria from water. Appl Environ Microbiol, 63: 547-552.
O'Brien, R.J., Geiter, L.J. and Snider, D.E. Jr. (1987). The epidemiology of nontuberculous

mycobacterial diseases in the United States. Results from a national survey. Am Rev

RespirDis, 135: 1007-1014.
Olano, J.P., Holmes, H. and Woods, G.L. (1998). Evaluation of the MycoAKT Latex

Agglutination test for rapid diagnosis of Mycobacterium avium complex infection. Diag

Microbiol Infect Dis, 30: 71-74.
Palacios, J.J., Ferro, J., Ruiz Palma, N. et al. (1999). Fully automated liquid culture system

compared with Lowenstein-Jensen solid medium for rapid recovery of mycobacteria

from clinical samples. Eur J Clin Microbiol Infect Dis, 18: 265-273.
Pelletier, P.A., Du Moulin, G.C. and Stottmeier, K.D. (1988). Mycobacteria in public water

supplies: comparative resistance to chlorine. Microbiol Sci, 5(5): 147-148.
Peters, M., Muller, C, Rusch-Gerdes, S. et al. (1995). Isolation of atypical mycobacteria

from tap water in hospitals and homes: is this a possible source of disseminated MAC

infection in AIDS patients? / Infect, 31(1): 39-44.
Petroff, S.A. (1915). A new and rapid method for the isolation and cultivation of tubercle

bacilli directly from the sputum and feces./ Exp Med, 21: 38-42.
Plikaytis, B.B., Plikaytis, B.D., Yakrus, M.A. et al. (1992). Differentiation of slowly

growing Mycobacterium species, including Mycobacterium tuberculosis, by gene

amplification and restriction fragment length polymorphism analysis./ Clin Microbiol,

30(7): 1815-1822.
Portaels, F. and Pattyn, S.R. (1982). Growth of mycobacteria in relation to the pH of the

medium. Ann Microbiol (Paris), 133B: 213-221.
Portaels, F., De Muynck, A. and Sylla, M.P. (1988). Selective isolation of mycobacteria

from soil: a statistical analysis approach./ Gen Microbiol, 134: 849-855.

170

The Mycobacterium avium complex

Roggenkemp, A., Hornef, M.W., Masch, A. et al. (1999). Comparison of MB/BacT and
BACTEC 460 TB systems for recovery of mycobacteria in a routine diagnostic labora-
tory. / Clin Microbiol, 37(11): 3711-3712.

Rohner, P., Ninet, B., Metral, C. et al. (1997). Evaluation of the MB/BacT system and com-
parison to the Bactec 460 system and solid media for isolation of mycobacteria from
clinical specimens./ Clin Microbiol, 35(12): 3127-3131.

Roth, A., Reischl, U., Streubel, A. et al. (2000). Novel diagnostic algorithm for identifica-
tion of mycobacteria using genus-specific amplification of the 16S-23S rRNA gene
spacer and restriction endonucleases./ Clin Microbiol, 38(3): 1094-1104.

Saito, H., Tomioka, H., Sato, K. et al. (1990). Identification of various serovar strains of
Mycobacterium avium complex by using DNA probes specific for Mycobacterium
avium and Mycobacterium intr a cellular e. J Clin Microbiol, 28: 1694-1697.

Schulze-Robbecke, R. (1993). Mycobacteria in the environment. Immun Infekt,
21(5): 126-131.

Schulze-Robbecke, R. and Buchholtz, K. (1992). Heat susceptibility of aquatic mycobacte-
ria. Appl Environ Microbiol, 58(6): 1869-1873.

Schulze-Robbecke, R, Janning, B. and Fischeder, R. (1992). Occurrence of mycobacteria in
biofilm samples. Tuber Lung Dis, 73(3): 141-144.

Sniadack, D.H., Ostroff, S.M., Karlix, M.A. et al. (1993). A nosocomial pseudo-outbreak
of Mycobacterium xenopi due to a contaminated potable water supply: lessons in pre-
vention. Infect Control Hosp Epidemiol, 14: 636-641.

Thompson, D.E. (1994). The role of Mycobacteria in Crohn's disease. / Me d Microbiol,
41: 74-94.

Thorel, M.F., Krichevsky, M. and Levy-Frebault, V.V. (1990). Numerical taxonomy of
mycobactin-dependent mycobacteria, emended description of Mycobacterium avium,
and description of Mycobacterium avium subsp. avium subsp. Nov., Mycobacterium
avium subsp. paratuberculosis subsp. Nov., and Mycobacterium avium subsp. sil-
vaticum subsp. Nov. Int J Syst Bacteriol, 40: 254-260.

Thoresen, O.F. and Saxegaard, F. (1991). Gen-Probe rapid diagnostic system for the
Mycobacterium avium complex does not distinguish between Mycobacterium avium
and Mycobacterium paratuberculosis. J Clin Microbiol, 29: 625-626.

Tizard, M.L.V., Bull, T., Millar, D. et al. (1998). A low G + C content genetic island in
Mycobacterium avium subsp. paratuberculosis and M. avium subsp. silvaticum with
homologous genes in Mycobacterium tuberculosis. Microbiology, 144: 3413-3423.

Twort, F.W. (1911). A method for isolating and growing the lepra bacillus of man and the
bacillus of Johne's disease in cattle. Vet], 3(67): 118-120.

Twort, F.W. and Ingram, G.L.Y. (1912). A method for isolating and cultivating the
Mycobacterium enteritidis chronicae pseudotuberculosae bovis, Johne and some experi-
ments on the preparation of a diagnostic vaccine for pseudotuberculous enteritis of
bovines. Vet J, 68: 353-365.

Von Rehn, C.F., Maslow, J.N., Barber, T.W. et al. (1994). Persistent colonisation of potable
water as a source of Mycobacterium avium infection in AIDS. Lancet, 343: 1137-1141.

Wallace, R.J. (1987). Nontuberculous mycobacteria and water: a love affair with increas-
ing clinical importance. Infect Dis Clin North Am, 1(3): 677-686.

Wallace, R.J. Jr, Brown, B.A. and Griffith, D.E. (1998). Nosocomial outbreaks/pseudo-out-
breaks caused by nontuberculous mycobacteria. Ann Rev Microbiol, 52: 453-490.

Wolinsky, E. (1992). Mycobacterial diseases other than tuberculosis. Clin Infect Dis,
15: 1-10.

Woods, G.L., Fish, G., Plaunt, M. et al. (1997). Clinical evaluation of Difco ESP culture
system II for growth and detection of mycobacteria./ Clin Microbiol, 35: 121-124.

171

Salmonella

Basic microbiology

Salmonella, as a group, are facultatively anaerobic, non-spore-forming, Gram-
negative bacilli. On average they are 2-5 |xm long and 0.8-1.5 fjim wide.
Motility, aided by peritrichous flagella, is a fundamental criterion for the identi-
fication of Salmonella. However, a large number of non-motile strains have been
isolated from the clinical environments. Somatic and flagellar antigens closely
relate salmonella serotypes to each other, with many strains showing diphasic
variation. A large number of Salmonella also express type 1 fimbriae.

Salmonella constitute a very pathogenic collection of species that cause
infections to a large diverse collection of animals. Salmonella predominantly
give rise to enteritis and to typhoid-like diseases.

Origin

The group Salmonella is now considered to comprise two species, namely
S. enterica and Salmonella bongori. There are six subspecies of S. enterica, the
most important of which is S. enterica subsp. enterica (subspecies I) that

Bacteriology

includes the typhoid and paratyphoid bacilli. Members of the other five sub-
species (II-VI), found in the natural environment, are primarily parasites of
cold-blooded animals.

Salmonella possesses two sets of antigens; a flagellar or H antigen and a
heat stable polysaccharide known as the somatic or O antigen. Some
Salmonella produce a surface polysaccharide, one example is Salmonella
typhi where the Vi antigen is very important in its identification. The antigenic
structure of any salmonella is expressed as an antigenic formula that is made
up of three parts. These are the O antigens, the phase 1 H antigens and the
phase 2 H antigens.

Early identification of Salmonella species involved description of the disease
it caused or the host with which the serotype was associated, this of course led
to major problems. These problems were overcome by the introduction of a
new system where each new type of Salmonella was named after the place in
which it was first isolated. The first published table of Salmonella serotypes
contained some 20 entires. This was increased in 1995 to 2399 (Popoff et al.,
1995).

With the introduction of more modern taxonomic techniques, Le Minor et al.
(1982a, b) have suggested that all serotypes of Salmonella probably belonged
to one DNA-hybridization group within which seven subgroups were identi-
fied. The seven DNA-hybridization subgroups were designated as subspecies:
S. enterica subspp. enterica, salamae, arizonae, diarizonae, houtenae, bongori
and indica. However, the DNA-DNA hybridization studies of Le Minor
and colleagues (Le Minor etal., 1982a, 1986) suggested that DNA subgroup V
(S. enterica subsp. bongori) had evolved significantly from the other six sub-
species. Results from multilocus enzyme electrophoresis (MLEE) studies sup-
ported that observation and the elevation of S. enterica subsp. bongori to the
level of species as S. bongori was proposed (Reeves et al., 1989).

Metabolism and physiology

Salmonellae are facultative anaerobes. They are catalase-positive, oxidase-
negative and ferment glucose and mannitol to produce acid or acid and gas.
This also seems true for sorbitol. While S. arizonae is able to ferment lactose,
this is the exception rather than the rule. As a group, Salmonella are able to
ferment sucrose but rarely adonitol and overall do not form indole. They also
do not hydrolyse urea or deaminate phenylalanine, but usually form H 2 S
on triple sugar iron agar and use citrate as sole carbon source. They form
lysine and ornithine decarboxylases, exceptions to this include S. paratyphi A
and S. typhi. Salmonellae yield negative Voges-Proskauer and positive methyl
red tests and do not produce cytochrome oxide. Salmonellae are not able to
deaminate tryptophan or phenylalanine and are usually urease and indole
negative.

174

Salmonella

However, for a more detailed identification of Salmonella, isolates are
generally serotyped, especially for epidemiological investigations. As mentioned
previously, typing of Salmonella is based on the recognition of bacterial surface
antigens - the thermostable polysaccharide cell wall or somatic ('O') antigens
and the thermolabile flagella proteins or C H' antigens. It is also possible to
subtype Salmonella serotypes on the basis of phage typing. These subdivisions
of Salmonella are made possible by plasmid profiling, ribotyping or pulsed
field gel electrophoresis (PFGE) of DNA fragments.

Clinical features

There are three clinically distinguishable forms of salmonellosis that occur in
humans. These include gastroenteritis, enteric fever and septicaemia. Gastro-
enteritis is an infection of the colon which usually occurs 18-48 hours after
ingestion of Salmonella. Gastroenteritis is characterized by diarrhoea, fever
and abdominal pain. The infection is usually self-limiting, lasting 2-5 days.

Enteric fever is most often caused by S. typhi (typhoid fever) and the para-
typhoid bacilli, S. paratyphi A, B, and C. Enteric fever from S. typhi is more
prolonged and has a higher mortality rate than paratyphoid fever. Symptoms
include sustained fever, diarrhoea, abdominal pain and may involve fatal liver,
spleen, respiratory and neurological damage. Typhoid fever symptoms persist
for 2-3 weeks. Enteric fevers from other than S. typhi have a shorter incubation
period, 1-10 days, compared to 7-14 days for typhoid fever, and the symptoms
are less severe.

Chills, high remittent fever, anorexia and bacteraemia characterize
Salmonella septicaemia. Organisms may localize in any organ in the body and
produce focal lesions resulting in meningitis, endocarditis, pneumonia or
osteomyelitis. The incubation period for human salmonellosis is usually
between 12 and 36 hours, but where people have consumed a large number of
cells, incubation periods of 6 hours or less have been recorded. However, it
has been documented in some Salmonella outbreaks that 12 days have elapsed
between consumption of contaminated material and clinical illness.

Pathogenicity and virulence

Most serotypes of Salmonella are pathogenic in mammals and birds and
generally belong to subsp. 1. When infected with Salmonella the disease may
present in three ways. First, a few host-adapted serotypes cause systemic dis-
ease in their hosts. When the manifestations of systemic disease are mainly sep-
ticaemic, as is usually the case with typhoid and paratyphoid bacilli in
humans, the clinical picture is one of enteric fever with an incubation period of

175

Bacteriology

10-20 days, but with outside limits of 3 and 56 days depending on the infecting
dose (Mandal, 1979). Diarrhoea, starting 3-4 days after onset of fever and lasting,
on average, 6 days, may occur in 50% of cases of typhoid fever and is more
common in younger, than in older children or adults in whom intestinal symp-
toms may be absent or insignificant (Roy et aL, 1985). Second, certain other
serotypes - Blegdam, Bredeney, Choleraesuis, Dublin, Enteritidis, Panama and
Virchow in humans and Gallinarum in adult fowl - are also invasive but tend
to cause pyaemic infections and to localize in the viscera, meninges, bones,
joints and serous cavities. Third, most other salmonellae, the ubiquitous
serotypes found in a number of animal species, tend to cause an acute, but
mild, enteritis with a short incubation period of 12-48 h, occasionally as long
as 4 days.

Local abscesses and even pyaemia may develop occasionally as a late compli-
cation of a diarrhoeal disease. Serotypes that usually cause enteritis in healthy
adults may cause septicaemia or pyaemic infections in young or elderly patients.
The incidence of laboratory-confirmed salmonellosis in AIDS patients is 20-100
times that in the general population (Angulo and Swerdlow, 1995). Recurrent
non-typhoidal salmonella septicaemia (RSS) has been included as an AIDS-
defining illness since 1987. Typhimurium, generally associated with enteritis in
humans, may give rise to more severe disease in other hosts. Again, the typhoid
bacillus, often considered to cause mild and atypical infections in children, will
give rise to severe infections with high mortality in children whose previous
health and nutrition have been poor (Scragg et al. 9 1969).

Convalescent patients may continue to excrete salmonellae, usually in the
faeces or urine but also from other sites after pyaemic infections, even after
treatment. Asymptomatic persons may also harbour salmonellae unknow-
ingly. Both remain potentially infectious for weeks or months and some
become life-long excreters.

The factors responsible for the virulence of salmonellae are still being
researched and are far from being fully understood - progress remains slow.
Because salmonellae are able to invade, survive and replicate within eukaryotic
cells this forms an important virulence mechanism for successful infection.
Aside from this, salmonellae possess several virulence factors that contribute to
the disease process; these virulence-promoting factors are generally conserved
among species, as are the mechanisms by which bacteria interact with the host
cell. Such factors facilitate entry into (invasion of) non-phagocytic cells, survival
in the intracellular environment and replication within host cells.

After oral ingestion, provided a sufficient number of Salmonella cells have
been injected, disease is initiated by the adhesion of Salmonella onto the epithe-
lial lining of the intestines. The next stage in the disease process is penetration of
the intestinal epithelium. It is well documented that invasive salmonellae inter-
act with the well-defined apical microvilli of epithelial cells and disrupt the
brush border. Once these cells have been penetrated, the bacteria are enclosed in
membrane-bound vacuoles within the host cytoplasm. Virulent salmonellae
survive and multiply within these vacuoles, eventually lysing their host cells and
disseminating to other parts of the body. Salmonellae also depolarize epithelial

176

Salmonella

barriers by affecting the integrity of tight junctions between cells, which has the
result of promoting significant cytotoxic damage to epithelial cells.

Induction of protein synthesis de novo is necessary for invasion and is regu-
lated by the microenvironment (low oxygen), growth phase or the epithelial
cell surface. Only viable, metabolically active salmonellae can adhere to host
cell surfaces. Adherence is followed almost immediately by internalization into
host cells. The bacterial genes necessary for salmonellae to enter eukaryotic
cells have yet to be fully characterized (Finlay et al. 9 1992; Guiney et aL, 1995).

Internalization of salmonellae through the apical surface of epithelial cells
is associated with disruption of the eukaryotic brush border. After a lag period
of about 4 hours, invading organisms begin to multiply within the vacuoles of
host epithelial cells. For Typhimurium, it has been demonstrated that this
intracellular replication is essential for pathogenicity. Salmonellae remain
within membrane-bound inclusions and their survival appears to involve
blockage of phagosome-lysosome fusion events.

Treatment

The treatment for Salmonella can be divided into three areas namely: enteric
fever, gastroenteritis and salmonella bacteraemia. In enteric fever many patients
are treated with oral chloramphenicol. With very ill patients intravenous treat-
ment may be necessary. With the use of chloramphenicol comes the problem of
bone marrow toxicity and the emergence of plasmid-mediated resistance.
Because of this, the drug of choice is ciprofloxacillin for adult typhoid. With
gastroenteritis, the management requires fluid and electrolyte replacement and
control of nausea, vomiting and pain. The role of antibiotics is limited but
is necessary where a patient has an increased risk of bacteraemia and salmonella
invasion. This is particularly important in children under 3 months of age,
ulcerative colitis and patients who are immunocompromised. Patients with sal-
monella bacteriaemia also require antimicrobial treatment with chlorampheni-
col, co-trimoxazole, and high doses of ampicillin or ciprofloxacin.

Survival in the environment

Reservoirs of Salmonella are domestic and wild animals, including poultry,
swine, cattle, birds, dogs, rodents, tortoises, turtles and cats. Humans also serve
as a reservoir, e.g. convalescent carriers and those with asymptomatic infections.
The occurrence of chronic carriers is rare in humans, but is common in birds
and animals. Infection from Salmonella occurs through ingestion of food, milk
or water contaminated with faeces from infected hosts or by ingestion of the
infected meat products. S. typhi and S. paratyphi are not widely distributed in

177

Bacteriology

nature with colonization only occurring in humans. Human infection with
these organisms indicates exposure to human faeces. Contrary to this, non-
typhoidal Salmonella spp. are widely distributed in nature and closely associ-
ated with animals.

The most common source of transmission of Salmonella seems to be the
consumption of contaminated poultry and meat products. Contamination of
meat products occurs particularly when exposed to faecal matter during
slaughter. Once the meat is contaminated, improper storage or undercooking
allows Salmonella to proliferate. When Salmonella are excreted in faeces, con-
tamination of food and water permits transmission of the infection to
humans. Person-to person, faecal-oral transmission does occur and has been a
problem in health care facilities traced to inadequate hand washing.

Studies undertaken to investigate the survival of enteric pathogens in aqua-
tic environments have shown that Salmonella may enter a Viable but non-
culturable' physiological state. The vast majority of detection and enumeration
techniques require culturing Salmonella using selective media. As a result if one
just relies on culturable techniques for the detection of viable organisms a gross
underestimation of the true extent of Salmonella viability in the environment
may be realized. Salmonellae are frequently isolated in polluted waters and can
persist in high-nutrient waters. Wastewater treatment reduces but does not
completely eliminate Salmonella. Reported Salmonella levels in non-chlorinated
wastewater effluents range from 1 to 1100/cells per ml.

With the exception of S. typhi, water is not often implicated as the vehicle in
outbreaks of human infection. Nevertheless, faecal contamination of ground-
water or surface waters, and insufficient treatment or inadequate disinfection
of drinking water have been identified as causes of waterborne outbreaks
of salmonellosis. However, food is a much more important source of exposure
to Salmonella infection. It seems though that contaminated water is important
indirectly when it is used in food processing or preparation.

Natural water courses are important in the transmission of infection between
herds of food animals and therefore play a role in zoonotic transmission.

Survival in water

Salmonellae are associated with faecal pollution, and may be found in any water
due to contamination. They have been isolated from sewage-polluted surface
waters (fresh, estuarine and marine) and groundwater. A significant factor in
some outbreaks has been contamination of well water (subsequently consumed
untreated) by sewage, runoff from agricultural land, leaking domestic drains,
and seepage from septic tanks. Most Salmonella serotypes are capable of pro-
longed survival in water and may be able to grow in heavily polluted water in
the warmer months. Standard disinfection used in drinking water treatment
procedures, in light of the documented evidence, are active against salmonellae,
although there is some evidence that they are more resistant to inactivation than
coliforms. However, absence of coliforms and E. coli from treated drinking

178

Salmonella

water should not be relied upon as adequate assurance that salmonellae will be
absent. Regrowth, in the form of a biofllm, of salmonellae in potable water dis-
tribution systems is thought to be possible. If Salmonella are evident in drinking
water biofilms, it is probable they may survive for long periods due to protec-
tion from disinfection within the biofllm matrix.

In drinking water there are no permissible levels for Salmonella. This is pri-
marily due to the problem that so many waterborne pathogens are difficult to
detect or enumerate in water and the methods are often impractical to use for
routine monitoring. In general, because of these problems, both the US
Environmental Protection Agency (US EPA) and World Health Organization
have adopted total coliform, faecal coliform and Escherichia coli microbio-
logical standards for drinking water. It is hoped that these organisms will indi-
cate the efficacy of a water-treatment plant, distribution system integrity, and
recent faecal contamination. However, the viable but non-culturable concept
needs addressing more extensively.

Methods of detection

Salmonellae are facultative anaerobes and grow readily on ordinary media.
Salmonellae are known to grow over a wide temperature and pH range,
7-48°C and pH 4-8, respectively (Baird-Parker, 1991). Most salmonellae are
prototrophic growing readily on minimal salts medium incorporating an
appropriate carbon source. Auxotrophic strains of prototrophic serotypes are
quite common, e.g. around 7% of Typhimurium strains (Duguid et al., 1975).
These auxotrophic serotypes will grow on a minimal medium supplemented
with appropriate growth f actor (s). On general laboratory agar some salmonel-
lae, particularly Paratyphi B, produce mucoid colonies. These have been shown
to develop at low temperature, low humidity and high osmolarity (Anderson
and Rogers, 1963). It is well documented that the presence of a mucoid surface
layer on the cell surface, the M antigen of Kauffmann, inhibits O and H agglu-
tinability. Because the M antigen is identical in all salmonella serotypes and is
like the colanic acid materials of other Enterobacteriaceae, it is unimportant
diagnostically.

If Salmonella is being isolated from mixed bacterial samples (i.e. faeces, food
and environmental), selective enrichment is needed. Within the diagnostic labora-
tory three selective enrichment media, which are commonly used, are tetra-
thionate broth, selenite broth, and Rappaport-Vassiliadis (RV) medium. Colonies
of Salmonella forming on this agar are generally circular with a smooth surface
and an even edge, or flat with an uneven surface and serrated edge. While the
temperature required for the optimal growth of Salmonella is 37°C, they
are able to grow at temperatures ranging from 10 to 43°C.

After primary isolation on selective media, presumptive Salmonella isolates
can be tested with commercial identification systems or screened with triple
sugar iron agar, urea broth, and lysine iron agar. Any isolate that has a

179

Bacteriology

biochemical pathway for Salmonella should then be tested with commercial
polyvalent O group, H, and Vi antisera.

With environmental samples, a large sample volume usually should be exam-
ined (1 litre or more). Concentration of the organisms can be accomplished using
Moore swabs, membrane filtration, diatomaceous earth, or large-volume sam-
plers. Although salmonellae can be recovered readily from water, sewage and
related environments they may be present in low numbers and may be sub-
lethally injured by exposure to such environments. It is normal practice for
water samples to be incubated in c pre-enrichment' media, such as buffered pep-
tone water (BPW) or lactose broth, for 18-24 hours at 37°C, although some
studies using naturally contaminated sewage samples have shown 43 °C to be
superior. Volumes of pre-enrichment media are added to a selective broth such
as Rappaport-Vassiliadis (RV), selenite or tetrathionate broths. Following select-
ive culture the broths are plated onto selective/diagnostic agars. As with the
liquid media, there are many agars commercially available. It is common prac-
tice for two media to be used and most laboratories will choose two from xylose
lysine deoxycholate (XLD), deoxycholate citrate (DCA) or brilliant green (BG)
agars. Newer plating media such as Rambach and XLT4 agars have been shown
to improve isolation rates of Salmonella.

Salmonella-like colonies on selective agars need to be confirmed by biochem-
ical testing, and serology and phage-typing where appropriate. It may take up to
5 days to complete sample examination. This delay has led to the investigation
of more rapid methods of Salmonella detection. These tend to be genome-based
and make use of techniques such as the polymerase chain reaction (PCR) where
specific sections of Salmonella DNA are targeted.

Epidemiology of waterborne outbreaks

With the exception of host-specific salmonellae, such as S. typhi in man and
S. gallinarum in poultry, salmonellosis is regarded as a zoonosis. Human illness
is usually the result of the consumption of contaminated foods or milk,
although contaminated drinking and recreational waters have been implicated.
Salmonella outbreaks have also occurred when contaminated water has been
used to cool cans of food after heat treatment. The number of bacteria required
to cause infection in humans will vary depending on the individuals at risk, their
age, general health and the presence of underlying disease. The vast majority of
waterborne outbreaks of salmonellosis are classified as acute gastrointestinal ill-
ness of unknown aetiology. Waterborne outbreaks usually involve poor-quality
source water, inadequate treatment, or contamination of the distribution system
(e.g. cross-connections) (Angulo et al., 1997). Large outbreaks of waterborne
salmonellosis have not been reported and would suggest that these organisms
are not a major problem as a waterborne pathogen in correctly treated water.
Despite this, salmonellosis represents a major communicable world-wide disease
problem. The annual occurrence of typhoid fever is estimated at 17 million

180

Salmonella

cases, with 600 000 deaths. In contrast to typhoid fever, with humans being the
sole source of the organism, animals and animal products are major sources of
other Salmonella. Large numbers of Salmonella may be present in contaminated
surface water and waste treatment plant influents and effluents. Studies have
shown that 80% of activated sludge effluents and 58% of contaminated surface
waters may contain Salmonella.

Risk assessment

Health effects: occurrence of illness, degree of morbidity and mortality, prob-
ability of illness based on infection:

• Three clinically distinguishable forms of salmonellosis occur in humans.
These include gastroenteritis, enteric fever and septicaemia.

• Salmonella enteritidis causes gastroenteritis, with fever, cramps and diar-
rhoea. This is the most common infection in developed countries.

• Enteric fever is most often caused by S. typhi (typhoid fever) and the
paratyphoid bacilli, S. paratyphi A, B and C. Enteric fever from S. typhi is
more prolonged and has a higher mortality rate than paratyphoid fever.
Symptoms include sustained fever, abdominal pain and may involve fatal
liver, spleen, respiratory and neurological damage.

• Death rates for typhoid fever range from 12 to 30%.

• Salmonella septicaemia is characterized by chills, high remittent fever,
anorexia and bacteraemia.

• Two per cent of cases result in chronic arthritis.

• According to the US Centers for Disease Control and Prevention, 1.4 mil-
lion cases of salmonellosis occur yearly in the USA.

• The number of bacteria required to cause infection in humans will vary
depending on the individuals at risk, their age, general health and the pre-
sence of underlying disease.

Exposure assessment: routes of exposure and transmission, occurrence in
source water, environmental fate:

• Salmonella spp. have been isolated and cultured from the soil, water,
wastes, plants and the normal flora of animals. At present there are some
2000 Salmonella serotypes that are human pathogens.

• Studies have shown that 80% of activated sludge effluents and 58% of con-
taminated surface waters may contain Salmonella.

• Most Salmonella serotypes are capable of prolonged survival in water and
may be able to grow in heavily polluted water in the warmer months.

• Regrowth of Salmonella in distribution systems is thought to be possible,
especially if levels of assimilable organic carbon are high.

181

Bacteriology

• Bacterial survival in aquatic ecosystems is affected by numerous factors,
including the presence of protozoa, antibiosis, organic matter, algal toxins,
dissolved nutrients, ultraviolet light, heavy metals and temperature.

• Many kinds of domestic and wild animals serve as reservoirs of Salmonella,
including poultry, swine, cattle and birds. Humans with asymptomatic
infections can also serve as reservoirs.

• Infection from Salmonella occurs through ingestion of food, milk or water
contaminated with faeces from infected hosts or by ingestion of the infected
meat and egg products or handling animals (e.g. pet turtles and lizards).
Consumption of contaminated poultry and meat products is the most
common source of transmission. A person-to-person transmission occurs,
but not as frequently.

• Faecal contamination of groundwater or surface waters, and insufficient
treatment of drinking water have been identified as causes of waterborne
outbreaks of salmonellosis. However, food is a much more important source
of exposure.

Risk mitigation: drinking-water treatment, medical treatment:

• Disinfection is generally effective on Salmonella. Chlorine residuals of at
least 0.2mg/l are sufficient to handle Salmonella in the distribution system.

• In enteric fever, many patients are treated with oral chloramphenicol.
However, the drug of choice is ciprofloxacin for adult typhoid.

• With gastroenteritis, the management requires fluid and electrolyte replace-
ment and control of nausea, vomiting and pain. Antibiotics are usually not
necessary.

• Patients with Salmonella bacteraemia also require antimicrobial treatment
with chloramphenicol, co-trimoxazole and high doses of ampicillin or
ciprofloxacin.

• An increasing number of Salmonella enteritidis isolates have shown antibi-
otic resistance.

References

Anderson, E.S. and Rogers, A.H. (1963). Slime polysaccharides of the Enterobacteriaceae.
Nature (London), 198: 714-715.

Angulo, F.J. and Swerdlow, D.L. (1995). Bacterial enteric infections in persons infected
with human immunodeficiency virus. Clin Infect Dis, 21(Suppl. 1): S84-S93.

Angulo, F.J. et al. (1997). A community waterborne outbreak of salmonellosis and the effect-
iveness of a boil water order. Am J Public Hltb, 87(4): 580-584.

Baird-Parker, A.C. (1991). Foodborne Salmonellosis, Lancet Review of Foodborne Illness.
London: Edward Arnold, pp. 53-61.

Duguid, J.P., Anderson, E.S. et al. (1975). A new biotyping scheme for Salmonella
typbimurium and its phylogenetic significance./ Med Microbiol, 8: 149-166.

Finlay, B.B., Leung, K.Y. et al. (1992). Salmonella interactions with the epithelial cell. ASM
News, 58: 486-489.

182

Salmonella

Guiney, D.G., Fang, EC. et al. (1995). Biology and clinical significance of virulence plas-

mids in Salmonella serovars. Clin Infect Dis, 21(Suppl. 2): S146-S151.
Le Minor, L., Veron, M. and Popoff, M. (1982a). Taxonomie des Salmonella. Ann

Microbiol (Paris), 133B: 223-243.
Le Minor, L., Veron, M. and Popoff, M. (1982b). Proposition pour une nomenclature des

Salmonella. Ann Microbiol (Paris), 133B: 245-254.
Le Minor, L., Popoff, M.Y. et al. (1986). Individualisation d'une septieme sous-espece de

Salmonella. Ann Microbiol (Paris), 137B: 211-217.
Mandal, B.K. (1979). Typhoid and paratyphoid fever. Clin Gastroenterol, 8: 715-735.
Popoff, M.Y., Bockemuhl, J. and Hickman-Brenner, F.W. (1995). Supplement 1994 (no.

38) to the Kauffmann-White scheme. Res Microbiol, 146: 799-803.
Reeves, M.W., Evins, G.M. et al. (1989). Clonal nature of Salmonella typhi and its genetic

relatedness to other salmonellae as shown by multilocus enzyme electrophoresis, and

proposal of Salmonella bongori comb. Nov. / Clin Microbiol, 27: 313-320.
Roy, S.K., Speelman, P. et al. (1985). Diarrhea associated with typhoid fever./ Infect Dis,

151: 1138-1143.
Scragg, J., Rubidge, C. and Wallace, H.L. (1969). Typhoid fever in African and Indian

children in Durban. Arch Dis Child, 44: 18-28.

183

Shigella

Basic microbiology

Shigellae are Gram-negative, non-motile rods. As a group they do not produce
gas from carbohydrates. They reside presently in the family Enterobacteriaceae
and show very similar characteristics to Escherichia coli. Unlike other mem-
bers in the Enterobacteriaceae group, Shigellae are non-lactose fermenting on
MacConkey agar or desoxycholate citrate agar after a period of incubation of
24 hours.

Shigellae, by definition, are non-motile and do not decarboxylate lysine or
hydrolyse arginine. Except for S. sonnei, certain serovars within the other species,
and certain strains within these serovars, they do not produce gas from glucose
or decarboxylate ornithine, nor do they use sodium acetate or produce indole
from tryptophan.

Origin and taxonomy

Kiyoshi Shiga first documented shigella in 1898 (Shiga, 1898). However, there
is evidence that Ogata (1892) in Japan and Chantemesse and Widal (1888) in

Bacteriology

France may have actually isolated Shigella much earlier. The genus name
Shigella was published in 1919 by (Castellani and Chalmers, 1919). Numerous
publications on Shigella occurred after this date with a significant paper on its
grouping in 1954 (Enterobacteriaceae Subcommittee, 1954; Rowe and Gross,
1981).

Shigella, based on both biochemical and serological evidence, can be classi-
fied into four major serological groups, namely: Group A, Shigella dysente-
riae, which includes at least ten serotypes; Group B, Shigella flexneri, includes
six serotypes; Group C, Shigella boydii, which includes 15 serotypes and
Group D Shigella sonnei, which includes only one serotype. The type species
is S. dysenteriae 1.

Metabolism and physiology

Shigella are facultatively anaerobic organisms that grow poorly under anaerobic
conditions. The optimal temperature for the growth of Shigella is 37°C, in media
containing 1% peptone as a carbon and nitrogen source. Shigellae are killed at a
temperature of 55°C within 1 hour (Rowe, 1990).

While shigellae are able to tolerate extremely acid conditions (pH 2.5) for
short periods they prefer to be grown at neutral or slightly alkaline pH (pH
7.0-7.4).

Clinical features

Apart from chimpanzees and monkeys, bacillary dysentery is specifically a
human disease characterized by a type of diarrhoea in which the stools con-
tain blood and mucus. In extreme cases it becomes associated with heavy
inflammation of the colonic mucosa. Shigellosis, principally a self-limiting
disease in healthy adults, has been known to cause fatalities, particularly in
young children (Salyers and Whitt, 1994).

Shigellae are transmitted by the direct faecal-oral route with infected indi-
viduals typically excreting 10 5 -10 9 shigellae per gram of wet faeces, with
symptomless carriers excreting 10 2 -10 6 per gram (Thomas, 1955; Dale and
Mata, 1968). Because of this food has the potential to be contaminated
through the soiled fingers of patients or carriers. The transfer of shigellae by
flies breeding on faeces has been established as a very important transmission
route during some outbreaks.

In a study of endemic shigellosis in Bangladesh, 2.1% of children 5 years of
age and under, were found to be asymptomatic carriers (Hossain et ah, 1994).
Similar results were reported from Mexico where 55% of infants, 2 years of age
and under, infected with Shigella were asymptomatic (Guerrero et al. 9 1994).

186

Shigella

The severity of disease due to Shigella depends on the virulence of the infect-
ing strain, aside from the host's immune system. The disease caused by S. son-
net tends to be mild and of short duration, whereas that caused by S. flexneri
tends to be more severe. S. boydii and S. dysenteriae produce disease of vary-
ing severity but S. dysenteriae has often caused epidemics of severe infections.

The infective dose for Shigella is generally quite low when compared to other
pathogens such as E. colt and Vibrio cholerae. The median infective dose (ID 50 )
for Shigella is around 10 4 (Dupont et al., 1972) in healthy adults compared
to Vibrio cholerae, which is around 10 7 or higher. Shaughnessy et al. (1946)
found that a dose of 10 8 organisms of Shigella flexneri were required to induce
disease in volunteers who had previously ingested 2g of sodium bicarbonate.

The incubation period following ingestion of Shigella ranges from 36 to
72 hours, but can be as short as 12 hours, with frank dysentery appearing
within 2 days.

Symptoms of shigellosis usually develop suddenly, often as an abdominal colic,
followed by watery diarrhoea often accompanied by fever and malaise. Some
individuals do go onto develop abdominal cramps, tenesmus and the frequent
passage of small volumes of stool (bloody mucus). Symptoms of dysentery can
last for about 4 days. In severe cases up to 10 days has been documented. Patients
who are recovering from infection often continue to excrete Shigella for a month
following infection. There is some documented evidence that some patients can
still excrete Shigella for over a year following recovery (Du Pont et al., 1970).
This carriage may be more important under conditions of poor hygiene.

Shigella is known to produce an exotoxin, initially described as a neurotoxin,
but it also has a fluid transuding effect on the intestinal mucosa. To date the role
of the toxin in the pathogenesis of dysentery is uncertain. Shigella dysenteriae
type 1 is responsible for many cases of haemolytic uraemic syndrome (HUS), first
described in 1955 (Gasser et al., 1955), accompanying outbreaks of dysentery
and in some parts of the world is one of the commonest forms of acute renal
failure in children. Although it is an invasive disease, Shigella usually do not
reach tissue beyond the lamina propria and, therefore, they very rarely cause
bacteraemia or systemic infections except under very special circumstances
(Struelens et al., 1985).

Pathogenicity and virulence

The two main virulence factors in Shigella are their invasive techniques and
toxigenic properties, however, the exact role of the toxin produced by Shigella
has yet to be defined.

As with most enteric pathogens, the first step in the invasion of host cells by
shigellae, as demonstrated in HeLa cells, is attachment. In the case of Shigella
this process does not seem to be mediated through interaction with a specific
receptor (Clerc and Sansonetti, 1987). There is, however, mounting evidence

187

Bacteriology

which suggests that the invasion plasmid antigen (Ipa) D may be involved as
an adhesive. The proteins, IpaB and IpaC, encoded by genes located on the
virulence plasmid, have been found on the bacterial surface suggesting some
importance in the pathogenicity of Shigella (Menard et al. 9 1994). It has been
suggested that these two proteins are essential for phagocytosis. It is thought
that the proteins rupture phagocytic vesicles allowing for intercellular spread.
This intercellular spread (ICS), mediated by two proteins, IcsA (also called
VirG; Bernardini et al. 9 1989) and IcsB, leads to cell death and inflammatory
response in the host cells. The process of cell death, induced by these proteins
is unknown to date, but seems to be independent of the production of Shiga
toxin (Salyers and Whitt, 1994).

S. dysenteriae is known to produce a Shiga toxin, encoded on the chromo-
some, which belongs to the family of A1/B5 toxins. This Shiga toxin, whose
main action is on blood vessels, is only released during cell lysis and not
actively secreted from the cell (Rowe, 1990).

Treatment

Most cases of shigella dysentery, due to Shigella sonnei, are mild and do not
require any antibiotic treatment. Hydration of the patient using oral salt
rehydration is generally the best treatment regimen. If antibiotics need to be
administered ampicillin, tetracycline, co-trimoxazole, or ciprofloxacin are an
appropriate choice for treatment. Antibiotics are usually administered because
they shorten the duration of illness and decrease the relapse rate (Hruska, 1991).

The first reported case of multiple-drug-resistant Shigella occurred in 1956
in Japan. To date, Shigella is acquiring resistance against many different
drugs, constituting a major concern. For example, between 1979 and 1982
the percentage of resistant S. sonnei strains isolated in a hospital in Madrid
increased from 39.6 to 97.9% for ampicillin, from 34.4 to 96.9% for
co-trimoxazole (SXT), from 6.3 to 18.0% for tetracycline and from 1.6 to
15.1% for chloramphenicol (Lopez-Brea et aL, 1983). The first SXT-resistant
strain of S. dysenteriae type 1 in Bangladesh was isolated in 1982 (Shahid
et aL, 1985) with SXT resistance also evident in Finnish travellers (Heikkila
etaL, 1990).

Antibiotic therapy for the treatment of shigellosis should always be based
on proper antibiotic susceptibility testing. If this is not possible, blind therapy
should include a quinolone as the majority of strains are susceptible.

Survival in the environment

Despite the public health significance of Shigella their presence and also per-
sistence in the environment is not well documented when compared to other

188

Shigella

species in the family Enteriobacteriaceae. Humans are the only important
reservoir of Shigella. Excretion of Shigella in stools is highest during the acute
phase of dysenteric illness. During this phase the environment is contaminated
and the organisms can survive for weeks in cool and humid locations (Rowe
and Gross, 1981). They survive for 5-46 days when dried on linen and kept
in the dark, and for 9-12 days in soil at room temperature (Roelcke, 1938).
Although shigellae tolerate a low pH (<3) for short periods, they soon perish,
but remain alive for days if specimens are kept alkaline and are prevented
from drying (Rowe, 1990).

Of all Shigella, S. sonnei appears to be more resistant to harsh conditions
when compared to S. dysenteriae and S. flexneri. It has been shown that
Shigella sonnei can survive for over 3 hours on ringers and up to 17 days on
wooden toilet seats (Huchinson, 1956). In a study by McGarry and Stainforth
(1978), Shigella dysenteriae was shown to survive for up to 17 days in
biogas plant effluent (11-28°C) but less than 30 hours in the biogas plant
itself (14-24°C).

Shigella may be spread in aerosol droplets, possibly by flushing toilets and
spray irrigation systems. A study by Newson (1972) showed that by flushing
a suspension (10 10 cells) of Shigella sonnei an aerosol of about 39 bacteria/mm 3
air was produced and that the Shigella could be recovered from the splashes
and could survive for up to 4 days.

The effectiveness of the sewage treatment plant in inactivating Shigella is
limited. However, research has shown that Shigella removal is very similar to
that of E. colt.

Survival in water

Shigella can be found in surface waters and also within contaminated drinking
water. This is highly significant as a mode of transmission in developing coun-
tries. As a rule drinking water will not contain Shigella unless it is untreated or
if there are problems with the water-treatment process (Green et al. 9 1968).

There have been studies on the survival of Shigella in water (Feachem et al.,
1980). From this and a number of other studies it is found that survival of
Shigella depends upon concentrations of other bacteria, nutrients, oxygen and
temperature. Within clean water, survival times are less than 14 days at tem-
peratures >20°C but in waters of less than 10°C they have been shown to
survive for weeks. The half-life of Shigella was found to be 24 hours. Talayeva
(1960) found that Shigella flexneri survived for up to 21 days in clean river
water at a temperature of 19-24°C, up to 47 days in autoclaved river water,
up to 9 days in well water, up to 44 days in autoclaved tap water and up to 6
days in polluted well water. Shrewsbury and Barson (1957) found that Shigella
dysenteriae could survive for between 2.5 and 29 months in sterile but
faecally contaminated water at 21°C. Hendricks (1972) reported that Shigella
flexneri was also able to multiply in sterilized river water.

189

Bacteriology

Infections with Shigella spp. are often acquired by drinking water contamin-
ated with human faeces or by eating food washed with contaminated water.
Food-borne outbreaks of shigellosis occur, especially in the tropics and less
frequently in the developed countries (Coultrip et al. 9 1977). In addition to
contamination from faeces by food-handlers who have poor hygiene, flies
may sometimes act as vectors in tropical countries (Khalil et ah, 1994).

There is evidence for Shigella infection acquired by swimming in sewage-
contaminated recreational waters (Rosenberg et al., 1976; Makintubee et al.,
1987; Sorvillo etaL, 1988). In developed countries infections are usually asso-
ciated with recent travel (Parsonnet et al., 1989; Luscher and Altwegg, 1994)
to countries with insufficient sanitary facilities.

Outbreaks of Shigella have occurred in day-care centres associated with
direct person-to-person transmission by the faecal-oral route, which is facili-
tated by the low infectious dose (10-200 organisms) necessary to induce
infection. Shigellae are the most often cause of laboratory-acquired infections
(Aleksic etal, 1981; Grist and Emslie, 1989).

Natural disasters and wars are frequently associated with shigellosis
outbreaks and mass encampments become breeding places, causing a high
incidence of illness and fatalities (Centers for Disease Control, 1994b; Sharp
etal.,1995).

Methods of detection

In pure cultures Shigella form circular, glistening, translucent or slightly opaque
colonies on nutrient agar. They are able to grow on blood agar but with no
haemolysis. The colony size varies with small colonies in S. dysenteriae, but
most Shigella serovars produce colonies of 1-2 mm after 18-24 hours of
incubation at 37°C. S. sonnei may dissociate in smooth and larger, flatter colo-
nies showing an irregular edge, often referred to as phase 1 colonies. 'Phase 2'
colonies are associated with the loss of the 120 000-140 000 kDa virulence
plasmid and a change of the antigenic specificity. Phase 2 strains still grow homo-
geneously in broth culture but they may partially sediment after boiling at 100°C
or autoagglutinate in 3.5% NaCl solution (Rowe, 1990; Bockemuhl, 1992). On
Leifson's deoxycholate citrate agar, Salmonella-Shigella agar, or xylose-lysine-
deoxycholate and on MacConkey agar, shigellae grow with colourless, translu-
cent, smooth colonies of 1-2 mm after 18-24 hours of incubation at 37°C. After
prolonged incubation (>48 hours) growth of S. sonnei becomes pinkish due to
delayed fermentation of lactose and sucrose.

For the isolation of shigellae, a freshly passed stool specimen during the acute
stage of illness is the material of choice. If present, mucus or blood-stained por-
tions should be selected for culture and, if desired, for microscopic examination
for the presence of faecal leucocytes. If the specimens cannot be cultured within
2-4 hours, they should be preserved in a transport medium.

190

Shigella

Shigellae are easily overgrown by the concomitant aerobic intestinal flora.
Enrichment of stool specimens in Gram-negative broth (Hajna) or selenite
broth for about 6 hours at 37°C can be tried.

After growth on an appropriate agar, Shigella spp. are identified by bio-
chemical reactions, combined with agglutination in group- or serovar-specific
antisera for shigellae.

Epidemiology and waterborne outbreaks

A number of outbreaks, primarily food related, have been due to Shigella.
These have included poi (Lewis et al., 1972), tuna (Bowen, 1980) and con-
taminated salads. In fact, during 1961-1975 there were 10 648 cases (72 out-
breaks) of shigellosis reported in the USA, principally due to contaminated
salads (Black et aL, 1978). The contamination of food is possibly the most
important route of transmission of Shigella (Barrel and Rowland, 1979).

S. dysenteriae 1 has caused major epidemics in Central America (1969-70),
Bangladesh (1972) and East Africa (since 1991), whereas S. boydii is mainly
prevalent in South Asia and the Middle East. In Europe and North America
S. sonnei is by far the predominant, followed by S. flexneri (Aleksic et al.,
1987; Lee et al., 1991; Centers for Disese Control, 1994a) and infections due
to S. boydii and S. dysenteriae are almost exclusively imported by travellers
or foreign-born citizens returning from visits to their home countries (Aleksic
etal, 1987).

In the UK and the rest of Europe, Shigella dysenteriae^ common during the
First World War, is rare. Between 1920 and 1930 both Shigella flexneri and
Shigella sonnei were endemic and of approximately equal incidence, but by
1940 Shigella sonnei had become dominant, increasing in incidence annually
to a peak of over 49 000 (99% of all shigellae notified) in 1956. The incidence
of Sonne dysentery then declined steadily in the UK to an annual average of
about 3000 notified cases between 1970 and 1990. However, the numbers
rose sharply in 1991 when there were several widespread community out-
breaks and continued to rise to a peak of 17 000 cases in 1992. The incidence
has since fallen but there were still more than 4550 cases in 1995. Infections
due to other shigellae, usually imported, have remained constant during the
last decade at about 800-900 a year.

Similar changes have taken place in the USA. Up to 1968 Shigella flexneri
and Shigella sonnei were equally common but Shigella sonnei now accounts
for 65% of cases and Shigella flexneri for about 30%.

In tropical areas shigellosis is endemic. It has been estimated that 5 million
cases of shigellosis require hospital treatment. Of these 5 million cases about
600 000 die every year.

There is general agreement in the literature that the maintenance of endemic
shigellosis has little or no relationship to water quality, but that it is strongly

191

Bacteriology

related to water availability and associated hygienic behaviour. However,
there will always be specific exceptions to this; for instance, Sultanov and
Solodovnikov (1977) considered that the maintenance of dysentery was due
to the widespread use of polluted surface water for domestic purposes.

Some epidemics of bacillary dysentery are waterborne. An outbreak of
2000 cases of shigellosis due to Shigella sonnet occurred in 1966 in Scotland
when the chlorination plant on the town's water supply broke down (Green
et al., 1968). During 1961-75, 38 waterborne outbreaks of shigellosis were
reported in the USA (Black et aL 9 1978). Most of these outbreaks involved
semi-public or individual water systems and were usually the result of inad-
equate or interrupted chlorination of water contaminated by faeces. Such
water-borne epidemics are usually dramatic but they can be terminated very
quickly when the water supply is adequately treated.

Risk assessment

Health effects: occurrence of illness, degree of morbidity and mortality, prob-
ability of illness based on infection:

• Shigella can be classified into four major serological groups. Group A, Shigella
dysenteriae, Group B, Shigella flexneri, Group C, Shigella boydii, and
Group D Shigella sonnei, which includes only one serotype. Shigella sonnei
accounts for most cases of dysentery in the developed world.

• Shigella infection is characterized by watery or bloody diarrhoea, abdom-
inal pain, fever, and malaise. Symptoms of dysentery can last for about
4 days but, in severe cases, up to 10 days has been documented. Shigellosis
is principally a self-limiting disease in otherwise healthy adults; it has been
known to cause fatalities, particularly in malnourished infants.

• Shigella dysenteriae type 1 is responsible for many cases of haemolytic
uraemic syndrome.

• The severity of disease depends on the virulence of the infecting strain,
aside from the host's immune system. The disease caused by S. sonnei tends
to be mild and of short duration whereas that caused by S. flexneri tends to
be more severe.

• Children and infants generally suffer a higher rate of infection and more
severe disease course.

Exposure assessment: routes of exposure and transmission, occurrence in
source water, environmental fate:

• Shigellosis is specifically a human disease that is transmitted by the direct or
indirect faecal-oral route.

• Infections with Shigella spp. are often acquired by drinking water con-
taminated with human faeces or by eating food washed with contaminated

192

Shigella

water. Transfer of shigellae by flies has been very important during some
outbreaks.

• Helped by the low infectious dose (10-200 organisms) necessary to induce
infection, person-to-person contact in day-care centres, institutions, and
other places where hygiene may not be the best have resulted in outbreaks.

• Shigella can be found in surface waters and also within contaminated
drinking water. This is significant as a mode of transmission in developing
countries.

• Within clean water, survival times are less than 14 days at temperatures
>20°C but, in waters of less than 10°C, they have been shown to survive
for weeks. They do not survive at acid pH.

• S. sonnei appears to be more resistant to detrimental environmental condi-
tions than S. dysenteriae and S. flexneri. Shigella dysentaeriae has been
shown to survive for at least 6 days in septic tank effluent.

Risk mitigation: drinking-water treatment, medical treatment:

• Drinking water treatment that includes disinfection is sufficient to remove
Shigella. Waterborne outbreaks have generally resulted from inadequate
treatment.

• Most cases of Shigella dysentery (i.e. S. sonnei) are mild and do not require
any antibiotic treatment. Treatment by means of hydration using an oral
salt rehydration is all that is required. Ampicillin, co-trimoxazole, tetra-
cycline or ciprofloxacin are appropriate antibiotic choices.

References

Aleksic, S., Bockemuhl, J. and Aleksic, V. (1987). Serologisch-epidemiologische Unter-

suchungen an 908 Shigella -Stammen von Patienten in der Bundesrepublik Deutschland,

1978-1985. Bundesgesundheitsblatt, 30: 207-210.
Barrell, R.A.E. and Rowland, M.G.M. (1979). Infant foods as a potential source of diar-

rhoeal illness in rural West Africa. Trans Roy Soc Trop Med Hyg, 73: 85-90.
Bernardini, M.L., Mounier, J. et al. (1989). Identification of icsA, a plasmid locus of

Shigella flexneri that governs intra- and intercellular spread through interaction with

F-actin. Proc Natl Acad Sci USA, 86: 3867-3871.
Black, R.E., Craun, G.F. and Blake, PA. (1978). Epidemiology of common-source

outbreaks of shigellosis in the United States, 1961-1975. Am J Epidemiol, 108: 47-52.
Bockemuhl , J. (1992). Enterobacteriaceae. In Mikrobiologische Diagnostik, Burkhardt, F.

(ed.). Stuttgart: Thieme Verlag, pp. 119-153.
Bowen, G.S. (1980). An outbreak of shigellosis among staff members of a large urban

hospital. In Epidemic Intelligence Service 29th Annual Conference. Atlanta, Georgia:

Centers for Disease Control, p. 46.
Castellani, A. and Chalmers, A.J. (1919). Manual of Tropical Medicine, 3rd edn.

New York: Williams Wood &c Co.
Centers for Disease Control. (1994a). Summary of notifiable diseases, United States 1994.

MMWR, 43: 54.
Centers for Disease Control. (1994b). Health status of displaced persons following civil

war - Burundi, December 1993-January 1994. MMWR, 43: 701-703.

193

Bacteriology

Chantemesse, A. and Widal, F. (1888). Sur les microbes de la dysenterie epidemique. Bull

Acad Med Ser 3,19: 522-529.
Clerc, P. and Sansonetti, P.J. (1987). Entry of Shigella flexneri into HeLa cells: evidence for

directed phagocytosis involving actin polymerization and myosin accumulation. Infect

Immun, 55:2681-2688.
Coultrip, R.L., Beaumont, W. and Siletchnik, M.D. (1977). Outbreak of shigellosis - Fort

Bliss, Texas. MMWR, 26: 107-108.
Dale, D.C. and Mata, L.J. (1968). Studies of diarrheal disease in Central America. XI.

Intestinal bacterial flora in malnourished children with shigellosis. Am J Trop Med Hyg,

17: 397-403.
Du Pont, H.L., Gangarosa, E.J., Reller, L.B. et al. (1970). Shigellosis in custodial institu-
tions. Am J Epidemiol, 92: 172-179.
DuPont, H.L., Hornick, R.B. et al. (1972). Immunity in shigellosis. II. Protection induced

by live oral vaccine or primary infection. / Infect Dis, 125: 12-16.
Enterobacteriaceae Subcommittee Reports. (1954). Int Bull Bacteriol Nomencl Taxon,

4: 1-94.
Feachem, R.T.G.A., Bradley, D.J., Garelick, H. et al. (1980). Health aspects of excreta and

sullage management: A state of the art review. Appropriate technology for water supply

and sanitation, vol. 2. Washington, DC: The World Bank, Transportation, Water and

Telecommunications Department.
Gasser, C, Gautier, E. et al. (1955). Hamolytisch-uramische Syndrome: bilaterale

Nierenrindennekrosen bei akuten erworbenen hamolytischen Anamien. Schweiz Med

Wochenschr, 85: 905-909.
Green, D.M., Scott, S.S., Mowat, D.A.E. et al. (1968). Waterborne outbreak of viral

gastroenteritis and Sonne dysentery./ Hyg, 66: 383-392.
Grist, N.R. and Emslie, J.A.N. (1989). Infections in British clinical laboratories, 1986-7.

J Clin Pathol, 42: 677-681.
Guerrero, L., Calva, J.J. et al. (1994). Asymptomatic Shigella infections in a cohort of

Mexican children younger than two years of age. Pediatr Infect Dis J, 13: 597-602.
Heikkila, E., Siitonen, A. et al. (1990). Increase of trimethoprim resistance among Shigella

species. / Infect Dis, 161: 1242-1248.
Hendricks, C.W. (1971). Enetric bacterial metabolism of stream sediment eluates. Can J

Microbiol, 17: 551-556.
Hossain, M.A., Hasan, K.Z. and Albert, M.J. (1994). Shigella carriers among non-diarrhoeal

children in an endemic area of shigellosis in Bangladesh. Trop Geogr Med, 46: 40-42.
Hruska, J.F. (1991). Gastrointestinal and intraabdominal infections. In A Practical

Approach to Infectious Diseases, 3rd edn, Reese, R.E. and Betts, R.F. (eds). Boston:

Little, Brown, pp. 305-356.
Huchinson, R.I. (1956). Some observations on the method of spread of Sonne dysentery.

Month Bull Ministr Hlth Public Hlth Lab Serv, 15: 110-118.
Khalil, K., Lindblom, G.B. et al. (1994). Flies and water as reservoirs for bacterial

enteropathogens in urban and rural areas around Lahore, Pakistan. Epidemiol Infect,

113: 435-444.
Lee, L.A., Shapiro, C.N. et al. (1991). Hyperendemic shigellosis in the United States:

a review of surveillance data for 1967-1988. / Infect Dis, 164: 894-900.
Lewis, J.N., Loewenstein, M.S., Guthrie, L.C. et al. (1972). Shigella sonnei outbreak on the

island of Maui. Am J Epidemiol, 96: 50-58.
Lopez-Brea, M., Collado, L. et al. (1983). Increasing antimicrobial resistance of Shigella

sonnei. J Antimicrob Chemother, 11: 598.
Liischer, D. and Altwegg, M. (1994). Detection of shigellae, enteroinvasive and enterotox-
igenic Escherichia coli using the polymerase chain reaction (PCR) in patients returning

from tropical countries. Mol Cell Probes, 8: 285-290.
Makintubee, S., Mallonee, J. and Istre, G.R. (1987). Shigellosis outbreak associated with

swimming. JAM A, 236: 1849-1852.
McGarry, M.G. and Stainforth, J. (eds) (1978). Compost, fertilizer and Biogas. Production

from human and farm wastes in the People's Republic of China. Ottawa: International

Development Research Centre.

194

Shigella

Menard, R., Sansonetti, P. and Parsot, C. (1994). The secretion of the Shigella flexneri Ipa

invasins is activated by epithelial cells and controlled by IpaB and IpaD. EMBO J,

13: 5293-5302.
Newson, S.W.B. (1972). Microbiology of hospital toilets. Lancet, 2: 700-703.
Ogata, M. (1892). Zur atiologie der Dysenteric Zentralbl Bakteriol Parasitenkd

Infektionskr Hyg, 11: 264-272.
Parsonnet, J., Greene, K.D. et al. (1989). Shigella dysenteriae type 1 infections in US trav-
ellers to Mexico. Lancet, 2: 543-545.
Roelcke, K. (1938). Uber die Resistenz verschiedener Ruhrkeime. Z Hyg Infektionskr,

120:307-314.
Rosenberg, M.L., Hazlet, K.K., Schaefer, J. et al. (1976). Shigellosis from swimming.

JAMA, 236: 1849-1852.
Rowe, B. (1990). Shigella. In Topley &C Wilson's Principles of Bacteriology, Virology and

Immunity, 8th edn, vol. 2, Parker, M.T. and Collier, L.H. (eds). London: Edward Arnold,

pp. 455-468.
Rowe, B. and Gross, R.J. (1981). The genus Shigella. In The Prokaryotes, vol. II,

Starr, M.P., Stolp, H. et al. (eds). Berlin, Heidelberg and New York: Springer- Verlag,

pp. 1248-1259.
Salyers, A.A. and Whitt, D.D. (1994). Bacterial Pathogenesis: a Molecular Approach.

Washington, DC: ASM Press.
Shahid, N.S., Rahaman, N.M. et al. (1985). Changing pattern of resistant Shiga bacillus

(Shigella dysenteriae type 1) and Shigella flexneri in Bangladesh. / Infect Dis, 152:

1114-1119.
Sharp, T.W., Thornton, S.A. et al. (1995). Diarrheal disease among military personnel dur-
ing Operation Restore Hope, Somalia, 1992-1993. Am J Prop Med Hyg, 52: 188-193.
Shaughnessy, D.J., Olson, R.C., Bass, K. et al. (1946). Experimental human bacillary

dysentery. JAM A, 132: 362-368.
Shiga, K. (1898). Uber den Dysenteriebacillus (Bacillus dysentericus). Zentralbl Bakteriol

Parasitenkd Infektionskr Hyg, 24: 817-828, 870-874, 913-918.
Shrewsbury, J.F.D. and Barson, G.J. (1957). On the absolute viability of certain pathogenic

bacteria in a synthetic well water. / Pathol Bacteriol, 74: 215-220.
Sorvillo, F.J., Waterman, S.H. et al. (1988). Shigellosis associated with recreational water

contact. Am J Prop Med Hyg, 38: 613-617.
Struelens, M.J., Patte, D. et al. (1985). Shigella septicemia: prevalence, presentation, risk

factors, and outcome./ Infect Dis, 152: 784-790.
Sultanov, G.V. and Solodovnikov, Y.P. (1977). Significance of water factor in epidemiology

of dysentery. Zh Mikrobiol Epidemiol Immunobiol, no. 6 (June), 99-101.
Talayeva, J.G. (1960). Survival of dysentery bacteria in water according to the results of

a reaction with haptenes. Microbiol Immunol, 4: 314-320.
Thomas, S. (1955). The numbers of pathogenic bacilli in faeces in intestinal diseases.

/ Hyg, 53:217-224.

195

Vibrio cholerae

Basic microbiology

Vibrio are Gram-negative, short rods (0.5 by 1.3-3 fjim) which are often curved
or comma shaped. They are non-sporulating, non-capsulated, facultative anaer-
obes, catalase-positive and motile by means of a single polar flagellum. In
liquid media all vibrios show vigorous darting motility. Most species are
oxidase-positive and reduce nitrates to nitrites.

By far the most significant of all vibrios is V. cholerae, which is 0.5-0.8 |xm
by 1.5-2.5 |xm in size and the cause of cholera. To date, V. cholerae have been
classified into 206 c O' serogroups (Yamai etaL 9 1997). Initially V. cholerae was
divided into Ol strains and non-Ol strains, however, there is now evidence of
an 0139 strain which will be discussed in more detail below. The Ol strains
form two biotypes, namely classical and El Tor. These are further subdivided
into three serotypes, namely Inaba, Ogawa and Hikojima (which is rare). It is
probable that these three serotypes, over many years, have undergone genetic
switching and possibly constitute variants of the same strain. In fact, this
serotype conversion has been shown to take place both in vivo and in vitro.

The first six cholera pandemics have been caused by the classical biotype,
whereas the seventh pandemic was caused by the El Tor biotype. The first
reports of a new epidemic of Vibrio cholerae occurred in 1993. At first, the

Bacteriology

organism responsible for the outbreak was referred to as non-Ol V. cholerae.
It was later established that the organism belonged to a new serogroup, 0139
Bengal (Shimada et al. 9 1993).

There are other potentially pathogenic Vibrio, aside from V. cholerae, such
as V. parahaemolyticus which is a major cause of food-borne illness in South-
East Asia, particularly Japan. Other species of Vibrio occasionally isolated
from humans include V. alginolyticus and V. vulnificus. V. alginolyticus is a
halophile and termed biotype 2 of V. parahaemolyticus. Vibrio vulnificus is a
highly invasive vibrio species affecting immunocompromised persons who have
consumed seafood. Other vibrios of human significance include V. damsela,
associated with wound infections, V kollisae, associated with diarrhoea,
V. mimicus, associated with gastroenteritis and V. fluvialis which is a cause
diarrhoea and fever.

Origin and taxonomy

Pacini first described the infection caused by V. cholerae in 1854 and it was
described later by Koch in 1884 (Koch, 1884). Vibrio cholerae Ol, biotype El
Tor was first detected in 1934 in Indonesia, however, prior to this it was isol-
ated but not truly identified in Sinai in 1905. It remained pandemic in Asia
until the 1960s and was detected in 1970 in Russia and South Korea. The first
case of Vibrio cholerae in the Americas occurred in Peru in 1991 spreading to
other countries in South America within weeks. Until 1992 the toxigenic Ol
serogroup had been associated with cholera epidemics and pandemics. The
non-Ol serogroup was mainly associated with extraintestinal infections and
limited outbreaks of gastroenteritis.

To date there are presently 36 species in the genus Vibrio with 12 species of
these potentially pathogenic to humans.

Metabolism and physiology

Vibrios have a requirement for salt, the concentration of which ranges for the
different species (Baumann et al., 1984). This difference provides a means of
separating pathogenic vibrios into the non-halophilic species, consisting of
V. cholerae and V. mimicus, which grow on nutrient agar, and the halophilic
species that require a salt supplement in the growth media.

Vibrios are able to grow over a wide temperature range (20 to >40°C). They
grow better in alkaline conditions, although most species of Vibrio grow between
a pH range of 6.5 and 9.0.

Vibrios catabolize D-glucose anaerobically via the Embden-Meyerhof path-
way, producing formic, lactic, acetic, succinic acids, ethanol and pyruvate

198

Vibrio cholerae

(Baumann et al. 9 1984). D-glucose and many other sugars are transported
into Vibrio internally via the phosphoenolpyruvate:carbohydrate phospho-
transferase system. In the case of D-glucose it is transported and subsequently
phosphorylated to glucose-6-phosphate (Sarker et al. 9 1994).

Clinical features

V. cholerae causes infections ranging from asymptomatic to very serious, pro-
fuse watery diarrhoea known as 'rice water stools' (watery, colourless stools
with a fishy odour and flecks of mucus). Cholera can also be characterized by
a sudden onset of effortless vomiting, which leads to rapid and severe dehy-
dration and possibly death within 1-5 days. Under these conditions, rehydra-
tion therapy is needed. If death does occur this arises because of fluid and
electrolyte imbalance. The incubation period is short and the clinical signs of
cholera develop within 0.5-5 days after infection.

The reservoirs of V. cholerae are asymptomatic human carriers and diseased
people who shed the microorganisms in their faeces. In fact convalescent and
asymptomatic individuals may excrete 10 2 -10 5 V. cholerae/g faeces whereas
an active case excretes 10 6 -10 9 /ml of 'rice water stool'.

Infective doses of V. chloreae are high in healthy individuals. Reports of 10 8
classical V. cholerae in water produced diarrhoea in 50% of adult volunteers
(Hornick et al., 1971) and 10 11 organisms produced 'rice water stools'. With
the addition of 2 g of sodium bicarbonate the ID 50 was found to be 10 4 for
diarrhoea and 10 8 for cholera-like diarrhoea. Although more than 10 8
V. cholerae cells are required to induce infection and diarrhoea the adminis-
tration of sodium bicarbonate (NaHCC^) reduces the infectious dose to less
than 10 4 organisms (Cash et al., 1974; Levine et al., 1988).

Pathogenicity and virulence

Infection due to V. cholerae begins with the ingestion of contaminated water
or food. Vibrio then colonizes the epithelium of the small intestine using the
toxin-coregulated pili (Taylor et al. 9 1987). Other colonization factors such as
the different haemagglutinins, accessory colonization factor, and core-encoded
pilus are all thought to play a role in the adhesion process. Once adhered
the enterotoxin is produced. In the small intestine V. cholerae produces an
enterotoxin known as the cholera toxin (CT). CT consists of one A subunit
(holotoxin, MW 27.2 kDa, two polypeptide chains linked by a disulphide
bond) and five B subunits. It is the B subunit that is involved in attaching
the toxin to a ganglioside receptor on the villi cell wall and also crypts in the

199

Bacteriology

intestine. The B subunit enters the host cell membrane forming a hydrophilic
trans-membrane channel. This allows subunit A, which is toxic, to enter the
cytoplasm. Once in the cytoplasm the toxin causes the transfer of adenosine
diphosphoribose (ADP ribose) from nicotinamide adenine dinucleotide
(NAD) to a regulatory protein that is responsible for the generation of intra-
cellular cyclic adenosine monophosphate (cAMP). Over-activation of cAMP
occurs due to activation of adenylate cyclase that then causes inhibition of
the uptake of Na + and Cl~ ions and water. Overall, there is a net outflow
of water across the mucosal cells and ultimately there is extensive loss of
electrolytes and water.

Cholera enterotoxin (CT) was first suggested by Robert Koch in 1884, how-
ever, its actual existence was not confirmed until 1959 (De, 1959). Another
factor, which is thought to contribute to the disease process, is haemolysin/
cytolysin (Honda and Finkelstein, 1979). Haemolysin has been shown to cause
accumulation of bloody fluid in ligated rabbit ileal loops. Other toxins pro-
duced by V. cholerae include the shiga-like toxin, a heat-stable enterotoxin
(Takeda etal., 1991), sodium channel inhibitor (Tamplin etal., 1987), thermo-
stable direct haemolysin-like toxin (Nishibuchi et al., 1992), and a non-
membrane-damaging cytotoxin (Saha and Nair, 1997).

V. cholerae Ol

As mentioned, V. cholerae Ol produces a potent and sometimes lethal cyto-
tonic enterotoxin, known as cholera toxin (CT). The CT seems to be both
structurally and functionally related to the heat-labile enterotoxin of E. coli
(Spangler, 1992).

The structural genes encoding both toxin subunits (ctxA and ctxB) have
been identified. The ctxB operon is located on a portion of the bacterial chromo-
some termed the core region (Ottemann and Mekalanos, 1994). In classical
V. cholerae strains two copies of the ctx element are widely separated on the
chromosome; for El Tor strains multiple copies are tandemly arranged
(Mekalanos, 1983). The B subunit serves to bind the toxin to the cell receptor
and the A subunit provides the toxigenic activity intracellularly after proteo-
lytic cleavage into two peptides, Al and A2 (Kaper et aL 9 1995). The Al pep-
tide is the active portion of the molecule acting as an ADP-ribosyltransferase
from NAD to a G protein, named Gs. Activation of Gs results in increased
intracellular levels of cAMP, which ultimately leads to protein kinase acti-
vation, protein dephosphorylation, altered ion transport and diarrhoeal disease
(Kaper et al., 1995). In vitro CT acts as a cytotonic (non-lethal) enterotoxin
causing rounding of Yl adrenal cells or CHO cell elongation. Removal of CT
from culture supernatant or preincubation of CT with antitoxin to CT causes
Yl and CHO cells to retain their original cell morphology.

Expression of CT is regulated by a 32 kDa integral membrane protein called
ToxR (Ottemann and Mekalanos, 1994). Certain amino acids, osmolarity and

200

Vibrio cholerae

temperature help to regulate ToxR (gene toxR) expression (Parsot and
Mekalanos, 1990; Ottemann and Mekalanos, 1994). In addition to CT, ToxR
also regulates several other factors. One of these is the toxin co-regulated pilus,
TcpA (gene tcpA), a 20.5 kDa protein that makes up the major subunit of the
V. cholerae pilus (Taylor et al., 1987). A (3-haemolysin is expressed by most V.
cholerae Ol El Tor strains. The gene (hlyA) encodes for a mature 84 kDa pro-
tein with haemolytic and cytolytic activity (Rader and Murphy, 1988).

V. cholerae 0139

V. cholerae 0139, is genetically similar to V. cholerae Ol El Tor (Albert,
1994). There are, however, a number of differences between V. cholerae Ol
and 0139. In 0139 there is evidence of a polysaccharide capsule and LPS
(Waldor et al., 1994). The structure of the 0139 capsule is composed of one
residue each of N-acetylglucosamine, N-acetylquinovosamine, galacturonic
acid and galactose and two molecules of 3,6-dideoxyxylohexose. LPS of
0139 contains colitose, glucose, 1-glycero-d-manno-heptose, fructose, glu-
cosamine and quinovosamine in its polysaccharide (Hisatsune et al., 1993). In
Ol LPS perosamine is present but in 0139 strains it is absent.

V. cholerae non-Ol

Non-Ol V. cholerae cause mild diarrhoea which is often bloody and in
extreme cases it can be severe. Non-Ol V. cholerae have also been reported in
wound infections, meningitis and bacteraemia. However, unlike Ol and
0139 V. cholerae non-Ol strains lack the capability to cause epidemic and
pandemic cholera. However, in some adult volunteers some strains have been
shown to produce mild to moderate gastroenteritis. The non-Ol strains are
documented as surviving better than V. cholerae Ol in a wide range of foods.
It has also been documented that some rare non-Ol strains (<4%) possess
CT. A common phenotypic feature to almost all non-Ol isolates is the pro-
duction of a p-haemolysin.

Treatment

Oral administration of fluid and electrolytes is necessary for an individual
who has cholera. The formula of a rehydration solution is sodium chloride,
3.5 g, potassium chloride, 1.5 g, sodium citrate, 2.9 g and glucose, 20 g all
dissolved in 1 litre of clean drinking water. In extreme cases, to reduce the

201

Bacteriology

excretion of V. cholerae, patients may be given tetracycline in order to reduce
the chances of cross-contamination and environmental contamination.

Survival in the environment

V. cbolerae has been shown to survive for contradictory time periods in envir-
onmental waters (Feachem et ah, 1981). The toxigenic Ol strains of Vibrio
cholerae have been shown to survive in aquatic environments for years, pos-
sibly residing in a biofilm. It is probable that these environmental biofilms
function as a reservoir for V cholerae; research in this area is warranted.
While the Ol serogroup of Vibrio cholerae has frequently been isolated from
aquatic environments most do not produce the cholera toxin (CT).

When present in the environment, V. cholerae are impossible to grow using
conventional culture techniques. Work by Colwell et al. (1994) found that
V. cholerae Ol enters a state of dormancy in response to nutrient deficiency,
high levels of saline, and low temperatures. Vibrio cholerae have also been found
in association with a wide range of aquatic life, including cyanobacteria
(Islam etal. 9 1989) and diatoms (Skeletonema costatum) (Martin and Bianchi,
1980) to name but a few.

Survival in water

Water is important in the transmission of cholera, with properly treated pub-
lic water supplies not generally considered to be a risk factor. In fact V cholerae
can survive longer in the environment than other faecal organisms suggesting
a public health concern when issues of risk of disease are based on the coli-
form index. Vibrio cholerae has been isolated from surface water and drink-
ing water and has been shown to be viable from one hour to 13 days in these
environments (Pesian, 1965).

It is probable that V cholerae exists in the marine environment in several
forms. These include a free-living state, particularly during elevated water tem-
peratures and nutrient concentrations; an epibiotic phase, association of vibrios
with specific substrates, e.g. chitin of shellfish, and the Viable but non-culturable'
state (Colwell and Huq, 1994). It is thought that Vibrio form microvibrio that
have altered morphologies resulting in a decrease in size and metabolic require-
ments/activities. These are formed under adverse conditions (Hood et aL, 1984).

Methods of detection

All environmental samples suspected of containing V. cholerae should be
transported to the laboratory at 4-1 0°C, inside a sterilized container and
processed within 6 hours (Donovan and van Netten, 1995). For the isolation
and detection of V. cholerae, specifically from water and the environment, a
qualitative enrichment medium of alkaline peptone water (APW) is often used.

202

Vibrio cholerae

However, a number of nutrient-rich modifications of APW, such as blood-
APW and egg-APW have been documented (Donovan and van Netten, 1995).
Following enrichment samples are plated onto thiosulphate-citrate-bile-salts-
sucrose agar (TCBS), a selective differential medium, or taurocholate-tellurite-
gelatine agar and incubated at 37°C for 18-24 hours. In the case of TCBS it is
known to suppress the growth of Gram-positive bacteria, pseudomonads, coli-
forms and aeromonads (Kobayashi et aL 3 1963). Other media that have also
been designed for the selective isolation or differentiation of Vibrio species
have included thiosulphate-chloride-iodide agar (Beazley and Palmer, 1992)
and polymyxin-mannose-tellurite agar (Shimada et al. 9 1990).

Following the incubation of TCBS agar plates at the appropriate tempera-
tures all suspected colonies of V. cholerae strains are identified by means of
biochemical tests. These tests are those that are used for the identification of
members of the Enterobacteriaceae and Vibrionaceae families.

For the effective identification of Vibrio colonies, following growth on agar,
the genus can be separated into two major groups based upon their ability to
utilize sucrose. Those colonies that are able to ferment sucrose form yellow
colonies, indicative of the possible presence of V. cholerae, V. alginolyticus,
or V^ fluvialis and green (sucrose-negative) colonies are observed when V. para-
haemolyticus, V. vulnificus or V. mimicus are present.

Once V. cholerae has been identified it is then important to establish the cor-
rect strain. For the identification of V. cholerae Ol or 0139 an agglutination
test is essential.

Blood agar appears to be a useful medium in the isolation, recognition and
identification of members of the family Vibrionaceae. Many Vibrio species
produce zones of (3-haemolysis on blood agar plates after overnight incuba-
tion (35-37°C). For V. cholerae, haemolysis of sheep red blood cells has been
traditionally used as one of several tests to distinguish the two biotypes of
V. cholerae Ol. The classic biotype is non-haemolytic while the El Tor biotype
is haemolytic; non-Ol V. cholerae strains are usually haemolytic.

Molecular diagnostic tests have now been developed for both clinical and
environmental monitoring of V. cholerae Ol and 0139. Polymerase chain
reaction (PCR), using primer pairs corresponding to the genes of the rfb com-
plex, which encode the O antigen, have been designed for the detection of Ol
(Hoshino et al. 9 1998) and 0139 (Albert et al, 1997). These primers have
been developed to detect V. cholerae from stool specimens. There are also spe-
cific probes for the detection of the A and B subunit genes of CT (Wright
et al. 9 1992). Molecular epidemiological techniques such as restriction frag-
ment length polymorphism of the enterotoxin gene have been used to study
outbreaks of V. cholerae strains (Yam et al., 1991).

Epidemiology of waterborne outbreaks

It seems that contaminated food is one of the most prevalent modes of trans-
mission for V. cholerae. If we consider the USA, most cases of cholera have

203

Bacteriology

been associated with the consumption of partially or undercooked seafood,
such as shellfish and oysters. However, water as both a direct and an indirect
means of transportation of V. cholerae is of great significance, particularly in
the developing world. Contaminated drinking water is a vector for V. cholerae
in areas of the world that do not practise drinking water disinfection, or
where treated water is susceptible to post-treatment contamination.

The source of Vibrio is usually the faeces of carriers or patients with
cholera. However, there are reported cases of cholera being acquired from the
natural water environments.

Cholera is considered an infection of over-crowding where poor standards
of hygiene are prevalent. Therefore the prevention of waterborne outbreaks of
V. cholerae could be brought about by good sanitation practices, including
protection of water resource quality, sewage treatment and effective treatment
of water supplies.

Risk assessment

Health effects: occurrence of illness, degree of morbidity and mortality, prob-
ability of illness based on infection:

• The most significant Vibrio pathogen is Vibrio cholerae, though there are
several other pathogenic species including Vibrio parahaemolyticus and
Vibrio vulnificus.

• V. cholerae causes infections ranging from asymptomatic to deadly. Illness
is marked by profuse watery diarrhoea. Sometimes there is a sudden onset
of vomiting, which leads to rapid and severe dehydration and possibly
death from electrolyte imbalance within 1-5 days.

• Some cholera species are associated with wound infections, bacteraemia
and meningitis.

• Illnesses occur rarely in developed countries.

Exposure assessment: routes of exposure and transmission, occurrence in
source water, environmental fate:

• The reservoirs of V. cholerae are asymptomatic human carriers and sick
people who shed the microorganisms in their faeces. Cholera is transmitted
through contaminated food or water. Person-to-person transmission is
unlikely because of the high infective dose.

• Infective doses of V. cholerae are high in healthy people. Reports of 10 8
organisms of classical V. cholerae in water produced diarrhoea in 50% of
adult volunteers.

• The ability of vibrios to survive in the environment varies. It is possible that
V. cholerae can survive for years in an aquatic environment, possibly in
biofilms. More information is needed.

204

Vibrio cholerae

• Vibrio cholerae has been found in surface and drinking water in areas
where the disease is endemic and has been shown to be viable from 1 hour
to 13 days in these environments. They survive better in saline water. They
can also survive for long periods in low-acid food.

Risk mitigation: drinking-water treatment, medical treatment:

• The vibrios are susceptible to chlorination, so countries with adequately
disinfected water supplies are not susceptible to outbreaks. The prevention
of waterborne outbreaks of V. cholerae depends on adequate sanitation
practices, including sewage treatment, protection of water resource quality,
and effective processing of water supplies.

• Oral administration of fluid and electrolytes is necessary for a person with
cholera. Adjunct antibiotic treatment can sometimes be helpful.

References

Albert, M.J. (1994).Vibno cholerae 0139.J Clin Microbiol, 32: 2345-2349.

Albert, M.J. et al. (1997). Rapid detection of Vibrio cholerae 0139 Bengal from stool spe-
cimens by PCR./ Clin Microbiol, 35: 1663-1665.

Baumann, P., Furniss, A.L. and Lee, J.V. (1984). Bergeys Manual of Systematic Bacteriology,
vol. 1, Krieg, N.R. and Holt, J.G. (eds). Baltimore: Williams & Wilkins, pp. 518-538.

Beazley, W.A. and Palmer, G.G. (1992). TCI - a new bile free medium for the isolation of
Vibrio species. Austr J Med Sci, 5: 25-27.

Cash, RA. et al. (1974). Response of man to infection with Vibrio cholerae. I. Clinical,
serologic, and bacteriologic responses to a known inoculum. / Infect Dis, 129: 45-52.

Colwell, R.R. and Huq, A. (1994). Vibrio cholerae and Cholera: Molecular to Global
Perspectives. Washington, DC: ASM Press, pp. 117-135.

Colwell, R.R. et al. (1994). Ecology of pathogenic vibrios in Chesapeake Bay. In Vibrios in
the Environment, Colwell, R.R. (ed.). New York: Wiley, pp. 367-387.

De, S.N. (1959). Enterotoxicity of bacteria-free culture nitrate of Vibrio cholerae. Nature,
183: 1533-1534.

Donovan, T.J. and van Netten, P. (1995). Culture media for the isolation and enumeration
of pathogenic Vibrio species in foods and environmental samples. Int J Food Microbiol,
26:77-91.

Feachem, R., Miller, C. and Drasar, B. (1981). Environmental aspects of cholera epidemi-
ology. II. Occurrence and survival of Vibrio cholerae in the environment. Trop Dis Bull,
78: 865-880.

Finkelstein, RA. and Lospalluto, J.J. (1969). Pathogenesis of experimental cholera: prepar-
ation and isolation of choleragen and choleragenoid. / Exp Med, 130: 185-202.

Hisatsune, K., Kondo, S. et al. (1993). O-antigenic lipopolysaccharide of Vibrio cholerae
0139 Bengal, a new epidemic strain for recent cholera in the Indian subcontinent.
Biochem Biophys Res Commun, 196: 1309-1315.

Honda, T. and Finkelstein, RA. (1979). Purification and characterization of a haemolysin
produced by V. cholerae biotype El Tor: another toxic substance produced by cholera
vibrios. Infect Immun, 26: 1020-1027.

Hood, MA., Ness, G.E. et al. (1984). Vibrios in the Environment. New York: John Wiley,
pp. 399-409.

Hornick, R.B., Music, S.I., Wenzel, R. et al. (1971). The broad street pump revis-
ited: response of volunteers to ingested cholera vibrio. Bull NY Acad Med, 47:
1181-1191.

205

Bacteriology

Hoshino, K. et al. (1998). Development and evaluation of a multiplex PCR assay for rapid
detection of toxigenic Vibrio cholerae Ol and 0139. FEMS Immunol Med Microbiol,
20:201-207.

Islam, M.S., Drasar, B.S. and Bradley, D.J. (1989). Attachment of toxigenic Vibrio cholerae
Ol to various freshwater plants and survival with a filamentous green alga,
Rbizoclonium fontanum. J Trop Med Hyg, 92: 396-401.

Kaper, J.B., Morris, J.G. Jr and Levine, M.M. (1995). Cholera. Clin Microbiol Rev,
8: 48-86.

Koch, R. (1884). An address on chlorea and its bacillus. Br Med J, August 30: 403-407,
453-459.

Kobayashi, T.S. et al. (1963). A new selective medium for vibrio group on a modified
Nakanishi's medium (TCBS agar medium). /#/? / Bacteriol, 18: 387-392.

Levine, M.M. et al. (1988). Volunteers studies of deletion mutants of Vibrio cholerae Ol
prepared by recombinant techniques. Infect Immun, 56: 161-167.

Martin, Y.P. and Bianchi, M.A. (1980). Structure, diversity and catabolic potentialities of
aerobic heterotropic bacterial population associated with continuous cultures of nat-
ural marine phytoplankton. Microb Ecol, 5: 265.

Mekalanos, J.J. (1983). Duplication and amplification of toxin genes in Vibrio cholerae.
Cell, 35: 253-263.

Mekalanos, J.J. (1985). Cholera toxin: genetic analysis, regulation, and role in pathogen-
esis. Curr Top Microbiol Immunol, 118: 97-118.

Nishibuchi, M. et al. (1992). Enterotoxigenicity of Vibrio parahaemolyticus with and with-
out genes encoding thermostable direct hemolysin. Infect Immun, 60: 3539-3545.

Ottemann, K.M. and Mekalanos, J.J. (1994). Vibrio cholerae and Cholera: Molecular to
Global Perspectives. Washington, DC: ASM Press, pp. 177-187.

Pacini, F. (1854). Osservazionemicroscopiche e deduzioni pathologiche sul Cholera
Asiatico. Gaz Med Ital Toscana Firenza, 6: 405-412.

Parsot, C. and Mekalanos, J.J. (1990). Expression of ToxR, the transcriptional activator of
the virulence factors in Vibrio cholerae, is modulated by the heat shock response. Proc
Natl Acad Sci USA, 9898-9902.

Pesian, T.P. (1965). Studies on the viability of El Tor vibrios in contaminated foodstuffs,
fomites and in water. In Proceedings of the cholera research symposium. PHS PUB,
1328: 317-332.

Rader, A.E. and Murphy, J.R. (1988). Nucleotide sequences and comparison of the
hemolysin determinants of Vibrio cholerae El Tor RV79(Hly+) and RV79(Hly— ) and
Classical 569B(Hly-). Infect Immun, 56: 1414-1419.

Saha, P.K. and Nair, G.B. (1997). Production of monoclonal antibodies to the non-
membrane damaging cytotoxin (NMDCY) purified from Vibrio cholerae 026 and
distribution of NMDCY among strains of Vibrio cholerae and other enteric bacteria
determined by monoclonal-polyclonal sandwich enzyme-linked immunosorbent assay.
Infect Immun, 65: 801-805.

Sarker, R.I., Ogawa, W. et al. (1994). Characterization of a glucose transport system in
Vibrio parahaemolyticus. J Bacteriol, 176: 7378-7382.

Shimada, T.E. et al. (1993). Outbreaks of Vibrio cholerae non-Ol in India and Bangladesh.
Lancet, 341: 1347.

Shimada, T., Sakazaki, R. et al. (1990). A new selective, differential agar medium for isol-
ation of Vibrio cholerae Ol: PMT (polymyxin-mannose-tellurite) agar. Jpn J Med Sci
Biol, 43: 37-41.

Spangler, B.D. (1992). Structure and function of cholera toxin and the related Escherichia
coli heat-labile enterotoxin. Microbiol Rev, 56: 622-647.

Takeda, T. et al. (1991). Detection of heat-stable enterotoxin in a cholera toxin gene-
positive strain of V. cholerae Ol. FEMS Microbiol Lett, 80: 23-28.

Tamplin, M.L. et al. (1987). Sodium channel inhibitors produced by enteropathogenic
Vibrio cholerae and Aeromonas hydrophila. Lancet, i: 975.

Taylor, R.K., Miller, V.L. et al. (1987). Use of phoA gene fusions to identify a pilus colon-
ization factor coordinately regulated with cholera toxin. Proc Natl Acad Sci USA,
84:2833-2837.

206

Vibrio cholerae

Waldor, M.K., Colwell, R.R. and Mekalanos, J.J. (1994). The Vibrio cholerae 0139
serogroup antigen includes an O -antigen capsule and lipopolysaccharide virulence
determinants. Proc Natl Acad Sci USA, 91: 11388-11392.

Wright, A.C. et al. (1992). Development and testing of a nonradioactive DNA oligonu-
cleotide probe that is specific for Vibrio cholerae cholera toxin. / Clin Microbiol,
30: 2302-2306.

Yam, W.C. et al. (1991). Restriction fragment length polymorphism analysis of Vibrio
cholerae strains associated with a cholera outbreak in Hong Kong. / Clin Microbiol,
28: 1058-1059.

Yamai, S. et al. (1997). Distribution of serogroups of Vibrio cholerae non-Ol non-0139
with specific reference to their ability to produce cholera toxin and addition of novel
serogroups. J Jap Assoc Infect Dis, 71: 1037-1045.

207

Yersinia

Basic microbiology

Yersinia comprises three important species of human significance. Each one is
essentially an animal parasite known to infect man. Yersinia enter ocolitica is
found in wild and also domestic animals and known to cause gastroenteritis.
Yersinia pseudotuberculosis is a parasite of rodents and is known to infect
man occasionally. Yersinia pestis is the agent responsible for the plague.

Yersinia are oxidase-negative, catalase-positive, straight, Gram-negative
rods (or coccobacilli) often measuring 0.8-3.0 fjim by 0.8 |xm. They are facul-
tative anaerobes, predominantly mesophilic though all exhibit growth at low
temperatures (e.g. 0-4°C). Yersinia are facultative anaerobes that ferment
glucose in addition to other sugars without the production of gas. Yersinia are
motile at 22-30°C, but not at 37°C.

As Yersinia possess the ability to grow under extreme ranges in tempera-
ture they are well-adapted to survival in the environment. The species of
Yersinia of most significance to water transmission is Yersinia enteroco-
litica which will be dealt with in this chapter primarily as it is excreted
in faeces. Information on Y. pseudotuberculosis is also documented as a
comparison.

Bacteriology

Origins

Yersinia, derived from the French bacteriologist Alexander Yersin, was first
isolated in Hong Kong in 1894 (Gyles and Thoen, 1993). Yersinia was not
isolated in the USA until 1923 and not recognized as a human pathogen until
its first human case in 1963. To date Yersinia is responsible for about 1% of
acute cases of gastroenteritis in Europe and parts of America.

Initially members of the genus Yersinia were included in the genus Fasteurella.
Yersinia was removed from this group in the late 1960s despite requests to
change this in 1954 (Bercovier and Mollaret, 1984).

To date the genus Yersinia is now classified as genus XI of the family
Enterobacteriaceae. Included in the genus Yersinia are three significant human
pathogens, Y pestis, Y pseudotuberculosis, after inoculation of guinea-pigs with
material isolated from a skin lesion of a child who died of meningitis, and

Y enterocolitica. Y enterocolitica was first described in 1939 under the name
Bacterium enter ocoliticum (Cover and Aber, 1989), later Fasteurella pseudo-
tuberculosis rodentium and then Fasteurella X and finally Y enter ocolitica in
1964 (Bottone, 1977).

Yersinia is classified into 11 species, of these Y enter ocolitica and

Y pseudotuberculosis are human pathogens, whereas the other species, with
the exception of Y ruckeri which is a fish pathogen, are environmental, non-
pathogenic organisms (Romalde, 1993).

Yersinia enterocolitica possess 50 serotypes of which 03, 08 and 09 are
the most significant to humans. Y enterocolitica was initially divided into five
biogroups based on a number of biochemical reactions namely, indole pro-
duction, hydrolysis of aesculin and salicin, lactose oxidation, acid from
xylose, trehalose, sucrose, sorbose and sorbitol, o-nitrophenyl-(3b-d-galactopyra-
noside (ONPG), ornithine decarboxylase, Voges-Proskauer reaction and nitrate
reduction (Wauters, 1981). Following modification of the classification of Yer-
sinia there are now six biotypes. These include biogroup 1 which was divided
into two groups, namely Group 1A, which include the environmental strains that
lack virulence plasmids and pyrazinamidase, aesculin and p-d-glucosidase posi-
tive. Biogroup IB are found to be aesculin, pyrazinamidase and (3-d-glucosidase
negative and belong to one of the following serogroups: 0:4, 0:8, 0:13, 0:18,
O:20orO:21.

Metabolism and physiology

Yersinia are facultative anaerobes. Acid, but not gas, is produced from ^-glucose
and polyhydroxylalcohols. Fructose, galactose, maltose, mannitol, mannose,
N-acetylglucosamine and trehalose are fermented. Y enterocolitica produces
acetoin when incubated at 28°C but not at 37°C and produces catalase, but
not oxidase.

210

Yersinia

Clinical features

The most common clinical presentation in Y enter ocolitica infections, but
rare in Y pseudotuberculosis infections, is acute enteritis. Y. enter ocolitica is
also known to cause severe gastroenteritis. Other disease presentations less
frequently associated with infection by these agents include systemic disorders,
focal abscesses of liver, spleen, kidney and lung, erythema nodosum, polyarthri-
tis, Reiter's syndrome, myocarditis, pneumonia, meningitis, conjunctivitis,
septicaemia, pustules, osteomyelitis; and local manifestations such as cellulitis
and wound infections and panophthalmitis (Bottone, 1992) with endocarditis
(Urbano-Marquez et al., 1983), pericarditis (Lecomte et al., 1989) and osteitis
(Fisch et al. 9 1989). In children infection symptoms often include fever, diar-
rhoea, abdominal pain and vomiting.

Y pseudotuberculosis infections in humans present as severe typhoid-like
illness often with purpura, fever and enlargement of the spleen and liver. It is
frequently a cause of mesenteric lymphadenitis which often simulates sub-
acute appendicitis and is most common in children and young adults. Terminal
ileitis is infrequently seen in Y. pseudotuberculosis infection, but is very com-
mon and often severe in Y enter ocolitica infections. Cases of fatal septicaemia
caused by Y pseudotuberculosis have been reported.

Y enter ocolitica have been reported as a special hazard when blood donors
have a bacteraemia. This is due to the fact that the organisms can grow dur-
ing cold temperature storage in blood and cause severe infection in transfu-
sion recipients (Arduino et al., 1989).

Pathogenesis and virulence

Pathogenicity of Yersinia species is aided by a 70-75 kb plasmid (pYV) and
two additional plasmids. These plasmids are known to control four major viru-
lence factors in this genus. These factors include the excreted antiphagocytic
proteins (Yops), proteins involved in processing and excretion of the Yops
(Ysc), regulatory proteins (Lcr) and adhesin/invasin proteins. Enterotoxins
are also thought to play a role in virulence, and are documented as causing
disease in humans.

In virulent Y enter ocolitica, the presence of a 72 kb low calcium response
plasmid is very important. Other plasmids in Y enter ocolitica encode for eight
or more Yersinia outer-membrane proteins. Of significance are the Yops, protein
1 (Tl'), which is associated with resistance to killing by serum, hydrophobicity,
autoagglutination in fluid media, and production of a fibrillar adhesin which
enhances attachment of the bacteria to epithelial cells of the host. Many of the
other Yops function to inhibit phagocytosis by mammalian hosts (Forsberg
et al., 1994). Yops produced by Y enterocolitica include YadA, Yops B, C, D, E
and H. As for Y pseudotuberculosis, YopE and YopH are essential for full

211

Bacteriology

virulence (Sodeinde et aL, 1988; Brubaker, 1991). Chromosomally generated
virulence factors include an invasin and a second factor that allows specific
invasion of various cell types (Miller and Falkow, 1988). Researchers have
established that M cells in the Peyer's patches are the primary target cells
of invading yersiniae, principally Y. enter ocolitica. Y. enter ocolitica and
Y. pseudotuberculosis colonization of these surfaces and invasion through
mucosal surfaces is based principally on toxin Yst. This toxin is possibly
responsible for the production of diarrhoea but to date it has not been deter-
mined. An exotoxin, YPM, has also been identified from Y. pseudotuberculosis.
This toxin acts as a superantigen by activation of specific human T cells in the
presence of immune cells bearing MHC class II molecules. Another invasion
factor in Yersinia aside from invasin is Ail. Ail (product of the Ail gene) is of
significance during adherence, invasion and serum resistance in Y. enterocolit-
ica (Wachtel and Miller, 1995). Invasin and YadA bind to integrins in intes-
tinal tissue (Pepe and Miller, 1993; Skurnik et al., 1994). YadA has also been
shown to bind to fibronectin (Tertti et al., 1992).

Once Yersinia has colonized the mucosal surfaces it has to avoid phago-
cytosis. The ability to avoid phagocytosis is under the control of basically 11
Yops (Straley et al., 1993). Yops function by interfering with signal transduc-
tion of the phagocytes or directly attacking the host cells. This will inhibit
phagocytosis by macrophages (Straley et al., 1993).

Yersinia have been found to produce both a siderophore, which is active at
low temperature, and an iron storage system which is induced at 37°C (Gyles
and Thoen, 1993).

Yersinia species produce a fibrillar protein, specifically in pathogenic
serotypes of Y. enter ocolitica. This protein is called Myf and is composed
of MyfA (a 21kDa protein), MyfB (a chaperone) and MyfC (an outer-
membrane protein) (Cornelis, 1994; Iriarte and Cornelis, 1995). It is probable
that these proteins, together with Yst, aid in the adhesion of Yersinia to a
host cell.

Another virulence mechanism of Yersinia is possibly the production of
stress proteins that help in its survival within phagocytes. It is probable that
the stress proteins neutralize toxic products produced by host cells.

Treatment

Y. enter ocolitica is found to be sensitive to many antibiotics. These include,
and are by no means exhaustive, aminoglycosides, chloramphenicol and tetra-
cycline. Treatment is only given with severe infections and tetracycline is usu-
ally the antibiotic of choice. Penicillin is not used to treat Y. enter ocolitica
infection as strains are resistant. Treatment of Y pseudotuberculosis septi-
caemia requires the use of tetracycline or ampicillin.

212

Yersinia

Environment

Both Y enter ocolitica and Y pseudotuberculosis can survive for long periods
in the environment due to their requirements for minimal nutrition and their
ability to remain metabolically active at extremes of temperature. Most of the
environmental isolates of Y. enter ocolitica are not pathogenic and are often
called atypical Y. enterocolitica-like organisms (Y. intermedia, Y frederiksenii
and Y. kristensenii) (Lassen, 1972; Kapperud, 1977; Langeland, 1983; Aleksic
and Bockem, 1988; Hellberg and Lofgren, 1988). Little is truly known about
the occurrence and survival of Y enter ocolitica in the environment and
research on its ecology is needed. Work by Dominowska and Malottke (1971)
studied the survival of Y enter ocolitica in water and found that the average
time for the survival in unfiltered surface water was 38 days in spring and
7 days in summer. Conversely in filtered water it was found that the bacteria
survived for 197 days in spring and 184 days in the summer. In the laboratory,
Yersinia was found to survive, following an initial incoculum of 10 3 cfu/ml,
for about 7 days in tap water and for 28 days in lake water. Back in 1972,
Y enter ocolitica was isolated from 10 of 50 drinking water supplies in
Norway (Lassen, 1972). They have also been isolated from wells (Schiemann,
1978), lakes and streams (Kapperud, 1977). Y enter ocolitica has also been
isolated from soil in California and Germany (Botzler, 1979, 1987). Most of
the German strains were not serotypable using serotyping reagents available
at the time; the one Y enter ocolitica that could be serotyped was 0:6,33,
not commonly involved in human infection. The California soil isolates
were of serotypes 0:4,32; 0:5, 0:17 and O:20, also not associated in human
infections.

Wild animals such as foxes, hares and shrews form the natural reservoir of
Yersinia enter ocolitica. Domestic animals such as sheep, cattle, pigs and dogs
have also been shown to harbour the bacterium. As with many water organisms
oysters and mussels have been shown to harbour Yersinia enter ocolitica.

Y pseudotuberculosis is found in a large array of mammalian and avian
hosts. Many of these infections are subclinical or chronic; the infected hosts
shed the organisms into the environment over long periods of time. A single
report has incriminated water as the source of infection of more than 200
humans with Y pseudotuberculosis serotypes lb and 4b in Japan (Fukushima
etaU 1989).

Water

Y enter ocolitica are quite common contaminants of water supplies. The presence
of Y enter ocolitica in drinking water, within the USA, was first reported by
Botzler and colleagues (Botzler et al., 1976). Lassen (1972) and Wauters (1972)
were the first to document cases in Europe. Y enter ocolitica has also been

213

Bacteriology

isolated in wells (Highsmith et aL 9 1977). In this study the strains that were
serotypable (5 of 14 isolates) were of types not commonly involved in human dis-
ease. Wetzler et al. (1978) reported numerous isolates from a variety of water
sources and wastewaters in Washington State.

Methods of detection

Yersinia are able to grow at temperatures ranging from 4 to 43° C and over a
pH range of 4-10, although the optimum pH is 7.2-7.4. Y pseudotuberculo-
sis and Y enter ocolitica will grow in up to 5% salt and all species can grow
on nutrient agar. Haemolysis is not observed when Yersinia is grown on blood
agar and most Yersinia will grow on MacConkey medium, although growth
of Y pseudotuberculosis is variable. Y pseudotuberculosis will grow as small
grey to black colonies on tellurite medium.

Infection by Yersinia is diagnosed by isolating the organism from lymph
nodes, blood or faeces on blood or MacConkey's agar. Cold enrichment of
cultures may enhance the isolation rates for Y pseudotuberculosis in heavily
contaminated specimens. This may be accomplished in tetrathionate broth or
selenite F broth or simply in physiological saline or a nutrient broth, all of
which are kept in the refrigerator at 4°C for up to 6 weeks with periodic sub-
culturing for recovery of the agents. Most isolates of the enteropathogenic
yersiniae grow well on MacConkey agar and eosin-methylene blue agar; their
growth is often inhibited or delayed on Salmonella-Shigella agar and deoxy-
cholate agar.

Among the more commonly used selective media for rapid identification of
the enterocolitica group of yersiniae are CIN agar (Schiemann, 1979) and
VYE agar (virulent Y enterocolitica agar) (Fukushima, 1987). Specimens not
grossly contaminated on an enriched medium, such as blood agar or brain-
heart infusion agar may facilitate recovery of Yersinia. Recovery (especially of
virulent organisms) may also be enhanced if incubation after the first 24 hours
is conducted at room temperature (24-28°C).

Identity of Yersinia is determined by biochemical test and the serotype is
confirmed by slide agglutination with rabbit antisera. Y enterocolitica and
Y pseudotuberculosis are motile, with peritrichous or paripolar flagella, when
grown at 22°C, but not at 37°C. It has been demonstrated biochemically and
by visualization under the electron microscope that at room temperature
strains of Y enterocolitica could go through a transition to a spheroplast type
L-form which then could revert back to an irregular-shaped structure with an
intact cell wall (Pease, 1979).

All species of Yersinia have a requirement of iron for growth. Virulent
strains of Y enterocolitica and Y pseudotuberculosis also have a requirement
for a low concentration of calcium in order to express some of the antigens
involved in the virulence of these species. The concentration of calcium in the

214

Yersinia

medium is critical for the expression of virulence factors of Y pseudotubercu-
losis and Y enter ocolitica.

Epidemiology

The primary method of transmission for Y enter ocolitica is via the faecal-oral
route but it may also occasionally occur via contaminated water sources
(Lassen, 1972; Saari and Quan, 1976; Saari and Jansen, 1977; Wetzler et al. 9
1978) and contaminated meats ( Aulisio et al. 9 1983) and poultry (DeBoer et al. 9
1982). Uncommon infections have resulted from the transfusion of contamin-
ated units of blood from bacteraemic, though apparently healthy, donors
(Jacobs et al. 9 1989). Several outbreaks of Y. enter ocolitica infections have
been traced to the consumption of pasteurized milk (Aulisio et al. 9 1983;
Shayegani et al. 9 1983; Tacket et al. 9 1984) and chocolate milk (Black et al. 9
1978) as well as of raw milk and cheese and contaminated foods (Aulisio etal. 9

1983) and meats (Doyle et al. 9 1981; Shayegani etal. 9 1983). Several outbreaks
of gastrointestinal illness have been traced to the ingestion of contaminated
foods such as raw and pasteurized milk (Moustafa et al. 9 1983; Tacket et al. 9

1984) and tofu (Aulisio et al. 9 1983).

Person-to-person transmission has not been well documented; the major
mode of transmission does seem to be the faecal-oral route. Domestic pets are
being increasingly associated with the occurrence of human disease (Gutman
et al. 9 1973; Fantasia et al. 9 1985; Fukushima et al. 9 1988).

Y enter ocolitica strains associated with human disease are mainly of serotypes
3 and 9. This is presently the case in Europe. In Japan and the USA other
serotypes are common.

Risk assessment

Health effects: occurrence of illness, degree of morbidity and mortality, prob-
ability of illness based on infection:

• Yersinia enter ocolitica causes gastroenteritis and is the most significant
Yersinia species related to water transmission.

• Yersinia is responsible for about 1% of acute cases of gastroenteritis in
Europe and parts of America.

• Y enter ocolitica causes mainly acute enteritis, but systemic infections, such
as bacteremia, joint pain, and rashes have occasionally resulted. It can also
cause inflammation of the mesenteric lymph glands, which is often con-
fused with appendicitis. The most common symptoms include fever, diar-
rhoea and abdominal pain.

215

Bacteriology

• Young children are most likely to become infected and ill, though pseudoap-
pendicitis is more often seen in older children and adults.

Exposure assessment: routes of exposure and transmission, occurrence in
source water, environmental fate:

• Y. enter ocolitica is spread through the faecal-oral route - most often by con-
taminated food - and sometimes by water. It is rarely transmitted person-to-
person. Contact with animals may be a route of transmission as well.

• Many wild and domestic animals are reservoirs for Y. enter ocolitica.

• Y. enter ocolitica can survive a long time in the environment, because it is
resistant to the extremes of temperature.

• The organisms have been found in surface and groundwater sources, but
little overall is known about their occurrence in the environment. Those
found in drinking water sources have mostly been strains non-pathogenic
to humans.

Risk mitigation: drinking-water treatment, medical treatment:

• Yersinia seems to be as susceptible to chlorine as £. coli, so adequately dis-
infected water supplies should control Y. enter ocolitica.

• Uncomplicated cases of diarrhoea due to Y. enter ocolitica are usually self-
limiting. However, in more severe infections, treatment may be required,
and Y. enter ocolitica is sensitive to many antibiotics. Tetracycline is usually
the antibiotic of choice.

References

Aleksic, S. and Bockem, H.L.J. (1988). Serological and biochemical characteristics of 416

Yersinia strains from well water and drinking-water plants in the Federal Republic of

Germany: lack of evidence that these strains are of public health importance. Zentralbl

Bakteriol Mikrobiol Hyg Reiche B, 185: 527-533.
Arduino, M.J., Bland, L.A. et al. (1989). Growth and endotoxin production of Yersinia

enterocolitica and Enterobacter agglomerans in packed erythrocytes. / Clin Microbiol,

27: 1483-1485.
Aulisio, C.C.G., Stanfleld, J.T. et al. (1983). Yersinioses associated with tofu consumption:

serological, biochemical, and pathogenicity studies of Yersinia enterocolitica isolates.

J FoodProt, 46:226-230.
Bercovier, H. and Mollaret, H.H. (1984). Bergey's Manual of Systematic Bacteriology,

vol. 1, Krieg, N.R. and Holt, J.G. (eds). Baltimore: Williams & Wilkins, p. 498.
Black, R.E., Jackson, R.J. et al. (1978). Epidemic Yersinia enterocolitica infection due to

contaminated chocolate milk. New Engl J Med, 298: 76-79.
Bottone, E.J. (1977). Yersinia enterocolitica: a panoramic view of a charismatic micro-
organism. Crit Rev Mir o bio I, 5: 211-41.
Bottone, E.J. (1992). The genus Yersinia (excluding Yersinia pestis). In The Prokaryotes,

2nd edn, Balows, A., Triipper, H.G. et al. (eds). New York: Springer- Verlag,

pp. 2862-2887.
Botzler, R.G. (1979). Yersiniae in the soil of an infected wapiti range. / Wildl Dis, 15:

529-532.

216

Yersinia

Botzler, R.G. (1987). Isolation of Yersinia enter ocolitica and Y. frederiksenii from forest

soil, Federal Republic of Germany. / Wildl Dis, 23: 311-313.
Botzler, R.G., Wetzler, T. and Cowan, A.B. (1976). Yersinia enter ocolitica and Yersinia-like

organisms isolated from frogs and snails. Bull Wildl Dis Assoc, 4: 110-115.
Botzler, R.G., Wetzler, T. et al. (1976). Yersiniae in pond water and snails. / Wildl Dis, 12:

492-426.
Brubaker, R.R. (1991). Factors promoting acute and chronic diseases caused by yersiniae.

Clin Microbiol Rev, 4: 309-324.
Cornelis, G.R. (1994). Yersinia pathogenicity factors. Curr Topics Microbiol Immunol,

192: 243-263 .
Cover, T.L. and Aber, R.C. (1989). Yersinia enter ocolitica. New Engl J Med, 321: 16-24.
DeBoer, E., Hartog, B.J. and Oosterom J. (1982). Occurrence of Yersinia in poultry prod-
ucts. / Food Prot, 45: 322-325.
Dominowska, C. and Matlotte, R. (1971). Survival of Yersinia in water samples originat-
ing from various sources. Bull Inst Marine Med Gdansk, 22: 173-182.
Doyle, M.P., Hugdahl, M.B. and Taylor, S.L. (1981). Isolation of virulent Yersinia entero-

colitica from porcine tongues. Appl Environ Microbiol, 42: 661-666.
Fantasia, M., Grazia Mingrone, M. et al. (1985). Isolation of Yersinia enter ocolitica bio-
type 4 serotype 03 from canine sources in Italy./ Clin Microbiol, 22: 314-315.
Fisch, A., Prazuck, T. et al. (1989). Hematogenous osteitis due to Yersinia enter ocolitica.

] Infect Dis, 160:554.
Forsberg, A., Rosqvist, R. and Wolf-Watz, H. (1994). Regulation and polarized transfer

of the Yersinia outer proteins (Yops) involved in antiphagocytosis. Trends Microbiol,

2: 14-19.
Fukushima, H. (1987). New selective agar medium for isolation of virulent Yersinia ente-

rocolitica. J Clin Microbiol, 25: 1068-1073.
Fukushima, H., Gomyoda, M. et al. (1988). Yersinia pseudotuberculosis infection

contracted through water contaminated by a wild animal. / Clin Microbiol, 26:

584-5S5.
Fukushima, H., Gomyoda, M. et al. (1989). Cat-contaminated environmental substances

lead to Yersinia pseudotuberculosis infection in children. / Clin Microbiol, 27:

2706-2709.
Gutman, L.T., Wilfert, CM. and Quan, T.J. (1973). Susceptibility of Yersinia enterocolit-

ica to trimethoprim-sulfamethoxazole./ Infect Dis, 128S: 538.
Gyles, C.L. and Thoen, CO. (1993). Pathogenesis of bacterial infections in animals, 2nd

edn. Ames, Iowa: Iowa State University Press, pp. 226-235.
Hellberg, B. and Lofgren, S. (1988). Gastrointestinal disturbances due to an established

bacterial contamination of a water distribution system. Vatten, 44: 277-281.
Highsmith, A.K., Feeley, J.C et al. (1977). Isolation of Yersinia enter ocolitica from well

water and growth in distilled water. Appl Environ Microbiol, 34: 745-750.
Iriarte, M. and Cornelis, G.R. (1995). MyfF, an element of the network regulating the syn-
thesis of fibrillae in Yersinia enter ocolitica.] Bacteriol, 177: 738-744.
Jacobs, J., Jamaer, D. et al. (1989). Yersinia enter ocolitica in donor blood: a case report and

review./ Clin Microbiol, 17: 1119-1121.
Kapperud, G. (1977). Yersinia enter ocolitica and Yersinia-like microbes isolated from

mammals and water in Norway and Denmark. Acta Pathol Microbiol Scand: Section B:

Microbiol Immunol, 85: 129-135.
Langeland, G. (1983). Yersinia enter ocolitica and Yersinia enterocolitica-like bacteria in

drinking-water and sewage sludge. Acta Pathol Microbiol Scand B, 91: 179-185.
Lassen, J. (1972). Yersinia enter ocolitica in drinking water. Scand] Infect Dis, 4: 125-127.
Lecomte, F., Eustache, M. et al. (1989). Purulent pericarditis due to Yersinia enter ocolitica.

] Infect Dis, 159:363.
Miller, V.L. and Falkow, S. (1988). Evidence for two genetic loci in Yersinia enter ocolitica

that can promote invasion of epithelial cells. Infect Immun, 56: 1242-1248.
Moustafa, M.K., Ahmed, A.A.-H. and Marth, E.H. (1983). Occurrence of Yersinia entero-

colitica in raw and pasteurized milk./ Pood Prot, 46: 276-278.

217

Bacteriology

Pease, P. (1979). Observations on L-forms of Yersinia enter ocolitica. J Med Microbiol,

12: 337.
Pepe, J.C. and Miller, V.L. (1993). Yersinia enter ocolitica invasin: a primary role in the ini-
tiation of infection. Proc Natl Acad Sci USA, 90: 6473-6477.
Saari, T.N. and Jansen, G.P. (1977). Waterborne Yersinia enter ocolitica in Wisconsin.

Abstr Bacteriol Proc, New Orleans.
Saari, T.N. and Quan, T.J. (1976). Waterborne Yersinia enter ocolitica in Colorado. Abstr

Bacteriol Proc, Atlantic City.
Schiemann, D.A. (1978). Isolation of Yersinia enter ocolitica from surface and well water in

Ontario. Can J Microbiol, 24: 1048-1052.
Schiemann, D.A. (1979). Synthesis of selective agar medium for isolation of Yersinia ente-

rocolitica. Can J Microbiol, 25: 1298-1304.
Shayegani, M., Morse, D. et al. (1983). Microbiology of a major foodborne outbreak of

gastroenteritis caused by Yersinia enter ocolitica serogroup 0:8. / Clin Microbiol, 17:

35-40.
Skurnik, M., Tahir, Y. et al. (1994). YadA mediates specific binding of enteropathogenic

Yersinia enter ocolitica to human intestinal submucosa. Infect Immun, 62: 1252-1261.
Sodeinde, O.A., Sample, A.K. et al. (1988). Plasminogen activator/coagulase gene of

Yersinia pestis is responsible for degradation of plasmid encoded outer membrane pro-
teins. Infect Immun, 56: 2743-2748.
Straley, S.C., Skrzypek, E. and Piano, G.V.E. (1993). Yops of Yersinia spp. pathogenic for

humans. Infect Immun, 61: 3105-3110.
Tacket, CO., Narain, J.P. et al. (1984). A multistate outbreak of infections caused by

Yersinia enter ocolitica transmitted by pasteurized milk. JAMA, 251: 483-486.
Tertti, R., Skurnik, M. et al. (1992). Adhesion protein YadA of Yersinia species mediates

binding of bacteria to fibronectin. Infect Immun, 60: 3021-3024.
Urbano-Marquez, A., Estruch, R. et al. (1983). Infectious endocarditis due to Yersinia

enter o colitica. ] Infect D is, 148: 940.
Wachtel, M.R. and Miller, V.L. (1995). In vitro and in vivo characterization of an ail

mutant of Yersinia enter ocolitica. Infect Immun, 63: 2541-2548.
Wauters, G. (1972). Souches de Yersinia enter ocolitica isolees de Peau. Rev Perm Ind

Ailment, 7: 18.7.
Wauters, G. (1981). Antigens of Yersinia enter ocolitica. In Yersinia enter ocolitica. Boca

Raton: CRC Press, pp. 41-53.
Wetzler, T.F., Rea, J.T. et al. (1978). Yersinia enter ocolitica. in waters and waste waters.

Presented at 106th Annual Meeting, American Public Health Association, Los Angeles,

CA, October 18, 1978.

218

Part 3

Protozoa

Acanthamoeba spp

Basic microbiology

Several species of the free-living amoebae Acanthamoeba (Phylum Sarco-
mastigophora, Order Amoebida, Family Acanthamoebidae), which are com-
monly found in soil and water, have been implicated in human infections and
disease, including Acanthamoeba culbertsoni, Acanthamoeba polyphaga,
Acanthamoeba castellanii, Acanthamoeba astronyxis, Acanthamoeba hatch-
etti, Acanthamoeba griffini, Acanthamoeba lugdenensis and Acanthamoeba
rhysodes (Bottone, 1993). These have been divided into three morphological
groups, I to III (Pussard and Pons, 1977). However, the taxonomy of the genus
is uncertain since isoenzymatic and genetic assays have shown similarities
among strains traditionally assigned to different species on the basis of morph-
ology (Kilvington and Beeching, 1996). Cyst morphology has been shown to
vary even within clonal populations of the species (Kilvington et al. 9 1991).
Additionally, the pathogenicity of some species is poorly established. Those
mainly associated with human disease are A. culbertsoni, which causes the
rare, but fatal granulomatous amoebic encephalitis (GAE) and A. polyphaga
and A. castellanii which are more usually associated with infections (keratitis)
of the eye (Visvesvara and Stehr-Green, 1990). GAE is not regarded as water-
related while keratitis is.

Protozoa

The life cycle of Acanthamoeba spp. is commonly referred to as 'simple',
comprising a feeding and dividing trophozoite stage and a resistant cyst stage
(Visvesvara, 1991). These amoebae favour aquatic habitats such as water, mud
and soil (Anon, 1989) where the active, aerobic trophozoites feed on bacteria,
fungi and other protozoa. The slender trophozoites (25-40 fjim) possess spine-
like processes called acanthopodia, which facilitate a slow gliding movement.
They divide by binary fission using a mitotic process, and do not possess a
flagellate stage. In adverse environments, including anaerobic conditions, the
trophozoites form dormant, resistant cysts. Cysts are wrinkled, double walled
and of a variety of shapes, measuring 10-30 (xm in diameter. The cyst comprises
an outer ectocyst and an inner endocyst, linked by a pore with a mucoid plug.

The transmission and epidemiology of disease caused by Acanthamoeba, par-
ticularly keratitis, is largely driven by host behaviour, mainly from improper
use and disinfection of contact lenses.

Origin of the organism

Acanthamoeba spp. were originally ascribed to the genus Hartmannella, but
subsequent studies of both the cyst and trophozoite forms showed distinct dif-
ferences between the two genera. Additionally, Hartmannella spp. are not patho-
genic to humans, whereas the pathogenicity of Acanthamoeba for humans
was suggested following experimental animal infections in 1958 (Culbertson
etal. 9 1958).

GAE was first recognized in humans in 1971 (Kenney, 1971), since when over
100 cases have been reported world-wide, mainly in patients immunosup-
pressed by chemotherapy, alcohol abuse, chronic disease and GAE has been
recorded as the primary cause of death in AIDS patients (Martinez, 1991). GAE
is thus regarded as a rare, fatal, opportunistic infection of immunocompromised
hosts. The main risk to humans from Acanthomoeba in terms of numbers of
cases is from keratitis. The first case of acanthamoeba keratitis was diagnosed
in a Texan farmer in 1973 (Jones et al. 9 1975). Suffering from ocular trauma
caused by straw fragments, he used tap water to rinse the affected eye. Two
further cases were reported in the UK (Nagington et al. 9 1974) and the condi-
tion was regarded as a rare opportunistic infection resulting from injury to the
eye (Ma et al. 9 1981) until the mid-1980s when increased reports prompted
examination and review of previous cases in the USA and UK. Of 208 cases
reported in the USA between 1973 and 1988, three-quarters occurred from 1985
(Visvesvara and Stehr-Green, 1990). Details of 189 cases were available and
showed that 160 (85%) were in contact lens wearers, particularly those using
soft contact lenses. A review of 72 consecutive cases between 1984 and 1992 in
the UK showed that 64 patients were contact lens wearers, 28 of which wore
disposable lenses (Bacon et al. 9 1993). This identified the need for better edu-
cation of contact lens wearers regarding cleaning and disinfection practices.

222

Acanthamoeba spp.

Clinical features

Acanthamoeba spp. cause two distinct diseases in humans: GAE affects the
central nervous system and keratitis affects the cornea of the eye. GAE has a
sudden onset and can last from 8 days to several months, invariably resulting
in death. Symptoms of GAE include fever, headache, seizures, meningitis and
visual abnormalities (Martinez, 1991).

A. polyphaga and A. castellanii have been associated with chronic granulo-
matous lesions of the skin, with or without secondary invasion of the CNS,
and infection of the eye (conjunctivitis) and cornea (keratoconjunctivitis).
Acanthamoeba keratitis is characterized by intense pain and ring-shaped infil-
trates in the corneal stroma, which can progress to hypopyon, scleritis, glaucoma
and cataract formation (Bacon et aL, 1993). Corneal perforation may occur.
One eye only is usually affected, although bilateral disease has been reported
(Auran et aL, 1987). As recognition of the disease is heightened, early diag-
nosis is prompted when infiltration of the superficial epithelium only may be
involved (Larkin et al., 1992). Symptoms may be non-specific and falsely
diagnosed as herpes simplex virus infection (Auran et al.> 1987).

Pathogenicity and virulence

Acanthamoeba are widespread in the environment and hence antibodies are
common in human sera (Bottone, 1993). Parasites have been found in the respira-
tory tract of healthy people, and subclincial, self-limiting infection may be
common in immune-competent hosts (Martinez, 1993). Immunity probably
involves both cell-mediated and humoral systems, and skin lesions caused
by Acanthamoeba may progress to invasion of the brain and meninges of
immunocompromised individuals and occasionally immune intact people,
causing GAE (Bottone, 1993; Martinez, 1993). The lower respiratory tract
may be a route of invasion through inhalation and adherence to mucosal sur-
faces. The trophozoites probably reach the meninges by haematogenous
spread from the site of primary infection. Although cases of GAE have been
reported in immunocompetent hosts, immunosuppressed individuals are at
greater risk, including those undergoing chemotherapy, AIDS patients, drug
abusers and alcoholics.

Infection with A. polyphaga and A. castellanii can occur in healthy people
and rarely cause GAE but, more commonly, leads to conjunctivitis and kerato-
conjunctivitis, the latter resulting in blindness. These infections occur primarily
in wearers of soft contact lenses. Although the mechanism of disease within the
cornea is unclear, cytopathic enzymes including a variety of proteases (Mitro
et al.y 1994), such as collagenase (He et aL, 1990), and proteolytic activity are
probably important.

223

Protozoa

Treatment

There is no treatment for GAE and it is invariably fatal. If untreated, Acan-
thamoeba keratitis can lead to permanent blindness. However, management is
difficult since the cysts are resistant to most antimicrobials at concentrations
tolerated by the cornea and a prolonged course of treatment is often required to
achieve elimination. Treatment includes ketoconazole, slotrimazole and propa-
midine isethionate (Martinez, 1993). Chlohexidine gluconate and polyhexa-
methylene biguanide (PHMB) are effective, particularly PHMB administered
following prompt diagnosis.

Survival in the environment

Acanthamoeba are ubiquitous in the natural and man-made environment,
including tap water, swimming, hydrotherapy and spa pools (De Jonckheere,
1987; Kilvington et aL 9 1990) and have even been isolated from eyewash
stations (Paszilo-Kolua et aL, 1991). Acanthamoeba spp. have a growth tem-
perature range of 12-45°C, and all pathogenic species have an optimum of
30°C, but most also grow at 37°C, although this is uncertain for members
of morphological group II. Therefore, during laboratory diagnosis, growth
at both 32°C and 37°C is recommended. The cysts are highly resistant to a
number of environmental pressures, including drying, freezing to — 20°C,
moist heat at 56°C and 50ppm free chlorine (Kilvington, 1989; Kilvington
and Price, 1990), and have been found in nearly all soil and aquatic environ-
ments (Page, 1988).

Survival in water

Trophozoites are killed by saline concentrations >1%, although the cysts can
survive and have been detected in seawater. Commercial contact lens disinfect-
ant solutions based on chlorhexidine, hydrogen peroxide or moist heat will
kill Acanthamoeba cysts providing the correct time and temperature (where
relevant) exposure conditions are met (Ludwig et al., 1986; Davies et al.,
1988). Cysts can be removed from lenses by commercial lens-cleaning agents
(Kilvington and Larkin, 1990). A survey of contact lens cases belonging to
healthy wearers showed that 43% contained >10 6 ml/l viable bacteria and
seven contained Acanthamoeba (Larkin et aL, 1990). It is therefore likely that
infection can be acquired through primary contamination of the lens storage
case, which becomes contaminated from rinsing lenses in tap water or non-
sterile saline solutions or through wearing lens while bathing or swimming in

224

Acanthamoeba spp.

lakes or ponds (Visvesvara, 1993). Non-contact lens-associated keratitis clearly
also occurs in the minority of cases and is associated with ocular trauma or
environmental contamination. Despite cyst survival in 50ppm free chlorine,
and detection in swimming pools, no infection has been reported to have been
directly acquired in a chlorinated swimming pool according to USA data.

Methods of detection

Clinical diagnosis of GAE is by microscopic examination of histological sections.
Examination at high power can help avoid misidentification as macrophages or
Entamoeba histolytica. Diagnosis of eye infections with Acanthamoeba spp.
is by similar examination, direct smears or culture of eye and skin lesion
scrapings, swabs or aspirates on non-nutrient agar seeded with suitable Enter-
obacter spp. (including Escherichia coli and Klebsiella aero genes) (Martinez,
1993). Trophozoites can be observed following Gram or Giemsa stains (Bottone,
1993) and show sluggish motility. Cysts can be observed by a variety of stains
including calcofluor, which stains the chitin and cellulose in the cyst wall
(Bottone, 1993). This stain has also been used to demonstrate Acanthamoeba
on contact lens surfaces (Johns et ah 9 1991).

Acanthamoeba has been isolated from a variety of environments, including
soil, natural and man-made aquatic environments (including chlorinated pools),
and potable water supplies (Kilvington and White, 1994). The organism is
readily cultured on non-nutrient agar plates seeded with Escherichia coli from
environmental samples using prior concentration by sedimentation or filtration
(Kilvington et ah 9 1990). Prior concentration of the water sample by centrifuga-
tion or by membrane filtration and elution of captured cells may be required
before culture on an £. coli lawn at 32°C and 37°C. Once growth on seeded
bacterial lawns is established, strains can be adapted to axenic (bacteria-free)
culture media. For detection, trophozoite plaques are subcultured from the
£. coli lawn to Page's saline in a microtitre plate (Anon, 2002), incubated at
32°C for 1-3 hours and examined microscopically.

Cyst morphology within a species varies with cultural conditions and so
other characteristics are used for species identification, including isoenzyme
profiles (Costas and Griffiths, 1980) and genetic analyses (Bolger et ah 9 1982;
McLaughlin et ah 9 1988). Since there is difficulty in the use of morphological
criteria for identification of isolates, molecular methods have been developed.
Many of these are based on the polymerase chain reaction for amplification
prior to identification (Vodkin et ah 9 1992) and differentiation of pathogenic
species (Howe et al. 9 1997). Since Acanthamoeba can be cultured to produce
a relatively pure suspension from both clinical and environmental samples,
eukaryotic-specific primers can also be used with confidence. However, having
to produce a culture prior to PCR can add a number of days to the analysis
process. In their studies, Schroeder et ah (2001) developed a genus-specific

225

Protozoa

PCR to amplify DNA from all known Acanth amoeba 18s rDNA genotypes,
that did not amplify DNA from closely related organisms, making it particu-
larly applicable to environmental samples. Amplicons can then be used for
further analysis for genetic relatedness.

Epidemiology

The study of the epidemiology of cases of Acanth amoeba keratitis in the USA in
the late 1980s (Visvesvara and Stehr-Green, 1990) identified that using home-
made saline solution to rinse soft contact lenses was a major risk factor for
disease. This practice was enabled by the then current sale of saline tablets and
distilled water precisely to create home-made solutions. Prior contamination of
the distilled water or the saline solution due to atmospheric exposure put users
at risk of infection, and the practice has been halted by the banning of the sale
of saline tablets and distilled water for this purpose. This has led to a decrease
in reported cases of disease in the USA (Visvesvara, 1993). In the UK cases are
also predominantly among contact lens wearers and have risen since the mid-
1980s, even at specialist units which have a history of expert diagnosis. In a
case control study to identify risk factors for Acanth o amoeba keratitis in
contact lens wearers, Radford and colleagues (1995) showed that failure to
disinfect daily-wear soft contact lenses, and the use of chlorine release lens
disinfection systems, were major risk factors.

Risk assessment

There appears to be no risk to healthy individuals from the ingestion of Acan-
thamoeba. However, it is recommended that immunocompromised persons
boil drinking water before consumption. The health risk from Acanthamoeba
spp. for people with intact immune systems lies not with consumption of water
but through exposure via lesions, particularly contact lens wearers who account
for the vast majority of cases in the USA and UK (Visvesvara and Stehr-Green,
1990; Bacon et al., 1993). Recognized risk factors are poor hygiene practices such
as washing and or storing lenses in non-sterile solutions or tap water, and inad-
equate disinfection. In non-contact lens wearers, infection usually follows trauma
of the eye with environmental contamination: A. polyphaga and A. castellanii
are most frequently identified in environmental samples (Page, 1988).

Health effects: occurrence of illness, degree of morbidity and mortality,
probability of illness based on infection:

• The types of Acanthamoeba mainly associated with human disease are A. cul-
bertsoni, which causes the rare but fatal granulomatous amoebic encephalitis
(GAE) and A. polyphaga and A. castellanii, which are more usually associated

226

Acanthamoeba spp.

with eye infections (keratitis). GAE is not regarded as water-related while
keratitis is.

• Acanthamoeba keratitis is characterized by intense pain and ring-shaped
infiltrates in the corneal stroma, which can progress to hypopyon, scleritis,
glaucoma and cataract formation. Corneal perforation may occur. One eye
only is usually affected, although bilateral disease has been reported.

• Based on sera studies, people are commonly exposed to Acanthamoeba, but
rarely affected. Parasites have been found in the respiratory tract of healthy
people, and subclinical, self-limiting infection may be common in immuno-
competent hosts.

• There appears to be no risk to healthy individuals from the ingestion of
Acanthamoeba.

Exposure assessment: routes of exposure and transmission, occurrence in
source water, environmental fate:

• Acanthamoeba can enter the skin through a cut, wound, or through the
nostrils.

• The transmission of keratitis caused by Acanthamoeba, is largely driven by
risky behaviour in contact lens wearers. Recognized risk factors for keratitis
are poor hygiene practices such as washing and/or storing contact lenses in
non-sterile solutions or tap water and inadequate disinfection of lenses.
Infection can also be acquired through primary contamination of the lens
storage case or through wearing lenses while swimming in lakes or ponds.

• Non-contact lens-associated keratitis occurs rarely and is associated with
ocular trauma or environmental contamination.

• Acanthamoeba are ubiquitous in the natural and man-made environment,
including tap water and swimming and spa pools. The amoebae favour
aquatic habitats such as water, mud and soil.

• The cysts are highly resistant to environmental forces, including drying, freez-
ing, heat, and 50ppm free chlorine.

Risk mitigation: drinking-water treatment, medical treatment:

• Prevention of Acanthamoeba keratitis in contact lenses is key; commercial
lens-cleaning agents can remove cysts from lenses.

• If untreated, Acanthamoeba keratitis can lead to permanent blindness.
Management is difficult since the cysts are resistant to most antimicrobials at
concentrations tolerated by the cornea. A prolonged course of treatment is
often required. Propamidine isethionate, chlohexidine gluconate, polyhexa-
methylene biguanide are effective, particularly when administered follow-
ing prompt diagnosis.

Future implications

Prevention of GAE is undetermined. Prevention of keratitis can be achieved
through public health messages aimed at those groups identified as at risk and

227

Protozoa

through rapid diagnosis of infection in these groups and following ocular
trauma.

Protozoa such as Acanthamoeba can support the intracellular growth of
other organisms pathogenic to humans, and have implications for their sur-
vival, resistance to disinfection regimens, ecology and dissemination and even
their increased virulence following passage through protozoa.

References

Anon. (1989). Isolation and identification of Giardia cysts, Cryptosporidium oocysts and

freeliving pathogenic amoebae in water etc. In Methods for the Examination of Waters

and Associated Materials. London: HMSO.
Anon. (2002). Isolation and identification of Acanthamoeba species, PHLS SOP W12,

Issue 2. Colindale, London: PHLS HQ.
Auran, J.D., Starr, M.B. and Jakobiec, F.A. (1987). Acanthamoeba keratitis: a review of the

literature. Cornea, 6: 2-26.
Bacon, A.S., Frazer, D.G., Dart, J.K.G. et al. (1993). A review of 72 consecutive cases of

Acanthamoeba keratitis, 1984-1992. Eye, 7: 719-725.
Bolger, S.A., Zarley, CD., Burianek, L.L. et al. (1982). Interstrain mitochondrial DNA

polymorphism detected in Acanthamoeba by restriction endonuclease analysis. Mol

Biochem Parasitol, 8: 145-163.
Bottone, E.J. (1993). Free-living amebas of the genera Acanthamoeba and Naegleria: an

overview and basic microbiological correlates. Mount Sinai J Med, 60: 260-270.
Costas, M. and Griffiths, A.J. (1980). The suitability of starch gel electrophoresis of

esterases and acid phosphotases for the study of Acanthamoeba taxonomy. Arch

Prostisterk, 123: 2727-2729.
Culbertson, C.D., Smith, J.W. and Minner, J.R. (1958). Acanthomoeba: observations on

animal pathogenicity. Science, 127: 1506.
Davies, D.J.G., Anthony, Y., Meakin, B.J. et al. (1988). Anti-acanthamoeba activity of

chlorhexidine and hydrogen peroxide. Trans Br Contact Lens Assoc, 5: 80-82.
De Jonckheere, J.F. (1987). Epidemiology. In Amphizic Amoebae, Human Pathology,

Rondanelli, E.G. (ed.). Padua, Italy: Piccin Nuova Libraria, pp. 127-147.
He, Y.G., Niederkorn, J.Y., McCulley, J.P. et al. (1990). In vivo and in vitro collagenolytic

activity of Acanthamoeba castellani. Ophthalmol Vis Sci, 31: 2235-2240.
Howe, D.K., Vodkin, M.H., Novak, R.J. et al. (1997). Identification of two genetic mark-
ers that distinguish pathogenic and non-pathogenic strains of Acanthamoeba. Parasitol

Res, 83: 345-348.
Johns, K.J., Head, W.S., Robinson, R.D. et al. (1991). Examination of the contact lens with

light microscopy; an aid in diagnosis of Acanthamoeba keratitis. Rev Infect Dis, 13:

S425.
Jones, D.B., Visvesvara, G.S. and Robinson, N.M. (1975). Acanthamoeba polyphaga and

Acanthamoeba uveitis associated with a fatal meningitis. Trans Ophthalmol Soc UK,

95: 221-232.
Kenney, M. (1971). The Micro-Kolmer complement fixation test in routine screening for

soil ameba infection. Hlth Lab Sci, 8: 5-10.
Kilvington, S. (1989). Moist-heat disinfection of pathogenic Acanthamoaeba cysts. Lett

Appl Microbiol, 9: 187-189.
Kilvington, S. and Larkin, D.F.P. (1990). Acanthamoeba adherence to contact lenses and

removal by cleaning agents. Eye, 4: 589-593.
Kilvington, S. and Price, J. (1990). Survival of Legionella pneumophila with Acanthamoeba

polyphaga cysts following chlorine exposure./ Appl Bacteriol, 68: 519-525.

228

Acanthamoeba spp.

Kilvington, S. and White, D.G. (1994). Acanthamoeba: biology, ecology and human dis-
ease. Rev Med Microbiol, 5: 12-20.

Kilvington, S., Beeching, J.R. and White, D.G. (1991). Differentiation of Acanthamoeba
strains from infected corneas and the environment using restriction endonuclease diges-
tion of whole cell DNA./C/m Microbiol, 29: 310-314.

Kilvington, S., Larkin, D.F.P., White, D.P. et al. (1990). Laboratory investigation of
Acanthamoeba keratitis./ Clin Microbiol, 28: 2722-2725.

Larkin, D.F.P., Kilvington, S. and Dart, J.K.G. (1992). Treatment of Acanthamoeba kera-
titis with polyhexamethylene biguanide. Ophthalmology, 99: 185-191.

Larkin, D.F.P., Kilvington, S. and Easty, D.L. (1990). Contamination of contact lens stor-
age cases by Acanthamoeba and bacteria. Br J Ophthalmol, 74: 133-135.

Ludwig, I.H., Meiser, D.M., Ritherford, I. et al. (1986). Susceptibility of Acanthamoeba to
soft contact lens disinfection systems. Invest Ophthalmol Vis Sci, 27: 626-628.

Ma, P., Willaert, E., Juechter, K.B. et al. (1981). A case of keratitis due to Acanthamoeba
in New York and features of 10 cases. / Infect Dis, 143: 662-667.

Martinez, A.J. (1991). Infections of the central nervous system due to Acanthamoeba.
Rev Infect Dis, 13: S399-402.

Martinez, A.J. (1993). Free-living amebas: infection of the central nervous system. Mount
Sinai] Med, 60: 271-278.

McLaughlin, G.L., Brabdt, F.H. and Visvesvara, G.S. (1998). Restriction fragment length
polymorphism of the DNA of selected Naegleria and Acanthamoeba amebae. / Clin
Microbiol, 26: 1655-1658.

Mitro, K., Bhagavathiammai, A., Zhou, O.-M. et al. (1994). Partial characterisation of the
proteolytic secretions of Acanthamoeba polyphaga. Exp Parasitol, 78: 377-385.

Nagington, J., Watson, P.G., Playfair, T.J. et al. (1974). Amoebic infections of the eye.
Lancet, ii: 1537-1540.

Page, F.C. (1988). A new key to freshwater and soil gymnamoebae. Cumbria: The Fresh
water Biological Association. The Ferry House, Ambleside, Cumbria.

Paszilo-Kolua, C, Yamamoto, H., Shahamat, M. et al. (1991). Isolation of amoebae and
Pseudomonas and Legionella spp. from eyewash stations. Appl Environ Microbiol, 57:
163-167.

Pussard, M. and Pons, R. (1977). Morphologie de la paroi kystique et taxonomie du genre
Acanthomoeba (Protozoa, Amoebida). Protistologica, 13: 557-598.

Radford, C.F., Woodward, E.F.G. and Stapleton, F. (1993). Contact lens hygiene compli-
ance in a university population. / Brit Contact Lens Assn, 16: 105-111.

Schroeder, J.M., Booton, G.C., Hay, J. et al. (2001). Use of subgenic 18s ribosomal DNA
PCR and sequencing for genus and genotype identification of Acanthamoebae from
humans with keratitis and from sewage sludge./ Clin Microbiol, 39: 1903-1911.

Visvesvara, G.S. (1991). Classification of Acanthamoeba. Rev Infect Dis, 13: S369-S372.

Visvesvara, G.S. (1993). Epidemiology of infections with free-living amebas and laboratory
diagnosis of microsporidiosis. Mount Sinai J Med, 60: 283-288.

Visvesvara, G.S. and Stehr-Green, J.K. (1990). Epidemiology of free-living amoeba infec-
tions./ Protozool, 37: 25S-33S.

Vodkin, M.H., Howe, D.K., Visvesvara, G.S. etal. (1992). Identification of Acanthamoeba
at the generic and specific levels using the polymerase chain reaction. / Protozool, 39:
378-385.

229

Balantidium coli

Basic microbiology

Balantidium coli is a large, ciliated protozoan (Phylum Ciliophora, Order
Trichostomatida, Family Balantidiidae) which, while having a world-wide
distribution, is a rare cause of human infection (Arean and Koppisch, 1956). Not
only is it the only ciliated protozoan pathogenic to man, it is also the largest
protozoan parasite of humans. B. coli can cause severe, life-threatening colitis
and transmission of infection is via the cyst stage shed in faeces. Following
ingestion, the cysts excyst in the small intestine and subsequently trophozoites
reside in the lumen of the large intestine where they feed on bacteria.
Trophozoites (30-150 |Jim by 25-120 fjim) replicate by lateral transverse binary
fission, and it is also speculated that conjugation may occur, but while this has
been observed in culture it has never been demonstrated in nature (Sargeaunt,
1971). At the anterior end of the trophozoite is a mouth (cytosome) that leads
into the cytopharynx, which occupies about a third of the parasite length. At the
anterior end is the anus (cytophage). The trophozoite is binucleate. Some
trophozoites invade the wall of the colon and multiply. Trophozoites undergo
encystation to produce infective cysts, which are up to 30-200 |xm by
20-120 [Jim. The cysts also contain a micro- and macro-nucleus. The cysts are
shed in the faeces and are responsible for transmission of infection. While

Protozoa

remaining trophozoites may disintegrate in the gut lumen, they can be readily
detected in stools during acute infection.

The prevalence of human infection is generally low, but higher in tropical and
subtropical regions, particularly where the principal animal reservoir, pigs, are
raised. For example, human balantidiasis is most common in the Philippines,
but is also reported in Central and South America, Iran and Papua New Guinea
(Barnish and Ashford, 1989), although prevalence is rarely above 1%. Other
animal reservoirs exist including rodents and non-human primates. B. coli is not
host-specific but may not be readily transmitted between some hosts since adjust-
ment to the symbiotic flora of the new host appears to be required. However,
the transmission of human-derived isolates to piglets and monkeys has been
demonstrated (Yang et al., 1995). Once adapted to the new host, it can flourish
and become a serious pathogen, particularly in humans. Biological features of
B. coli affect its transmission and epidemiology, particularly in the ecological
relationship between hosts and the survival of cysts.

Clinical features

Although most infections are asymptomatic, acute balantidiasis manifests as
persistent diarrhoea, abdominal pain, weight loss, tenesmus, nausea, vomiting
and occasionally dysentery which may resemble amoebiasis (Baskerville, 1970).
Chronic infection and disease also occurs, characterized by intermittent diar-
rhoea and occasional blood in the stools. Symptoms of balantidiasis can be
severe in debilitated people. Misdiagnosis can result in failure to treat the infec-
tion and subsequent case fatality. Rapid progression, with fever and prostration
leads to death, usually due to peritonitis from colonic perforation. Rare cases
of balantidial appendicitis have been reported (Dodd, 1991).

Pathogenicity, virulence and causation

The pig is regarded as the primary host for B. coli, in which it is a commensal
organism and rarely associated with the mucosa. However, in man, B. coli pro-
duces ulcers ranging from superficial to deep, and associated dysentery. Only
rarely are other tissues, such as the liver, invaded. On invasion and penetration
of the distal ileal and colonic mucosa and submucosa, the trophozoite causes
mucosal inflammation and ulceration. The enzyme hyaluronidase, produced by
the parasite, is thought to facilitate this penetrative invasion. Inflammation is
caused by liberation of other products by the parasite and possibly by the
recruitment of mucosal inflammatory cells, particularly neutrophils. Studies
of trophozoite populations from pigs with balantidiasis, by cytophotometric
studies, have shown differences in nucleic acids from populations infecting
asymptomatic pigs (Skotarczak and Zielinski, 1997).

232

Balantidium coli

Treatment

Balantidiasis is treatable by antibiotic therapy, including tetracycline,
iodoquinol, bacitracin, ampicillin, paromomycin and metronidazole (Garcia-
Laverde and DeBonilla, 1975). In fulminant disease surgery may be required.

Survival in the environment and in water

B. coli has been detected in sewage sludge (Amin, 1988) and in water storage
tanks in different parts of Hyderabad, India (Jonnalagagga and Bhat, 1995).
The cysts can survive for several weeks in moist conditions, such as pig faeces
or water, but are rapidly destroyed in dry heat and by pH <5. While the cysts
are resistant to levels of chlorination used to treat drinking water, they are
killed by boiling. The impact of water-treatment processes on removal of
B. coli has not been measured, but due to its large size it is likely that standard
treatment involving coagulation and filtration would be effective.

Methods of detection

Although trophozoites rapidly disintegrate they are more frequently detected
than cysts in stools during acute infection and therefore should be sought for
diagnosis. Trophozoites are oval in shape and about 17fjim X 15 |xm, and
covered in cilia that propel the organism within the intestinal lumen. Shedding
in stools is intermittent and therefore repeat stools should be collected and
examined immediately or preserved prior to examination. Trophozoites can
also be detected in material from the margins of ulcers seen in the rectum by
sigmoidoscopy. Parasites from swine faeces were examined for autofluores-
cence. Cysts of B. coli have been shown to emit light after excitation with UV
light, which may facilitate diagnosis (Daugschies et aL, 2001). While culture
methods are available for B. coli, these are more suited to research than diag-
nostic laboratories (Clark and Diamond, 2002).

Epidemiology

Human infection is generally regarded as a zoonosis, transmitted by faecal-
oral contact with swine faeces. Waterborne epidemics have occurred in areas
of poor sanitation and, although they do not suffer enteric disease, swine are
an important reservoir. A large epidemic involving over 100 human cases of

233

Protozoa

balantidiasis followed a severe typhoon which caused gross contamination of
ground and surface water supplies by pig faeces (Walzer et al. 9 1973).

Person-to-person transmission may occur, particularly if personal hygiene is
difficult. In a survey of residents of psychiatric institutions in Italy, jB. coli was
detected in the stools of 97/234 (40.8%) (Giacometti et aL 9 1997) and parasitic
infections were associated with diarrhoea and other gastrointestinal symptoms.

Risk assessment

The risk of transmission from swine to humans appears to be greatest not only
where these reservoir animals are kept but also where sanitation is poor. A cross-
sectional study of the prevalence of JB. colt in pigs on a Danish research farm
showed the prevalence of infection increased from 57% in suckling piglets to
100% in most pig groups >4 weeks old (Hindsbo et al. 9 2000). However, no
human cases have been published in Denmark indicating that either the strain
was not infectious for man or that proper control measures are effective in
preventing zoonotic spread.

Surveys of water supplies for B. coli have been rarely reported, but in a survey
of stored drinking water in Hyderabad City, India, 61/232 samples indicated
the presence of pathogenic parasites which include protozoans (cysts of
Giardia lamblia 9 Entamoeba histolytica, Balantidium coli) and nematode eggs
(Enterobius vermicularis 9 Ascaris lumbricoides 9 Trichuris trichiura) 9 rhabditi-
form and filariform larvae and adult stages of Strongy hides stercoralis and
Enterobius vermicularis) (Jonnalagagga and Bhat, 1995). Interestingly, hand
washings from food handlers also showed the presence of pathogenic para-
sites, although the original water used for such washings were free from
contamination.

Overall risk assessment

Health effects: occurrence of illness, degree of morbidity and mortality, prob-
ability of illness based on infection:

• Balantidium coli occurs world-wide, but is a rare cause of human infection.

• Although most infections are asymptomatic, acute balantidiasis manifests as
persistent diarrhoea, abdominal pain, weight loss, tenesmus, nausea, vomit-
ing, and occasionally dysentery which may resemble amoebiasis.

• Chronic infection and disease also occurs, characterized by intermittent
diarrhoea and occasional blood in the stools.

• Symptoms of balantidiasis can be severe in debilitated people. jB. coli can
cause severe, life-threatening colitis.

• The prevalence of human infection is generally low, but higher in tropical
and subtropical regions, particularly where the principal animal reservoir,
pigs, are raised. Prevalence is rarely above 1%.

234

Balantidium coli

Exposure assessment: routes of exposure and transmission, occurrence in
source water, environmental fate:

• Infective cysts are excreted in faeces of humans and pigs. The route of trans-
mission is faecal-oral. Person-to-person transmission may occur.

• Though pigs are the primary animal reservoir, other animal reservoirs include
rodents and non-human primates.

• Human infection is generally regarded as a zoonosis, transmitted by faecal-
oral contact with swine faeces.

• Waterborne epidemics have occurred in areas of poor sanitation.

• B. coli has been detected in sewage sludge and in water storage tanks.

• The cysts can survive for several weeks in moist conditions, such as water,
but are rapidly destroyed in dry heat and by pH <5.

Risk mitigation: drinking-water treatment, medical treatment:

• Like other protozoan cysts, B. coli cysts are resistant to levels of chlorination
used to treat drinking water, however, they are killed by boiling.

• The impact of water-treatment processes on removal of B. coli has not been
measured, but because they are large (30-200 |xm by 20-120 fJim), it is
likely that standard treatment involving coagulation and filtration would
be effective.

• Balantidiasis is treatable by antibiotic therapy, including tetracycline,
iodoquinol, bacitracin, ampicillin, paromomycin and metronidazole.

Future implications

Given that B. coli has a cyst stage and can survive in water, it is possible that
waterborne transmission can occur following contamination by faeces from
infected reservoir hosts or from infected people. However, where the prevalence
is low the likelihood is reduced and control can be achieved by good sanitation.

References

Amin, O.M. (1988). Pathogenic micro-organisms and helminths in sewage products,

Arabian Gulf, country of Bahrain. Am J Public Hlth, 78: 314-315.
Arean, V.M. and Koppisch, E. (1956). Balantidiasis: a review and report of cases. Am J

Pathol, 32: 1089-1115.
Barnish, G. and Ashford, R.W. (1989). Occasional parasitic infections of man in Papua

New Guinea and Irian Jaya (New Guinea). Ann Prop Med Parasitol, 83: 121-135.
Baskerville, L. (1970). Balantidium colitis. Report of a case. Am J Dig Dis, 15: 727-731.
Clark, C.G. and Diamond, L.S. (2002). Methods for cultivation of luminal parasitic pro-

tists of clinical importance. Clin Microbiol Rev, 15: 329-341.

235

Protozoa

Daugschies, A., Bialek, R., Joachim, A. et al. (2001). Autofluorescence microscopy for the

detection of nematode eggs and protozoa, in particular Isospora suis, in swine faeces.

Farasitol Res, 87: 409-412.
Dodd, L.G. (1991). Balantidium coli infestation as a cause of acute appendicitis (Letter).

J Infect Dis, 163: 1392.
Garcia-Laverde, A. and de Bonilla, L. (1975). Clinical trials with metronidazole in human

balantidiasis. Am J Trop Med Hyg, 24: 781-783.
Giacometti, A., Cirioni, O., Balducci, M. et al. (1997). Epidemiologic features of intestinal

parasitic infections in Italian mental institutions. Eur J Epidemiol, 13: 825-830.
Hindsbo, O., Nielsen, C.V., Andreassen, J. et al. (2000). Age-dependent occurrence of the

intestinal ciliate Balantidium coli in pigs at a Danish research farm. Acta Vet Scand, 41:

79-83.
Jonnalagagga, P.R. and Bhat, R.V. (1995). Parasitic contamination of stored water used for

drinking/cooking in Hyderabad. SE Asian J Trop Med Public Hltb, 26: 789-794.
Sargeaunt, P.G. (1971). The size range of Balantidium coli. Trans Roy Soc Trop Med Hyg,

65:428.
Skotarczak, B. and Zielinski, R. (1997). A comparison of nucleic acid content in

Balantidium coli trophozoites from different isolates. Folia Biol, 45: 121-124.
Walzer, P.D., Judson, F.N., Murphy, K.B. et al. (1973). Balantidiasis outbreak in Truk.

Am J Trop Med Hyg, 22: 33-41.
Yang, Y., Zeng, L., Li, M. et al. (1995). Diarrhoea in piglets and monkeys experimentally

infected with Balantidium coli isolate from human faeces. / Trop Med Hyg, 98: 69-72.

236

Cryptosporidium spp

Basic microbiology

Cryptosporidium is an obligate intracellular parasite (Phylum Apicomplexa,
Order Eucoccidiida, Family Cryptosporidae). It therefore requires host cells
in which to continue and finish its life cycle, which is completed in a single
host. Although currently classified as an eimeriid coccidian, there is mounting
phenotypic and molecular phlyogenetic evidence for closer relationship with
the gregarines and reclassification has been proposed but not yet adopted
(Tenter et al., 2002). Many Cryptosporidium species have been confirmed by
genetic analyses (Table 18.1) and they infect a wide range of hosts. Although
the majority of human disease is caused by Cryptosporidium parvum (syn.
C. parvum genotype 2, 'cattle' or C C) and Cryptosporidium hominis (syn.
C. parvum genotype 1, 'human' or C H') (Fayer et al. 9 2000; Morgan-Ryan
et al., 2002), other species are also detected, albeit more rarely, in both
immunocompetent and immunocompromised patients (Fayer et al., 2000;
Chalmers et al., 2002a) (Table 18.1).

Human infection with Cryptosporidium occurs following the ingestion and
excystation of oocysts in the small intestine, when four motile sporozoites are
released. Five further developmental events follow: merogony, gametogony,

Protozoa

Table 18.1 Currently accepted species of Cryptosporidium

Cryptosporidium Original
species host

Mean oocyst
sizes |jim (range)
in original host

Confirmed in humans by genetic analysis?
Immunocompetent Immunocompromised

Intestinal species

C. parvum

Mice

5.0 X 4.5

(3.8-6.0 X 3.0-5.3)

Yes

C. hominis

Man

4.9 X 5.2

(4.4-5.4 x 4.4-5.9)

Yes

C. felis

Cat

4.6 X 4.0

(3.2-5.1 X 3.0-4.0)

Yes

C. canis

Dog

5.0 X 4.7

(3.7-5.9 X 3.7-5.9)

Yes

C. wrairi

Guinea-pig

5.4 X 4.6

(4.8-5.6 X 4.0-5.0)

C. nasorum

Fish

4.3 X 3.3

(3.5-4.7 X 2.5-4.0)

C. saurophilum

Lizard

5.0 X 4.7

(4.4-5.6 X 4.2-5.2)

Gastric species

C. muris

Mice

7.0 X 5.0

(6.5-8.0 X 5.0-6.5)

C. andersoni

Cattle

7.4 X 5.5

(6.6-8.1 X 5.0-6.5)

C. serpentis

Snakes

6.2 X 5.3

(5.6-6.6 X 4.8-5.6)

Multi-site species

C. meleagridis

Turkeys

5.2 X 4.6

(4.5-6.0 X 4.2-5.3)

Yes

C. baileyi

Chickens

6.2 X 4.6

(5.6-6.3 X 4.5-4.8)

Yes

Yes
No

From:Tyzzer, 1910; Chalmers etal., 1994; Arrowood, 1997; Fayer etal., 1997, 2000, 2001; Koudela and Modry, 1998;
Lindsay etal., 2000; Morgan-Ryan etal., 2002.

fertilization, oocyst wall formation and sporogony (Current and Long, 1983;
Current and Hayes, 1984; Current and Reese, 1986). Excystation is initiated
by reducing conditions and exposure to secretions such as bile salts and pan-
creatic enzymes, although Cryptosporidium oocysts do excyst in warm aque-
ous solutions, which may enable infection of extraintestinal sites including the
respiratory tract, conjunctiva of the eye, and reproductive system. Sporozoites
infect the apical portion of epithelial cells (usually enterocytes in the small
intestine) when the anterior end adheres to their surface and becomes sur-
rounded by microvilli, uniquely occupying an intracellular but extracytoplas-
mic location in the host cell. Cycles of asexual and then sexual multiplication
follow culminating in the production of immature oocysts. These mature and
sporulate inside the host and oocysts containing four infectious sporozoites

238

Cryptosporidium spp.

are shed in the faeces. These oocysts are thick-walled, resistant to many envir-
onmental pressures, and are the only exogenous stage of the life cycle. Thin
walled oocysts are not detected in faeces but initiate a cycle of autoinfection
within the host (Current and Reese, 1986), causing prolonged disease in those
unable to clear the infection. Direct faecal-oral host-to-host transmission can
occur because the parasites are immediately infective upon release in the fae-
ces. Equally important is the robust nature of the oocysts which contributes to
indirect transmission following contamination of the food or water environ-
ment and of fomites.

Cryptosporidium has a world-wide distribution with greater numbers of
cases of cryptosporidiosis reported among children than adults. The majority
of human infections are with either of two species: C. hominis is the anthro-
ponotic genotype that is largely restricted to humans, and C. parvum is the
zoonotic genotype that causes both human and animal disease (Fayer et al.,
2000; Morgan-Ryan et ai, 2002).

Thus the detection of C. hominis is indicative of a human source of infec-
tion or contamination and C. parvum of either an animal or a human source.
Application of genotyping techniques has also led to the characterization of
additional Cryptosporidium spp. and genotypes, and it is has become clear
that other species are also found infecting both immunocompetent and
immunocompromised patients (Fayer et al., 2000; Chalmers et al., 2002a).

A number of biological features of Cryptosporidium affect its transmission
and epidemiology:

• the life cycle does not require dual or multiple hosts

• the oocyst stage is shed in a fully sporulated state so direct transmission can
occur between hosts

• autoinfection enables persistent disease in immunocompromised hosts

• there is a large livestock and human reservoir

• the thick-walled oocysts are resistant to a wide range of pressures and can
survive for long periods in the environment

• the infectious dose is low and so small numbers of contaminating organ-
isms are significant

• hosts can shed large numbers of oocysts

• there is a lack of specific drug therapy to clear infection efficiently.

Of additional hindrance to risk assessment, environmental testing is difficult
since sampling for and detection of oocysts is problematic, and viability/infec-
tivity assessment currently inadequate from such samples.

Origin of the organism

The first person to recognize, describe and name Cryptosporidium was
E.E. Tyzzer, who, in 1907, published the asexual, sexual and oocyst stages of

239

Protozoa

a parasite he frequently found in the gastric glands and faeces of laboratory
mice (Tyzzer, 1907). He proposed the murine gastric isolate Cryptosporidium
muris, the type strain (Tyzzer, 1910) and, in 1912, published a description
of a new, smaller species found in the small intestine of laboratory mice and
rabbits, which he called C. parvum (Tyzzer, 1912). Tyzzer 's remarkable
observations, including the proposal of autoinfection, established the life
cycle of the parasite, which electron microscopy has served to confirm with
the only additional observation of the extracellular developmental stages
(merozoites and microgametes). In 1929, Tyzzer also described endogenous
stages of Cryptosporidium in chicken caecal epithelium (Tyzzer, 1929).

In 1955 a new species, Cryptosporidium meleagridis, was reported causing
illness and death in young turkeys (Slavin, 1955) and, in 1971, a report was
published where Cryptosporidium was associated with bovine diarrhoea
(Panciera etaL, 1971). While this stimulated veterinary investigations of the para-
site, human cases were not identified until 1976 when two reports were pub-
lished. One described an otherwise healthy 3-year-old girl with symptoms of
vomiting, watery diarrhoea and abdominal pains (Nime et aL 9 1976). Diagnosis
was made by rectal biopsy, and the patient recovered after 2 weeks' illness. In
contrast, the other described a severely dehydrated immunosuppressed patient
with chronic watery diarrhoea (Meisel et al., 1976). Diagnosis was by histo-
logical examination of jejunal biopsy. The patient recovered following with-
drawal of immunosuppressive treatment and subsequent restoration of T-cell
function.

It was not until the 1980s that the role of Cryptosporidium in human disease
and its impact on human health really began to be recognized. Contributing
to the emergence of Cryptosporidium as a human pathogen were the AIDS
epidemic and consequent increase in the number of immunocompromised
individuals unable to clear Cryptosporidium infections, and the occurrence
of a number of waterborne outbreaks of disease in developed countries. There
was clearly an inconsistency in the perception of the parasite as an oppor-
tunist zoonotic infection and its occurrence in primarily urban, male AIDS
patients (Casemore and Jackson, 1984). Improved laboratory methods by
veterinary workers for the detection of oocysts in faeces led to increased ascer-
tainment and recognition of the parasite in clinical laboratories, and important
epidemiological studies during the early 1980s showed that cryptosporidiosis
also occurred in otherwise healthy subjects, particularly children (Casemore
et al., 1985). Widespread reporting of microbiological results to disease sur-
veillance schemes contributed to the recognition of Cryptosporidium as a
cause of acute, self-limiting gastroenteritis in the general population and of
potentially fatal infection in the immunocompromised.

Clinical features

A range of clinical features characterize cryptosporidiosis, varying in severity
from asymptomatic carriage to severe, life-threatening illness. The predominant

240

Cryptosporidium spp.

feature of cryptosporidiosis is watery, sometimes mucoid, diarrhoea with
dehydration, weight loss, anorexia, abdominal pain, fever, nausea and vomiting.
The incubation period has been reported from outbreaks as a mean of 7 days
from exposure (range 1-14 days) (MacKenzie et al. 9 1995). In experimental
human infection diarrhoeal symptoms appeared at mean 9 days and median 6.5
days in subjects with no serological evidence of previous infection (DuPont et al. 9
1995) and median 5 days (range 3-12 days) in subjects with evidence of prior
infection (Chappell et al. 9 1999).

In immunocompetent patients symptoms usually last for about 1-2 weeks.
The duration of diarrhoeal illness in sporadic cases who visited primary
health care in the UK and submitted a faecal specimen has been reported as
mean 9 days, median 7 days (mode 7 days, range 1-90 days) (Palmer and
Biffin, 1990). However, a recent study in Melbourne, Australia showed that
similarly selected patients reported symptoms of mean 22 days (range 1-100
days) and in Adelaide mean 19 days (range 2-120 days) (Robertson et al. 9
2002). In experimental infection, duration of gastrointestinal illness in sub-
jects without evidence of prior exposure to Cryptosporidium was 6.5 days
(DuPont et al. 9 1995) and in subjects with prior exposure 3.1 days (Chappell
et al. 9 1999). It is likely that cases identified through passive surveillance rep-
resent the more severe end of the spectrum of disease in that they experienced
symptoms which prompted them to seek primary health care and to submit a
faecal specimen for testing.

Oocysts may continue to be shed in the faeces following cessation of diar-
rhoea for 7 days (range 1-15 days) (Jokipii and Jokipii, 1986). Asymptomatic
carriage has been reported in natural and experimental infections (Checkley
et al. 9 1997; Chappell et al. 9 1999). While the diarrhoeic phase may pose a
greater risk of onward transmission due to unpredictable faecal release, risks
from asymptomatic shedders have not been fully evaluated, but may be signifi-
cant since hygiene precautions may be more relaxed.

Immunocompetent patients are able to resolve cryptosporidiosis spon-
taneously, albeit after prolonged diarrhoea compared with many other gas-
trointestinal illnesses. However, in patients with impaired cell-mediated
immunity, particularly with reduced lymphocyte and CD4 T cell counts of
<200/mm 3 , gastrointestinal symptoms are chronic, severe, debilitating and
indeed life threatening. In a study of HIV patients, fulminant infection, where
patients passed 2 litres of watery diarrhoea per day, occurred with CD4
counts of <50 cells/mm 3 (Blanshard et al. 9 1992). Prior to the introduction of
highly active antiretroviral therapy (HAART) in 1996, cryptosporidiosis
affected 10-15% of AIDS patients, causing death in 50% of cases (Clifford etal. 9
1990). Colonization throughout the gastrointestinal tract has also been reported
in patients with primary immunodeficiencies, such as common variable
immunodeficiency, hypogammaglobulinaemia, severe combined immunodefi-
ciency, X-linked hyper IgM syndrome (CD40 ligand deficiency) or gamma
interferon deficiency, in secondary immunodeficiencies due to HIV/AIDS,
organ transplantation and immunosuppressive drugs, haematological malig-
nancies and anti-cancer chemotherapy (Farthing, 2000). Extraintestinal and

241

Protozoa

extra-abdominal cryptosporidiosis has been reported and resulted in pancrea-
titis, chronic cholangiopathy, sclerosing cholangitis, cirrhosis, cholangiocarci-
noma and respiratory involvement (Farthing, 2000).

Chronic health effects of cryptosporidiosis include Reiter's syndrome
(Shepherd et aL, 1988). This is a reactive arthritis that has been identified
following bouts of diarrhoea caused by a number of aetiological agents, and
has been linked to cryptosporidiosis in children (Cron and Sherry, 1995).
Long-term effects of childhood cryptosporidiosis have been measured, par-
ticularly in children in developing countries. For example, in Guinea-Bissau
undernourished children below 3 years of age suffered significant weight loss
and impaired growth, which was not followed by subsequent catch-up
growth (Molbak et aL, 1997). However, in the investigation of gastrointest-
inal infection, it is often hard to unravel the elements of the malnutrition-
infection cycle and the immune consequences of malnutrition. Studies in Peru
have shown that a measurable effect occurred in the growth of children who
were not severely or acutely malnourished following cryptosporidial infec-
tions, even in the absence of diarrhoea (Checkley et al. 9 1998). While catch-up
growth was reported in older children this was age-related, and reduced in
younger children. However, this did not occur in infants who were under
5 months at the acquisition of Cryptosporidium.

Pathogenicity

Cryptosporidium diarrhoea is caused largely by intestinal transport defects
resulting from infection of enterocytes in the small intestine (Clark and
Sears, 1996). Blunting of the microvilli and villous tip damage causes mal-
absorption of sodium ions, but there is also evidence for increased secretion
of chlorine ions stimulated by cell infection or infiltration of inflammatory
cells into the lamina propria. The production of alpha interferon by infiltrat-
ing macrophages and release of inflammatory mediators, such as exogenous
prostaglandin E 2 production, may stimulate chlorine ion secretion. Add-
itionally, disruption of the epithelium due to hyperplasia of the intestinal
crypt cells, cell defects and cell death can lead to increased paracellular per-
meability. However, the exact role of inflammatory cells, enteric nerves,
cytokines or hormones in the pathogenesis of Cryptosporidium diarrhoea are
as yet unknown.

Speculation of a cholera-like toxin has been made because of the profuse
nature of the watery diarrhoea experienced by some patients and the detec-
tion of enterotoxin-like activity in the filtered faecal supernate from
C. parvum-'miected calves (Guarino et aL 9 1994). However, studies have failed
to differentiate between parasite-derived enterotoxin and host-derived factors
and, in further studies, between other possible causes of diarrhoea.

242

Cryptosporidium spp.

Virulence

The infectious dose has been shown in human infectivity studies to be low,
with infection occurring following challenge of subjects without evidence of
prior infection with 30 oocysts (DuPont et al. 9 1995). The median infective
dose in that study was 132 oocysts. Although not statistically significant,
higher doses of oocysts resulted in occurrence of one or more gastrointestinal
symptoms, shorter incubation periods and longer duration of illness. In similar
studies of volunteers with prior exposure, infection and diarrhoea were asso-
ciated with higher challenge doses and the ID 50 was over 20-fold higher
(1880) than that in seronegative subjects (Chappell et al. 9 1999).

Human infectivity studies have thus far been undertaken using C. parvum
isolates, and where the infectivity of different isolates has been compared,
variation in ID 50 , attack rates and duration of diarrhoea has been observed
(Okhuysen etal., 1999). The median infectious dose of three different C. parvum
isolates ranged from nine to 1042 oocysts. The apparently limited host range
of C. hominis indicates that this may be a human-adapted species with differ-
ing infectivity for humans from C. parvum.

Although there are differences in pathogenicity and infectivity between isol-
ates and differences in antigenic profile and in host immunoreactivity have
been observed, the molecular basis for this is poorly understood in Crypto-
sporidium. Factors responsible for the initiation, establishment and perpetu-
ation of infection are poorly defined. Potential candidate molecules for virulence
factors include those involved in locomotion, adhesion, fusion, invasion, and
the investigation of heat shock proteins, toxins and host cell apoptosis
and have been reviewed by Okhuysen and Chappell (2002). Genome surveys
and the development of cell culture methods, particularly long-term mainten-
ance, will assist in identifying and determining the relative importance of
expressed and deleted virulence factors.

Causation

The clinical consequences of C. parvum infection are directly linked to the
immune function of the host. Autoinfective oocysts and recycling of type 1
meronts are probably features contributing to persistent disease in immuno-
compromised patients who are not repeatedly exposed to environmentally
resistant oocyst forms. AIDS patients have been most studied, particularly in
the USA and UK, and have shown a significant relationship between immuno-
suppression evaluated by total lymphocyte and CD4 counts, with fulminant
infection occurring in patients with a CD4 count of <50 cells/mm 3 (Blanshard
et al. 9 1992). Median time of survival of this group of patients was 5 weeks. In
developing countries, particularly sub-Saharan Africa, the relationship between

243

Protozoa

CD4 count and severe infection is less clear with more profound pathology and
clinical manifestations, and differing background enteropathy suggests that
other host-related factors may be important in determining the outcome of
infection (Kelly et aL, 1997).

Avoidance of exposure to the parasite is imperative among high-risk groups.
In the UK, patients whose T-cell function is compromised and those with specific
T-cell deficiencies are advised by the Department of Health to boil and then
cool all their drinking water (including that used for making ice), which may
have helped limit the number of infections in these groups. In the USA, guid-
ance created by the USEPA and CDC also recommends boiling, but alterna-
tives include filtering through absolute 1 fjim filters.

Treatment

There is no established curative therapy for cryptosporidiosis. Management
of diarrhoea can be supported by fluid replacement and electrolyte balance.
While cryptosporidiosis in otherwise healthy patients may be self-limiting,
immunocompromised patients and those in poor health or suffering malnu-
trition are at high risk of severe illness. Many of the drugs investigated for
efficacy against Cryptosporidium have been the ones used in the treatment of
infection with eimeriid coccidians. However, differences between these organ-
isms and Cryptosporidium are being recognized and may explain the appar-
ent insensitivity to anti-coccidial drugs.

In the absence of reliably effective curative chemotherapy, suppression
of proliferation of the parasite has been reported, particularly with paro-
momycin (White et al. 9 1994; Flanigan et al. 9 1996), albendazole and nita-
zoxanide in HIV-negative children (Rossignol et al. 9 1998, 2001; Amadi
et al. 9 2002) and paromomycin in combination with azythromycin (Smith
et al. 9 1998), although many clinical and other trials have been inconclusive
(Clinton Wight et al. 9 1994; Hoepelman, 1996; Theodos et al. 9 1998; Hewitt
et al. 9 2000). This may in part be due to the varying reactions in different
patient groups: paromomycin has been used to suppress the parasite in HIV
patients, while nitazoxanide only showed efficacy in HIV-negative children.
Additionally, the Cryptosporidia isolated in drug trials are rarely character-
ized. Multi-drug regimens that reduce viral load and increase CD4 T lympho-
cytes, such as highly active antiretroviral therapy (HAART) prescribed to
AIDS patients in developed countries, have led to the resolution of symptoms
and reduction of severe health effects of cryptosporidiosis in this patient
group. Where HAART is available AIDS-related cryptosporidiosis is a
decreasing problem. However, this treatment is not available in developing
countries, and rapid relapse has been reported after discontinuation of
HAART (Carr et aL 9 1998; Maggi et al. 9 2000). Additional approaches to
treatment of cryptosporidiosis, particularly for vulnerable groups of patients,

244

Cryptosporidium spp.

have little supporting evidence of success but include administration of hyper-
immune bovine colostrums, and new approaches to be explored further
include glutamane, blocking prostaglandin and ligand/receptor blocking.

Survival in the environment

Cryptosporidium oocysts have been detected in a variety of environmental
matrices including farmyard manure, leachate, slurry, and soil (Kemp et aL,
1995), as well as various water matrices (see below) and foodstuffs (Rose
and Slifco, 1999). Oocysts have been detected in sewage discharge from
treatment works and data show that neither removal nor inactivation by pri-
mary and secondary treatment can be guaranteed (Robertson and Gjerde,
2000). If the integrity of the thick, two-layered oocyst wall is maintained,
Cryptosporidium oocysts are robust and resistant to a variety of environmen-
tal pressures particularly under cool, moist conditions. Rose and Slifco (1999)
have reviewed the survival of Cryptosporidium under various food preserva-
tion conditions. Extremes of temperature and reduction in water activity
including freeze-drying, temperatures above 60°C and below — 20°C for
30 minutes will kill Cryptosporidium (Anderson, 1985), as will low tempera-
ture, long time and high temperature, short time pasteurization conditions
(Harp et aL, 1996). Blewett (1989) demonstrated, using excystation studies,
that oocysts are killed by exposure to moist heat at 60°C and above for 5
minutes. Survival in manure heaps and slurry stores is adversely affected
by the pH, temperature and ammonia generated (Bukhari et aL, 1995;
Jenkins et aL, 1998). Although many common disinfectants used on farms,
in hospitals or veterinary surgeries have little effect on Cryptosporidium,
both hydrogen peroxide and ammonia inactivate oocysts (Blewett, 1989;
Casemore et aL, 1989).

However, survival of oocysts, for example in food processing procedures,
and the efficacy of disinfectants proposed for water treatment such as
ozone or UV, is poorly evaluated due to lack of reliable, routine methods
to assess infectivity and viability (Casemore and Watkins, 1999). Addition-
ally, the oocysts and isolates used in survival or disinfection studies have
often been poorly characterized in terms of source, age, storage, purification
method and viability. While the gold standard might be human infectivity,
animal models are available but the mouse model is not applicable to
C. hominis isolates. Additionally, animal infectivity is not suitable for routine
assessment. Alternative methods include vital dyes, excystation and cell cul-
ture. In vitro assays tend to overestimate viability when compared with a
mouse model (Clancy et aL, 2000) and further evaluation of cell culture
with different isolates is currently required. It is important that a valid assess-
ment of viability and infectivity is made for the proper evaluation of disinfec-
tion interventions.

245

Protozoa

Survival in water

The occurrence of Cryptosporidium oocysts in environmental waters depends
on the land use within the catchment (since greater concentrations of oocysts
are detected in waters receiving animal and sewage discharges), climate (since
rainfall can influence the numbers of oocysts in surface waters and temperature
affects their survival) and community factors such as watershed management
(Rose et al. 9 2002). In a study undertaken in New Zealand, the highest sample
prevalence of Cryptosporidium was in areas of intensive livestock farming
(Ionas et al. 9 1998). In an on-farm study in the UK, Cryptosporidium was
detected in surface waters throughout the year but with the highest frequency
and maximum concentrations during the autumn and winter, coinciding with
calving and peaks in wildlife populations, but not in that study with rainfall or
slurry spreading (Bodley-Tickell et al. 9 2002). The authors also studied a small
pond at the top of a catchment which was not under the influence of livestock,
in which the only source of oocysts detected could have been wildlife. Other
studies, however, have detected a link between climatic factors and (oo)cyst
concentrations, and noted the distribution in the number of human cases often
follows a seasonal (rainfall-related) pattern (Rose et al. 9 2002). Urban waters
are also vulnerable to contamination from diverse sources, including sewage
outfalls. In La Palta, Argentina, much of the endemic cryptosporidiosis is prob-
ably caused by the high level of contamination through the discharge of raw
sewage into the city's main water source, the La Palta river, despite conven-
tional, but probably overloaded, treatment (Basualdo et al. 9 2000).

Cryptosporidium occurs frequently in raw waters world-wide. In the USA,
surface waters were monitored under the EPA's Information Collection
Rule between 1996 and 1998. Data show that 20% of the 5858 samples
contained Cryptosporidium oocysts, providing a mean national estimate of
2 oocysts/1001 water, although the sampling was based on relatively low
volumes of water (median 3 1) (Messner and Wolpert, 2000). While water moni-
toring data from a range of countries have previously been summarized and
shown that the mean number of wastewater samples containing Crypto-
sporidium oocysts was 62%, the mean for surface waters was 46% and the
mean number of drinking waters was 24% (Smith and Rose, 1998), further
data from additional countries are now available. For example, in Russia,
23/87 (26%) raw river waters throughout the European Russian region
were found to contain Cryptosporidium oocysts (Egorov et al. 9 2002). In
Japan, one study showed that all 13 raw water samples entering a treatment
works contained Cryptosporidium oocysts (Hashimoto et al. 9 2002), while
more widespread sampling showed that 74/156 river water samples contained
Cryptosporidium (Ono et al. 9 2001). Ground waters have traditionally been
considered protected from contamination but those under the influence of sur-
face waters or other routes of contamination are vulnerable and oocysts have
been detected in such waters and outbreaks have occurred in communities sup-
plied from such water sources (Hancock et al. 9 1998; Willocks et al. 9 1998).

246

Cryptosporidium spp.

Numbers of oocysts detected and, indeed, frequency of detection will
depend on sampling strategies and methods of oocyst recovery and detection.
In the UK, continuous monitoring of treated water is legally required at those
treatment plants considered as being at risk of having Cryptosporidium in the
treated water, based on a structured risk assessment under the Water Supply
(Water Quality) (Amendment) Regulations 1999 and Water Supply (Water
Quality) Regulations 2000. Monitoring is based on continual sampling at a
rate of 40 1/hour, using foam filters, with a treatment standard of 1 oocyst/10 1.
Data show that between April 2000 and June 2002, oocysts were detected in
5305/109704 samples from 166/204 sites. However, 91% of the detections
were at less than 10% of the treatment standard, which was violated in just
seven samples from two sites (Peter Marsden, Drinking Water Inspectorate,
personal communication). The methods used for this monitoring do not
include viability assessment (in fact, methods for oocyst recovery may them-
selves affect viability, and fixing oocysts on microscope slides renders them
non-viable). Although Cryptosporidium oocysts experience die-off under envir-
onmental pressures, this is relatively slow and variable. Survival of up to 176
days has been reported in drinking water or river water with inactivation of
89-99% of the oocyst population (Robertson et al. 9 1992). Oocysts are resist-
ant to chlorine at levels used to treat both drinking water and in swimming
pools (Rose et al. 9 1997; Carpenter et al. 9 1999) and so there is no defence
should oocysts breach physical barriers present in full water treatment (floc-
culation, sedimentation/flotation and filtration) or if such removal treatment
is absent. Chlorine dioxide has shown some efficacy against Cryptosporidium,
inactivating about 90% of oocysts, but the most successful treatments that
may be of use in large-scale treatment are ozone and UV (Peeters et al. 9 1989).
Medium and low pressure UV light are capable of inactivating Cryptosporidium
oocysts. Relatively low doses of 9mJ/ml achieve >3 log reduction in viability
(Bukhari et al. 9 1999; Craik et al. 9 2001) and although repair of UV-damaged
DNA has been reported, oocysts did not recover infectivity (Shin et al. 9 2001).
By producing free radicals that affect the permeability of the oocyst wall and
its DNA, ozone has proven to be highly effective against Cryptosporidium,
particularly at lower temperatures (Peeters et al. 9 1989; Rennecker et al. 9
1999). Sequential inactivation has been demonstrated using ozone followed
by free chlorine (Li et al. 9 2001). Efficacy tests for the treatment of small vol-
umes of drinking water have shown that iodine disinfection is not effective
(Gerbae**/., 1997).

Methods of detection

Despite the absence of specific therapy to eliminate infection, there is a need
for differential diagnosis to prevent inappropriate application of drug therapy,

247

Protozoa

to support disease surveillance and outbreak investigations and ensure appro-
priate public health measures are in put in place to prevent onward transmis-
sion. Clinical diagnosis is most commonly by identification of oocysts in
stools by examination of stained smear on microscope slides (Arrowood,
1997). Acid-fast stains and auramine phenol stains are most commonly used
in clinical laboratories, although increased sensitivity and specificity can be
achieved by use of immunofluorescence microscopy. Enzyme immunoassays
and other commercial test kits are available (Petry, 2000) and widely used in
private laboratories.

The oocysts of many Cryptosporidium species are indistinguishable by
microscopy, and the antibodies in test kits are not generally species specific
(Anon, 1997a). The detection of Cryptosporidium oocysts by these methods
should therefore only be reported as 'Cryptosporidium spp.' and molecular
methods are required for species determination. These have contributed
greatly to our understanding of the heterogeneity of the organism and its
epidemiology. Characterization of isolates using DNA amplification-based
methods is advantageous over phenotypic methods since relatively few organ-
isms are required (Gasser and O'Donoghue, 1999). Genotyping methods are
largely PCR-based and have included analysis of repetitive DNA sequences,
randomly amplified polymorphic DNA (RAPD), direct polymerase chain
reaction (PCR) with DNA sequencing, and PCR restriction fragment length
polymorphism (RFLP) analysis (Clark, 1999; Morgan et al. 9 1999). The two
distinct C. parvum genotypes, now recognized as separate species (C. hominis
and C. parvum) have been consistently differentiated at a variety of gene
loci, including Cryptosporidium oocyst wall protein (COWP), ribonuclease
reductase, 18S rDNA (syn. small subunit ribosomal RNA), internal tran-
scribed rDNA spacers (ITS1 and ITS2), acetyl-CoA synthetase, dihydrofolate
reductase-thymidylate synthase (dhfr-ts), thrombospondin-related adhesive
proteins (TRAP-C1 and TRAP-C2), the a and (3 tubulin and 70kDa heat
shock protein (hsp70) (Fayer et al. 9 2000). It is evident that some primer pairs
are more species specific than others, such as those for TRAP-C2 which are
specific for C. parvum and C. hominis (Elwin et al. 9 2001), while others
amplify related protozoan parasites, and that some PCR-RFLPs differentiate
species/genotypes more readily than others (Sulaiman et al., 1999). While
PCR-RFLP is widely used for characterization, and allows many specimens
to be analysed and compared, only bases at the restriction enzyme sites are
examined. Sequence analysis provides the 'gold standard' since all the bases
within the target sequence at the locus are examined. The importance of
sequence confirmation of RFLP patterns was illustrated by Chalmers et al.
(2002b) who identified a novel RFLP pattern, similar to C. hominis, in the
COWP gene of isolates from sheep, but sequence data clearly differentiated
the isolate. Therefore, careful primer selection and PCR product analysis are
required for detection and characterization, particularly from environmental
specimens where a wide range of Cryptosporidia and other organisms may be
present. It must also be noted that oocyst recovery and genotyping methods
have yet to be standardized.

248

Cryptosporidium spp.

Detection of Cryptosporidium oocysts in non-clinical samples involves
concentration of the oocysts from large volumes of the sample matrix (usually
by centrifugation, filtration or flocculation), separation from the sample
matrix (by density gradient centrifugation or immunomagnetic separation)
and detection by immunofluorescence microscopy (Anon, 1999a,b). The
development of immunomagnetic separation techniques has contributed
greatly to the improvement of methods for detection in environmental matri-
ces, including water, slurry and food, and enhanced our evaluation of food-
borne risks. Using IMS/IFA and improved washing procedures, Robertson
and Gjerde (2000) reported mean recovery efficiencies for Cryptosporidium
of 44% from iceberg lettuce, 48% from green lollo lettuce, 41% from Chinese
leaves, 38% from Autumn salad mix, 39% strawberries and 22-35% from
bean sprouts.

The methods for the detection of Cryptosporidium in non-clinical samples
based on immunodetection generally do not differentiate C. parvum from
other Cryptosporidium species, nor do they differentiate live from dead oocysts.
While the PCR-based genotyping analyses described above indicate the pres-
ence of a strain that may cause human/animal infection, they cannot deter-
mine if infectivity is maintained. The 'gold standard' is host infectivity studies,
which have been undertaken in humans and animals, but this is obviously not
practical for routine assessment or during disinfection studies. While animal
models for C. parvum exist, there is no practical animal model for C. hominis.
Cell culture methods have been used to determine viability, but application of
cell culture for disinfection efficacy trials requires further evaluation for sensi-
tivity and specificity (Gasser and O'Donoghue, 1999). Viability assessment as
a surrogate for infectivity has been undertaken in vitro using vital dyes and
excystation, but these do not consistently correlate with in vivo studies
(Clancy et al., 2000). The viability of C. parvum oocysts has been determined
by induction of hsp70 and subsequent detection of messenger RNA by RT-
PCR but has not been developed as a quantitative assay, and may result in the
detection of residual mRNA (Gasser and O'Donghue, 1999). Fluorescence
in situ hybridization (FISH), specific for C. parvum, has been developed for
detection and viability assay (Vesey et aL 9 1998) and can combine target-
specific detection with confirmation and viability estimation in one test.
However, specificity and sensitivity, as well as sampling difficulties must be
taken into account if Cryptosporidium viability assays are to be applied to
environmental samples.

Epidemiology

The epidemiology of cryptosporidiosis differs between developed and develop-
ing countries: in some developing countries, infection is common in infants
aged less than 1 year, while in developed countries incidence is most common
in 1-5 year olds, with a secondary peak in young adults (Palmer and Biffin,

249

Protozoa

1990). The peak in adult disease is rarely seen in developed countries where
more frequent exposure may generate greater immunity. During waterborne
outbreaks, a relative increase in adult cases is often seen (Meinhardt et al. 9
1996). As with many gastrointestinal infections, cryptosporidiosis is more
frequently reported in boys than girls, but adult males and females are gener-
ally affected with equal frequency. Several studies world-wide have shown
seasonal peaks in reported disease (Casemore, 1990), particularly in spring
and autumn, which do not necessarily both occur in any one locality, nor
recur year by year. They coincide generally with agricultural activities such as
muck spreading, lambing and calving, with maximal rainfall, and following
foreign travel.

Substantial outbreaks of cryptosporidiosis have been caused by a variety
of sources of infection and routes of transmission, and reflect the ubiquity of
the organism. These have included consumption of drinking water from both
surface and ground water sources contaminated with human sewage and
animal manure, contaminated natural and man-made recreational waters,
animal contact during farm visits, person-to-person spread within institutions
and day-care centres including children's nurseries, and food-borne out-
breaks (Casemore et al. 9 1997; Rose et al. 9 1997). Person-to-person spread
in households and institutions is important and has been estimated at 60%.
Contact with another person with diarrhoea, particularly children, has
been identified as a risk factor in case control studies (Puech et al. 9 2001;
Robertson et al. 9 2002) and has been demonstrated in outbreaks (Hannah and
Riordan, 1988).

In developed countries, foreign travel is frequently associated with illness,
attributable to exposure of naive subjects to new risks and isolates, often as a
result of poor hygiene. The prevalence of Cryptosporidium varies throughout the
world and is higher in developing countries (6.1% diarrhoeic patients; 1-5.2%
non-diarrhoeic) than developed countries (2.2% diarrhoeic patients; 0.2% non-
diarrhoeic) (Guerrant, 1997). In developing countries cryptosporidiosis rarely
occurs in indigenous adult populations. In neonates in these countries there
appears to be an association with cryptosporidiosis and bottle-feeding, pos-
sibly due to a combination of contaminated water supplies and the absence
of any protective effect of breast-feeding (Casemore et al. 9 1997). By contrast,
in developed countries, although more cases are reported in children than
adults, cryptosporidiosis is also a disease of adults.

The application of molecular tools to collections of Cryptosporidium isol-
ates has further elucidated the epidemiology of human cryptosporidiosis, and
shown that regional and seasonal differences exist, probably reflecting differ-
ing exposures and behaviours (McLauchlin et al. 9 2000; Anon, 2002). For
example, regional differences may reflect urban/rural or human/zoonotic
cycles of transmission. Seasonal differences in the UK may be linked to animal
husbandry and reproduction, resulting in a spring increase in human
C. parvum infections. An increase in C. hominis in the late summer is linked
to reports of recent foreign travel during the summer holidays. This, however,

250

Cryptosporidium spp.

is worthy of further investigation to identify more precisely the risks during
foreign travel. Furthermore, species variations are observed when the data
are analysed by countries visited (Anon, 2002). Thus far, little is known of
the epidemiology of non-C. parvum, non-C. hominis infections in humans, of
which C. meleagridis predominates.

Analysis of outbreak samples has confirmed that urban transmission is not
restricted to C. hominis but can occur with C. parvum. For example, in an out-
break associated with an indoor swimming pool in England, where the likely
source of contamination was human faecal material, C. parvum was con-
firmed in 34/41 cases (Anon, 2000a). The identification of the species causing
human illness has also been of benefit during outbreaks in the form of the pro-
vision of advice regarding appropriate control measures. For example, during
an outbreak in Belfast, Northern Ireland, epidemiologically linked to the
drinking water supply, the source of contamination was initially assumed to
be livestock since the source water arose in a rural area and flowed through an
aqueduct under agricultural land prior to distribution. However, PCR-RFLP
of the COWP gene identified C. hominis in the human cases indicating human
sanitation failure as the source of contamination and infection (Glaberman
et al. 9 2002). Inspection of the aqueduct showed a point of ingress of domestic
sewage and remedial action was taken.

Food-borne illness has been attributed to Cryptosporidium and food
items associated with outbreaks of illness by descriptive and analytical
epidemiology include fresh-pressed apple juice (Millard et al. 9 1994; Anon,
1997b), chicken salad (Besser-Wick et al. 9 1996) and improperly pasteurized
milk (Gelletli et al. 9 1997). Infected food handlers have been linked to
two outbreaks (Quinn et al. 9 1998; Quiroz et al. 9 2000) and in one case isol-
ates from cases of illness and the food handler were indistinguishable by
PCR-RFLP (Quiroz et al. 9 2000). Additionally, surveys of raw foods have
demonstrated oocysts in molluscan shellfish in Ireland (Chalmers et al. 9
1997), Hawaii (Johnson et al. 9 1995), Chesapeake Bay, USA (Fayer et al. 9
1998; Graczyk et al. 9 1999) and Spain (Gomez-Bautista et al. 9 2000), raw
market vegetables in Costa Rica (Monge and Chinchilla, 1996) and Peru
(Ortega et al 9 1997).

The sources of oocysts on foodstuffs may be direct faecal contamination,
infected food handlers, wastewater used in irrigation or water used in food
processing. Uncooked produce requires proper washing to remove surface
contamination. The impact of Cryptosporidium on food-borne illness is prob-
ably underestimated, as is the impact of food-borne protozoal disease gener-
ally. A substantial number of outbreaks of gastrointestinal illness are not
attributed to an aetiological agent, and the lack of sensitive methods for the
detection of cryptosporidial oocysts in food matrices may have contributed to
this. The introduction of immunomagnetic separation has facilitated greater
recovery of oocysts from the sample matrix, but variable recoveries highlight
the need for improved methods for detection of Cryptosporidium from foods
(Robertson and Gjerde, 2000).

251

Protozoa

Recreational activities, such as those undertaken in natural recreational
waters and man-made swimming pools, have been associated with outbreaks
of cryptosporidiosis. The parasite has been increasingly recognized as a problem
in swimming pools since the first two reported outbreaks associated with
swimming pools occurred in 1988. One of these outbreaks was in Doncaster,
UK where plumbing defects were identified allowing ingress of sewage into
the circulating pool water (Joce et al. 9 1991). The other outbreak was in Los
Angeles at a pool where one of the filters was inoperative (Sorvillo et al. 9
1992). Sources of Cryptosporidium in natural waters can be human or ani-
mal, but in swimming pools human sources predominate, either from breach
of the pool water by sewage or faecal accidents in the pool. A survey of 54
pools in Germany identified Cryptosporidium oocysts in 16/94 (17%) sam-
ples of filter back wash water from swimming pools where cryptosporidiosis
had not been reported, further demonstrating the widespread nature of the
organism (Marcic et al. 9 2000). Toddler pools in particular were among the
positive pools. In New South Wales, Australia, a case-control study identified
swimming at a public pool and swimming in a dam, river or lake as associated
with having cryptosporidiosis (Puech et al. 9 2001). Contact with a person
with diarrhoea was identified as a risk factor in rural areas, and swimming in
a public pool in urban areas. In the UK, 18 outbreaks of cryptosporidiosis
associated with swimming pools were reported for the 10 years from 1989 to
1999, but seven were during 1999 alone (Anon, 2000b). The reasons for the
apparent increase may be genuine or as a result of improved outbreak investi-
gation as well as increased awareness of swimming pools as a risk factor for
cryptosporidiosis. Since pool water filtration systems were not designed with
awareness of Cryptosporidium, it is likely that they are not efficient at its
removal and the main public health measure is to keep faecal material and
hence Cryptosporidium out of the pool. This can be achieved by public edu-
cation, encouraging people with diarrhoea not to swim, improved pool facil-
ities (pre-swim showers, toilets and hand washing) and policies on dealing
with faecal accidents available to all pool operators. Outbreaks of crypto-
sporidiosis have not been linked to seawater, but oocysts have been detected
in marine waters (Johnson et al. 9 1995; Fayer et al. 9 1998) and marine
mammals (Johnson et al. 9 1995).

Despite the many sources and routes of transmission of Cryptosporidium,
drinking water has attracted the greatest attention since the first waterborne
outbreak of cryptosporidiosis was documented in 1985 in the USA
(D' Antonio etal. 9 1985). Traditionally, groundwater sources of drinking water
are considered protected from contamination, but in the spring of 1997 an
outbreak associated with a deep borehole supply occurred in North Thames,
England (Willocks et al. 9 1998) and has implications for the understanding of
water quality for such supplies. Drinking waterborne outbreaks of crypto-
sporidiosis have been the largest in terms of numbers of human cases, and
expose all members of the community who use potable water. The largest docu-
mented waterborne outbreak occurred in Milwaukee in April 1993, and was
first detected because of the high level of absenteeism in schools and among

252

Cryptosporidium spp.

staff at local hospitals (MacKenzie et al. 9 1995). Although 739 cases were
confirmed by laboratory diagnosis of Cryptosporidium in a stool specimen,
a telephone survey of households was undertaken and estimated, from the
numbers of people with watery diarrhoea, that the extent of the outbreak was
403 000 cases.

Drinking waterborne outbreaks of cryptosporidiosis have been associated
with both C. parvum and C. hominis. Sporadic and outbreak cases, descrip-
tive data and epidemiological evidence have demonstrated the possibility
of human sewage and animal sources of contamination in source waters
(Casemore, 1998). These can come from direct breaches into the water or via
agricultural or natural run-off and effluent such as abattoir waste. In a recent
outbreak in Northern Ireland, subtyping methods also illustrated the presence
of the same C. hominis subtype in the human cases and in the suspected
(human) source of contamination (Glaberman et al. 9 2002). The application
of genotyping techniques is providing further information about the role of
animals in human cryptosporidiosis and has demonstrated that microscopical
detection of oocysts alone from possible sources is not proof of source of
infection (Chalmers et al. 9 2002b). The public health significance of oocysts
when detected in environmental samples must be investigated and one study
has shown that a variety of species and genotypes were detected in surface and
wastewaters, some of which are not known to be infectious for humans (Xiao
et al. 9 2001a). This has implications for both monitoring and for outbreak
investigations.

Private water supplies (i.e. supplies not managed by a water company) are
often in rural areas, inadequately protected from grazing animals and surface
run off, and many do not receive any treatment. Risks from private water sup-
plies may be underestimated since clusters of cases may be within families
and not reported as outbreaks and, indeed, people who have been drinking
the water for years may have generated some immunity. However, visitors
may be at risk and with increasing diversification of commercial enterprise in
the countryside are likely to be an increasing population.

Epidemiological studies to identify exposures during outbreaks can be
affected by underlying immunity in the community generated by frequent
exposure (Meinhardt et al. 9 1996) and may be a particular problem in analyt-
ical studies of outbreaks associated with surface waters (Hunter and Quigley,
1998). Public health measures to control drinking-water outbreaks of crypto-
sporidiosis include changes in the source of the water provided and notice to
boil water for consumption (Hunter, 2000). Boil water notices themselves
impact on the community with increased risk of scald injuries (Mayon- White
and Frankenberg, 1989), increased energy demands, problems for local indus-
tries who use mains water in food manufacturing, and for services such as
hospitals. There may also be an adverse effect on tourism. Risk of disease is
only reduced if the notice is put in place while pathogens are still present in
the water, and if compliance occurs. Non-compliance was reported in over
50% people in the target area during one outbreak (O'Donnell et al. 9 2000).
Implicit to the imposition of a boil water notice must be the criteria for lifting

253

Protozoa

it. As a result of water-related illness, and activities to control it, loss of
public confidence in the water supply can be substantial.

Risk assessment

The potential risk of waterborne cryptosporidiosis lies with the biological
properties of the organism, particularly the robust oocyst, and environmental
and climatic factors affecting introduction of oocysts into source waters and
their survival, water treatment affording removal and disinfection and com-
munity factors in the exposed population. The exact significance of environ-
mental oocysts for human health depends upon their viability and infectivity
for humans, and protective immunity afforded by prior exposure in the popu-
lation. Mathematical models for quantitative risk assessment of waterborne
cryptosporidiosis have been explored in a number of different ways. Dose
response models have estimated the number of oocysts required to cause ill-
ness after infection, which show similarity to volunteer infectivity studies,
except at very low doses (Teunis et al. 9 1999; Messner et al., 2001) and demon-
strate a near linear dose-response curve. This negates the need for further
modelling doses for risk prediction since an arithmetic mean exposure is suf-
ficient. However, it remains to be seen whether this assumption will hold true
as further genetic analyses reveal more about the population structure of
Cryptosporidium. There is evidence for recombination within species which
could lead to the generation of more virulent genotypes. Transmission
models, incorporating the multiple pathways involved in the transmission of
Cryptosporidium, have been used to estimate risk from drinking water (Teunis
and Havelaar, 1999; Chick et al. 9 2001) and Chick and colleagues addition-
ally explored the effect of secondary transmission. Routes of exposure for
consideration in mathematical risk assessment models are illustrated with the
UK regulatory framework by Gale (2002), since the aim of risk assessment is
to identify worthwhile interventions. It is important, however, to note that
risk assessment models must acknowledge variability (the intrinsic hetero-
geneity of a variable) and uncertainty (ignorance caused by methodological
limitations). For Cryptosporidium there is considerable variation in the num-
bers of oocysts in source waters and in their removal by water treatment, and
uncertainty particularly impacts on accurate measurement leading to best esti-
mates of oocyst numbers and distribution in raw waters, removal processes
and performance, disinfection performance, and detection parameters in
treated waters (recovery, infectivity and viability) (Gale, 2001).

The further differentiation of subtypes within Cryptosporidium genotypes
provides additional resolution for epidemiological investigations (Glaberman
et al. 9 2002), and information about the infectivity of 'strains' for humans.
A variety of tracking tools are being investigated and evaluated for further

254

Cryptosporidium spp.

segregation including the application of microsatellite typing (Caccio, 2000;
Blasdall et aL 9 2001). Sequence analysis of small double-stranded extra
chromosomal RNAs in C. parvum (Xiao et al. 9 2001b) and of a highly poly-
morphic gene encoding a 60kDa glycoprotein (Strong et al. 9 2000), analysis
of single strand conformation polymorphisms and mutation scanning (Gasser
et al. 9 2003) also offer potential as tracking tools.

Despite many attempts to correlate oocyst detection and counts in water
with both traditional and novel indicator organisms, no reliable surrogate for
Cryptosporidium has been widely accepted. Recently, there has been better
identification of risks leading to the detection of oocysts in water supplies,
and event-based sampling is being investigated. However, approaches to moni-
toring and legislation differ world-wide. The USA Environment Protection
Agency's (EPA) surface water treatment rule under the National Primary
Drinking Water Regulations as of 1 January 2002, for systems using surface
water or groundwater under the direct influence of surface water requires dis-
infection or filtration to meet the criterion of 99% removal/inactivation.
Methods are prescribed by the EPA methods 1622 (Cryptosporidium) (Anon,
1999a) and 1623 (Cryptosporidium and Giardia) (Anon, 1999b) for the
assessment of the occurrence of these parasites in raw surface source waters.
These specify the testing of 10-litre volumes of water and the use of capsule
filtration, immunomagnetic separation and immunofluorescence antibody
staining with confirmation through vital dye inclusion and differential inter-
ference contrast microscopy. Alternative oocyst recovery methods such as
membrane filtration, vortex flow and continuous flow centrifugation are per-
mitted with appropriate evaluation and quality control.

In Australia, following the Sydney water crisis during which increased num-
bers of oocysts were detected in the water supply but no rise in the number of
cases of cryptosporidiosis in the community was detected, a risk-based frame-
work has been developed, assessing the systems in place from catchment to
tap (Fairley et al. 9 1999). This is in line with current WHO revisions of guide-
lines for drinking water incorporating source-to-customer risk assessment. By
contrast, in England and Wales, the Water Supply (Water Quality) (Amendment)
Regulations 1999 came in to force in June 1999 and were replaced, in January
2001, by the Water Supply (Water Quality) Regulations 2000. Water under-
takers must conduct risk assessments with respect to Cryptosporidium on all
water-treatment works and set out the results of the assessment. Sites with a
'significant risk' classification, based on consideration of the source water,
catchment characteristics and treatment provided, are obliged, under the regu-
lations, to treat the water to ensure that the standard is maintained. This
'treatment standard', i.e. an average of less than 1 oocyst in 10 litres of treated
water supplied, measured by continuous sampling of at least 40 litres of water
per hour, must be met and compliance demonstrated by continual monitoring
and reporting of results. Significant risk works do not have to be monitored if
all particles >1 |xm are continuously removed. The USA approach monitors
the removal of Cryptosporidium from the water while the UK approach

255

Protozoa

measures what is in the treated water and it is a criminal offence to breach this
treatment standard. While this is an operational treatment standard, not a
health-related standard, the implications for public health of Cryptosporidium
in water are hard to ignore.

It has been suggested, from risk assessment models that a threshold value of
10-30 oocysts per 1001 be applied to drinking water for action to prevent an
outbreak of disease (Haas and Rose, 1995). However, it is very difficult at
present to set a health-related standard for Cryptosporidium in drinking
water in routine samples because the significance of oocysts in water samples
has not been defined. Using current monitoring methods, there is no species
differentiation and host specificity cannot be established, viability or infective
dose is not measured, and therefore infectivity for humans is not established.
Additionally, the status of the herd immunity of the local population, gener-
ated by prior exposure, will influence the significance of the numbers of
oocysts for causing waterborne disease. It has been shown that individuals
with prior exposure vary in their responses to re-infection (Okhuysen et al.,
1998). The relationship between oocyst counts in water supplies and cases
of illness has not been established. In the past samples were collected retro-
spectively and it was rare to have samples taken prior to the outbreak. An
exception is one outbreak where 34 oocysts were detected per 10 litres (Anon,
1999c). Conversely, high oocyst counts have been demonstrated without
subsequent outbreaks of illness. While justification of the cost of continuous
monitoring has been questioned (Fairley et aL 9 1999), it is desirable that
oocysts are kept out of the drinking water supply, and while it is impossible to
characterize and assess the viability and infectivity of oocysts in water, treat-
ment-related standards will also help improve water quality standards gener-
ally. The data also contribute to the historical picture for that water supply
and trends in oocyst counts are probably more important than individual
numbers. The epidemiology of cryptosporidiosis has been advanced by the
identification and characterization of subtypes, which has also led to import-
ant developments in identifying sources of infection and action for preven-
tion. However, the public health significance of oocysts in environmental
samples depends on their viability and infectivity for humans, and the rela-
tionship between oocysts numbers and health risk needs to be evaluated in
this context.

Overall risk assessment

Health effects: occurrence of illness, degree of morbidity and mortality, prob-
ability of illness based on infection:

• Cryptosporidium has a world-wide distribution with more cases of
cryptosporidiosis reported among children than adults.

256

Cryptosporidium spp.

• The primary feature of cryptosporidiosis is profuse watery, sometimes
mucoid, diarrhoea, which can be accompanied by dehydration, weight loss,
anorexia, abdominal pain, fever, nausea and vomiting. In immunocompe-
tent patients symptoms usually last for about 1-2 weeks.

• Cryptosporidium causes acute, self-limiting gastroenteritis in the general
population and potentially fatal infection in the immunocompromised.

• Some people who are infected have no symptoms of the disease.

• Individuals with prior exposure vary in their responses to re-infection. It is
thought that low levels of exposure are protective against infection and illness.

Exposure assessment: routes of exposure and transmission, occurrence in
source water, environmental fate:

• Routes of exposure have included drinking water from both surface and
groundwater sources contaminated with human sewage and animal
manure, contaminated natural and man-made recreational waters, animal
contact, person-to-person spread within institutions and day-care centres,
and food-borne outbreaks. Person-to-person spread in households and
institutions is important and has been estimated at 60%.

• Groundwater under the influence of surface water has been known to become
contaminated, and waterborne outbreaks have resulted from groundwater
drinking water sources.

• Cryptosporidium occurs frequently in raw waters world-wide. The occur-
rence of Cryptosporidium oocysts in environmental waters depends on the
source water's susceptibility to animal manure and sewage, climate (since
rainfall can influence the numbers of oocysts in surface waters and tempera-
ture affects their survival) and community factors such as watershed man-
agement. There is uncertainty associated with the actual occurrence,
because it is difficult to test for the oocysts.

• Because of the thick oocyst wall, Cryptosporidium oocysts are robust and
resistant to a variety of environmental pressures particularly under cool,
moist conditions.

Risk mitigation: drinking-water treatment, medical treatment:

• It is difficult to determine how effective drinking water treatment is because
of the uncertainty associated with analytical methods to estimate oocyst
numbers, infectivity and viability.

• Oocysts are resistant to chlorine at levels used to treat both drinking water
and swimming pools.

• The most successful treatments that may be of use in large-scale treatment
are ozone and UV.

• There is no current curative medical treatment for cryptosporidiosis.
Dehydration from diarrhoea can be treated with fluid replacement and elec-
trolyte balance. While cryptosporidiosis in otherwise healthy patients may
be self-limiting, immunocompromised patients and those in poor health or
suffering malnutrition are at high risk of severe illness or death.

257

Protozoa

Among high-risk groups, such as the immunocompromised, avoiding expos-
ure to the parasite is imperative. Immunocompromised patients are advised
to boil their water or use appropriate filtration devices.

Future implications

It is clear from the study of reported outbreaks of cryptosporidiosis that
emerging risks are still being identified. For example, swimming pool-associ-
ated outbreaks have attracted much attention, particularly since information
regarding best practice for control and prevention have previously been lack-
ing and are currently under refinement. Issues of control of nosocomial infec-
tion, and advice for immunocompromised individuals are constantly under
review: is boiling water the best advice? Guidance for food manufacturers
using water during product processing is relatively new (see for example
Anon, 2000c). The use of new water sources/supplies in water-poor areas
(which are ever increasing) may pose a threat to human health.

Application of risk assessment methodologies to Cryptosporidium may
require new definitions of disease, since infection without diarrhoea can have
long-term health effects. Additionally, unknown or unaccounted factors may
impact on risk assessment, such as the varying infectivity of isolates for
humans, effects of cross-immunity, possible type variation in survival and the
effect of natural waters on survival and disinfection of oocysts. Better mark-
ers of viability/infectivity are required for adequate assessment. Development
and application of typing techniques provides the potential for better identifi-
cation of point sources of contamination but better understanding of the
dynamics of genetic exchange is required to complement progress in risk
assessment.

References

Amadi, B., Mwlya, M., Musuku, J. et al. (2002). Effect of nitazoxanide on mortality in
Zambian children with cryptosporidiosis: a randomised conrolled trial. Lancet, 360:
1375-1380.

Anderson, B.C. (1985). Moist heat inactivation of Cryptosporidium sp. Am J Public Hlth,
75: 1433-1444.

Anon. (1997a). Cryptosporidium and water: a public health handbook. Working Group
on Waterborne Cryptosporidiosis. CDC, Atlanta, Georgia.

Anon. (1997b). Outbreaks of Escherichia coli 0157 infection and cryptosporidiosis asso-
ciated with drinking unpasteurised apple-cider - Connecticut and New York, October
1996. MMWR, 46:4-8.

Anon. (1999a). Method 1622: Cryptosporidium in water by filtration/IMS/FA. United States
Environmental Protection Agency.

258

Cryptosporidium spp.

Anon. (1999b). Method 1623: Cryptosporidium and Giardia in water by filtration/

IMS/FA. United States Environmental Protection Agency.
Anon. (1999c). Outbreak of cryptosporidiosis in the North West of England. CDR Wkly,

9: 175-178.
Anon. (2000a). Surveillance of waterborne disease and water quality: January to June

2000, and summary of 1999. CDR Wkly, 10: 319-322.
Anon. (2000b). Review of outbreaks of cryptosporidiosis in swimming pools and advice on

proceedings of strategic workshop on viability tests and genetic typing. Final report to

DEFRA: Drinking Water Inspectorate Foundation for Water Research, Marlow, Bucks,

UK (http://www.fwr.org/).
Anon. (2000c). Water quality for the food industry: management and microbiological

issues. Guideline no. 27. Campden and Chorleywood Research Association Group.
Anon. (2002). The development of a national collection for oocysts of Cryptosporidium .

Final report to DEFRA: Drinking Water Inspectorate. Foundation for Water Research,

Marlow, Bucks, UK (http://www.fwr.org/).
Arrowood, M.J. (1997). Diagnosis of Cryptosporidium and cryptosporidiosis.

In Cryptosporidium and Cryptosporidiosis , Fayer, R. (ed.). Boca Raton: CRC Press,

pp. 43-46.
Basualdo, J., Pezzani, B., de Luca, M. et al. (2000). Screening of the municipal water sys-
tem of La Palta, Argentina, for human intestinal parasites. IntJ Hyg Environ Hlth, 203:

177-182.
Besser-Wick, J.W., Forfang, J., Hedberg, C.W. et al. (1996). Foodborne outbreak of diar-

rhoeal disease associated with Cryptosporidium parvum - Minnesota 1995. MM WR,

45:783.
Blewett, D.A. (1989). Disinfection and oocysts. In Cryptosporidiosis: Proceedings of the First

International Workshop, Angus, K. and Blewett, D.A. (eds). Edinburgh: Animal Disease

Research Association, pp. 107-115.
Blanshard, C, Jackson, A.M., Shanson, D.C. et al. (1992). Cryptosporidiosis in HIV-

seropositive patients. Q J Med, 85: 813-823.
Blasdall, S.A., Ongerth, J.E. and Ashbolt, N.J. (2001). Differentiation of Cryptosporidium

parvum subtypes by a novel microsatellite-telomere PCR with PAGE. Proceedings of

Cryptosporidium from Molecules to Disease, 7-12 October 2001, Esplanade Hotel

Fremantle, Western Australia. Murdoch University, Perth.
Bodley-Tickell, A.T., Kitchen, S.E. and Sturdee, A.P. (2002). Occurrence of Cryptosporidium

in agricultural surface waters during an annual farming cycle in lowland UK. Water

Res, 36: 1880-1886.
Bukhari, Z., Smith, H.V., Humphreys, S.W. et al. (1995). Comparison of excretion and

viability patterns in experimentally infected animals: potential for release of viable

C. parvum oocysts into the environment. In Protozoan Parasites and Water, Betts, W.B.,

Casemore, D.P., Fricker, C. et al. (eds). London: The Royal Society of Chemistry,

pp. 188-191.
Bukhari, Z., Clancy, J.L., Hary, T.M. et al. (1999). Medium pressure UV for oocysts

inactivation. J A WWA, 91: 86-94.
Caccio, S., Homan, W., Camilli, R. et al. (2000). A micro satellite marker reveals popula-
tion heterogeneity within human and animal genotypes of Cryptosporidium parvum.

Parasitology, 120: 237-244.
Carpenter, C, Fayer, R., Trout, J. et al. (1999). Chlorine disinfection of recreational water

for Cryptosporidium parvum. Emerg Infect Dis, 5: 579-584.
Carr, A., Marriott, D., Field, A. et al. (1998). Treatment of HIV-1 -associated microsporidio-

sis and cryptosporidiosis with combination antiretro viral therapy. Lancet, 351:

256-261.
Casemore, D.P. (1990). Epidemiological aspects of human cryptosporidiosis. Epidemiol

Infect, 104: 1-28.
Casemore, D.P. (1998). Cryptosporidium and the safety of our water supplies. Commun

Dis Public Hlth, 4: 218-219.
Casemore, D.P. and Jackson, B. (1984). Hypothesis: cryptosporidiosis in human beings is

not primarily a zoonosis./ Infect, 9: 153-156.

259

Protozoa

Casemore, D.P. and Watkins, J. (1999). Review of disinfection and associated studies on

Cryptosporidium. London: Report to DEFRA: DWI.
Casemore, D.P., Sands, R.L. and Curry, A. (1985). Cryptosporidium species a 'new' human

pathogen. / Clin Pathol, 38: 1321-1336.
Casemore, D.P., Blewett, D.A. and Wright, S.E. (1989). Cleaning and disinfection of equip-
ment for gastro-intestinal flexible endoscopy. Gut, 30: 1156.
Casemore, D.P., Wright, S.E. and Coop, R.L. (1997). Cryptosporidiosis - human and ani-
mal epidemiology. In Cryptosporidium and Cryptosporidiosis, Fayer, R. (ed.). Boca

Raton: CRC Press, pp. 65-92.
Chalmers, R.M., Sturdee, A.P., Casemore, D.P. et al. (1994). Cryptosporidium muris in

wild house mice (Mus musculus): first report in the UK. Eur J Protistol, 30:

151-155.
Chalmers, R.M., Sturdee, A.P., Mellors, P. et al. (1997). Cryptosporidium parvum in envir-
onmental samples in the Sligo area, Republic of Ireland: a preliminary report. Lett Appl

Microbiol, 25,380-384.
Chalmers, R.M., Elwin, K., Thomas, A. and Joynson, D.H.M. (2002a). Unusual types of

Cryptosporidia are not restricted to immunocompromised patients. / Infect Dis, 185:

270-271.
Chalmers, R.M., Elwin, K., Reilly, W.J. et al. (2002b). Cryptosporidium in farmed animals:

the detection of a novel isolate in sheep. Int J Parasitol, 32: 21-26.
Chappell, C.L., Okhuysen, P.C., Sterling, C.R. et al. (1999). Infectivity of Cryptosporidium

parvum in healthy adults with pre-existing anti-C. parvum serum immunoglobulin G.

Am J Prop Med Hyg, 60: 157-164.
Checkley, W., Epstein, L.D., Gilman, R.H. et al. (1998). Effects of Cryptosporidium

parvum in Peruvian children: growth faltering and subsequent catch up growth. Am J

Epidemiol, 148: 497-506.
Chick, S.E., Koopman, J.S., Soorapanth, S. et al. (2001). Infection transmission system

models for microbial risk assessment. Sci Total Environ, 274: 197-207.
Clancy, J.L., Bukhari, Z., McCuin, R.M. et al. (2000). Cryptosporidium viability and

infectivity methods. Denver, CO: AWWA Research Foundation.
Clark, D.P. (1999). New insights into human cryptosporidiosis. Clin Microbiol Rev,

12: 554-63.
Clark, D.P. and Sears, C.L. (1996). The pathogenesis of cryptosporidiosis. Parasitol Today,

12:221-225.
Clifford, C.P., Crook, D.W.M., Conlon, C.P. et al. (1990). Impact of waterborne

outbreak of cryptosporidiosis on AIDS and renal transplant patients. Lancet, i:

1455-1456.
Clinton-Wight, A., Chappell, C.L., Sikander Hayat, C. et al. (1994). Paromomycin for

cryptosporidiosis in AIDS: a prospective double-blind trial./ Infect Dis, 170: 419-424.
Craik, S.A., Weldon, D., Finch, G.R. et al. (2001). Inacitivation of Cryptosporidium parvum

oocysts using medium and low pressure ultraviolet radiation. Water Sci Res, 35:

1387-1398.
Cron, R.Q. and Sherry, D.D. (1995). Reiter's syndrome associated with cryptosporidial

gastroenteritis./ Rheumatol, 22: 1962-1963.
Current, W.L. and Hayes, T.B. (1984). Complete development of Cryptosporidium in cell

culture. Science, 224: 603-605.
Current, W.L. and Long, P.L. (1983). Development of human and calf Cryptosporidium in

chicken embryos. / Infect Dis, 148: 1108-1113.
Current, W.L. and Reese, N.C. (1986). A comparison of endogenous development of three

isolates of Cryptosporidium in suckling mice./ Protozool, 33: 98-108.
D'Antonio, R.G., Winn, R.E., Taylor, J.P. et al. (1985). A waterborne outbreak of

cryptosporidiosis in normal hosts. Ann Intern Med, 103: 886-888.
DuPont, H.L., Chappell, C.L., Sterling, C.R. etal. (1995). The infectivity of Cryptosporidium

parvum in healthy volunteers. New Engl J Med, 332: 855-859.
Egorov, A., Paulauskis, J., Petrova, L. et al. (2002). Contamination of water supplies with

Cryptosporidium parvum and Giardia lamblia and diarrhoeal illness in selected Russian

Cities. Int J Hyg Environ Hlth, 205: 281-289.

260

Cryptosporidium spp.

Elwin, K., Chalmers, R.M., Roberts, R. et al. (2001). The modification of a rapid method

for the identification of gene-specific polymorphisms in Cryptosporidium parvum, and

application to clinical and epidemiological investigations. Appl Environl Microbiol,

67-.55S1-55S4.
Fairley, C.K., Sinclair, M.I. and Rizak, S. (1999). Monitoring drinking water: the receeding

zero. MJA, 171: 397-398.
Farthing, M.J.G. (2000). Clinical aspects of human cryptosporidiosis. In Cryptosporidiosis

and Microsporidiosis, Petry, F. (ed.). Basel: Karger, 6, 50-74.
Fayer, R., Speer, C.A. and Dubey, J.P. (1997). The general biology of Cryptosporidium .

In Cryptosporidium and Cryptosporidiosis, Fayer, R. (ed.). Boca Raton: CRC Press,

pp. 1-41.
Fayer, R., Graczyk, T.K., Lewis, E.J. et al. (1998). Survival of infectious Cryptosporidium

parvum oocysts in seawater and Eastern oysters (Crassostrea virginica) in the

Chesapeake Bay. Appl Environ Microbiol, 64: 1070-1074.
Fayer, R., Morgan, U. and Upton, SJ. (2000). Epidemiology of Cryptosporidium: trans-
mission, detection and identification. Int J Parasitol, 30: 1305-1322.
Fayer, R., Trout, J.M., Xiao, L. et al. (2001). Cryptosporidium canis n. sp from domestic

dogs. / Parasitol, 87: 1415-1422.
Flanigan, T.P., Ramratnam, B., Graeber, C. et al. (1996). Prospective trial of paromomycin

for cryptosporidiosis in AIDS. Am J Med, 100: 370-372.
Gale, P. (2001). Developments in microbiological risk assessment for drinking water.

J Appl Microbiol, 91: 191-205.
Gale, P. (2002). Using risk assessment to identify future research requirements. JAWWA,

September: 30-38.
Gasser, R.B. and O'Donoghue, P. (1999). Isolation, propagation and characterization of

Cryptosporidium. Int J Parasitol, 29: 1379-1413.
Gasser, R.B., Abs El-Osta, Y.G. and Chalmers, R.M. (2003). An electrophoretic analysis of

genetic variability within Cryptosporidium parvum from imported and autochthonous

cases of human cryptosporidiosis in the United Kingdom. Appl Environ Microbiol,

69:2719-2730.
Gelletli, R., Stuart, J., Soltano, N. et al. (1997). Cryptosporidiosis associated with school

milk. Lancet, 350: 1005-1006.
Gerba, C.P., Johnson, D.C. and Hasan, M.N. (1997). Efficacy of iodine water purification

tablets against Crypto sp or dium oocysts and Giardia cysts. Wilderness Environ Med,

8: 96-100.
Glaberman, S., Moore, J.E., Lowery, C.J. et al. (2002). Three drinking-water-associated

cryptosporidiosis outbreaks, Northern Ireland. Emerging Infect Dis, 8: 631-633.
Gomez-Bautista, M., Ortega-Mora, L.M., Tabares, E. et al. (2000). Detection of infectious

Cryptosporidium parvum oocysts in mussels (Mytilus galloprovincialis) and cockles

(Cerastoderma edule). Appl Environ Microbiol, 66: 1866-1870.
Graczyk, T.K., Fayer, R., Lewis, E.J. et al. (1999). Cryptosporidium oocysts in Bent mus-
sels (Iscbadium recurvum) in the Chesapeake Bay. Parasitol Res, 85: 518-520.
Guarino, A., Canani, R.B., Pozio, E. et al. (1994). Enterotoxic effect of stool supernatant

of Cryptosporidium-infected calves on human jejunum. Gastroenterology, 106: 28-34.
Guerrant, R.L. (1997). Cryptosporidiosis: An emerging and highly infectious threat.

Emerging Infect Dis, 3: 51-57.
Haas, C. and Rose, J. (1995). Developing an action level for Cryptosporidium. JAWWA,

87: 81-84.
Hancock, CM., Rose, J.B. and Callahan, M. (1998). The prevalence of Cryptosporidium

and Giardia in US groundwaters. JAWWA, 90: 58.
Hannah, J. and Riordan, T. (1988). Case to case spread of cryptosporidiosis: evidence from

a day nursery outbreak. Public Hltb, 102: 539-544.
Harp, J.A., Fayer, R., Pesch, B.A. et al. (1996). Effect of pasteurisation on infectivity of

Cryptosporidium parvum oocysts in water and milk. Appl Environ Microbiol, 62:

2866-2868.
Hashimoto, A., Kunikane, S. and Hirata, T. (2002). Prevalence of Cryptosporidium oocysts

and Giardia cysts in the drinking water supply in Japan. Water Res, 36: 519-526.

261

Protozoa

Hewitt, R.G., Yiannoutsos, C.T., Higgs, E.S. et al. (2000). Paromomycin: no more effect
than placebo for treatment of cryptosporidiosis in patients with advanced human
immunodeficiency virus infection. AIDS clinical trial group. Clin Infect Dis, 31:
1084-1092.

Hoepelman, A.L. (1996). Current therapeutic approaches to cryptosporidiosis in immuno-
compromised patients. / Antimicrob Chemother, 37: 871-880.

Hunter, P.R. (2000). Advice on the response from public and environmental health to the
detection of cryptosporidial oocysts in treated drinking water. Commun Dis Public
Hlth, 3: 24-27.

Hunter, P.R. and Quigley, C. (1998). Investigation of an outbreak of cryptosporidiosis
associated with treated surface water finds limits to the value of case control studies.
Commun Dis Public Hlth, 1: 234-238.

Ionas, G., Learmonth, J.J., Keys, E.A. et al. (1998). Distribution of Giardia and Crypto-
sporidium in natural water systems in New Zealand - a nationwide survey. Water Sci
Technol, 38: 57-60.

Jenkins, M.B., Bowman, D.D. and Ghiorse, W.C. (1998). Inactivation of Cryptosporidium
parvum oocysts by ammonia. Appl Environ Microbiol, 64: 784-788.

Joce, R.E., Bruce, J., Kiely, D. et al. (1991). An outbreak of cryptosporidiosis associated
with a swimming pool. Epidemiol Infect, 107: 497-508.

Johnson, D.C., Reynold, K.A., Gerba, C.P. et al. (1995). Detection of Giardia and Crypto-
sporidium in marine waters. Water Sci Technol, 5-6: 439-442.

Jokipii, L. and Jokipii, A.M.M. (1986). Timing of symptoms and oocyst excretion in
human cryptosporidiosis. New Engl J Med, 315: 1643-1647.

Kelly, P., Davies, S.E., Mandanda, B. et al. (1997). Enteropathy in Zambians with HIV
related diarrhoea: Regression modelling of potential determinants of mucosal damage.
Gut, 41: 811-816.

Kemp, J.S., Wright, S.E., Coop, R.L. et al. (1995). Protozoan, bacterial and viral pathogens,
farm wastes and water quality protection. Final report to MAFF (CSA 2064).

Koudela, B. and Modry, D. (1998). New species of Cryptosporidium (Apicomplexa:
Cryptosporidiidae). Fol Parasitol, 45: 93-100.

Li, H., Finch, G.R., Smith, D.W. et al. (2001). Sequential inactiviation of Cryptosporidium
parvum using ozone and chlorine. Water Res, 35: 4339-4348.

Lindsay, D.S., Upton, S.J., Owens, D.S. et al. (2000). Cryptosporidium andersoni n. sp.
(Apicomplexa: Cryptosporiidae) from Cattle, Bos taurus. J Eukaryotic Microbiol, 47:
91-95.

MacKenzie, W.R., Schell, W.L., Blair, K.A. et al. (1995). Massive outbreak of waterborne
Cryptosporidium infection in Milwaukee, Wisconsin: recurrence of illness and risk of
secondary transmission. Clin Infect Dis, 21: 57-62.

McLauchlin, J., Amar, C, Pedraz-Diaz, S. et al. (2000). Molecular epidemiological analy-
sis of Cryptosporidium spp. in the United Kingdom: results of genotyping
Cryptosporidium spp. in 1705 fecal samples from humans and 105 fecal samples from
livestock animals./ Clin Microbiol, 38: 3984-3990.

Maggi, P., Larocca, A.M., Quarto, M. et al. (2000). Effect of antiretroviral therapy on
cryptosporidiosis and microsporidiosis in patients infected with human immunodefi-
ciency virus type 1. Eur J Clin Microbiol Infect Dis, 19: 213-217.

Marcic, A., Potyka, J., Siegfriedt, D. et al. (2000). Toddlers and small children: a source for
Cryptosporidia in swimming-pools. HRWM conference, Paris 2000.

Mayon-White, R.T. and Frankenberg, R.A. (1989). Boil the water. Lancet, ii: 216.

Meinhardt, P.L., Casemore, D.P. and Miller, K.B. (1996). Epidemiologic aspects of
human cryptosporidiosis and the role of waterborne transmission. Epidemiol Rev, 18:
118-136.

Meisel, J.L., Perera, D.R., Meligro, B.S. et al. (1976). Overwhelming watery diarrhoea
associated with Cryptosporidium in an immunosuppressed patient. Gastroenterology,
70: 1156-1160.

Messner, M.J. and Walpert, R.L. (2000). Occurrence of Cryptosporidium in the national
drinking water sources - ICR data analysis. In Water Quality Technology Conference
Proceedings. Denver, CO: AWWA, 2000.

262

Cryptosporidium spp.

Messner, M.J., Chappell, C.L. and Okhuysen, P.C. (2001). Risk assessment for

Cryptosporidium: a hierarchical Bayesian analysis of human dose response data. Water

Res, 35: 3934-3940.
Millard, P., Gensheimer, K., Addis, D. et al. (1994). An outbreak of cryptosporidiosis from

fresh-pressed apple cider. JAMA, 272: 1592-1596.
Molbak, K., Andersen, M., Aaby, P. et al. (1997). Cryptosporidium infection in infancy

as a cause of malnutrition: A community study from Guinea-Bissau, West Africa. Am J

Clin Nutr, 65: 149-152.
Monge, R. and Chinchilla, M. (1996). Presence of Cryptosporidium oocysts in fresh vege-
tables. / Food Protect, 59: 202-203.
Morgan, U.M., Xiao, L., Fayer, R. et al. (1999). Variation in Cryptosporidium: towards a

taxonomic revision of the genus. Int J Parasitol, 29: 1733-1751.
Morgan-Ryan, U.M., Fall, A., Ward, L.A. et al. (2002). Cryptosporidium bominis n. sp.

(Apicomplexa: Cryptosporidiidae) from Homo sapiens. Eukaryotic Microbiol, 49:

433-440.
Nime, F.A., Burek, J.D. and Page, D.L. (1976). Acute enterocolitis in a human being

infected with the protozoan Cryptosporidium. Gastroenterology, 70: 592-598.
O'Donnell, M., Piatt, C. and Aston, R. (2000). Effect of a boil water notice on behaviour

in the management of a water contamination incident. Commun Dis Public Hltb,

3: 56-59.
Okhuysen, P.C. and Chappell, C.L. (2002). Cryptosporidium virulence determinants - are

we there yet? Int J Parasitol, 32: 517-525.
Okhuysen, P.C, Chappell, C.L., Sterling, C.R. et al. (1998). Susceptibility and serologic

response of healthy adults to reinfection with Cryptosporidium parvum. Infection and

immunity, 66: 441-443.
Okhuysen, P.C, Chappell, C.L., Crabb, J.H. et al. (1999). Virulence of three distinct

Cryptosporidium parvum isolates for healthy adults./ Infect Dis, 180: 1275-1281.
Ono, K., Tsuji, H., Rai, S.K. et al. (2001). Contamination of river water by Cryptosporidium

parvum oocysts in Western Japan. Appl Environ Microbiol, 67: 3832-3836.
Ortega, Y.R., Roxas, C.R., Gilman, R.H. et al. (1997). Isolation of Cryptosporidium

parvum and Cyclospora cayetanensis from vegetables collected in markets in an

endemic region in Peru. Am J Prop Med Hyg, 57: 683-686.
Palmer, S.R. and Biffin, A. and the Public Health Laboratory Service Study Group. (1990).

Cryptosporidiosis in England and Wales: prevalence and clinical and epidemiological

features. Br Med J, 30: 774-777.
Panciera, R.J., Thomassen, R.W. and Garner, F.M. (1971). Cryptosporidial infection in a

calf. Vet Pathol, 8: 479-484.
Peeters, J.E., Mazas, E.A., Masschelein, W.J. et al. (1989). Effect of disinfection of drink-
ing water with ozone or chlorine dioxide on survival of Cryptosporidium parvum

oocysts. / Appl Environ Microbiol, 55: 1519-1522.
Petry, F. (2000). Laboratory diagnosis of Cryptosporidium parvum infection. In Crypto-
sporidiosis and Microsporidiois, Petry, F. (ed.). Basel: Karger, 6: 33-49.
Puech, M.C, McAnulty, J.M., Lesjak, M. et al. (2001). A statewide outbreak of crypto-
sporidiosis in New South Wales associated with swimming at public pools. Epidemiol

Infect, 126: 389-396.
Quinn, K., Baldwin, G., Stepak, P. et al. (1998). Foodborne outbreak of cryptosporidiosis-

Spokane, Washington, 1997. MMWR, 47: 565-567.
Quiroz, E.S., Bern, C, MacArthur, J.R. et al. (2000). An outbreak of cryptosporidiosis

linked to a food handler. / Infect Dis, 181: 695-700.
Rennecker, J.L., Marinas, B.J., Owens, J.H. et al. (1999). Inactivation of Cryptosporidium

parvum oocysts with ozone. Water Res, 33: 2481-2488.
Robertson, L.J. and Gjerde, B. (2000). Isolation and enumeration of Giardia cysts, Crypto-
sporidium oocysts, and Ascaris eggs from fruits and vegetables. / Food Protect, 63:

775-778.
Robertson, L.J., Campbell, A.T. and Smith, H.V (1992). Survival of Cryptosporidium

parvum oocysts under various environmental pressures. Appl Environ Microbiol, 55:

1519-1522.

263

Protozoa

Robertson, L.J., Paton, C.A., Campbell, A.T. etal. (2000). Giardia cysts and Cryptosporidium

oocysts at sewage treatment works in Scotland, UK. Water Res, 34: 2310-2322.
Robertson, B., Sinclair, M.I., Forbes, A.B. et al. (2002). Case-control studies of sporadic

cryptosporidiosis in Melbourne and Adelaide, Australia. Epidemiol Infect, 128:

419-431.
Rose, J.B. and Slifco, T.R. (1999). Giardia, Cryptosporidium, and Cyclospora and their

impact on foods: a review./ Food Protect, 62: 1059-1070.
Rose, J.B., Lisle, J.T. and LeChevallier, M. (1997). Waterborne cryptosporidiosis: inci-
dence, outbreaks, and treatment strategies. In Cryptosporidium and cryptosporidiosis,

Fayer, R. (ed.). Boca Raton: CRC Press, pp. 93-109.
Rose, J.B., Huffman, D.E. and Gennaccaro, A. (2002). Risk and control of waterborne

cryptosporidiosis. FEMS Microbiol Rev, 26: 113-123.
Rossignol, J.F., Hidalgo, H., Feregrino, M. etal. (1998). A double-blind placebo-controlled

study of nitazoxanide in the treatment of cryptosporidial diarrhoea in AIDS patients in

Mexico. Trans Roy Soc Trop Med Hyg, 92: 663-666.
Rossignol, J.F., Ayoub, A. and Ayers, M.S. (2001). Treatment of diarrhoea caused by

Cryptosporidium parvum: a prospective, randomised, double-blind, placebo-controlled

study of nitazoxanide. / Infect Dis, 184: 103-106.
Shepherd, R.C., Sinha, G.P., Reed, C.L. et al. (1988). Cryptosporidiosis in the West of

Scotland. Scot Med J, 33: 365-368.
Shin, G.A., Linden, K.G., Arrowood, M.J. et al. (2001). Low-pressure UV inactivation and

DNA repair potential of Cryptosporidium parvum oocysts. Appl Environ Microbiol,

67: 3029-3032.
Slavin, D. (1955). Cryptosporidium meleagridis (sp. nov.)./ Comp Pathol, 65: 262-266.
Smith, H.V. and Rose, J.B. (1998). Waterborne cryptosporidiosis: current status. Parasitol

Today, 14: 14-22.
Smith, N.H., Cron, S., Valdez, L.M. et al. (1998). Combination drug therapy for crypto-

sporidiois in AIDS. / Infect D is, 178: 900-903.
Sorvillo, F.J., Fujioka, K., Nahlen, B. et al. (1992). Swimming-associated cryptosporidiosis.

Am J Public Hltb, 82: 742-744 .
Strong, W.B., Gut, J. and Nelson, R.G. (2000). Cloning and sequence analysis of a highly

polymorphic Cryptosporidium parvum gene encoding a 60-kilodalton glycoprotein and

characterization of its 15- and 45-kilodalton zoite surface antigen products. Infect

Immun, 68: 4117-4134.
Sulaiman, I.M., Xiao, L. and Lai, A.A. (1999). Evaluation of Cryptosporidium parvum

genotyping techniques. Appl Environ Microbiol, 65: 4431-4435.
Tenter, A.M., Barta, J.R., Beveridge, I. et al. (2002). The conceptual basis for a new classi-
fication of the coccidian. Int J Parasitol, 32: 595-616.
Teunis, P.F.M. and Havelaar, A.H. (1999). Cryptosporidium in drinking water: evaluation

of the ILSI/RSI quantitative risk assessment framework. Report no. 284 550 006.

The Netherlands: National Institute of Public Health and the Environment, Bilthoven,

1999.
Teunis, P.F.M. , Nagelkerke, N.J.D. and Haas, C.N. (1999). Dose response models for infec-
tious gastroenteritis. Risk Analysis, 19: 1251-1260.
Theodos, CM., Griffiths, J.K., D'Onfro, J. et al. (1998). Efficacy of nitazoxanide against

Cryptosporidium parvum in cell culture and in animal models. Antimicrob Ag Chemother,

42: 1959-1965.
Tyzzer, E.E. (1907). A sporozoan found in the peptic glands of the common mouse. Proc

Soc Exp Med, 5: 12-13.
Tyzzer, E.E. (1910). An extracellular coccidium, Cryptosporidium muris (gen.et sp. nov.) of

the gastric glands of the common mouse. / Med Res, 23: 487-509.
Tyzzer, E.E. (1912). Cryptosporidium parvum (sp. nov.), a coccidium found in the small

intestine of the common mouse. Arch protistenkd, 26: 394-412.
Tyzzer, E.E. (1929). Coccidiosis in gallinaceous birds. Am J Hyg, 10: 269-383.
Vesey, G., Ashbolt, N., Fricker, E.J. et al. (1998). The use of a ribosomal RNA targeted

oligonucleotide probe for fluorescent labeling of viable Cryptosporidium parvum

oocysts. / Appl Microbiol, 85: 429-440.

264

Cryptosporidium spp.

White, A.C., Chappell, C.L., Hayat, C.S. et al. (1994). Paromomycin for cryptosporidiosis

in AIDS: a prospective, double blind trial./ Infect Dis, 170: 419-424.
Willocks, L., Crampin, A., Milne, L. et al. (1998). A large outbreak of cryptosporidiosis

associated with a public water supply from a deep chalk borehole. Comm Dis Public

Hlth, 1: 239-243.
Xiao, L., Singh, A., Limor, J. et al. (2001a). Molecular characterisation of Cryptosporidium

oocysts in samples of raw surface water and wastewater. Appl Environ Microbiol, 67:

1097-1101.
Xiao, L., Limor, J., Bern, C. et al. (2001b). Tracking Cryptosoridium parvum by sequence

analysis of small double-stranded RNA. Emerging Infect Dis, 7: 141-145.

265

Cyclospora cayetanensis

Basic microbiology

Cyclospora is a genus of obligate intracellular coccidian protozoan parasites
(Phylum Apicomplexa, Order Eucoccidiorida, Family Eimeriidae). Although
there are many species in the genus, the only one believed to infect humans is
Cyclospora cayetanensis (Ortega et al., 1994). The life cycle is initiated by the
sporulation of spherical oocysts (8-10 |xm) after a period of maturation in the
environment (see below), when two sporocysts develop within the oocyst,
each containing two sporozoites (Ortega et al., 1993). When the mature
oocysts are ingested, the sporozoites excyst and invade the enterocytes of the
small intestine (Bendall et aL, 1993; Sun et aL, 1996; Ortega et aL, 1997).
Reproduction follows, involving asexual and sexual stages, resulting in the
production of immature oocysts (Ortega et al., 1997). Endogenous stages
have been identified in the jejunum and duodenum and inhabit an intracyto-
plasmic parasitopherous vacuole (Bendall et aL, 1993; Sun et #/., 1996).
Asexual stages (trophozoites, type 1 meronts with eight to twelve merozoites
and type II meronts with four merozoites) and sexual stages (gametocytes)
have been detected in the same patient, demonstrating that the life cycle can
be completed in a single host (Ortega et aL, 1997).

Since the oocysts are unsporulated when they are shed in the faeces, they
are non-infectious at this stage. However, they mature and sporulate during

Protozoa

environmental exposure and become infectious. C. cayetanensis infection has
not been confirmed in any hosts other than humans nor has the organism been
experimentally transmitted to other hosts (Eberhard et al. 9 2000). Based on
comparative sequence analysis of small subunit ribosomal DNA, the genus
Cyclospora is most closely related to Eimeria, which are also host-species spe-
cific and complete their life cycle in a single host (Relman et aL 9 1996).

Distribution of the parasite is world-wide but infection appears to be
endemic throughout the tropics (Soave, 1996). In developing countries, the
epidemiology and clinical presentation vary according to social factors. About
30% naturally infected children in the shanty towns of Lima, Peru experience
diarrhoea of short duration, and symptomatic infection is rarely reported in
adults (Ortega et al. 9 1993). Upper class Peruvians, tourists and expatriates in
Nepal report more disease than the local population and diarrhoea can last
for over 1 month (Taylor et al. 9 1988). In developed countries, cases are most
frequently among people reporting foreign travel. However, these data may be
skewed by the use of foreign travel as a selection criterion for laboratory test-
ing (Cann et al. 9 2000). Despite this, cases have indeed been reported among
people who have not travelled outside the UK, USA and Germany. Although
the exact routes of transmission of C. cayetanensis have yet to be elucidated,
food- and waterborne disease have both been reported.

A number of features of the biology of C. cayetanensis affect its epidemi-
ology and transmission. The oocysts are shed in the faeces unsporulated and
require a period of maturation in the environment. This means that direct
person-to-person transmission is unlikely. However, the oocysts are robust and
can survive for long periods. Transmission is therefore likely to occur through
food and water, contaminated by human faeces.

Origin of the organism

Since the genus was created in 1881 for parasites found in myriapods
(Schneider, 1881), Cyclospora spp. have been described in many animal hosts
(Levine, 1982). Although not identified as such at the time, the first reports of
C. cayetanensis in humans were in 1979 when Ashford reported three cases
that occurred in the previous 2 years. Working in Papua New Guinea, he
recorded unsporulated oocysts in the faeces of two ill people and one well
person, which took some days to sporulate and thus be recognized as a coccidian
protozoan (Ashford, 1979). He postulated that the organism was Isospora, but
observed that until sporulated, the bodies could be mistaken for fungal spores.
Over the next decade few reports were made, but included those of an
'unsporulated, coccidian body but a fungal spore could not be ruled out'
(Soave et al. 9 1986) and ' Crypto sp >or dium muris-like objects' (Naranjo et al. 9
1989), and c cyanobacterium-like or coccidian-like body' (Long et al. 9 1991).
The latter was to be adopted as the reporting nomenclature, and coccidian-like

268

Cyclospora cayetanensis

bodies (CLBs) 8-10 |Jim were identified world-wide using acid-fast staining and
autofluorescing under UV light. The morphology of sporulated oocysts was
described in detail by Ortega et al. (1993) and this description provided evi-
dence for classification as Cyclospora. The morphological characteristics,
patient symptoms and apparent failure of conventional antimicrobial
chemotherapy linked Cyclospora to CLBs reported throughout the world, and
was reinforced by sporulation studies of various isolates (Ortega et al., 1993).
The species name cayetanensis was proposed and adopted, referring to the
location of the studies at the Universidad Peruana Cayetano Heredia, Lima,
Peru (Ortega et al., 1994). Increased use of acid-fast stains for the detection
of Cryptosporidium in stool specimens during the 1980s facilitated better
recognition of C. cayetanensis in primary diagnostic laboratories. Widespread
outbreaks in North America during the 1990s, often associated with the con-
sumption of raspberries imported from Guatemala, raised the profile of this
organism and the disease it causes.

Clinical features

The incubation period for cyclosporiasis has been determined from outbreaks,
and has a mean of 7 days (range 2-11 days) (Soave, 1996). The onset of diar-
rhoeal illness is usually abrupt, although it can be preceded by a prodromal ill-
ness of several days of flu-like symptoms (Soave, 1996). Stools are frequent and
sometimes explosive, and other symptoms include anorexia, nausea, vomiting,
abdominal bloating and cramps, weight loss, fatigue, low grade fever and body
aches. Fatigue is often reported to be profound. Although self-limiting in
immunocompetent patients, illness is often prolonged, and the duration of diar-
rhoea averages 5 days to 15 weeks in untreated patients (Brown and Rotschafer,
1999). Symptoms can be relapsing-remitting, including alternate diarrhoea and
constipation (Soave et al., 1998). Therefore, patients may not have diarrhoea at
the time of presentation to medical practitioners.

Oocyst shedding has been reported 3 months after initial detection but this
could represent prolonged infection or intermittent re-infection (Eberhad et al.,
1999a). Asymptomatic infection occurs in areas where C. cayetanensis is
endemic (Madico et al., 1997), and can provide an unevaluated source for the
transmission of disease.

Cyclosporiasis has occurred among immunocompromised patients, predomin-
antly those with HIV/AIDS. In endemic countries, the prevalence is higher in
AIDS patients than in diarrhoeic patients without AIDS (Chacin-Bonilla et al.,
2001), and prolonged, severe illness with a mean duration of 4 months has been
reported (Pape et al., 1994). There is also some evidence for biliary tract infec-
tion in AIDS patients, leading to acalculous cholangitis and cholecystitis
(Sifuentes-Osornio et al., 1995; de Gorgolas et al., 2001), including histological
evidence in the gallbladder epithelium and acalculous cholecystitis in a patient

269

Protozoa

with HIV who required a cholecystectomy (Zar et al. 9 2001). Therefore
HIV/AIDS patients are considered as having an elevated risk of cyclosporiasis.
Although symptoms usually resolve with parasite eradication, inflammatory
changes may persist. A myelin-like material has been identified as a marker for
persistent inflammation but requires further definition (Connor et al. 9 1999).
Reiter's syndrome, a triad of ocular inflammation, inflammatory oligoarthritis
and sterile urethritis that usually occurs several weeks after a triggering infec-
tion (Konttinen et al. 9 1988), has been reported following cyclosporiasis in a
31-year-old male in the USA (Connor et al. 9 2001). Cyclospora infection has
also been proposed as a trigger for Guillain-Barre syndrome (Richardson
et al. 9 1998). Post-infectious complications could be minimized by prompt,
effective treatment with appropriate anti- Cyclospora drugs.

Pathogenicity and virulence

Impaired absorption of D-xylose (Shlim et al. 9 1991; Connor et al. 9 1993)
implies involvement of the proximal small intestine and endogenous stages have
been identified in the epithelial cells of the jejunum and duodenum (Bendall
et al. 9 1993; Sun et al. 9 1996; Ortega et al. 9 1997). How C. cayetanensis causes
diarrhoea is not fully understood, but invasion of epithelial cells by micro-
organisms releases cytokines which in turn activate and recruit phagocytes from
the blood (Powell, 1995). Phagocytes release factors including histamine and
prostaglandin and platelet aggregating factors that increase intestinal secretion
of chloride and water and inhibit absorption (Ciancio and Chang, 1992).
Varying degrees of villous blunting, atrophy, crypt hyperplasia and inflamma-
tion of the lamina propria have been reported (Bendall et al. 9 1993; Connor
et al. 9 1993). In a controlled study, inflammatory changes caused by T cells, pro-
teases and oxidants secreted by mast cells were also associated with C. cayeta-
nensis infection (Connor et al. 9 1993). While intestinal inflammation can be
severe, numbers of oocysts present in stools often appear to be low. The path-
ology of the infection has been likened to tropical sprue for which C. cayeta-
nensis has been suggested as a trigger organism (Bendall et al. 9 1993). The
infectious dose is not known but, based on evidence from similar parasites such
as Cryptosporidium, and the high attack rate during outbreaks, even those asso-
ciated with foods consumed in small quantities, it is likely to be low.

Causation

The immune response to C. cayetanensis infection plays a critical role in the
clinical course and outcome but has not been characterized. Although

270

Cyclospora cayetanensis

immunocompromised patients, particularly those with AIDS, appear to har-
bour larger numbers of parasites than immunocompetent hosts, the prophy-
lactic use of trimethoprim sulphamethoxazole against Pneumocystis carinii
has probably contributed to the low prevalence of C. cayetanensis in AIDS
patients in North America and Europe, while a high prevalence has been
reported in Haiti where this prophylaxis is not widely available (Pape et al.,
1994). Antibodies have been detected in patients and shown to increase dur-
ing convalescence (Long et al. 9 1991), while others failed to detect them in
convalescent sera (Clark and Mclntyre, 1997). In contrast, a 10-fold increase
in serum IgM has been reported in convalescent patients compared with acute
phase sera from the same patients (Wurtz, 1994). That protective immunity
may be achieved is indicated by studies in Peru. Children who live in the
shanty towns of Lima and have poor sanitation may have more than one
infection episode, which is usually mild or asymptomatic (Bern et aL 9 2002a),
but infection and illness are rarely detected in those over 11 years of age
(Madicio et al. 9 1997). This contrasts with prolonged and more severe illness
reported in adults from more wealthy areas with good sanitation. Immuno-
logically naive adults experience more severe disease than those in whom
infection and immunity are established early in life in endemic areas. Thus
adults from non-endemic areas are at higher risk of cyclosporiasis than adults
from endemic areas.

Treatment

Although cyclosporiasis is generally self-limiting, the duration of illness is
prolonged and there is the possibility of chronic sequelae. Therefore differen-
tial diagnosis and treatment are desirable. Rehydration may be necessary and
can be undertaken orally, but antidiarrhoeals are ineffective. Trimethoprim
sulphamethoxazole provides rapid, effective anti -parasitic treatment (Soave
and Johnson, 1995), first demonstrated in a double-blind placebo controlled
trial in Nepal (Hoge et al. 9 1995), and is effective in both immunocompetent
and immunocompromised patients (Pape et al. 9 1994). Although there is no
recommended alternative treatment for sulph-allergic patients and treatment
with trimethoprim alone is ineffectual (Brown and Rotschafer, 1999),
ciprofloxacin, while not as effective as trimethoprim sulphamethoxazole, is
acceptable (Verdier et al., 2000).

Survival in the environment

Although Cyclospo ra-like bodies and C. cayetanensis have been detected in
environmental samples, including water, wastewater and foods, methods lack

271

Protozoa

sensitivity and few prevalence studies have been undertaken (see below). The
development of methods to assess the viability/infectivity of C. cayetanensis
has been hampered by the lack of an animal model, and cell culture methods
have yet to be established, although propagation has been reported in HCT-8
and Henle 407 cell monolayers (Miliotis et al. 9 1997). Surrogate methods such
as the inclusion/exclusion of vital dyes used for Cryptosporidium have not
been developed for Cyclospora. The ability to sporulate, with subsequent
excystation, has been used as a viability indicator (Ortega et al. 9 1994).
Excystation can be induced in vitro by exposure to bile salts, sodium tauro-
cholate and mechanical pressure (Ortega et al. 9 1994). The rate of sporulation
is probably influenced by environmental factors. For example, oocyst suspen-
sions in 2.5% potassium dichromate kept at 25°C and 32°C showed 20%
sporulation by day 5 and complete sporulation between 7 and 13 days
(Ortega et al. 9 1993). Suspensions maintained at 37°C only showed contrac-
tion and darkening of central mass after 5 days. However, the age of the
organisms on initiation of the experiment may affect the results. Although
data are sparse, it appears that sporulation does not occur following exposure
to -20°C for 24 hours or to 60°C for 1 hour thus rendering oocysts non-
infectious (Smith et al. 9 1997). However, an outbreak has been epidemiologi-
cally linked with the consumption of a cake containing raspberries at a
wedding, the raspberry rilling having been stored frozen, although the freezer
temperature was just -3.3°C (Ho et al. 9 2002). More rapid methods for
assessing viability, such as electrorotation (which also determines sporulation
state) have been explored (Dalton et al. 9 2001) using a purified suspension of
oocysts. However, applicability to parasites recovered from food, water or
environmental sources has yet to be demonstrated.

If Cyclospora behaves in a similar way to related parasites, and using infor-
mation from vehicles of infection implicated in outbreaks, a moist environment
is more likely to encourage survival than a dry one. Given the problems in
estimating the survival of C. cayetanensis due to the lack of an animal model
or surrogate methods, and paucity of material for experimentation, Eimeria
acervulina has been used as a surrogate organism to test decontamination
treatments (Lee and Lee, 2001). Chick-feeding experiments with E. acervulina
showed that freezing to — 18°C and heat treatment at >80°C for 60 minutes
rendered oocysts non-infectious. Gamma irradiation was completely effective
at 1.0 kGy or higher. Toxoplasma gondii has also been suggested as a surrogate
(Kniel et al. 9 2002), but the oocysts are highly infectious for humans and could
present a health and safety risk to laboratory personnel.

Survival in water

That a waterborne outbreak occurred following consumption of drinking
water containing acceptable levels of chlorine (Rabold et al. 9 1994) indicates
that C. cayetanensis is resistant to levels of chlorine used to treat potable
water. CLBs, C. cayetanensis oocysts and DNA have been detected in water

272

Cyclospora cayetanensis

Table 19.1 Detection of Cyclospora cayetanensis in water and wastewater

Study
area

Sample
type

Recovery
method

Detection
method

Results

Reference

Chicago,

Municipal

Not stated

Microscopy

CLBs seen

Wurtz et al.

USA

drinking water
from a surface
water supply
(Lake Michigan)

(1994)

Utah, USA

Sewer drain
pipe effluent

Not stated

Microscopy

CLBs seen

Hale et al.
(1994)

Nepal

Chlorinated
municipal
water supply

Membrane
filtration

Light
microscopy

Cyclospora
oocysts seen

Rabold etal.
(1994)

Lima, Peru

Wastewater

Envirocheck

Cyclospora

Sturbaum

filter capsules
and Haniffin

epifluorescence
microscopy

oocysts seen
and DNA

etal. (1998)

polypropylene
cartridge filters

and PCR

amplicons
detected

Guatemala

River water

Calcium

carbonate

flocculation

Microscopy

Cyclospora
oocysts seen

Bern etal.
(1999)

Nepal

Sewage water
Drinking water

Centrifugation

Microscopy

Cyclospora
oocysts seen
in sewage
water

Sherchand
etal. (1999)

and wastewater samples (Table 19.1) but difficulties in the detection of this
parasite present a challenge to the analytical laboratory (see below).

Methods of detection

Clinical diagnosis is based upon the detection of oocysts in faeces (Eberhard
et al., 1997). The oocysts are spherical and 8-10 |xm in diameter, and show vari-
able staining and a granular appearance with acid-fast stains, particularly the
modified Ziehl Neelsen stain used to detect Cryptosporidium oocysts. Other
suitable stains include Kinyoun and safranin. Not all oocysts present in a speci-
men take up the stain using cold acid-fast stains and it has been suggested that
conventional heating can improve the uptake of the stain (Jayshree et ah, 1998).
Microwave heating of faecal smears improves the uptake of a safranin-based
stain (Visvesvara et ah, 1997). Oocysts can also be detected by phase- or
interference contrast examination of a wet mount, and particular advantage
can be made of their ability to autofluoresce: when examined using blue epi-
illumination (using a 450-490 nm dichroic mirror exciter filter) they exhibit
green autofluorescence, and when examined using a 365 nm dichroic mirror fil-
ter they display blue autofluorescence. In a comparative study of detection
methods, using modified acid-fast stain as the gold standard, wet mount

273

Protozoa

sensitivity was 75%, safranin O was 30% and auramine rhodamine was 23%
(Pape et al. 9 1994), while the use of autofluorescence improves the detection
over conventional examination of wet mounts (Berlin etal. 9 1998).

Oocysts are often present in low numbers even in samples received at the
laboratory from clinically ill patients (Eberhard et al. 9 1999a), and therefore
concentration by the formalin-ethyl acetate technique may be required.
However, this is often difficult due to resource constraints in primary testing
laboratories. Other problems with diagnosis and surveillance include selective
screening in the first place, the relatively non-descript appearance of unsporu-
lated oocysts which may be overlooked or passed off as fungal spores, mis-
identification in acid-fast stains due to variable staining and incorrect ID,
often mistakenly as Cryptosporidium, due to failure to take into account the
size of the oocysts. Pseudo-outbreaks have been reported (Anon, 1997) and
laboratory proficiency in identification is relatively poor (Cann et al. 9 2000),
resulting in under ascertainment. Demonstration of sporulation and excysta-
tion, with observation of two sporozoites from within each of two sporocysts
confirms the diagnosis, but sporulation can take 7-13 days when stored at
room temperature in 2.5% potassium dichromate.

Demonstration of parasite DNA by PCR has been applied as a research tool.
To differentiate from other Cyclospora spp. or closely related Eimeria spp.,
DNA sequence analysis, restriction fragment length polymorphisms, or PCR
with mismatched primers are required (Relman et al. 9 1996). Suggestions that
Cyclospora should be considered a member of the genus Eimeria on the basis
of molecular analyses (Pieniazek and Herwaldt, 1997) remain unresolved, but
using structural and ultrastructural definitions according to zoological nomen-
clature the current taxonomy remains. A molecular diagnostic assay for iden-
tifying C. cayetanensis in faeces was developed by Yoder et al. (1996), based on
a nested PCR amplifying a segment of the 18S rDNA gene. While this assay
does not differentiate between Cyclospora and Eimeria, only the former is
found in humans. Molecular characterization differentiates Cyclospora spp.
found in non-human primates from those found in humans (Eberhard et ah,
1999b; Lopez et al. 9 1999) but multiple sequences of the intervening tran-
scribed spacer region 1 (ITS1) have been detected in human isolates indicating
the presence of either multiple clones in single clinical isolates or variability
within single clones (Adam et al. 9 2000; Olivier et al. 9 2001).

Methods for the recovery of Cyclospora from water and similar sample
matrices have often been based on those developed to detect Cryptosporidium.
For example, membrane filtration and light microscopy were used to detect
Cyclospora in chlorinated water supplied to homes of cases during an outbreak
in Nepal in 1994 (Rabold et al. 9 1994). The Envirocheck capsule and Haniffin
polypropylene cartridge filters followed by UV epifluorescence microscopy and
molecular tools were used to detect Cyclospora in wastewater in Peru
(Sturbaum et al. 9 1998). A nested PCR amplifying a 294 bp portion of the
18s rDNA gene was used with RFLP fragment digestion to differentiate
Cyclospora from Eimeria in water samples (Yoder et al. 9 1996). One of eight
samples was PCR-positive while four samples were microscopy-positive. The

274

Cyclospora cayetanensis

calcium carbonate flocculation method (Vesey et al., 1993) has also been suc-
cessfully used to isolate C. cayetanensis from river (Bern et al., 1999) and
wastewater (unpublished observation).

Cyclospora has been detected in prospective studies of fruit and vegetables
from markets in Peru, Egypt and Nepal. A variety of vegetables were collected
in three sample rounds in a peri-urban slum in Lima, Peru where C. cayeta-
nensis is endemic, and prepared by washing steps and centriflgation and
detected by wet mount observation, acid-fast staining and autofluorescence
(Ortega etal., 1997). None of 35 samples collected in a preliminary study were
positive, 2/110 (1.8%) vegetables collected at the end of the high incidence sea-
son were positive, and 1/62 vegetables collected at the beginning of the high
incidence season were positive. The vegetables concerned were Yerba buena,
huacatay and lettuce. However, experimental inoculation showed that the
recovery was just 13-15% and scanning electron microscopy demonstrated the
presence of oocysts on vegetables after washing. Cyclospora oocysts have also
been detected on lettuce in Egypt (Abou el Naga, 1999) and green leafy vege-
tables in Nepal (Sherchand et al., 1999). The use of antibody-coated paramag-
netic beads has improved the recovery of Cryptosporidium and Giardia cysts
from various matrices but have not been developed (due to the absence of suit-
able antibodies) for Cyclospora (Rose and Slifco, 1999). In an attempt to
improve recovery efficiencies, Robertson et al. (2000) used lectin-coated para-
magnetic beads. Oocyst recoveries from mushrooms, lettuce and raspberries
were around 12% with no significant difference with or without the beads,
although microscopy was greatly facilitated by their use. Recovery from bean
sprouts remained very low at 4%. Studies of food items are important in eluci-
dating the transmission of Cyclospora since many foods are produced using
poor-quality and wastewaters during, for example, irrigation.

Molecular techniques have been applied to the detection of Cyclospora in
food and are particularly useful where the nature of the food reduces the util-
ity of microscopy. Jinneman et al. (1999) used an oligonucleotide-ligation
assay that differentiates between Cyclospora and Eimeria. Sensitivity was 19
oocysts in an optimized template format and 25 oocysts on the addition of
raspberry extract. An ITS region was used by Adam et al. (2000), which also
provides genotype data of epidemiological value. Implicated foods have been
tested and confirmed by oocyst detection and PCR in basil (Lopez etal., 2001)
and PCR only in raspberry filling in a wedding cake (Ho et al., 2002).

Epidemiology

Most of the epidemiological data have been generated by studies in Peru,
Haiti, Guatemala and Nepal where cyclosporiasis is endemic. Cyclosporiaisis
has a marked but geographically variable seasonality. In Lima, Peru, which
has a desert, coastal location with minimal rainfall, prevalence is highest

275

Protozoa

during the warmer summer months of December to May (Madico et al. 9
1997) and cases are rarely detected in the cooler June to November period
(Bern et al. 9 2002a). In Nepal, which is at altitude, cases cluster prior to and
during the warm monsoon season (usually May to August) but decrease
before the monsoon ends (Shlim et al. 9 1991; Hoge et al. 9 1993, 1995). In
Guatemala prevalence of infection increases in May and peaks in June, coin-
ciding with the beginning of the rainy season (Bern et al. 9 1999). In Haiti,
peak prevalence is during the dry, cool winter months of January to March
(Eberhard et al. 9 1999a). Oocyst viability and infectivity may be affected by
ambient temperature and humidity, while transmission may be favoured by
rainfall. The varying combination of these factors may account for the differ-
ing geographical picture. However, few survival studies have been undertaken
because of the lack of a good model for viability or infectivity.

Surveillance in Guatemala has shown that 126/5552 (2.3%) specimens
screened over a 1-year period from patients submitting faecal specimens to two
health centres contained C. cayetanensis, while in patients without gastroen-
teritis the prevalence was 1.1% (Bern et al. 9 1999). The detection rate peaked
in June when 6.7% specimens were positive. Oocysts were detected more fre-
quently among children aged 1.5-9 years and among people with gastroenter-
itis, but rarely detected in children <18 months of age. Incidence was
significantly higher among HIV patients screened at a local clinic. Another
study found oocysts in the stools of 1.5% subjects, although the presence or
absence of symptoms was not stated (Pratdesaba et al. 9 2001). In Haiti, preva-
lence of the organism in cohorts of mothers and children in a rural community
peaked at 16% in March, but the organism was not detected in every month
(Eberhard et al. 9 1999a). Infection was more common in children than adults,
and there was no statistical difference in detection rates between diarrhoeic
and non-diarrhoeic specimens. In a study of diarrhoeic children in Nepal, 6/50
(12%) children aged <18 months were infected with Cyclospora, whereas none
of 74 over this age were (Hoge et al. 9 1995). In a cross-sectional study in a peri-
urban shanty town in Lima, Peru the prevalence of C. cayetanensis infection in
children under 18 years was 1% and was highest in children aged 2-4 years
(Madico et al. 9 1997). Infection was most prevalent during the summer when
3-4% of children were infected. One-third of infections were apparently asymp-
tomatic. The prevalence was lower than that reported during previous studies
commencing some 4 years earlier in the same area (Ortgea et al. 9 1993), but
sanitary conditions had improved. Due to the seasonal variation in incidence,
longitudinal surveillance data and studies are required fully to understand the
epidemiology of C. cayetanensis carriage and disease.

The observed age relationship may be due to waning maternal antibodies or
increased environmental exposure coincident with weaning. Additionally, the
development of acquired immunity may occur, and concurring with this is evi-
dence that length of stay and therefore possibly increased exposure among
travellers to Nepal has been associated with decreased symptoms (Hoge et al. 9
1993). Thus there are consistent themes in the epidemiology of endemic
cyclosporiasis: marked seasonality, high prevalence in children compared with

276

Cyclospora cayetanensis

adults and higher rates in those with than those without gastrointestinal
symptoms (Bern et al., 2002a).

The first case report suggesting food-borne transmission of Cyclospora was
in an airline pilot flying between Port-au-Prince, Haiti and New York in 1995
(Connor and Shlim, 1995). During the 1990s many outbreaks of cyclosporiasis
were reported in North America associated with the consumption of fresh pro-
duce, including raspberries, other berries, mesclun lettuce, basil and fruit salad
(Herwaldt, 2000). An outbreak has also been reported in Germany impli-
cating mixed whole leaf salad inclusive of fresh herbs sourced from whole-
salers in France and Italy (Brockman et al., 2001). However, two outbreaks,
one in the USA and one in Nepal, have been associated with the consumption
of drinking water.

The first outbreak of cyclosporiasis to be associated with drinking water
was in 1990 among staff at a hospital in Chicago, USA (Huang et al., 1995).
At the time organisms identified in stools of cases were described as algal-like
but later identified as C. cayetanensis. Nine out of 14 resident house staff and
1/7 other staff who met the case definition had stools positive for Cyclospora.
All either lived in the house-staff accommodation or had attended a party
there. Epidemiological investigations implicated the water supply but oocysts
were not detected in water samples (Kocka et al. 9 1991). The second outbreak
was among British soldiers and their dependents in June 1994 in Pokhara,
Nepal (Rabold etaL, 1994). Twelve out of 14 at risk became ill and C. cayeta-
nensis was identified in 6/8 specimens from clinical cases. The water supply
was a mixture of river water and municipal water, mixed and stored in a tank.
The water was chlorinated to microbiologically acceptable levels in the tank
and supplied to homes in a sealed pipe. Structures morphologically resem-
bling C. cayetanensis were found in a 21 sample of water from the tank.
Although the mode of contamination was not identified, it is plausible that the
river water became contaminated from a human source.

Previously, studies in Nepal revealed a cluster of more than 50 laboratory con-
firmed cases of cyclosporiasis, mainly among expatriate visitors, between May
and November during 1989 (Hoge et al., 1993). No further cases were detected
until May the following year when 85 cases were confirmed among visitors and
6/184 local people. Cases were detected until October, and the first cases each
year coincided with the monsoon period. In 1992 a case control study among
travellers and expatriates at two outpatient clinics in Kathmandu identified
drinking untreated water as a risk factor for cyclosporiasis and oocyst-like bod-
ies were also detected in the tap water from the household of one case (Hoge
et al., 1993). Cases of illness were more likely to have drunk reconstituted milk
than controls. Although significant, only 28% cases had drunk untreated tap
water and other routes of transmission may also have been involved. An out-
break centred on a golf club in New York in June 1995 was epidemiologically
linked to drinking water from water coolers but incomplete information meant
that a food-borne route could not be ruled out (Carter etaL 9 1996).

Clusters of cases have also been reported, including a family cluster in Peru
where four family members reported drinking from the same canal (Zerpa et al.,

277

Protozoa

1995). CLBs were also detected in faeces from a symptomatic duck but infection
was unconfirmed. In a case control study in Guatemala of cases largely recruited
from patients presenting at the health centres, risk factors identified in multi-
variate analysis included drinking untreated water in the 2 weeks before illness
(OR = 4.2, 95% CI = 1.4-12.5) and contact with soil among children <2
years old (OR = 19.8, 95% CI = 2.2-182) (Bern et al. 9 1999).

Individual case reports have also linked cases to waterborne sources but again
cannot rule out other exposures. These have included a man in Utah, USA who
cleaned up his basement when it became flooded with sewage, and where CLBs
were also detected in the effluent from the sewer drain pipe (Hale et aL 9 1994);
a child who swam in Lake Michigan, and water from there to the Chicago
municipal supply contained presumptive CLBs (Wurtz, 1994). The consumption
of well water was implicated in a case in Massachussetts, USA (Ooi et al. 9 1995).
Indigenous cyclosporiasis in developed countries needs closer examination.
One study in the USA has linked infection to gardening and working with soil
(Koumans et al. 9 1996).

Prospective sampling of wastewater in sewage lagoons in Lima, Peru which
receive waste from shanty towns where Cyclospora is endemic, confirmed the
presence of C. cayetanensis oocysts by microscopy and PCR (Sturbaum et al. 9
1998). The water from these lagoons is used to irrigate pasture land, corn-
fields and trees, while water from similar lagoons in other parts of Lima is
used to irrigate vegetable crops. It is possible that the use of water contamin-
ated with human faeces in the production of crops is one of the routes by
which C. cayetanensis is transmitted. In a study of river water in Guatemala,
2/30 (7%) specimens from two different rivers contained Cyclospora oocysts
(Bern et al. 9 1999). Interestingly, the river water was positive during May,
coinciding in the seasonal rise in human cases which peaks in June, which also
coincides with the spring raspberry harvest.

Risk assessment

One of the problems of evaluating the waterborne risk presented by
Cyclospora is that many of the fundamental data are missing. The relative
insensitivity and lack of standardization of detection methods has made the
determination of exact modes of transmission difficult. However, there is both
epidemiological and microbiological evidence for a waterborne route. While
outbreaks in developed countries have been primarily caused by contamin-
ated produce, in some developing countries the main vehicle of infection is
drinking water and the use of contaminated water in crop production is a bio-
logically plausible route of contamination. Following the food-borne out-
breaks in North America during the 1990s, Chalmers and colleagues (2000)
evaluated the risk of similar exposures in the UK and identified different
importation patterns. However, such work needs to be revisited in the light of

278

Cyclospora cayetanensis

outbreaks elsewhere and the identification of broadening importation pat-
terns in terms of foodstuffs and countries of origin. Similarly, proper evalu-
ation of the proportions of imported versus indigenous cases of illness needs
to be undertaken.

However, sources of oocysts have not been fully identified: although CLBs
have been detected in poultry, ducks and non-human primates (Ashford et al. 9
1993; Zerpa et al. 9 1995; Garcia-Lopez et al. 9 1996; Smith et al. 9 1996) con-
vincing evidence for a non-human host of C. cayetanensis has not yet been
provided and humans appear to be the only host species. Although the source
of infection is therefore most likely to be humans, direct person-to-person
transmission is unlikely since a period of maturation in the environment is
required for oocysts to sporulate and become infective. Food and water are,
however, likely vehicles, and it is possible that contaminated water plays an
important role in food-borne cyclosporiasis.

Known risk factors for cyclosporiasis are consumption of contaminated
water or produce (particularly if eaten raw and of a type difficult to clean) and
environmental sources in association with avians in Guatemala (Bern et al. 9
1999) and Peru (Bern et al. 9 2002a). Association of infection with keeping
guinea-pigs and rabbits is unexplained, but could represent a marker for some
other identified risk factor, particularly since this occurred within migrant
families recently relocated from rural to urban areas. Such families come from
regions of lower endemnicity of cyclosporiasis and may not have developed
immunity. So far as drinking water is concerned, removal by conventional coagu-
lation and filtration should be equal to or greater than for Cryptosporidium
since the oocysts are twice as big. Chlorine resistance has not been evaluated but
is probably similar to that exhibited by Cryptosporidium.

In order to identify principal transmission routes and interventions, it is
essential that the sources of contamination are identified. However, to enable
this, methods for detection in food produce and water must be developed and
applied along with environmental studies at sites of production.

Overall risk assessment

Health effects: occurrence of illness, degree of morbidity and mortality, prob-
ability of illness based on infection:

• Cyclospora is distributed world-wide, but infection appears to be endemic
throughout the tropics. Outbreaks have been sporadically reported in
developed countries.

• The symptoms from Cyclospora infection include diarrhoea which is
sometimes explosive. Other symptoms include anorexia, nausea, vomiting,
abdominal bloating and cramps, weight loss, fatigue, low grade fever and
body aches. Although cyclosporiasis is generally self-limiting, the duration
of illness is prolonged and there is the possibility of chronic sequelae.

279

Protozoa

• Immunocompromised patients, predominantly those with HIV/AIDS are at

high risk of cyclosporiasis.

Exposure assessment: routes of exposure and transmission, occurrence in
source water, environmental fate:

• Infected persons excrete oocysts of Cyclospora in their faeces. Oocysts are
not immediately infectious after they are excreted and may require from
days to weeks to become infectious.

• Although the exact routes of transmission of C. cayetanensis are not yet
clear, outbreaks linked to contaminated water, as well various types of fresh
produce, have been reported in recent years.

• The insensitivity of current detection methods has made determining the
exact mode of transmission difficult. Although Cyclospora-like bodies and
C. cayetanensis have been detected in environmental samples, including
water, wastewater, and foods, few prevalence studies have been undertaken
because of the lack of sufficient detection methods.

• The Cyclospora oocysts are robust and can probably survive for long periods.

Risk mitigation: drinking-water treatment, medical treatment:

• Removal by conventional coagulation and filtration should be at least as
efficient as for Cryptosporidium, since the oocysts are twice as big.

• Chlorine resistance has not been evaluated, but a waterborne outbreak
occurred following consumption of chlorinated drinking water indicating
that C. cayetanensis is resistant to levels of chlorine used to treat potable
water.

• Trimethoprim sulphamethoxazole provides rapid, effective anti-parasitic
treatment for those with cyclosporiasis.

Future implications

The major risks are probably posed by inadequately treated water for con-
sumption and/or cultivation of fresh produce. Large volumes of foods cross
international borders from an ever-increasing range of countries of origin, but
generally the laboratory and epidemiological tools are lacking for investigation.
Removal from food as a control measure is problematic since many implicated
foods are difficult to wash and contain crevices in which oocysts can reside.
Oocysts have been shown to remain on foods after washing (Ortega et al. 9
1997) and although minimum time/temperature for inactivation is yet to be
determined, pasteurization and freezing should both be effective but may
damage the aesthetic nature of fresh produce. Future means of control may
include ionizing radiation, although this faces a lack of consumer acceptance
(Monk etal, 1995).

280

Cyclospora cayetanensis

References

Abou el Naga, I.F. (1999). Studies on a newly emerging protozoal pathogen: Cyclospora

cayetanensis. J Egyptian Soc Parasitol, 29: 575-586.
Adam, R.D., Ortega, Y.R., Gilman, R.H. et al. (2000). Intervening transcribed spacer

region 1 variability in Cyclospora cayetanensis. J Clin Microbiol, 38: 2339-2343.
Anon. (1997). Outbreaks of pseudo-infection with Cyclospora and Cryptosporidium -

Florida and New York, 1995. MMWR, 46: 354-358.
Ashford, R.W. (1979). Occurrence of an undescribed coccidian in man in Papua New

Guinea. Ann Trop Med Parasitol, 73: 497-500.
Ashford, R.W., Warhurst, D.C. and Reid, G.D.F. (1993). Human infection with cyanobac-

terium-like bodies. Lancet, 341: 1034.
Bendall, R.P., Lucas, A., Moody, A. et al. (1993). Diarrhoea associated with cyanobacterium-

like bodies: A new coccidian enteritis of man. Lancet, 341: 590-592.
Berlin, O.G., Peter, J.B., Gagne, C. et al. (1998). Autofluorescence and the detection of

Cyclospora oocysts. Emerging Infect Dis, 4: 127-128.
Bern, C., Hernandez, B., Lopez, M.B. et al. (1999). Epidemiologic studies of Cyclospora

cayetanensis in Guatemala. Emerging Infect Dis, 5: 766-774.
Bern, C., Ortega, Y., Checkley, W. et al. (2002a). Epidemiologic differences between cyclo-

sporiasis and cryptosporidiosis in Peruvian Children. Emerging Infect Dis, 8: 581-585.
Bern, C., Arrowood, M.J., Eberhard, M. et al. (2002b). Cyclospora in Guatemala: further

considerations./ Clin Microbiol, 40: 731-732.
Brockmann, S., Doller, C.R., Dreweck, C. et al. (2001). Cyclosporiasis in Germany. Euro

Surveillance Weekly, 5: 1.
Brown, G.H. and Rotschafer, J.C. (1999). Cyclospora: review of an emerging pathogen.

Pharmacotherapy, 19: 70-75.
Cann, K.J., Chalmers, R.M., Nichols, G. et al. (2000). Cyclospora infections in England

and Wales: 1993 to 1998. Commun Dis Public Hlth, 3: 46-49.
Carter, R.J., Guido, F., Jacquette, G. et al. (1996). Outbreak of cyclosporiasis associated

with drinking water (abstract). In Program of the 30th Inter science Conference on

antimicrobial agents and Chemotherapy, New Oreans 1996, p. 259.
Chacin-Bonilla, L., Estevez, J., Monsalve, F. et al. (2001). Cyclospora cayetanensis infec-
tions among diarrheal patients from Venezuela. Am J Trop Med Hyg, 65: 351-354.
Chalmers, R.M., Rooney, R. and Nichols, G. (2000). Foodborne outbreaks of cyclospori-
asis have arisen in North America. Is the United Kingdom at risk? Commun Dis Public

Hlth, 3: 50-55.
Ciancio, M. and Chang, E. (1992). Epithelial secretory response to inflammation. Ann NY

Acad Sci, 664: 210-221.
Clark, S.C. and Mclntyre, M. (1997). An attempt to demonstrate a serological immune

response in patients infected with Cyclospora cayetanensis. Br J Biomed Sci, 54: 73.
Connor, B.A., Shlim, D.R., Scholes, J.V. et al. (1993). Pathologic changes in the small

bowel in nine patients with diarrhoea associated with a coccidian-like body. Ann Int

Med, 119: 377-382.
Connor, B.A. and Shum, O.R. (1995). Foodborne transmission of Cyclospora. Lancet,

346: 1634.
Connor, B.A., Reidy, J. and Soave, R. (1999). Cyclosporiasis: clinical and histopathologic

correlates. Clin Infect Dis, 28: 1216-1222.
Connor, B.A., Johnson, E. and Soave, R. (2001). Reiter syndrome following protracted

symptoms of Cyclospora infection. Emerging Infect Dis, 7: 453-454.
Dalton, C, Goater, A.D., Pethig, R. et al. (2001). Viability of Giardia intestinalis cysts and

viability and sporulation state of Cyclospora cayetanensis determined by electrorota-

tion. Appl Environ Microbiol, 67: 586-590.
De Gorgolas, M., Fortes, J. and Guerrero, M.L.F. (2001). Cyclospora cayetanensis chole-
cystitis in a patient with AIDS. Ann Intern Med, 143: 166.
Eberhard, M.L., Pieniazek, N.J. and Arrowood, M.J. (1997). Laboratory diagnosis of

Cyclospora infections. Arch Pathol Lab Med, 121: 792-797.

281

Protozoa

Eberhard, M.L., Nace, E.K., Freeman, A.R. et al. (1999a). Cyclospora cayetanensis infec-
tions in Haiti: a common occurrence in the absence of watery diarrhoea. Am J Prop
Med Hyg 9 60: 584-586.

Eberhard, M.L., Da Silva, A.J., Lilley, B.G. et al. (1999b). Morphologic and molecular
characterisation of new Cyclospora species from Ethiopian monkeys: C. cercopitheci
sp.n., C. colobi sp.n. and C. papionis sp.n. Emerging Infect Dis, 5: 561-658.

Eberhard, M.L., Ortega, Y.R., Hanes, D. et al. (2000). Attempts to establish experimental
Cyclospora cayetanensis infection in laboratory animals./ Parasitol, 86: 577-582.

Garcia-Lopez, H.L., Rodriquez-Tovar, L.E. and Mdeina-De la Garza, C.E. (1996).
Identification of Cyclospora in poultry. Emerging Infect Dis, 2: 356-357.

Hale, D., Aldeen, W. and Carroll, K. (1994). Diarrhoea associated with cayeno bacteria-
like bodies in an immunocompetent host. An unusual epidemiological source. JAMA,
271: 144-145.

Herwaldt, B.L. (2000). Cyclospora cayetanensis: a review, focussing on the outbreaks of
cyclosporiasis in the 1990s. Clin Infect Dis, 31: 1040-1057.

Ho, A.Y., Lopez, A.S., Eberhart, M.G. et al. (2002). Outbreak of cyclosporiasis associated
with imported raspberries, Philadelphia, Pennsylvania, 2000. Emerging Infect Dis, 8:
783-788.

Hoge, C.W., Shlim, D.R., Rajah, R. et al. (1993). Epidemiology of diarrhoeal illness asso-
ciated with coccidian-like organism among travellers and foreign residents in Nepal.
Lancet, 341: 1175-1179.

Hoge, C.W., Shlim, D.R., Ghimire, M. etal. (1995). Placebo-controlled trial of co-trimoxazole
for cyclospora infections among travellers and foreign residents in Nepal. Lancet, 345:
691-693.

Huang, P., Weber, J. and Sosin, D.M. (1995). The first reported outbreak of diarrheal dis-
ease associated with Cyclospora in the United States. Ann Int Med, 123: 409-414.

Jayshree, R.S., Acharya, R.S. and Sridhar, H. (1998). Cyclospora cayetanensis- associated
diarrhoea in a patient with acute myeloid leukaemia./ Diarrhoeal Dis Res, 16: 254-255.

Jinneman, K.C., Wetherington, J.H., Hill, W.E. et al. (1999). An oligonucleotide-ligation
assay for differentiation between Cyclospora and Eimeria spp. polymerase chain reac-
tion amplification products. / Food Protect, 62: 682-685.

Kniel, K.E., Lindsay, D.S., Sumner, S.S. et al. (2002). Examination of attachment and
survival of Toxoplasma gondii oocysts on raspberries and blueberries./ Parasitol, 88:
790-793.

Kocka, F., Peters, C, Dacumos, E. et al. (1991). Outbreaks of diarrhoeal illness associated
with cyanobacteria (Blue-Green Algae)-like bodies - Chicago and Nepal, 1989 and
1990. MMWR, 40: 325-327.

Konttinen, Y.T., Nordstrom, D.C., Bergroth, V. et al. (1988). Occurrence of different ensu-
ing triggering infections preceding reactive arthritis: a follow-up study. Br Med J, 296:
1644-1645.

Koumans, E.H., Katz, D., Malecki, J.A. et al. (1996). Novel parasite and mode of trans-
mission: Cyclospora infection - Florida 1996. Annual Epidemic Intelligence Service
Conference, p. 4560.

Lee, M.B. and Lee, E.H. (2001). Coccidial contamination of raspberries: mock contamin-
ation with Eimeria acervulina as a model for decontamination treatment studies./ Food
Protect, 64: 1854-1857.

Levine, N.D. (1982). Taxonomy and life cycles of coccidian. In The Biology of the
Coccidian, Long, P.L. (ed.). London: Edward Arnold, pp. 1-3.

Long, E.G., White, E.H., Carmichael, W.W. et al. (1991). Morphologic and staining char-
acteristics of a cycanobacterium-like organism associated with diarrhoea. / Infect Dis,
164: 199-202.

Lopez, A.S., Dodson, D.R., Arrowood, M.J. etal. (2001). Outbreak of cyclosporiasis asso-
ciated with basil in Missouri in 1999. Clin Infect Dis, 32: 1010-1017.

Lopez, F.A., Manglicmot, J., Schmidt, T.M. et al. (1999). Molecular characterisation of
Cyclospora-like organisms from baboons. / Infect Dis, 179: 670-676.

Madico, G., McDonald, J., Gilman, R. et al. (1997). Epidemiology and treatment of
Cyclospora cayetanensis infection in Peruvian children. Clin Infect Dis, 24: 977.

282

Cyclospora cayetanensis

Miliotis, M.D., Hanes, D.E., Tall, B.D. et al. (1997). Food and Drug Administration Science
Forum Poster Abstract 1997 (http://www.cfsan.fda.gov/~frf/forum97/97L10.htm).

Monk, J., Beuchat, L. and Doyle, M. (1995). Irradiation inactivation of food-borne micro-
organisms. / Food Protect, 58: 197-208.

Naranjo, J., Sterling, C, Gilman, R. et al. (1989). Cryptosporidium-muris-like objects
from faecal samples of Peruvians (abstract 324). In Program and abstracts of the 38th
Annual Meeting of the American Society of Tropical Medicine and Hygiene (Honolulu)
10-14 December 1989, p. 243.

Olivier, C, van de Pas, S., Lepp, P.W. et al. (2001). Sequence variability in the first internal
transcribed spacer region within and among Cyclospora species is consistent with poly-
parasitism. Int J Parasitol, 31: 1475-1487.

Ooi, W.W., Zimmerman, S.K. and Needham, C.A. (1995). Cyclospora species as a gas-
trointestinal pathogen in immunocompetent hosts./ Clin Microbiol, 33: 1267-1269.

Ortega, Y.R., Sterling, C.R., Gilman, R.H. et al. (1993). Cyclospora species: a new proto-
zoan pathogen of humans. New Engl J Med, 328: 1308-1312.

Ortega, Y.R., Gilman, R.H. and Sterling, C.R. (1994). A new coccidian parasite
(Apicomplexa: Eimeriidae) from humans./ Parasitol, 80: 625-629.

Ortega, Y.R., Roxas, C.R., Gilman, R.H. et al. (1997). Isolation of Cryptosporidium
parvum and Cyclospora cayetanensis from vegetables collected in markets of an
endemic region in Peru. Am J Prop Med Hyg, 57: 683-686.

Pape, J.W., Verdier, R.I. and Boncy, M. et al. (1994). Cyclospora infection in adults infected
with HIV. Ann Int Med, 121: 654-657.

Pieniazek, N.J. and Herwaldt, B.L. (1997). Reevaluating the molecular taxonomy: is
human-associated Cyclospora a mammalian Eimeria species? Emerging Infect Dis,
3:381-383.

Powell, D. (1995). Approach to the patient with diarrhoea. In Textbook of Gastroenterology,
2nd edn, Yamada, T. (ed.). Philadelphia: JB Lippincott, pp. 820-824.

Pratdesaba, R.A., Gonzalez, M., Piedrasanta, E. et al. (2001). Cyclospora cayetanensis in
three populations at risk in Guatemala./ Clin Microbiol, 39: 2951-2953.

Rabold, J.G., Hoge, C.W. and Shim, D.R. (1994). Cyclospora outbreak associated with
chlorinated drinking water. Lancet, 344: 1360-1361.

Relman, D.A., Schmidt, T.M., Gajadhar, A. et al. (1996). Molecular phylogenetic analysis
of Cyclospora, the human intestinal pathogen, suggests that it is closely related to
Eimeria species. / Infect Dis, 173: 440.

Richardson, R.F., Remler, B.F., Katirji, B. et al. (1998). Guillain-Barre syndrome after
Cyclospora infection. Muscle Nerve, May: 669-671.

Robertson, L.J., Gjerde, B. and Campbell, A.T. (2000). Isolation of Cyclospora oocysts
from fruits and vegetables using lectin-coated paramagnetic beads. / Food Protect, 63:
1410-1414.

Rose, J.B. and Slifko, T.R. (1999). Giardia, Cryptosporidium and Cyclospora and their
impact on foods: a review./ Food Protect, 62: 1057-1070.

Schneider, A. (1881). Sur les psorospemies oviformes ou coccidies, especes nouvelles ou peu
connues. Arch Zool Exp Gen, 9: 387-404.

Sherchand, J.B., Cross, J.H., Jimba, M. et al. (1990). Study of Cyclospora cayetanensis in
health care facilities, sewage water and green leafy vegetables in Nepal. SE Asian J Trop
Med Public Hlth, 30: 58-63.

Shlim, D.R., Cohen, M.T., Eaton, M. et al. (1991). An alga-like organism associated with
an outbreak of prolonged diarrhoea among foreigners in Nepal. Am J Trop Med Hyg,
45:383-389.

Sifuentes-Osornio, J., Porrras-Cortes, G., Bendall, R.P. et al. (1995). Cyclospora cayeta-
nensis infection in patients with and without AIDS: biliary disease as another clinical
manifestation. Clin Infect Dis, 21: 1092-1097.

Smith, H.V., Paton, C.A., Girdwood, R.W.A. et al. (1996). Cyclospora in non-human pri-
mates in Gombe, Tanzania (letter). Vet Rec, 138: 528.

Smith, H.V., Paton, C.A., Mtambo, M.M. et al. (1997). Sporulation of Cyclospora sp.
Oocysts. Appl Environ Microbiol, 63: 1631-1632.

Soave, R. (1996). Cyclospora: an overview. Clin Infect Dis, 23: 429-437.

283

Protozoa

Soave, R. and Johnson, W.D. (1995). Cyclospora: conquest of an emerging pathogen.

Lancet, 345: 667-668.
Soave, R., Dubey, J.P., Ramos, L.J. et al. (1986). A new intestinal pathogen? (abstract).

Clin Res, 34: 533A.
Soave, R., Herwaldt, B.L. and Relman, D.A. (1998). Cyclospora. Infect Dis Clin N Am,

12: 1-12.
Sturbaum, G.D., Ortega, Y.R., Gilman, R.H. et al. (1998). Detection of Cyclospora cayeta-

nensis in wastewater. Appl Environ Microbiol, 64: 2284-2286.
Sun, T., Ilardi, C.F., Asnis, D. et al. (1996). Light and electron microscopic identification of

Cyclospora species in the small intestine: evidence of the presence of asexual life cycle

in a human host. Am J Clin Microbiol, 105: 216-220.
Taylor, D.N., Houston, R., Shlim, D.R. et al. (1988). Etiology of diarrhoea among travel-
ers and foreign residents in Nepal. JAMA, 260: 1245-1248.
Verdier, R.I., Fitzgerald, D.W., Johnson, W.D. et al. (2000). Trimethoprim-sulfamethoxazole

compared with ciprofloxacin for treatment and prophylaxis of Isospora belli and

Cyclospora cayetanensis infection in HIV-infected patients. A randomised, controlled

trial. Ann Int Med, 132: 885-888.
Vesey, G., Slade, J.S., Byrne, M. et al. (1993). A new method for the concentration of

Cryptosporidium oocysts from water./ Appl Bacteriol, 75: 82-86.
Visvesvara, G.S., Moura, H., Kovacs-Nace, E. et al. (1997). Uniform staining of

Cyclospora oocysts in fecal smears by a modified safranin technique with microwave

heating./ Clin Microbiol, 35: 730-733.
Wurtz, R. (1994). Cyclospora: a newly identified intestinal pathogen of humans. Clin

Infect Dis, 18:620-623.
Yoder, K.E., Sethabutr, O. and Relman, D.A. (1996). PCR-based detection of the intestinal

pathogen Cyclospora. In PCR Protocols for Emerging Infectious Diseases, a Supplement

to Diagnostic Molecular Biology: Principles and Applications, Persing, D.H. (ed.).

Washington, DC: ASM Press, pp. 169-176.
Zar, F.A., El-Bayoumi, E. and Yungbluth, M.M. (2001). Histologic proof of acalculous

cholecystitis due to Cyclospora cayetanensis. Clin Infect Dis, 33: 140-141.
Zerpa, R., Uchima, N. and Huicho, L. (1995). Cyclospora cayetanensis associated with

water diarrhoea in Peruvian patients. / Prop Med Hyg, 98: 325-329.

284

Entamoeba histolytica

Basic microbiology

Amoebic protozoa of the genus Entamoeba are members of the Phylum
Sarcodina, Order Amoebida, Family Endamoebidae. Although many amoebae
inhabit the human gastrointestinal tract (including Entamoeba histolytica.
Entamoeba coli, Entamoeba dispar, Entamoeba hartmanni, Entamoeba polecki,
Chilmastix mesnili, Endolimax nana, and Iodamoeba buetschlii), most are not
pathogenic. Entamoeba gingivalis inhabits the mouth and has been associated
with periodontal disease (Lyons et aL 9 1983), while E. polecki is of uncertain
pathogenicity (Levine and Armstrong, 1970). However, E. histolytica is the
predominant pathogenic amoeba and causes amoebic dysentery (amoebiasis)
(Brumpt, 1925; Sargeaunt et al. 9 1978).

£. histolytica is indistinguishable by microscopy from the non-pathogenic
E. dispar and the differentiation is rarely made in routine clinical microbiological
diagnoses. Historically, since these organisms could not be differentiated, both
were referred to as E. histolytica, and it has been estimated that about 10% of the
world's population are infected (Walsh, 1988). However, 90% of these infections

Protozoa

are asymptomatic and probably with the non-pathogenic E. dispar. Despite this
there remains a considerable burden of disease and more than 100000 deaths
occur annually from invasive amoebiasis, making it the third leading parasitic
cause of death in developing countries (Reed, 1992).

The life cycle of E. histolytica has been described by Dobell (1928) and
comprises an infective cyst form, metacyst, metacystic trophozoite, motile
feeding trophozoite and precyst stages. The cyst form (10-16 fJim), which
develops only in the intestinal tract, is shed in the faeces and is capable of
survival in food and water. The mature cysts, which contain four nuclei, are
transmitted to humans by the ingestion of faecally-contaminated food, water
or from body contact. The amoebae within the mature cyst are activated by
the neutral or alkaline environment in the small intestine, and separate from
the cyst wall which is digested by enzymes within the gut lumen. Rapid
nuclear and cytoplasmic division results in eight uninucleate trophozoites.
The trophozoites (20-40 |Jim) migrate to the large intestine where they multi-
ply by binary fission and feed on the bacteria of the intestinal flora and on cell
debris. Encystation is probably stimulated by the dehydrating luminal condi-
tions and cysts develop, each with one to four nuclei, which are passed in
the faeces. Although trophzoites may be shed during acute colitis, they are
not responsible for the spread of infection because they do not survive outside
the body and are destroyed by the low gastric pH. The robust cysts transmit
infection.

Trophozoites of non-pathogenic E. dispar colonize the gut lumen and the
infected host sheds cysts asymptomatically. E. histolytica can also cause
asymptomatic infection but trophozoites can invade the intestinal mucosa,
causing intestinal disease (dysentery), or travel via the blood stream to
extraintestinal sites including the liver, brain and lungs (Ravdin, 1995).

Although the distribution of E. histolytica is world-wide, there is a higher
prevalence of amoebiasis in developing countries and this appears to depend on
sanitation, age, crowding and socioeconomic status (Ravdin, 1988). Although
natural infections with indistinguishable organisms have been reported in
macaque monkeys and pigs (Hoare, 1962), and primates, dogs and cats also
shed morphologically similar amoebae, humans are the main reservoir of E.
histolytica. Transmission is usually from a chronically ill or asymptomatic cyst
shedder. Those at high risk in developed countries are travellers returning from
developing countries, immigrants, migrant workers, immunocompromised indi-
viduals and sexually active male homosexuals (Ravdin, 1988). Waterborne
transmission is commonly associated with faecally-contaminated water supplies
in developing countries.

A number of biological features of E. histolytica affect its transmission and
epidemiology. The robust cysts are resistant to gastric acid and once shed in
the faeces are immediately infective. Indeed, contact with an asymptomatic
carrier is regarded as the cause of most cases. However, environmental contam-
ination can lead to waterborne transmission, which is common in developing
countries where drinking water is untreated.

286

Entamoeba histolytica

Origin of the organism

A Russian clinical assistant first discovered organisms now classified as E. his-
tolytica in 1873, by observing large numbers of amoebic trophozoites in the
stools of a patient with bloody dysentery (Losch, 1875). However, it was sev-
eral decades before the concept that intestinal amoebae could cause disease
was generally accepted, principally since, while large numbers of amoebae
were often seen, disease was present in the minority of cases. Councilman and
LaFleur made descriptions of the clinical and pathological outcomes of infec-
tion in 1891. Genus names Amoeba, Endamoeba and Entamoeba have been
used in the past, and confusing and regularly changing taxonomy at the genus
and species level persisted throughout the 20th century (Clark, 1998). Unclear
pathogenicity was noted and differences between morphologically distinct
cysts of E. hartmanni, E. poleki and £. histolytica were observed (Burrows,
1959). While pathogenic and non-pathogenic Variants' of E. histolytica have
been proposed in the past, particularly by Brumpt in the 1920s on the basis of
clinical and experimental observations (Brumpt, 1925, 1928), the lack of
morphological differences hampered acceptance of this proposal.

Isoenzyme, antigenic and genetic studies demonstrate differences between
£. histolytica isolates (Sargeaunt et al., 1978; Strachan et aL, 1988; Tannich
et al., 1989; Clark and Diamond, 1991). While changing zymodeme patterns
suggested that genetic transfer occurred between pathogenic and non-pathogenic
strains, this is not supported by genetic evidence and two stable, separate species
were described by Diamond and Clark in 1993, proposing that the pathogenic
species was to retain the name E. histolytica and the non-pathogenic species
E. dispar. In 1997, the WHO recommended the acceptance of these two species
(Anon, 1997). However, lack of distinction between the two species has
certainly hampered the understanding, clinical management and epidemiology
of E. histolytica infections.

Clinical features

Infection with E. histolytica can range from asymptomatic infection and cyst pas-
sage, to acute amoebic rectocolitis, chronic non-dysenteric colitis and amoeboma
(Reed, 1992; Ravdin, 1995). Severe invasive disease affects up to 20% of patients.
Patients with acute amoebic colitis (dysentery) usually present with a 1-2 week
history of watery stools with blood or mucus, abdominal pain, tenesmus and
fever. Fulminant colitis mainly occurs in children, pregnant women and patients
on corticosteroids and is characterized by abdominal pain, profuse bloody diar-
rhoea and fever: mortality is over 50% (Li and Stanley, 1996). Chronic amoebic
colitis may be indistinguishable from irritable bowel disease (IBD) and must be
ruled out before treatment with corticosteroids for IBD commences. Amoeboma
(asymptomatic lesion or symptomatic dysentery with a tender mass) is a localized

287

Protozoa

chronic infection that occurs in the caecum or ascending colon and can be differ-
entiated from carcinoma by biopsy. Invasive extraintestinal amoebiasis depends
on the site infected and includes liver and brain abscesses, peritonitis, pleuropul-
monary abscess, cutaneous and genital amoebic lesions.

Pathogenicity and virulence

As its name suggests, £. histolytica has a lytic effect upon host tissue. Invasion
of the intestinal mucosa of the caecum and colon by trophozoites is initiated by
depletion of the protective mucus blanket and proteolytic disruption of tissue,
causing lysis and necrosis of host cells and characteristic flask-shaped lesions
(Tse and Chadee, 1991). Virulence factors, such as the galactose-specific bind-
ing lectin cause lesions and permit spread through the blood stream.
Trophozoites of invasive £. histolytica are resistant to complement-mediated
lysis which may facilitate extraintestinal invasion (Ravdin, 1990a). Invasion
of the liver occurs when trophozoites ascend the portal venous system and
cause hepatic necrosis and amoebic liver abscesses which contain proteinaceous
debris. Amoebic lysis of neutrophils releases toxic non-oxidative products that
contribute to the destruction of host tissue. Periportal inflammation can cause
liver enzyme abnormalities in the absence of demonstrable trophozoites.

Adherence of trophozoites to the mucosa, epithelial cells and host inflam-
matory cells is important in disease pathogenesis. In vitro studies have shown
that this is mediated by an adherence lectin, which acts in the presence of
extracellular calcium ions (Ravdin et al., 1988). Cytolytic activity is aug-
mented by phorbol esters and protein kinase activators (Weikel et al., 1988).
An ionophore-like protein has been identified in E. histolytica which induces
leaking of sodium, potassium and calcium ions and acts as a parasite defence
mechanism against ingested bacteria (Leippe etaL, 1994) and the bacterial gut
flora probably plays a role in virulence and ability to colonize a host. Several
strains of E. histolytica have been identified and cytotoxic haemolysins
encoded by plasmid DNA have been identified in the most pathogenic strains
(Jansson et aL, 1994).

Causation

Disease severity is increased in children, particularly neonates, pregnancy and
post-partum states, with the use of corticosteroids, and during malignancy
and malnutrition (Ravdin, 1988). Initial invasion of the mucosa is probably
not related to immunity but the severity of disease. This is indicated by the
exacerbation of intestinal amoebiasis by corticosteroid therapy, in infants and
pregnant women. Acquired immunity appears to have a protective effect since

288

Entamoeba histolytica

recurrences of amoebic colitis or liver abscesses in endemic areas are rare
(DeLeon, 1970) and the presence of serum antibodies is associated with a
lower rate of intestinal infection (Choudhuri etal., 1991), although asymptom-
atic infection is recurrent and serum antibodies are only present at low levels
when infection is with non-invasive amoebae. The role of secretory immuno-
globulins may be limited by the organism's ability to shed anti-amoebic anti-
bodies (Arhets et al. 9 1995), but cell-mediated immunity plays an important
role in limiting the extent of invasive amoebiasis and in protection of recur-
rence. It is not known whether infection with a non-pathogenic species con-
fers any protection from pathogenic species.

Treatment

Since primary testing laboratories rarely differentiate E. histolytica from
E. dispar, and research has shown that the vast majority of infections in Europe
and North America are in fact £. dispar (Sargeaunt, 1987), there is generally
no need for the administration of antiparasitic agents in patients without
symptoms of amoebiasis. Even in parts of the world where invasive amoebia-
sis is prevalent, such as Central America, southern Africa and India, E. dispar
infections still outnumber E. histolytica by 10:1 (Petri, 1996). WHO recom-
mends that unequivocal differential diagnosis is made (see below), or that
there is strong reason to suspect amoebiasis, before treatment (Anon, 1997).
If asymptomatic £. histolyica infection is detected by discriminatory tests, this
should also be treated because of the risk of progression to symptomatic infec-
tion and risk of transmission.

Luminal amoebicides, such as iodoquinol, paromomycin and diloxanide
furoate are effective against organisms in the intestinal lumen, but are not highly
effective against invasive disease. For intestinal disease a tissue amoebicide such
as metronizadole or tinidazole should be followed by a luminal amoebicide
because luminal parasites are not otherwise eliminated. For severe or refactory
disease dehydroemetine followed by iodoquinol, paromomycin or diloxanide
furoate are suitable. Non-surgical aspiration may be necessary for patients with
liver abscesses if they continue to be febrile. Treatment of laboratory confirmed
asymptomatic E. histolytica with luminal amoebicides is important in the con-
trol of the spread of infection, but chemoprophylaxis is never appropriate.

Survival in the environment and in water

Although trophozoites may be passed from the body in stools, they are rap-
idly destroyed and, furthermore, if ingested would not survive the gastric
environment. The protective cyst wall, however, ensures cyst survival, and the

289

Protozoa

use of human faeces as fertilizer is an important source of infection. In a study
of 107 'zir' stored drinking water and 11 tap water samples in a village in the
Nile Delta, 55% zir stored waters and 63% tap water samples contained
E. histolytica, although differentiation was not made between E. dispar
(Khairy e* a/., 1982)

Methods for evaluating the survival of Entamoeba cysts have been based on
chemical staining, in vitro culture (measuring total population kill) or on count-
ing excysted cysts from populations, on the assumption that unexcysted (intact)
cysts were not viable and therefore did not excyst (Stringer, 1972). However,
while there are other reasons for which excystation may not occur, this was con-
sidered unlikely significantly to alter results. Cysts can remain viable for as long
as 3 months depending on conditions, and survive at 4°C for 12 weeks from egg
slant cultures (Neal, 1974). Early studies showed that treatment at 68°C for
5 minutes and boiling for 10 seconds killed cysts (Boek, 1921; Mills et al.,
1925). Time/temperature studies by Myjak (1967) showed that cysts from cul-
ture were killed at 47°C for 25 minutes, 49°C for 11 minutes, 51°C for 3 min-
utes and 53°C for 1 minute. Cysts in faeces from infected carriers showed a
decrease in the time required to kill cysts with increasing temperature, although
higher temperatures (57°C for 1 minute) were required to achieve kill from all
five isolates tested. Cysts are destroyed by hyperchlorination or iodination
(Kahn and Visscher, 1975; Markell et aL, 1986) but are resistant to chlorine at
levels used in public water supplies.

Methods of detection

Since all E. dispar and the majority of £. histolytica infections are asymptomatic,
cysts may be detected in faeces without consequence for the patient. Diagnosis of
intestinal amoebiasis is by both signs and symptoms of the patient and micro-
scopical examination of stool or mucosal biopsy. If stools cannot be examined
fresh they should be preserved in polyvinyl alcohol prior to examination. Occult
blood is usually present in faeces, but presence or absence of faecal leucocytes is
non-contributory to the diagnosis due to the lytic effect of the organism. The for-
malin-ethyl acetate concentration method is most commonly used to concentrate
stools and, in addition to examination of wet preparations and stool concentrate,
permanent stains of fresh or preserved faecal specimens should be examined.
Two or three repeat stools taken over a 10-day period may be required for diag-
nosis. The presence of haematophagus trophozoites in faeces and other speci-
mens, or the detection of trophozoites in biopsy material, is highly predictive of
invasive E. histolytica (Gonzales-Ruiz et aL, 1994). Identification can be difficult
since confusion between Entamoeba and other intestinal protozoa, leucocytes or
macrophages in stool samples may occur: a pseudo-outbreak was reported in
California when one laboratory identified 38 cases in 3 months against a back-
ground about 1 per month (Garcia et al., 1985). This highlights the need for

290

Entamoeba histolytica

laboratory proficiency schemes. The issue of differentiation of £. histolytica cysts
from £. dispar is outwith the scope of most primary testing laboratories, and
when diagnosis is made by light microscopy, results should be reported as
'£. bistolytica/E. dispar' (Anon, 1997).

Differentiation of £. histolytica cysts from the non-pathogenic E. dispar is
necessary when symptoms of amoebiasis are present, unless there are other rea-
sons to suspect £. histolytica. Differentiation can only be undertaken by isoen-
zymatic, immunological or molecular analysis of isolates. Zymodemes,
representing isoelectrophoretic patterns of various enzymes (Sargeunt, 1987),
differentiate £. histolytica and E. dispar but this method requires high numbers
of organisms that can often only be generated by culture, which may present
problems of competition if both species or other species are present.
Cultivation can be in association with bacteria (Robinson, 1968) or axenically
(Diamond, 1968). Optimal growth is at 35-37°C, pH 7.0 with reduced oxygen
tension. However, culture can never exclude the presence of £. histolytica.

Serological tests can be helpful for cyst-positive, symptomatic patients,
where a positive result indicates current or remote £. histolytica infection,
since an antibody response is not usually elicited by £. dispar (Ravdin, 1990b;
Caballero-Salcedo, 1994). However, usefulness is limited for diagnosis of
acute infection in endemic areas where seroprevalence can be high (Caballero-
Salcedo, 1994). However, IgM does not persist in serum and may provide a
good target during the acute phase (Abd-Alla et al. 9 1998). Antibody detec-
tion is also useful in patients with extraintestinal disease where parasites are
generally not found on stool examination.

Detection in environmental samples requires recovery from the sample matrix
by filtration, followed by differentiation from other protozoan cysts. Membrane
filtration using 1.2 |xm, 47 mm diameter filters, followed by elution of the cap-
tured organisms from the filter, concentration by centrifugation and examin-
ation of the sediment is a long-established technique (Chang and Kabler, 1956).
Recovery is probably erratic since cysts may be lost in the process.

Genetic markers confirm the separation of the pathogenic species.
Techniques such as hybridization of genomic DNA with various gene probes
and PCR-based tests targeting various gene loci, including small subunit
rRNA genes, have been described (Clark and Diamond, 1991; Farthing et al. 9
1996; Troll et aL 9 1997). However, these are only undertaken in specialist test-
ing laboratories. Monoclonal antibodies directed at the galactose-specific
adherence lectin showed some epitopes present on pathogenic isolates and
absent in non-pathogenic isolates (Petri et al. 9 1990) and an ELISA kit has
been developed and marketed commercially that distinguishes between £. his-
tolytica and £. dispar on the basis of antigens detected directly in stools
(Haque et al. 9 1998). However, PCR offers greater sensitivity than ELISA
(Mirelman et al. 9 1997) and is therefore probably more appropriate for epi-
demiological studies. In-house immunological methods have also been
developed (Bhaskar et al. 9 1996).

Differentiation between the two species has assisted greatly in understand-
ing the epidemiology of amoebiasis, and is essential for subsequent control of

291

Protozoa

transmission, but there remains a need for simple, cheap differential diagnos-
tic tests to reduce unnecessary treatment. It is important also for prevalence
and epidemiological studies for risk factors for invasive disease.

Critical review of the epidemiology

Greater understanding of the epidemiology of amoebiasis has been achieved
by the differentiation of pathogenic from non-pathogenic isolates, since
E. dispar accounts for 90% of infections. Accurate prevalence data for E. his-
tolytica are rare and improved methods for specific detection, including those
appropriate for use in developing countries are required. In a study in the
Philippines using PCR, 7% of 1872 individuals carried E. dispar and 1% were
infected by E. histolytica (Rivera et al., 1998). Similar to serological studies in
other endemic areas, peak prevalence was detected in children aged 5-14 years.
In a study in Bangladesh, age-specific seroprevalence rose sharply in the
1-2 year olds, peaked at 14 years and showed an age-related decline there-
after (Hossain et al. 9 1983). Detection in stool specimens in the same study
showed the lowest rate in children <1 year, increasing with age. A high
proportion of urban adults were infected, although differentiation between
£. histolytica and E. dispar was not made.

Risk factors have been identified epidemiologically for both increased
prevalence and severity of disease (Ravidin, 1995). The groups at risk of increased
prevalence of amoebiasis in developed countries include promiscuous male
homosexuals, travellers and recent immigrants, institutionalized populations
and communal living. In areas of endemicity, increased prevalence is linked
to lower socioeconomic status including factors such as crowding and lack of
indoor plumbing. Increased severity of disease is linked to neonates, pregnancy
and post-partum, use of corticosteroids, malignancy and malnutrition. Water-
borne transmission is considered common in developing countries due to
faecal contamination and lack of water treatment. Other factors include the
use of human faeces and wastewater in agriculture (Bruckner, 1992). In a
study of the health effects of the agricultural reuse of urban wastewater in
Morocco, the combined risk of infection rate for Giardia and Entamoeba was
in the region of 41% (Amhmid and Bouhoum, 2000).

Amoebiasis, although more prevalent, is not restricted to developing coun-
tries. Outbreaks have been reported in long-term care facilities in the USA
(Nicolle et al., 1996) and one waterborne outbreak was reported in the UK in
the 1950s among service personnel. Transmission was thought to be via the
sewage system from personnel who probably acquired the infection while
serving overseas (Galbraith et al., 1987). Outbreaks have occurred world-
wide, including in the USA (Lippy and Waltrip, 1984), Scandinavia, Taiwan
(Lai et aL 9 2000) and Tblisi, Georgia (Kreidl et al., 2000). The cause of the
Taiwanese outbreak was suspected to be contamination of the water supply

292

Entamoeba histolytica

by patients who had visited an endemic area. In Tblisi, the drinking water
supply was also suspected.

Food-borne amoebiasis is possible, from directly or indirectly contaminated
produce. Protozoan parasites were detected in 52% of 500 fresh clinical stool
specimens collected over a 1-year period in Nigeria (Nzeako, 1992). Of those
infected, six (1%) were from the University of Nigeria community, 89 (18%)
were urban dwellers and 166 (33%) from rural areas. The highest incidence
of Entamoeba histolytica (28%) was found among the rural community.
Parasitic infections were seasonal, and started in April of each year (onset of
rainy season), peaked between July and August, and were the lowest between
November and March (dry season). The green vegetable Amaranthus viri-
dans, which gets polluted by the sewage oxidation pond at the locality, was
identified as the main vehicle of infection.

In a survey of vegetables in Costa Rica, Entamoeba histolytica cysts were
found in 6.2% (5/80) of cilantro leaves, in 2.5% (2/80) cilantro roots, in
3.8% (3/80) lettuce, in 2.5% (2/80) radish samples and at least a 2% inci-
dence of this amoeba was found in other vegetable samples (carrot, cucumber,
cabbage and tomatoes) (Monge and Arias, 1996). In a follow-up study, more
protozoan parasite-positive samples (including E. histolytica) were found dur-
ing the dry season, although the association was only significant (P < 0.05) in
radish (Raphanus sativus) and cilantro leaves (Monge et al., 1996).

Risk assessment

Differentiation between pathogenic Entamoeba and non-pathogenic species is
fundamental to the epidemiology and control of infection. The use of human
waste has been identified as an important source of human infection. Person-
to-person transmission is common, as are family clusters of cases, with
asymptomatic carriers responsible for disease spread and transmission since
exposure of susceptible hosts to asymptomatic pathogen-excreters places
them at risk of infection (Bruckner, 1982). If the proportion of the former in
the community is high, detection and thus control of the spread of infection
becomes difficult. However, if the majority of asymptomatic infections is with
a non-pathogenic species, and symptomatic and thus reported infection due to
a pathogenic species rare, then control is possible and transmission of the
pathogenic species can be controlled. Differentiation between pathogenic and
non-pathogenic species is clearly essential to risk assessment.

Although no differentiation was made in their study, Amhmid and
Bouhoum (2000) highlighted the potential health effects of the agricultural
reuse of urban wastewater in Morocco, and in a survey of sewage effluent,
parasites including E. histolytica were detected at high levels discharging into
the La Palta river, which is used for recreation and drinking water abstraction,
both with and without treatment.

293

Protozoa

Additional to risk assessment from exposure is the predisposition of the
host to infection and increased severity of disease.

Overall risk assessment

Health effects: occurrence of illness, degree of morbidity and mortality, prob-
ability of illness based on infection:

• Infection with Entamoeba histolytica dispar is common - up to 10% of
the world's population becomes infected. It is most common in developing
countries with poor hygiene. Outbreaks, however, have occurred world-wide.

• Infection with E. histolytica causes a disease called amoebiasis. Infection
can be asymptomatic or cause relatively mild intestinal upset and diar-
rhoea. Amoebic dysentery is a severe form of amoebiasis that causes stom-
ach pain, bloody stools and fever. Rarely, E. histolytica can invade other
parts of the body such as the liver, lungs, or brain.

• It is estimated that 1 in 10 people who are infected with E. histolytica
develop disease. Severe invasive disease affects up to 20% of cases.

• Disease severity is increased in children, particularly neonates, pregnancy
and post-partum states, with the use of corticosteroids and during malig-
nancy and malnutrition.

Exposure assessment: routes of exposure and transmission, occurrence in
source water, environmental fate:

• Infected people shed cysts in their faeces. The infection is transmitted by the
ingestion of faecally-contaminated food, water, or from person-to-person
contact.

• Waterborne transmission is commonly associated with faecally-contami-
nated water supplies in developing countries.

• Like other cysts, E. histolytica cysts can survive for a long time in the envir-
onment, though they are susceptible to heat (about 60°C for 1 minute or
boiling for 10 seconds).

Risk mitigation: drinking-water treatment, medical treatment:

• Cysts are destroyed by hyperchlorination or iodination but are resistant to
chlorine at levels used in public water supplies. Waterborne outbreaks have
occurred when potable water was contaminated with sewage. Because of
their size, standard sedimentation, flocculation and filtration should handle
cysts in treated drinking water.

• Amoebicides, such as iodoquinol, paromomycin, and diloxanide furoate
are effective against organisms in the intestinal lumen, but are not highly
effective against invasive disease. For intestinal disease, a tissue amoebicide
such as metronidazole or tinidazole should be followed by a luminal amoe-
bicide because luminal parasites are not otherwise eliminated.

294

Entamoeba histolytica

Future implications

WHO has identified E. histolytica as an organism to be controlled by vaccin-
ation (Anon, 1997), which would reduce the incidence of disease and control
the source of human infection.

From their study of the parasitological quality of zir stored water and tap
water in a village in the Nile Delta, Khairy and colleagues (1982) concluded that
simply supplying water via taps was not enough to ensure a safe drinking water
supply: sufficient taps are required, supplying water protected from pollution,
which in turn requires wastewater and sewage disposal. Safe storage of water
in rural areas also needs to be addressed. Similar conclusions were reached by
Feachem and colleagues (1983) who investigated the excreta disposal facilities
and intestinal parasitism in sub-Saharan Africa: the provision of improved water
supply methods and sanitation in individual houses or small clusters of houses
did not necessarily protect from infection where the overall faecal contamina-
tion of the environment was high.

References

Abd-Alla, M.D., Jackson, T.G. and Ravdin, J.I. (1998). Serum IgM antibody response to

the galactose-inhibitable adherence lectin of Entamoeba histolytica. Am J Trop Med

Hyg, 59:431-434.
Amhmid, O. and Bouhoum, K. (2000). Health effect of urban wastewater reuse in a peri-
urban area in Morocco. Environ Mgmt Hlth, 11: 263-269.
Anon. (1997). WHO/PAHO/UNESCO Report of a consultation of Experts on Amoebiasis,

Mexico City, Mexico, January 1997.
Arhets, P., Gounon, P., Sansonetti, P. et al. (1995). Myosin II is involved in capping and

uroid formation in the human pathogen Entamoeba histolytica. Infect Immun, 63:

4358-4367.
Bhaskar, S., Singh, S. and Sharma, M. (1996). A single-step immunochromatographic test

for the detection of Entamoeba histolytica antigen in stool samples. / Immunol Mthds,

196: 193-198.
Boeck, W.C. (1921). The thermal-death point of the human intestinal protozoan cysts.

Am J Hyg, 1:365-387.
Bruckner, DA. (1992). Amebiasis. Clin Microbiol Rev, 5: 356-369.
Brumpt, M.E. (1925). Etude sommaire de YEntomoebae dispar n. sp. Bull Acad Med

(Paris), 94: 942-952.
Brumpt, M.E. (1928). Differentiation of human intestinal amoebae with four-nucleated

cysts. Trans Roy Soc Trop Med Hyg, 22: 101-114.
Burrows, R.B. (1959). Morphological differentiation of Entamoeba hartmanni and

E. polecki from E. histolytica. Am J Trop Med Hyg, 8: 583-589.
Caballero-Salcedo, A. et al. (1994). Seroepidemiology of amebiasis in Mexico. Am J Trop

Med Hyg, 50:412-419.
Chang, S.L. and Kabler, P.W. (1956). Detection of cysts of Entamoeba histolytica in tap

water by the use of membrane niters. Am J Hyg, 64: 170-180.
Choudhuri, G., Prakash, V., Kumar, A. et al. (1991). Protective immunity to Entamoeba

histolytica infection in subjects with antiamoebic antibodies residing in a hyperendemic

zone. Scand J Infect Dis, 23: 771-776.
Clark, C.G. (1998). Entamoeba dispar, an organism reborn. Trans Roy Soc Trop Med

Hyg, 92: 361-364.

295

Protozoa

Clark, C.G. and Diamond, L.S. (1991). Ribosomal RNA genes of 'pathogenic' and

'non-pathogenic' Entamoeba histolytica are distinct. Mol Biochem Parasitol, 49: 297-302.
Clark, C.G. and Diamond, L.S. (1993). Entamoeba histolytica: an explanation for the

reported conversion of 'non-pathogenic' amebae to the 'pathogenic' form. Exp

Parasitol, 77: 456-460.
Councilman, W.T. and LaFleur, H.A. (1891). Amoebic dystentery. Johns Hopkins Hosp

Rep, 2: 395-548.
DeLeon, A. (1970). Prognostico tardio en el absceso hepatico amibiano. Arch Invest

Med (Mexico), 1(1): 205-206.
Diamond, L.S. (1968). Techniques of axenic cultivation of Entamoeba histolytica

Schaudinn, 1903 and Entamoeba-like amoeba./ Parasitol, 54: 1047-1056.
Diamond, L.S. and Clark, CD. (1993). A redescription of Entamoeba histolytica

Schaudinn, 1903 (emended Walker, 1911) separating it from Entamoeba dispar

Brumpt, 1925. J Eukaryotic Microbiol, 40: 340-344.
Dobell, C. (1928). Research on the intestinal protozoa of monkeys and man. Parasitology,

20: 357-412.
Farthing, M.J.G., Cavellos, A.M. and Kelly, P. (1996). Intestinal protozoa. In Mansons

Tropical Diseases, Farthing, M.J.G. (ed.)
Feachem, R.G., Guy, M.W., Harrison, S. et al. (1983). Excreta disposal facilities and intes-
tinal parasitism in urban Africa: preliminary studies in Botswana, Ghana and Zambia.

Trans Roy Soc Trop Med Hyg, 77: 515-521.
Galbraith, N.S., Barrett, N.J. and Stanwell-Smith, R. (1987). Water and disease after

Croydon./ Inst Water Environ Mngmt, 1: 7-21.
Garcia, L., Sorvillo, F., Epstein, M. et al. (1985). Epidemiological notes and reports.

Pseudo-outbreak of intestinal amoebiasis. California. MMWR, 34: 125-126.
Gonzalez-Ruiz, A., Haque, R., Aguirre, A. et al. (1994). Value of microscopy in the diagnosis

of dysentery associated with invasive Entamoeba histolytica.] Clin Microbiol, 47: 236-239.
Haque, R., Ali, I.K.M., Alkther, S. et al. (1998). Comparison of PCR, isoenzyme analysis,

and antigen detection for diagnosis of Entamoeba histolytica infection. / Clin

Microbiol, 36: 449-452.
Hoare, C.A. (1962). Reservoir hosts and natural foci of human protozoal infection. Acta

Trop, 19:281-317.
Hossain, M.M., Ljungstrom, I., Glass, R.I. et al. (1983). Amoebiasis and giardiasis in

Bangladesh: parasitological and serological studies. Trans Roy Soc Trop Med Hyg,

77: 552-554.
Jansson, A., Gillin, F., Kagardt, U. et al. (1994). Coding of hemolysins within the ribo-
somal RNA repeat on a plasmid in Entamoeba histolytica. Science, 263: 1440-1443.
Kahn, F.H. and Visscher, B.R. (1975). Water disinfection in the wilderness - a simple

method of iodination. West J Med, 122: 450-453.
Khairy, A.E.M., El Sebaie, O.E., Gawad, A.A. et al. (1982). The sanitary condition of rural

drinking water in a Nile Delta village. / Hyg, 88: 57-61.
Leippe, M., Andra, J. and Muller-Eberhard, H.J. (1994). Cytolytic and antibacterial activ-
ity of synthetic peptides derived from amoebapore, the pore-forming peptide of

Entamoeba histolytica. Proc Natl Acad Sci USA, 91: 2602-2060.
Levine, R.L. and Armstrong, D.E. (1970). Human infection with Entamoeba polecki.

Am J Clin Pathol, 54: 611-614.
Li, E. and Stanley, S.L. Jr (1996). Protozoa: Amebiasis. Gastroenterol Clin North Am,

25: 471-492.
Lippy, E.C. and Waltrip, S.C. (1984). Waterborne disease outbreaks - 1946-1980: a thirty

year perspective./ Am Waterworks Assoc, 76: 60-67.
Losch, F. (1975). MassenhafteEntwickelung von Amoben im Dickdarm. Arch Pathol Anat

Physiol Klin Med Rudolf Virchow, 65: 196-211.
Lyons, T., Scholten, T., Palmer, J.C. et al. (1983). Oral amoebiasis: the role of Entamoeba

gingivalis in periodontal disease. Quint Int, 14: 1245-1248.
Markell, E.K., Voge, M. and John, D.T. (1986). Medical Parasitology. Philadelphia:

WB Saunders Co.

296

Entamoeba histolytica

Mills, R.G., Bartlett, C.L. and Kessel, J.F. (1925). Am J Hyg, 5: 559-563.
Mirelman, D., Nuchamowitz, Y. and Stolarsky, T. (1997). Comparison of use of enzyme-
linked immunosorbent assay-based kits and PCR amplification of rRNA genes for

simultaneous detection of Entamoeba histolytica and E. dispar. J Clin Microbiol,

35: 2405-2407.
Monge, R. and Arias, M.L. (1996). Presence of various pathogenic microorganisms in

fresh vegetables in Costa Rica. Arch Latinoam Nutr, 46: 292-294.
Monge, R., Chinchilla, M. and Reyes, L. (1996). Seasonality of parasites and intestinal

bacteria in vegetables that are consumed raw in Costa Rica. Rev Biol Trop, 44:

369-375.
Myjak, P. (1967). The effect of temperature on the survival rate of Entamoeba histolytica

(Schaudinn, 1903) cysts in water. 18: 35-42.
Neal, R.A. (1974). Survival of Entamoeba and related Amoebae at low temperature. I.

Viability of Entamoeba cysts at 4 degrees C. Int J Parasitol, 4: 227-229.
Nicolle, L.E., Strausbaugh, L.J. and Garibaldi, R.A. (1996). Infections and antibiotic resist-
ance in nursing homes. Clin Microbiol Rev, 9: 1-17.
Nzeako, B.C. (1992). Seasonal prevalence of protozoan parasites in Nsukka, Nigeria.

/ Commun Dis, 24: 224-230.
Petri, W.A. (1996). Recent advances in amoebiasis. Crit Rev Clin Lab Sci, 33: 1-37.
Petri, W.A., Jackson, T.F., Gathiram, V. et al. (1990). Pathogenic and non-pathogenic

strains of Entamoeba histolytica can be differentiated by monoclonal antibody to the

galactose specific adherence lectin. Infect Immun, 58: 1802-1806.
Ravdin, J.I. (1988). Intestinal disease caused by Entamoeba histolytica. In Amebiasis:

Human Infection by Entamoeba histolytica, Ravdin, J.I. et al. (eds). New York:

Churchill Livingstone, pp. 495-509.
Ravdin, J.L. (1990a). Pathogenic mechanisms, human immune response, and vaccine

development. Clin Enatamoeba histolytica Res, 38: 215-225.
Ravdin, J.L. (1990b). Association of serum antibodies to adherence lectin with invasive

amebiasis and asymptomatic infection with pathogenic Entomoeba histolytica. J Infect

Dis, 162: 768-772.
Ravdin, J.I. (1995). Amebiasis. Clin Infect Dis, 20: 1453-1466.
Ravdin, J.L, Moreau, F., Sullivan, J. A. et al. (1988). The relationship of free intracellular

calcium ions to the cytolytic activity of Entamoeba histolytica. Infect Immun, 56:

1505-1512.
Reed, S.L. (1992). Amebiasis: an update. Clin Infect Dis, 14: 385-393.
Rivera, W.L., Tachibana, H. and Kanbora, H. (1998). Field study on the distribution of

Entamoeba histolytica and Entamoeba dispar in the Northern Philippines as detected

by the PCR. Am J Trop Med Hyg, 59: 916-921.
Robinson, G.L. (1968). The laboratory diagnosis of human parasitic amoebae. Trans Roy

Soc Trop Med Hyg, 62: 285-294.
Sargeaunt, P.G. (1987). The reliability of Entamoeba histolytica zymodemes in clinical

diagnosis. Parasitol Today, 3: 40-43.
Sargeaunt, P.G., Williams, J.E. and Grene, J.D. (1978). The differentiation of invasive and

non-invasive Entamoeba histolytica by isoenzyme electrophoresis. Trans Roy Soc Trop

Med Hyg, 72: 519-521.
Strachan, W.D., Spice, W.M., Chiodini, P.L. et al. (1988). Immunological differentiation

of pathogenic and non-pathogenic isolates of Entamoeba histolytica. Lancet, i:

561-563.
Stringer, R.P. (1972). New bioassay system for evaluating per cent survival of Entamoeba

histolytica cysts./ Parasitol, 58: 306-310.
Tannich, E., Horstmann, R.D., Knoblock, J. et al. (1989). Genomic DNA differs between

pathogenic and non-pathogenic Entamoeba histolytica. Proc Natl Acad Sci USA, 86:

5118-5122.
Troll, H., Marti, H. and Weiss, N. (1997). Simple differential detection of Entamoeba

hisolytica and Entamoeba dispar in fresh stool specimens by sodium acetate-acetic acid-
formalin concentration and PCR./ Clin Microbiol, 35: 1701-1705.

297

Protozoa

Tse, S.K. and Chadee, K. (1991). The interaction between intestinal mucus glycoproteins

and enteric infections. Parasitol Today, 7: 163.
Walsh, J.A. (1998). Prevalence of Entamoeba histolytica infection. In Amebiasis:

Human Infection by Entamoeba histolytica, Ravdin, J.I. (ed.). New York: Wiley,

pp. 93-105.
Weikel, C.S., Murphy, C.F., Orozco, M.E. et at. (1988). Phorbol esters specifically enhance

the cytolytic activity of Entamoeba histolytica. Infect Immun, 56: 1485-1491.

298

Giardia duodenalis

Basic microbiology

Giardia are bi-nucleate flagellated protozoa (Phylum Sarcomastigophora,
Order Diplomonadida, Family Hexamitidae). Although the taxonomy of the
genus is uncertain, three morphological types have been identified: Giardia
duodenalis (syn. lamblia, syn. intestinalis) which is found in humans and
many other mammals, birds and reptiles, G. muris which is found in rodents,
birds and reptiles and G. agilis found in amphibians (Filice, 1952). G. duode-
nalis is the only type known to infect humans and indeed most other domes-
ticated animals. Differentiation has been historically on the basis of size and
morphology of cysts and trophozoites, size of the ventral adhesive disc rela-
tive to the length of the cell, and the shape of median bodies (Kulda and
Nohynkova, 1996). Two further types have been described on the basis of
ultrastructural features: Giardia ardeae (Erlandsen et ah, 1990) and Giardia
psittaci (Erlandsen and Bemrick, 1987), and a further type proposed on the
basis of cyst morphology and ss rRNA sequence analysis (Feely, 1988;
van Keulen et al., 1998). DNA analyses have confirmed differences between
types (van Keulen et al., 1991): G. muris is distant from G. duodenalis, G.
ardeae is closer to G. duodenalis than to G. muris and G. microti is similar to
G. duodenalis genotypes. G. duodenalis appears to be a clade with multiple

Protozoa

Table 21 .1 Giardia duodenalis genotypes

Genotype

Subgroup Host range

Assemblage A All

(syn. 'Polish' or 'Group 1/2')

All

Assemblage B
(syn. 'Belgian' or 'Group 3')
Dog
Cat

Hoofed animal
Rat
Wild rodents

Humans, livestock, dogs, cats,

beavers, guinea-pigs, slow loris
Humans

Humans, slow loris, chinchillas, dogs,

beavers, rats, siamang
Dogs
Cats

Cattle, goats, pigs, sheep, alpaca
Domestic rats
Muskrats, voles

Adapted from Thompson, 2000

genotypes, variation having been demonstrated by isoenzyme and DNA analy-
ses (Meloni et al. 9 1995; Monis et aL, 1999). Genetic groups or 'assemblages'
of G. duodenalis appear to have varying host-specificity (Thompson et al.,
2000) (Table 21.1), providing valuable information to further understanding
of the epidemiology and sources of infection and provide better risk assess-
ment (see below). However, G. duodenalis assemblages are highly divergent,
possibly representing distinct species (Monis et al., 1999) and, increasingly,
genetic analyses illustrate the inadequacy of the current taxonomy within
the genus.

Giardia has a two-stage life cycle, consisting of flagellate trophozoites
(12-18 fjimby 5-7 fjim) and ellipsoid-shaped cysts (9-12 |xm). Infection is non-
invasive and follows the ingestion of cysts in contaminated water, food or by
direct faecal-oral transmission. Excystation occurs in the small intestine when
two trophozoites are released from each cyst. Excystation is probably stimu-
lated by the acid environment of the stomach and can be induced in vitro by
exposure to low pH (Bingham and Meyer, 1979), but has also been observed in
pH neutral aqueous solutions and in patients with achlorhydria (Boucher and
Gillin, 1990). Excystation has also been induced by 0.3 M sodium bicarbonate,
suggesting that pancreatic secretions also have a role (Feely et al., 1991).
Emergence of the trophozoites is stimulated by chymotrypsin, trypsin and pan-
creatic fluid (Boucher and Gillin, 1990). Trophozoites are either free within the
lumen of the duodenum or jejunum, or attached to the surface of mucosal ente-
rocytes by a ventral attachment disc (Inge et al., 1986). Adherence of tropho-
zoites is optimal at 37°C and dependent on structures including the lateral crest
and ventrolateral flange and a combination of forces within the gut lumen (Ceu
Sousa Goncalves etaL 9 2001). The trophozoites multiply by longitudinal binary
fission in the lumen of the proximal small intestine. As the parasites pass from
the small intestine through to the colon they mature to form resistant cysts,
which are shed in the faeces, and are often detected in non-diarrhoeic stools.
Encystation has also been induced in vitro under optimal conditions of bile

300

Giardia duodenalis

salts and glycocholate with myristic acid at pH 7.8 (Gillin et aL 9 1988). The cysts
contain four nuclei and cytokinesis results in the development of two binucle-
ated trophozoites immediately or shortly after being passed in the stool cysts can
be transmitted either directly or via a vehicle such as contaminated food or
water. Although trophozoites can also be shed in the faeces, they are not robust
and are non-infectious (Meyer and Jarroll, 1980).

Giardia, along with Entamoeba, is one of the few eukaryotes that relies on
fermentative metabolism of carbohydrates. Glucose is incompletely oxidized
to ethanol, acetate and CO2, and energy is obtained by substrate-level phos-
phorylation (Lindmark, 1980; Jarroll et al., 1981). Trophozoite metabolism is
affected by oxygen concentration and although minimal amounts of oxygen
alter the metabolic pathway, trophozoites metabolize under anaerobic or
microaerophilic conditions.

Giardia has a world-wide distribution but infection is more prevalent in
warm climates. In developed countries, detection rates of 2-5% have been
reported (Kappus et al. 9 1994). In developing countries, carriage rates of
20-30% have been detected and is particularly prevalent in children (Bryan
et aL, 1994). WHO figures show that some 200 million people in Asia, Africa
and Latin America have giardiasis and the annual incidence is 500 000 cases.
Many of the biological features of Giardia affect its transmission and epi-
demiology. Although transmission can occur between hosts, the life cycle is
simple and dual or multiple hosts are not required for completion. The cysts
are infective immediately or soon after being shed in the faeces, and direct
host-to-host transmission can occur. The infective dose in experimental infec-
tions was 10-25 cysts (Rendtorff, 1954).

Recognized routes of transmission to humans include faecal-oral spread
(including in child day-care centres and male homosexual contact) (Meyers
et al., 1977; Rauch et al. 9 1990), drinking and recreational waters (Hunter,
1997; Marshall et al., 1997), and consumption of contaminated food (Slifko
et al. 9 2000). Cysts are resistant to some degree to many environmental pres-
sures, enabling the survival and environmental transmission of the parasite.

Origin of the organism

Organisms now recognized as G. duodenalis were first described by
Leeuwenhoek in 1681, who, on inspection of his own faeces, described their
size, shape and morphology, noting their presence and link with his own diar-
rhoeal symptoms and dietary habits. The organism was formally described by
Lambl in 1859, who named it Cercomonas intestinalis. While a number of
genus and species names were proposed over the next 40 or so years, G. lam-
blia and G. enterica were proposed in 1915 and 1920 respectively (Kofoid and
Christensen, 1915, 1920). However, G. duodenalis and G. intestinalis are
more accurate taxonomically, although the former may be preferred since it
conforms with Filice's morphologically-based nomenclature. The classification

301

Protozoa

of Giardia remains controversial and difficult, however, since there are no well-
defined criteria for species-designation for clonal organisms in the same clade.
Early descriptions assigning species-status according to host overestimated the
number of species, while those based solely on morphological differences prob-
ably underestimate the number. It is unfortunate that some medical literature
has advocated the use of the term G. lamblia for Giardia isolated solely from
humans, while genetic studies are demonstrating some of the genotypes present
in humans are also found in animals (Thompson et al., 2000).

Although the cyst and trophozoite flagella and nuclei were described in the
19th century, the most complete description of Giardia morphology was made
by Filice in 1952. Dobell had identified the intestinal tract as the natural habi-
tat for growth and reproduction in 1932, but universal acceptance of Giardia
as a human pathogen took some time because many human infections are
asymptomatic and the parasite is non-invasive (Erlandsen and Meyer, 1984).
Clinical studies during the 1970s described the pathology of human infection
(Kulda and Nohynkova, 1978) and in 1981 Giardia was included in the WHO
list of parasitic pathogens (Anon, 1981). The recognition of waterborne out-
breaks of giardiasis in Europe and the USA during the 1970s and 1980s
(Craun, 1986) stimulated research into the epidemiology, pathogenesis and
treatment of the disease.

Clinical features

Clinical features vary widely in severity from asymptomatic carriage to severe
diarrhoea and malabsorption (Meyer and Jarroll, 1980). This variability may
be due to differences in pathogenicity of different isolates as well as differ-
ences in host responses (see below). In human volunteer studies, the prepatent
period was 6-10 days when fresh cysts were used and 13 days using stored
cysts (Rendtorff, 1954). In natural infections the pre-patent period can be
considerably longer, and acute giardiasis usually lasts 1-3 weeks. Symptoms
include diarrhoea, abdominal pain, bloating, flatulence, malaise, sulphurous
belching, nausea and vomiting. The acute phase is usually resolved spontan-
eously in immune-competent individuals, but in some cases can develop into
a chronic phase during which symptoms relapse in short recurrent bouts
(Wolfe, 1990). Immunodeficient patients, particularly those with hypogam-
maglobulinaemia, often suffer chronic disease (see below).

In chronic cases the symptoms are recurrent and, if untreated, malabsorption
and debilitation may occur. Untreated acute giardiasis lasts for at least 10 days and
can last for 3 months or even years. Cryptosporidium and Giardia have been asso-
ciated with persistent diarrhoea in children in developing countries (Lima et aL 9
2000), where a cycle of mainour ishment and continuing diarrhoea can be estab-
lished and lead to poor growth rates and a substantial risk of death (Guerrant
et al., 1992; Fraser et aL, 2000). Other chronic sequelae of Giardia infection also
include reactive arthritis (Tupchong et al., 2001).

302

Giardia duodenalis

Asymptomatic cases can shed cysts in their faeces and present an important
reservoir of infection, particularly in families, child-care settings and institu-
tions, and may impact on cases in the community (Polis et al. 9 1986).

Pathogenicity and virulence

Enteropathy appears, in animal models, to involve enterocyte and micro-
villous damage, villous atrophy and crypt hyperplasia, although normal
appearance and function is restored on clearance of the infection (Ferguson etal. 9
1990). Subsequent to brush border damage, reduction of the enzyme disac-
charidase may occur. Adherence of trophozoites to the surface of mucosal
enterocytes may provide sufficient suction to damage the microvilli, causing
these histopathological changes and interfering with nutrient absorption (Inge
et al. 9 1988). Trophozoites multiply rapidly and can become so numerous as
to create a physical barrier between the epithelial cells and the lumen, causing
brush border enzyme deficiencies and impaired absorption. A relationship has
been demonstrated between the amount of villous damage and malabsorption
in animal infection (Wright and Tomkins, 1978). Cytopathic substances such
as thiol proteinases and lectins may be released from the parasite and stimu-
lation of the host immune response causes release of cytokines and mucosal
inflammation, which have a pathological effect (Farthing, 1997). The role of
inflammatory cells in giardial diarrhoea is not understood, since polymor-
phonuclear leucocytes and eosinophils have not been consistently detected in
patients (Yardley, 1964; Brandborg et al. 9 1967).

Although rarely invasive, human infection with Giardia has been reported
invading tissues of the gallbladder and urinary tract (Meyers et al. 9 1977;
Goldstein et al. 9 1978) and has been demonstrated in a mouse model (Owen
etaL 9 1979). However, approximately 50% of Giardia infections are asymptom-
atic, and risk factors for disease involve both host and parasite factors.

Genotypic and antigenic variation is marked between Giardia isolates
(Nash et al. 9 1987) and antigenic variation is important not only in defining
immune response but also in the development of serological tests. Antigenic
variation occurs as a result of expression of different surface proteins in popu-
lations of trophozoites in axenic culture, animal models and human volun-
teers. Variation may arise from immune responses in the host, adaptation to
different environments or induced shift during excystation. The advantage for
the parasite is that it promotes evasion of the host immune defences and
enables survival in different environments. Ingestion of as few as 10-25 cysts
can cause human disease (Rendtorff, 1954), but strain variation has been
identified in variable infectivity results when two human isolates were used in
human volunteer studies (Nash et al. 9 1987). Thus it is thought that genetic
differences may confer virulence and variation within G. duodenalis and may
even explain some of the discrepant results obtained in infectivity/viability
studies.

303

Protozoa

Causation

Variation in manifestation of clinical symptoms includes the virulence of the
Giardia strain, the number of cysts ingested, the age and immunocompetence
of the host. Both antibodies and T cells appear to be required to control
Giardia infections and giardiasis has been associated with nodular lymphoid
hyperplasia, in which B and T cell functions are decreased. Interestingly, while
the incidence of giardiasis is no different in AIDS patients than in people with
an intact immune system, some such individuals may experience severe illness
and prolonged or combination therapy may be required to clear the infection.
More cases are, however, reported among hypogammaglobulinaemic patients
(Wright et aL 9 1977), who also appear to experience greater villous damage
(Ferguson et al., 1990). Their infections can be difficult to treat and may
require prolonged courses of therapy.

Treatment

Supportive treatment by fluid and electrolyte replacement is important in the
treatment of giardiasis, but therapeutic agents are available to clear the
infection and have been extensively reviewed by Gardner and Hill (2001).
The nitroimidazoles (metronidazole, tinidazole, ornidazole and secnidazole)
are the main agents used, while the nitrofuran compound furazolidone, the
aminoglycoside paromomycin and benzimidazoles (i.e. albendazole) are also
prescribed. Furazolidone is available in a liquid suspension and has therefore
been advocated for paediatric treatment. Mepacrine is an alternative when
nitroimidazole drugs are contraindicated or have failed. Treatment failure has
been reported with all the drugs commonly used to treat giardiasis and may
occur because of drug resistance, non-compliance with the treatment regimen,
re-infection or immunological deficiency. Vost-Giardia lactose intolerance
may cause an apparent recurrence of symptoms (Duncombe et al., 1978).
Treatment of asymptomatic carriers may be suggested for the prevention of
onward transmission, for example in households or in outbreak settings such
as institutions. In endemic areas this has to be balanced with the risk of
re-infection, although it may be desirable since clearance of the infection may
allow catch-up growth in children of marginal nutritional status.

Survival in the environment

Although the trophozoites are not robust, cysts survive in aqueous environments
and show greater resistance to UV, chlorine and ozone than bacteria or viruses
but, with the exception of UV, generally less resistance than Cryptosporidium.

304

Giardia duodenalis

Giardia cysts have been detected in a variety of environmental matrices,
including dairy farm runoff, sewage plant effluent and sewer outflows (States
et al., 1997). Removal can be achieved by clarification and filtration (Sykora
et aL 9 1991), although recycling of filter back wash water has been identified
as a means of reintroducing cysts into water supplies (States et aL 9 1997).
Giardia cysts have been detected in sewage and sewage-contaminated drink-
ing water in La Palta, Argentina (Basualdo et al., 2000) and studies in
Scotland have shown that, although cysts are removed by both activated
sludge and trickle filter treatment, they may occur in treated sewage
(Robertson et al., 2000). Payment et al. (2001) reported 75% removal from
urban wastewater treatment by physicochemical treatment.

While the studies described above are for detection of cysts, methods for
assessing viability are yet to be standardized. Cyst survival can either be meas-
ured as infectivity in a suitable host (usually mouse infectivity) or a surrogate
marker for viability (using vital or fluorogenic dyes, or in vitro excystation).
Although vital dye exclusion tests, particularly eosin, have been historically
applied, comparison tests found that eosin exclusion consistently underesti-
mated cyst viability compared with in vitro excystation (Bingham et al.,
1979). Comparison between eosin exclusion, in vitro excystation and experi-
mental infection has shown lack of correlation (Kasprzak and Majewska,
1983). However, there may be additional influences that may inhibit in vitro
culture that do not affect excystation, and workers recognize that the latter
may at least provide a simple method for viability determination (Hoff et al.,
1985). Propidium iodide (PI) has more recently been used to assess Giardia
viability, since the uptake of this vital dye indicates cell death or damage.
However, while good positive correlation was found between PI staining and
lack of excystation in the evaluation of cysts exposed to heat and quaternary
ammonium compounds, there was no correlation when cysts were exposed
to chlorine or monochloramines (Sauch et al., 1991). Similarly fluorogenic
dyes were found to overestimate viability compared to mouse infectivity when
cysts were treated with ozone, although good correlation was observed
between mouse infectivity and in vitro excystation (Labatiuk et al., 1991).
In vitro excystation was also found to have a good correlation with mouse
infectivity for assessing the viability of cysts treated with chlorine (Hoff et al.,
1985). Although Smith and Smith (1989) also found that PI consistently over-
estimated viability by underestimating dead cysts, a suitable combination of
PI and cyst morphology was suggested as a good marker for viability. Bukhari
et al. (1999) identified problems in the use of surrogate indicators of viability
compared with animal models and suggested that the latter be used as the
gold standard.

Detection of respiratory activity has been used to measure the effects of
drug susceptibility of trophozoites (Sousa and Poiares-Da-Silva, 1999), and
has been applied as a test to determine enzyme redox activity of intact cysts in
environmental isolates (Iturriaga et al., 2001). Electrorotation has also been
used to determine Giardia cyst viability, but application to environmental
samples has yet to be undertaken (Dalton et al., 2001). Amplification of total

305

Protozoa

RNA has been used as an indicator of Giardia viability (Mahbubani et al.,
1991). In a further refinement, test conditions have been used to induce the
synthesis of stress-induced proteins, such as heat-shock proteins, and PCR to
amplify the induced mRNA (Abbaszadegan et al., 1997). This method correl-
ated with PI staining, and since the half-life of mRNA is relatively short it is
unlikely that residual mRNA is detected.

Survival in water

Giardia cysts have been detected in surface waters in both urban and rural
areas. LeChavallier et al. (1991a) studied waters in 14 American states and
one Canadian province and found a greater proportion positive and higher
concentration of cysts in urban areas. The same authors also found that
17% filtered drinking water supplies in the same study areas also contained
Giardia cysts (LeChavallier et ah, 1991b). A Canadian study found Giardia
in one-third of surface waters in remote rural areas and also found cysts
in 17% drinking water supplies to one city, and in sewage (Roach et al.,
1993). Other Canadian studies have also found 21% raw waters and 18%
finished waters (Wallis et al., 1996) in 72 municipalities across Canada con-
tained cysts but with wide geographical (watershed/catchment) variation.
A detailed study in British Columbia, Canada found a high sample prevalence
of Giardia oocysts (Isaac-Renton et al., 1996), and was followed up by a long-
itudinal study of the effects of watershed management on parasite concen-
trations (Ong et al., 1996), which showed that similarly high prevalence of
Giardia cysts were detected, with a distinct seasonal variation, peaking in the
winter months. Sampling in relation to agricultural activity in the catchments
showed a significant increase in the number of cysts detected from sample
points below cattle ranching than above. However, genotypic and phenotypic
variation was observed between cattle, water and human isolates within the
catchment indicating a variety of sources. Groundwaters have been found to
contain Giardia cysts, and are vulnerable where they are under the influence
of surface water or other sources of contamination (Moulton-Hancock et al.,
2000).

Studies have also been undertaken of Giardia cysts in surface waters and
finished waters elsewhere. For example, in Russia, raw river water from the
Sheksna River was found to contain Giardia cysts at a mean of 2 oocysts/100 1
while finished water contained a mean of 1.6 per 10000 litres (Egorov et al.,
2002). In the same study 26/87 (30%) source surface waters throughout the
region contained Giardia cysts while they were detected in 5/70 (7%) finished
water samples. In Japan, Giardia cysts were detected in 12/13 raw water sam-
ples (92%) compared with 3/26 (12%) treated water samples (Hashimoto
etal., 2002).

While some studies have shown a lack of detection in finished waters (Rose
et al., 1991), others have consistently detected cysts in the region of 2-5%
samples (Hancock et al., 1996; Rosen et al., 1996). Observation of naturally

306

Giardia duodenalis

occurring cysts has shown a log removal of Giardia cysts. For example, 1.54
log removal based on low initial concentrations of cysts in the raw water was
reported by States et al. (1997). In contrast, a 3.26 log removal of Giardia
cysts was achieved following experimental seeding of a full-scale conventional
water treatment plant (Nieminski and Ongerth, 1995). The US EPA Surface
Water Treatment Rule, 1989, requires that a 3 log reduction (99.9%) is achieved
in Giardia cysts by a combination of removal and disinfection. Numbers
of cysts detected may vary not only due to natural fluctuations but also
according to the methods used for sampling and detection. Extreme runoff
conditions can result in increased parasite contamination and need to be
accounted for (Kistemann et al., 2002).

Giardia cysts have been shown to maintain viability for up to 3 months in
cold raw water sources and in tap water, with a range of 75-99.9% natural
die-off (deReigner et al., 1989). Although Giardia cysts can be inactivated by
chlorine they do show some resistance (Jarroll et al., 1981), and can survive
water treatment, particularly where chemical dosing levels are unreliable.
Giardia cysts have been shown to be sensitive to UV light at low doses under
laboratory experimental conditions since a 2 log inactivation was observed at
a 3mJ/cm 2 dose (Modifi et al., 2002) and 3 log inactivation at 20-40 mj/cm 2
(Campbell and Wallis, 2002), both studies using an animal model for
viability/infectivity assessment. In bench-scale trials, ozone has been evaluated
but it was shown that to achieve a 2 log inactivation a Ct value 2.4 times
that recommended under the SWTR was required (Finch et al., 1993).
Electroporation has been applied to both Cryptosporidium and Giardia to
increase the permeability of the (oo)cyst wall and reduce the parasite's resist-
ance to disinfection. Haas and Aturaliye (1999) reported that electroporation
enhanced the effect of free chlorine and permanganate on Giardia in phos-
phate buffer. However, further evaluation needs to be undertaken in low con-
ductivity waters and to optimize conditions for disinfection.

The effectiveness of water disinfection methods used to treat small volumes
of drinking water was assessed by excystation by Jarroll and colleagues
(1980), who found that, while saturated iodine, bleach, Globaline, tincture
of iodine were all effective in both cloudy and clear water at 20°C, in cloudy
water at 3°C saturated iodine failed to inactivate cysts and in clear water
at the same temperature all methods failed. Commercial phenol-based dis-
infectants (Pine Glo, ammonia and Dettol) have been shown to inactivate
99% Giardia cysts following a 1 -minute contact time while Pine Sol and
vinegar were less effective (Lee, 1992). Sufficient residual and/or contact time
are critical to the disinfection process, as was demonstrated by Ongerth
et al. (1989) who found that iodine-based disinfectants were more effective
than chlorine-based disinfectants. Although all failed to achieve 99.9% reduc-
tion in viability (measured by animal infectivity) after 30 minutes, iodine-
based disinfection was effective after 8 hours. Heating water to 70°C for
10 minutes was effective. However, Gerba et al. (1997) reported that iodine
tablets were effective in 'general case water' at pH 9 but relatively
ineffective at pH 5.

307

Protozoa

Methods of detection

Clinical diagnosis is by the identification of cysts or, during heavy infections,
trophozoites in faeces either by direct examination of wet mounts or follow-
ing concentration (Isaac-Renton, 1991). Organisms may also be detected in
duodenal biopsy material or aspirates in difficult cases. Histological or
immunofluorescent stains are also used to visualize the organisms and enzyme
immunoassays are available for the detection of antigens. Cysts may be
shed intermittently and so more than one stool may need to be examined for
diagnosis and screening of more than one stool can significantly increase
the chance of making a diagnosis (Cartwright, 1999). Screening of asympto-
matic contacts of cases is desirable to control the spread of infection in
some settings, for example during outbreaks in child-care centres and other
institutions.

Low numbers of cysts in stools and inexperience of laboratory technicians
can also contribute to poor sensitivity in microscopical detection (Wolfe,
1990). While some enzyme immunoassay kits are more sensitive than others,
many show a greater sensitivity than microscopy on examination of a single
stool specimen (Hanson and Cartwright, 2001) or when treatment is already
underway (Nash et aL, 1987; Knisley et al. 9 1989). However, antigenic
variation may result in negative results from some antibody-based methods
(Moss etaL, 1990).

Methods for detection in environmental samples require recovery of the
cysts from the sample matrix (usually by filtration), concentration (by
centrifugation and immunomagnetic separation) and staining with immuno-
fluorescent antibodies prior to inspection by UV microscopy. Large volumes
of samples are required for representative sampling and to recover low con-
centrations of parasites. Methods are detailed in the USEPA method 1623
(Anon, 1999). Development of alternative methods includes continuous flow
centrifugation which has improved recovery of waterborne pathogens from
water samples (Borchardt and Spencer, 2002) with recoveries over 90%
reported, and membrane dissolution procedures (McCuin et aL 9 2000).

The development of IMS and IFAs has improved the isolation and detection
of organisms such as Giardia and Cryptosporidium from water and environ-
mental samples, including food and enhanced our evaluation of water- and
food-borne risks. Using IMS/IFA and improved washing procedures,
Robertson and Gjerde (2000) reported mean recovery efficiencies for Giardia
of 71% from iceberg lettuce, 65% from green lollo lettuce, 69% from Chinese
leaves, 66% from an autumn salad mix, 62% strawberries and 4-42% from
bean sprouts, largely depending on the age of the sprouts tested. Fewer cysts
were recovered from older sprouts and therefore they recommend that fruit
and vegetables should be tested as fresh as possible.

Molecular methods have been applied for both the detection and character-
ization of Giardia in both clinical and environmental samples. The applica-
tion of electrophoresis-based and PCR-based methods for the investigation of

308

Giardia duodenalis

G. duodenalis isolates has demonstrated the phenotypic and genotypic
heterogeneity of the species and contributed much towards our understanding
of the epidemiology of the parasite. Methods have included pulsed field gel
electrophoresis, isoenzyme typing, sequence-based analyses of the glutamate
dehydrogenase, triosphosphate isomerase (tpi), elongation factor la and 18S
rRNA genes (Monis et al., 1999) and PCR/RFLP of the tpi gene locus
(Mclntyre et al. 9 2000). The sensitivity and specificity of the PCR/RFLP of the
tpi gene locus has been improved by Amar and colleagues (2002) to distin-
guish G. duodenalis assemblages A and B and the subgroups of assemblage A
in human clinical isolates. PCR-based methods eliminate the need for in vitro
or in vivo amplification of the parasite, and thus broaden the range of isolates
that can be analysed. Improved tools for molecular analyses and the investi-
gation of large, representative panels of isolates provide better data to identify
genetic divisions within the G. duodenalis group. These need to be comple-
mented by investigation of biotypic characteristics, for example, host range
and pathogenicity, virulence markers and drug sensitivity.

Two viruses have been identified in G. duodenalis. One is a 6.2 kb non-
enveloped double-stranded RNA virus (Wang and Wang, 1986) with a
100 kDa major capsid protein (Yu et al. 9 1995) and the other also has 6.2 kb
double-stranded RNA genome but encodes a smaller (95 kDA) protein (Tai
et al., 1996). These may serve as markers for strain variation.

Epidemiology

Human cases of giardiasis are reported world-wide in both adults and
children, but are more common where hygiene is poor and water treatment
inadequate. For example, on examination of stools submitted for ova and
parasites, 2-5% prevalence is observed in developed countries and 20-30%
in developing countries. Seasonality has been observed with peak incidence in
late summer in the UK, USA and Mexico. Person-to-person transmission
occurs and is frequent where hygiene is poor, largely because Giardia is infec-
tive immediately or shortly after being passed in the stools. Giardiasis is of
emerging importance, particularly since the number of outbreaks associated
with day-care centres has increased, and the organism poses a risk in long-term
care settings (Nicolle, 2001). In developed countries, many patients report for-
eign travel, probably reflecting exposure to poor hygiene, poorly treated water
and unfamiliar strains of the parasite. Water is increasingly recognized as a vehi-
cle for transmission and has caused concern among water providers in devel-
oped countries. CDC began surveillance for waterborne disease in 1971, and in
the USA Giardia is the most frequently implicated organism in outbreaks of
waterborne disease, although in the UK that accolade goes to Cryptosporidium.
Waterborne outbreaks have been associated with the treated drinking water
supply, indicating failure to remove/inactivate cysts.

309

Protozoa

In North America giardiasis is commonly referred to as c beaver fever' from
the perception that drinking untreated surface water contaminated by beavers
is a cause of illness. Indeed, strains isolated from human cases during an out-
break were shown to be indistinguishable from those isolated from beavers
(Isaac-Ren ton et al. 9 1993). However, meta analysis of studies in the USA has
shown that there was minimal evidence for an association between drinking
untreated water and a high incidence of giardiasis among back country recre-
ationalists (Welch, 2000), and that other recreational exposures such as con-
tamination of hands and hand-to-mouth contact while camping should be
explored and therefore proper hand washing prior to eating and drinking
would be an effective control measure.

Much of the information about waterborne giardiasis has arisen from the
study of outbreaks. While the relative contribution of the multiple possible
routes of transmission of the parasite is unknown, it has been estimated that
60% of all cases of giardiasis in the USA are water-related (Bennett et al. 9 1987).

By contrast to the predominance of waterborne outbreaks in the USA
caused by Giardia, in the UK relatively few waterborne outbreaks of giardia-
sis have been reported. In the summer of 1985, 108 laboratory-confirmed
cases were linked to consumption of the municipal water supply in the South
West (Jephcott et al. 9 1986). Cases were almost exclusively adults. The water
supply had been treated by filtration and chlorination. A case control study
showed a significant association with water consumption coinciding with the
dates the water main supplying the community was opened for routine work.
However, the symptoms of some cases anteceded the main outbreak and
could have represented other aetiologies or the source of the outbreak itself.
Reservoir samples showed no indication of pollution and ingress into the system
was suspected, although environmental samples were negative for Giardia.
The importance of laboratory screening of all stools for Giardia was illus-
trated by this outbreak.

In 1991/2 a private water supply in England supplying 260 people was
implicated in an outbreak involving 31 cases, five of which were laboratory
confirmed. Three Giardia cysts were detected per litre and faecal streptococci
and coliforms were also detected. As a control measure a boil water notice
was imposed. Inspection of the supply showed the spring collection chamber,
although covered, had animals grazing round it. The reservoir was chlorine
dosed twice weekly but filtration was installed only after the outbreak. The
identification of private water supplies as a route of transmission of giardiasis
is because of their general lack of appropriate treatment and their vulnerabil-
ity to faecal contamination.

A case control study was undertaken in Avon and Somerset, England over
a 12-month period from July 1992 to May 1993, to identify risk factors for
sporadic giardiasis, in particular the risk associated with recreational water use
and associations with foreign travel (Gray et al. 9 1994). Swimming, travel to
developing countries and accommodation type (camping/caravanning/staying
in holiday chalets) were identified as independent risk factors for giardiasis.
A larger study encompassing the same region of England was carried out for

310

Giardia duodenalis

12 months from April 1998 and excluded people who had travelled outside the
UK in the 3 weeks prior to the onset of illness in order to investigate risk fac-
tors for endemic sporadic giardiasis (Stuart et al. 9 2003). Swallowing water
while swimming, recreational fresh water contact, drinking treated tap water,
and eating lettuce were all associated with giardiasis.

Food-borne outbreaks of giardiasis are frequently related to infected food
handlers or their contact with an infected person. Interestingly, while cysts
may have been detected in subsequent screening, the food handlers and their
contacts are frequently asymptomatic (Rose and Slifco, 1999). For example,
the first report of food-borne transmission was of an outbreak in December
1979 when 29 cases occurred at a school in Minnesota, in which salmon was
identified as the vehicle of infection. The food, including the fish, had been
prepared by someone caring for a 1-year-old child who was excreting cysts.
Other vehicles of infection have included noodle salad, fruit salad, sand-
wiches, various salad items or raw vegetables (Rose and Slifco, 1999).
Transmission may also occur via fresh produce as a result of the use of uncom-
posted manure or human waste. Giardia has been detected in clams (Macoma
balthica and Macoma mitchelli) from Rhode River, a subestuary of
Chesapeake Bay (Graczyk et al. 9 1999). Not only may this pose a risk through
the consumption of undercooked shellfish, but it is possible that filter feeders
could serve as biological indicators and contribute to the sanitary assessment
of water.

Recreational waters clearly have a role in the epidemiology of giardiasis and
outbreaks have occurred linked to the use of both chlorinated swimming
pools and natural fresh recreational waters.

Molecular typing of isolates has indicated that an assumed potentially large
animal reservoir of the organism may have lower zoonotic potential than pre-
viously thought, although this was previously difficult to determine because
morphological differences between strains are lacking. Studies of albeit
limited numbers of human clinical isolates from sporadic cases in the UK
using PCR-RFLP of the tpi gene show that 21/33 (64%) were assemblage B,
9/33 (27%) were assemblage All, 3/33 (9%) contained both B and All and
none of the 33 were assemblage AI ( Amar et al. 9 2002). These results therefore
question the role of livestock animals as a major reservoir of human giardia-
sis in the UK. Sequence analysis of the 18S rDNA gene showed that human
infections in an endemic area of Western Australia, previously assumed to be
zoonotic, were different from dog isolates (Hopkins et al. 9 1997), although
there are some common genotypes in both humans, dogs and cats within
assemblages A and B (Thompson, 2000). Case control studies have generated
conflicting evidence for the role of both household pets and contact with
farmed animals: a study in the South East of England showed an association
with pigs, dogs and cats (Warburton et al. 9 1994), while other studies found
no such associations (Mathias et al. 9 1992; Dennis et al. 9 1993). Thus the role
of pets is not fully established, nor is the pathogenicity of this parasite for cats
and dogs. Interestingly, in the USA a vaccine is commercially available for the
prevention of clinical signs and reduction of cyst shedding in cats and dogs.

311

Protozoa

Further molecular epidemiological studies are required to understand fully the
reservoirs of human infection and their relative importance for public health.

Risk assessment

It is likely, from epidemiological and genetic evidence, that two cycles of infec-
tion with Giardia probably exist: human-human and zoonotic (Thompson,
2000). Control of direct transmission between humans can be made by imple-
mentation of normal hygiene practices and there is a higher risk of transmis-
sion where faecal-oral contamination occurs (e.g. child-care centres; male
homosexual practices). Sources of zoonotic Giardia need to be better defined
by more extensive molecular epidemiological studies and sample surveys, but
there is evidence that some species, such as the beaver, act as amplification
hosts for strains infective for humans (Isaac-Renton et al., 1993). Outbreak
investigations have demonstrated the importance of the animal reservoir but
the contribution to sporadic cases is unknown. Molecular epidemiology pro-
vides a powerful predictive tool for evidence of transmission and for determin-
ing sources of contamination and infection and investigations of the genomic
characteristics of G. duodenalis have contributed greatly to our understanding
of the epidemiology of giardiasis and contributes to risk assessment.

In a risk assessment model for Giardia, Rose and colleagues (1991) used
a priori prevalence data showing average numbers of cysts in surface waters
ranging from 0.33 to 104 cysts/1001 compared with 0.6-5 cysts/1001 in pris-
tine waters. Assuming a source water contaminated with 0.7-70 cysts/1001
they demonstrated that treatment achieving 3 to 5 log reduction in cysts
would be required to meet a risk target of less than one case giardiasis per
10000 population. However, such models have yet to consider the infectivity
of cysts for humans and their significance for public health. Although filtra-
tion is more effective for removal of Giardia cysts than Cryptosporidium
oocysts, not all water supplies are filtered and both organisms have a low
infectious dose.

Overall risk assessment

Health effects: occurrence of illness, degree of morbidity and mortality, prob-
ability of illness based on infection:

• Human cases of giardiasis are reported world-wide in both adults and chil-
dren, but are more common where hygiene is poor and water treatment
inadequate. In developed countries, detection rates of 2-5% have been
reported and between 2 and 20% of those infected will have symptoms.
It has been estimated that 60% of all cases of giardiasis in the USA are

312

Giardia duodenalis

water-related. In developing countries, carriage rates of 20-30% have been
detected and are particularly prevalent in children.

• Clinical features vary from asymptomatic carriage to severe diarrhoea and
malabsorption. Symptoms include diarrhoea, abdominal pain, bloating,
flatulence, malaise, sulphurous belching, nausea and vomiting. The acute
phase is usually resolved spontaneously in immune-competent individuals, but
in some cases can develop into a chronic phase during which symptoms relapse
in short recurrent bouts. Untreated acute giardiasis lasts for at least 10 days
and can last for 3 months or even years. Immunocompromised patients, par-
ticularly those with hypogammaglobulinemia, often suffer chronic disease.

• Variation in clinical symptoms includes the virulence of the Giardia strain,
the number of cysts ingested and the age and immunocompetence of the host.

Exposure assessment: routes of exposure and transmission, occurrence in
source water, environmental fate:

• Asymptomatic cases can shed cysts in their faeces and provide an important
reservoir of infection, particularly in families, child-care settings and
institutions.

• Recognized routes of transmission to humans include faecal-oral spread
through person-to-person contact, ingestion of drinking and recreational
waters, consumption of contaminated food and zoonotic transmission.

• Many animals carry the pathogenic species of Giardia and a number of water-
borne outbreaks have been associated with animal faecal contamination.

• The infective dose in experimental infections was 10-25 cysts, which makes
Giardia easily spread from person to person and of significance even if
detected in 'low' numbers.

• Giardia cysts have been found in surface waters in urban and rural areas,
groundwater, sewage and treated drinking water.

• Cysts are resistant to many environmental pressures, enabling the survival
and environmental transmission of the parasite. Cysts survive in water
environments and show greater resistance to UV, chlorine and ozone than
bacteria or viruses, but with the exception of UV, generally less resistance
than Cryptosporidium.

• Giardia cysts have been shown to maintain viability for up to 3 months in cold
raw water sources and in tap water, with a range of 75-99.9% natural die-off.

Risk mitigation: drinking-water treatment, medical treatment:

• Cyst removal can be achieved by clarification and filtration, although recyc-
ling of filter back wash water has been identified as a means of reintroduc-
ing cysts into water supplies. Proper filtration can remove cysts at levels of
99% or higher.

• With appropriate concentration and contact time, chemical disinfectants
can effectively inactivate Giardia cysts.

• Supportive treatment by fluid and electrolyte replacement is important in
the treatment of giardiasis. The nitroimidazoles (metronidazole, tinidazole,
ornidazole and secnidazole) are the main agents used, while the nitrofuran

313

Protozoa

compound furazolidone, the aminoglycoside paromomycin and benzimida-
zoles (i.e. albendazole) are also prescribed.

Future implications

Untreated water sources are vulnerable and treatment plants, particularly
those operating under conditions of, for example, low temperature, may not
be capable of consistently meeting annual risk targets of 1/10 000 for giardia-
sis set under the Surface Water Treatment Rule. Work by LeChevallier et al.
(1991a) suggests not. Therefore, additional treatment may be required to
prevent outbreaks occurring. Better identification of sources and routes of
transmission would help prevent waterborne incidents.

References

Abbaszadegan, M., Huber, M.S., Gerba, C.P. etal. (1997). Detection of viable Giardia cysts

by amplification of heat shock-induced mRNA. Appl Environ Microbiol, 63: 324-328.
Amar, C.F.L., Dear, P.H., Pedraza-Diaz, S. et al. (2002). Sensitive PCR-restriction fragment

length polymorphism assay for detection and genotyping of Giardia duodenalis in

human faeces./ Clin Microbiol, 40: 446-452.
Anon. (1981). Intestinal protozoan and helminthic infections. WHO Technical Report Ser,

58:666-671.
Anon. (1999). Method 1623: Cryptosporidium and Giardia in water by filtration/IMS/FA.

United States Environmental Protection Agency.
Basualdo, J., Pezzani, B., de Luca, M. etal. (2000). Screening of the municipal water system of

La Palta, Argentina, for human intestinal parasites. IntJ Hyg Environ Hltb, 203: 177-182.
Bennett, J.V., Homberg, S.D. and Rogers, M.F. (1987). Infectious and parasitic diseases.

Am J Prevent Med, 3: 102-114.
Bingham, A.K. and Meyer, E.A. (1979). Giardia excystation can be induced in vitro in

acidic solutions. Nature, 277: 301-302.
Bingham, A.K., Jarroll, E.L. and Meyer, E.A. (1979). Giardia sp.: Physical factors of excys-
tation in vitro, and excystation vs eosin exclusion as determinants of viability. Exp

Parasitol, 47:284-291.
Borchardt, M.A. and Spencer, S.K. (2002). Concentration of Cryptosporidium,

microsporidia and other water-borne pathogens by continuous separation channel

centrifugation. / Appl Microbiol, 92: 649-652.
Boucher, S.E.M. and Gillin, F.D. (1990). Excystation of in vitro-derived Giardia lamblia

cysts. Infect Immun, 58: 3516-3522.
Brabdborg, L.L., Tankersley, C.B., Gottlieb, S. et al. (1967). Histological demonstration of

mucosal invasion by Giardia lamblia in man. Gastroenterology, 52: 143-150.
Bryan, R.T., Pinner, R.W. and Berkelman, R.L. (1994). Emerging infectious diseases in the

United States. Ann NY Acad Sci, 740: 346-361.
Bukhari, Z., Hargy, T., Bolton, J. et al. (1999). Medium-pressure UV for oocyst inactiva-

tion. J AWWA, 91:86.
Campbell, A.T. and Wallis, P. (2002). The effect of UV irradiation on human-derived

Giardia lamblia cysts. Water Res, 36: 936-969.
Cartwright, C.P. (1999). Utility of multiple-stool-specimen ova and parasite examinations

in a high-prevalence setting./ Clin Microbiol, 37: 2408-2411.

314

Giardia duodenalis

Ceu Sousa Goncalves, C.A., Bairos, V.A. and Poiares-da-Silva, J. (2001). Adherence of
Giardia lamblia trophozoites to Int-407 human intestinal cells. Clin Diag Lab
Immunol, 8: 258-265.

Craun, G.F. (1986). Waterborne giardiasis in the United States 1965-1984. Lancet, ii:
513-514.

Dalton, C, Goater, A.D., Pethig, R. et al. (2001). Viability of Giardia intestinalis cysts
and viability and sporulation state of Cyclospora cayetanensis oocysts determined by
electrorotation. Appl Environ Microbiol, 67: 586-590.

Dennis, D.T., Smith, R.P., Welch, J.J. et al. (1993). Endemic giardiasis in New Hampshire:
a case control study of environmental risks. / Infect Dis, 167: 1391-1395.

deReigner, D.P., Cole, L., Schupp, D.G. et al. (1989). Viability of Giardia cysts suspended
in lake, river and tap water. Appl Environ Microbiol, 55: 1223-1229.

Duncombe, V.M., Bolin, T.D., Davies, A.E. et al. (1978). Histopathology in giardiasis: a
correlation with diarrhoea. Aust NZ J Med, 8: 392-396.

Egorov, A., Paulauskis, J., Petrova, L. et al. (2002). Contamination of water supplies with
Cryptospordiium parvum and Giardia lamblia and diarrhoeal illness in selected
Russian Cities. Int J Hyg Environ Hlth, 205: 281-289.

Erlandsen, S.L. and Meyer, E.A. (eds) (1984). Giardia and Giardiasis Biology, Pathogenesis,
and Epidemiology. New York: Plenum Press.

Erlandsen, S.L. and Bemrick, W.J. (1987). SEM evidence for a new species, Giardia
psittaci. J Parasitol, 73: 623-629.

Erlandsen, S.L., Bemrick, W.J., Wells, C.L. et al. (1990). Axenic culture and character-
isation of Giardia ardeae from the great blue heron (Ardea herodias). J Parasitol, 76:
717-724.

Farthing, M.J.G. (1997). The molecular pathogenesis of giardiasis./ Paediatr Gastroenterol
Nutr, 24: 79-88.

Feely, D.E. (1988). Morphology of the cyst of Giardia microi by light and electron
microscopy./ Protozool, 35: 52-54.

Feely, D.E., Gardner, M.D. and Hardin, E.L. (1991). Excystation of Giardia muris induced
by a phosphate-bicarbonate medium: localisation of acid phosphatase. / Parasitol, 77:
441-448.

Ferguson, A., Gillon, J. and Munro, G. (1990). Pathology and pathogenesis of the intes-
tinal mucosal damage in giardiasis. In Giardiasis, Meyer, E.A. (ed.). New York: Elsevier
Publishing Co, pp. 155-173.

Filice, F.P. (1952). Studies on the cytology and life history of a Giardia from the laboratory
rat. Univ California Publ Zool, 57: 53-146.

Finch, G.R., Black, E.K., Labatiuk, C.W. et al. (1993). Comparison of Giardia lamblia and
Giardia muris cyst inactivation by ozone. Appl Environ Microbiol, 59: 3674-3680.

Fraser, D., Bilenko, N., Deckelbaum, R.J. et al. (2000). Giardia lamblia carriage in Israeli
Bedouin infants: risk factors and consequences. Clin Infect Dis, 30: 419-424.

Gardener, T.B. and Hill, D.R. (2001). Treatment of Giardiasis. Clin Microbiol Rev, 14:
114-128.

Gerber, C.P., Johnson, D.C. and Hasan, M.N. (1997). Efficacy of iodine water purification
tablets against Cryptospordium oocysts and Giardia cysts. Wilderness Environ Med, 8:
96-100.

Gillin, F.D., Reiner, D.S. and Boucher, S.E. (1988). Small intestinal factors promote excys-
tation of Giardia lamblia in vitro. Infect Immun, 56: 705-707.

Goldstein, F.J., Thornton, J.J. and Szyldlowski, T. (1978). Biliary tract dysfunction in giar-
diasis. Am J Dig Dis, 23: 559-560.

Graczyk, T.K., Thompson, R.C., Fayer, R. et al. (1999). Giardia duodenalis cysts of geno-
type A recovered from clams in the Chesapeake Bay subestuary, Rhond River. Am J
Prop Med Hyg, 61: 526-529.

Gray, S.F., Gunnell, D.J. and Peters, T.J. (1994). Risk factors for giardiasis: a case control
study in Avon and Somerset. Epidemiol Infect, 113: 95-102.

Guerrant, R.L., Schorling, J.B., McAuliffe, J.F. et al. (1992). Diarrhoea as a cause and
effect of malnutrition: diarrhoea prevents catch-up growth and malnutrition increases
diarrhoea frequency and duration. Am J Prop Med Hyg, 47: 28-35.

315

Protozoa

Haas, C.N. and Aturaliye, D. (1999). Semi-quantitative characterisation of electroporation-

assisted disinfection processes for inactivation of Giardia and Cryptosporidium. J Appl

Microbiol, 86: 899-905.
Hancock, CM. etal. (1996). Assessing plant performance using MPA. JAWWA, 88: 24-30.
Hashimoto, A., Kunikane, S. and Hirata, T. (2002). Prevalence of Cryptosporidium

oocysts and Giardia cysts in the drinking water supply in Japan. Water Res, 36:

519-526.
Hoff, J.C., Rice, E.W. and Shaeffer, F.W. (1985). Comparison of animal infectivity and

excystation as measures of Giardia muris cyst inactivation by chlorine. Appl Environ

Microbiol, 50: 115-117.
Hopkins, R.M., Meloni, B.P., Groth, D.M. et al. (1997). Ribosomal RNA sequencing

reveals differences between the genotypes of Giardia isolates recovered from humans

and dogs living in the same locality./ Parasitol, 83: 44-51.
Hunter, P.R. (1997). Waterborne Disease Epidemiology and Ecology. Chichester: Wiley.
Inge, P.M.G., Edson, CM. and Farthing, M.J.G. (1986). Attachment of Giardia lamblia to

rat intestinal epithelial cells. Gut, 29: 795-801.
Isaac-Renton, J., Cordeiro, C, Sarafis, K. et al. (1993). Characterisation of Giardia duo-

denalis isolates from a waterborne outbreak./ Infect Dis, 167: 431-440.
Isaac-Renton, J., Moorehead, W. and Ross, A. (1996). Longitudinal studies of Giardia con-
tamination in two community drinking water supplies: cyst levels, parasite viability and

health impact. Appl Environ Microbiol, 62: 47-54.
Iturriaga, R., Zhang, S., Sonek, G.J. et al. (2001). Detection of respiratory enzyme activity

on Giardia cysts and Cryptosporidium oocysts using redox dyes and immunofluores-
cence techniques. / Microbiol Metb, 46: 19-28.
Jarroll, E.L., Bingham, A.K. and Meyer, E.A. (1980). Giardia cyst destruction: effectiveness

of six small-quantity water disinfection methods. Am J Trop Med Hyg, 29: 8-11.
Jarroll, E.L., Bingham, A.K. and Meyer, E.A. (1981). Effect of chlorine on Gardia lamblia

cyst viability. Appl Environ Microbiol, 41: 483-487.
Jarroll, E.L., Muller, P.J., Meyer, E.A. et al. (1981). Lipid and carbohydrate metabolism of

Giardia lamblia. Mol Biocbem Parasitol, 2: 187-196.
Jephcott, A.E., Begg, N.T. and Baker, LA. (1986). Outbreak of giardiasis associated with

mains water in the United Kingdom. Lancet, i: 730-732.
Kappus, K.D., Lundgren, R.G., Juranek, D.D. et al. (1994). Intestinal parasitism in the

United States: update on a continuing problem. Am J Trop Med Hyg, 50: 705-713.
Kasprzak, W. and Majewska, A.C (1983). Infectivity of Giardia sp. cysts in relation to

eosin exclusion and excystation in vitro. Tropmed Parasitol, 34: 70-72.
Kistemann, T., Classen, T., Koch, C et al. (2002). Microbial load of drinking water reser-
voir tributaries during extreme rainfall and runoff. Appl Environ Microbiol, 68:

2188-2197.
Knisley, C.V., Engelkirk, P.G., Pickering, L.K. et al. (1989). Rapid detection of Giardia

antigen in stool with the use of enzyme immunoassays. Am J Clin Pathol, 91: 704-708.
Kofoid, CA. and Christensen, E.B. (1915). On binary and multiple fission in Giardia muris

(Grassi). Univ California Publ Zool, 16: 30-54.
Kofoid, CA. and Christensen, E.B. (1920). A critical review of nomenclature of human

intestinal flagellates. Univ California Publ Zool, 20: 160.
Kulda, J. and Nohynkova, E. (1978). Flagellates of the human intestine and of intestines of

other species. In Protozoa of Veterinary and Medical Interest, Kreier, J.P. (ed.). New

York: Academic Press, pp. 69-104.
Labatiuk, C.W., Schaefer, F.W., Finch, G.R. et al. (1991). Comparison of animal infectivity,

excystation and fluorogenic dye as measures of Giardia muris cyst inactivation by

ozone. Appl Environ Microbiol, 57: 3187-3192.
Lambl, W. (1859). Mikroskopische untersuchungen der Darmexcrete. Vierteljabrsscbr

Prakst Heikunde, 61: 1-58.
LeChavallier, M.W., Norton, W.D. and Lee, R.G. (1991a). Occurrence of Giardia and

Cryptosporidium spp. in surface water supplies. Appl Environ Microbiol, 57: 2610-2616.
LeChevallier, M.W., Norton, W.D. and Lee, R.G. (1991b). Giardia and Cryptsporidium

spp. in filtered drinking water supplies. Appl Environ Microbiol, 57: 2617-2621.

316

Giardia duodenalis

Lee, M.B. (1992). The effectiveness of commercially available disinfectants upon Giardia

lamblia cysts. Can] Public Hltb, 83: 171-172.
Lindmark, D.G. (1980). Energy metabolism of the anaerobic protozoon, Giardia lamblia.

Mol Biocbem Parasitol, 1: 1-12.
Mahbubani, M.H., Bej, A.K., Perlin, M. et al. (1991). Detection of Giardia cysts by using

the polymerase chain reaction and distinguishing live from dead cysts. Appl Environ

Microbiol, 57: 597-600.
Marshall, M.M., Naumovitz, D., Ortega, Y. et al. (1997). Waterborne protozoan

pathogens. Clin Microbiol Rev, 10: 67-85.
Mathias, R.G., Riben, P.D. and Osei, W.D. (1992). Lack of an association between

endemic giardiasis and a drinking water source. Can] Public Hltb, 83: 382-384.
McCuin, R.M., Bukhari, Z. and Clancy, J.L. (2000). Recovery and viability of

Cryptosporidium parvum oocysts and Giardia intestinalis cysts using the membrane

dissolution procedure. Can] Microbiol, 46: 700-707.
Mclntyre, L., Hoang, L., Ong, C.S.L. et al. (2000). Evaluation of molecular techniques to

biotype Giardia duodenalis collected during an outbreak./ Parasitol, 86: 172-177.
Meyer, E.A. and Jarroll, E.L. (1980). Giardiasis. Am] Epidemiol, 111: 1-12.
Meyers, J.D., Kuharic, H.A. and Holmes, K.K. (1977). Giardia lamblia infection in homo-
sexual men. Br ] Vener Dis, 53: 54-55.
Modifl, A.A., Meyer, E.A., Wallis, P.M. et al. (2002). The effect of UV light on the inacti-

vation of Giardia lamblia and Giardia muris cysts as determined by animal infectivity

assay (P-2951-01). Water Res, 36: 2098-2108.
Monis, P.T., Andrews, R.H., Mayrhofer, G. et al. (1999). Molecular systematics of the para-
sitic protozoan Giardia intestinalis. Mol Biol Evoln, 16: 1135-1144.
Moss, D.M., Mathews, H.M., Visvesvara, G.S. et al. (1990). Antigenic variation of Giardia

lamblia in the faeces of Mongolian gerbils. / Clin Microbiol, 28: 254-257.
Moulton-Hancock, C., Rose, J.B., Vasconcelos, G.J. et al. (2000). Giardia and

Cryptosporidium occurrence in groundwater. JAW WA, 92: 117-123.
Nanson, K.L. and Cartwright, C.P. (2001). Use of an enzyme immunoassay does not

eliminate the need to analyse multiple stool specimens for sensitive detection of Giardia

lamblia.] Clin Microbiol, 39: 474-477.
Nash, T.E., Harrington, G.A., Losonsky, G.A. et al. (1987). Experimental human infec-
tions with Giardia lamblia. ] Infect Dis, 156: 974-984.
Nicolle, L.E. (2001). Preventing infections in non-hospital settings: long term care.

Emerging Infect Dis, 7: 205-207.
Ong, C., Moorehead, W., Ross, A. et al. (1996). Studies of Giardia spp. and

Cryptosporidium spp. in two adjacent watersheds. Appl Environ Microbiol, 62:

2798-2805.
Ortega, Y.R. and Adam, R.D. (1997). Giardia: overview and update. Clin Infect Dis, 25:

545-550.
Owen, R.L., Nemanic, P.D. and Stevens, D.P. (1979). Ultrastructural observations of

giardiasis in a murine model. Gastroenterology, 76: 757-769.
Payment, P., Plante, R. and Cejka, P. (2001). Removal of indicator bacteria, human enteric

viruses, Giardia cysts, and Cryptosporidium oocysts at a large wastewater primary

treatment facility. Can ] Microbiol, 47: 188-193.
Polis, M.A., Tuazon, C.U., Ailing, D.W. et al. (1986). Transmission of Giardia lamblia

from a day care center to the community. Am ] Public Hlth, 76: 1142-1144.
Rauch, A.M., Van, R., Bartlett, A.V. et al. (1990). Longitudinal study of Giardia lamblia in

a day care center population. Pediatr Infect Dis J, 9: 186-189.
Rendtorff, R.C. (1954). The experimental transmission of human intestinal protozoan para-
sites. II. Giardia lamblia cysts given in capsules. Am ] Hyg, 59: 209-220.
Robertson, L.J. and Gjerde, B. (2000). Isolation and enumeration of Giardia cysts,

Cryptosporidium oocysts, and Ascaris eggs from fruits and vegetables. / Food Protect,

63: 775-778.
Robertson, L.J., Paton, C.A., Campbell, A.T. et al. (2000). Giardia cysts and

Cryptosporidium oocysts at sewage treatment works in Scotland, UK. Water Res, 34:

2310-2322.

317

Protozoa

Rose, J.B., Gerba, C.P. and Jakubowski, W. (1991). Survey of potable water supplies for

Cryptosporidium and Giardia. Environ Sci Technol, 258: 1393-1401.
Rose, J.B., Haas, C.N. and Regli, S. (1991). Risk assessment and control of waterborne

giardiasis. Am J Public Hlth Assoc, 81: 709-713.
Rose, J.B. and Slifco, T.R. (1999). Giardia, Cryptosporidium and Cyclospora and their

impact on foods: a review./ Food Protect, 62: 1059-1070.
Rosen, J.S. et al. (1996). Development and Analysis of a National Protozoan Database.

Proceedings of the AWWA WQTC Boston, 1996.
Slifko, T.R., Smith, H.V. and Rose, J.B. (2000). Emerging parasite zoonoses associated

with food and water. Int J Parasitol, 30: 1379-1393.
Smith, A.L. and Smith, H.V. (1989). A comparison of fluorescein diacetate and propidium

iodide staining and in vitro cultivation for determining Giardia intestinalis cyst viabil-
ity. Parasitology, 99: 329-331.
Sousa, M.A. and Poiares-Da-Silva, J. (1999). A new method for assessing metronizadole

susceptibility of Giardia lamblia trophozoites. Antimicrob Agents Chemother, 43:

2939-2942.
States, S., Stadterman, K., Ammon, L. etal. (1997). Protozoa in river water: sources, occur-
rence, and treatment. J AWWA, 89: 74-83.
Sykora, J.L. et al. (1991). Distribution of Giardia cysts in wastewater. Water Sci Technol,

24:2187-2193.
Tai, J.H., Chang, S.C., Chou, C.F. et al. (1996). Separation and characterisation of two

related giardia viruses in the parasitic protozoan Giardia lamblia. Virology, 216:

124-132.
Thompson, R.C.A. (2000). Giardiasis as a re-emerging infectious disease and its zoonotic

potential. Int J Parasitol, 30: 1259-1267.
Thompson, R.C.A., Hopkins, R.M. and Homan, W.L. (2000). Nomenclature and genetic

groupings of Giardia infecting mammals. Parasitol Today, 16: 210-213.
Tupchong, M., Simor, A. and Dewar, C. (2001). Beaver fever - a rare cause of reactive

arthritis./ Rheumatol, 28: 683.
van Keulan, H., Campbell, S.R., Erlandsen, S.L. et al. (1991). Cloning and restriction

enzyme mapping of ribosomal DNA of Giardia duodenalis, Giardia ardeae and Giardia

muris. Mol Biochem Parasitol, 46: 275-284.
van Keulan, H., Feely, D., Macechko, M. et al. (1998). The sequence of Giardia small sub-
unit rRNA shows that voles and muskrats are parasitised by a unique species Giardia

microti. J Parasitol, 84: 294-300.
Wallis, P.M., Erlandsen, S.L., Isaac-Renton, J.L. et al. (1996). Prevalence of Giardia cysts

and Cryptosporidium oocysts and characterisation of Giardia spp. isolated from drink-
ing water in Canada. Appl Environ Microbiol, 62: 2789-2797.
Wang, A.L. and Wang, C.C. (1986). Discovery of a specific double-stranded RNA virus in

Giardia lamblia. Mol Chem Parasitol, 21: 269-276
Warburton, A.R.E., Jones, P.H. and Bruce, J. (1994). Zoonotic transmission of giardiasis:

a case control study. Commun Dis Rep, 4: R32-35.
Welch, T.P. (2000). Risk of giardiasis from consumption of wilderness water in North

America: a systematic review of epidemiological data. Int J Infect Dis, 4: 100-103.
Wolfe, M.S. (1990). Clinical symptoms and diagnosis by traditional methods. In

Giardiasis, Meyer, E.A. (ed.). New York: Elsevier Publishing Co, pp. 175-185.
Wright, S.G. and Tomkins, A.M. (1978). Quantitative histology in giardiasis. / Clin

Pathol, 31: 712-716.
Wright, S.G., Tomkins, A.M. and Ridley, D.S. (1977). Giardiasis: clinical and therapeutic

aspects. Gut, 18: 343-350.
Yardley, J.H., Yakano, J. and Hendrix, T.R. (1964). Epithelial and other mucosal lesions

of the jejunum in giardiasis: jejunal biopsy studies. Bull Johns Hopkins Hosp, 115:

389-406.
Yu, D., Wang, C.C. and Wang, A.L. (1995). Maturation of giardiavirus capsid protein

involves posttranslational proteolytic processing by a cysteine protease. / Virol, 69:

2825-2830.

318

Naegleria fowleri

Basic microbiology

Naegleria fowleri is an amoeboflagellate of the Phylum Sarcomastigophora,
Order Schizopyrenida, Family Vahlkampfiidae. This is a free-living protozoan,
found in aquatic and soil habitats, and the life cycle consists of a motile, feed-
ing trophozoite stage, non-feeding, non-replicating biflagellate stage and a
resistant cyst stage (Bottone, 1993). Trophozoites (10-15 fjim) excyst from the
cysts via pores in the cyst wall and are elongate in shape with rounded processes
called lobopodia. They feed off bacteria, including Escherichia coli. Trophozoites
differentiate to pear-shaped, motile biflagellates, which can revert to tropho-
zoites. The biflagellates encyst in adverse environmental conditions. The spher-
ical cysts are approximately 10 |xm in diameter, with a smooth double wall.
Infection in humans occurs following exposure to biflagellates in recreational
waters or cysts in soil or dust. Of the six species currently identified in the genus
Naegleria, N. fowleri is the primary human pathogen. Following inhalation of
biflagellates or cysts, trophozoites invade the nasopharyngeal mucosa, migrat-
ing through the olfactory nerve to invade the brain via the cribriform plate,
typically causing primary amoebic meningoencephalitis (PAM) which is invari-
ably fatal (Duma et al., 1969). Other species including N. australiensis may
also be pathogenic, but have not been identified to the same extent as N. fowleri
(Bottone, 1993).

Protozoa

Although cases of PAM are rare, they have occurred throughout the world
and are usually linked to swimming in contaminated water. The epidemiology
of PAM and transmission of N. fowleri is driven by its survival in the aquatic
environment and activity. Antibodies to Naegleria spp. are detected in human
sera (Cursons et al. 9 1980), indicating widespread exposure may occur. How-
ever, it is unknown why disease occurs when others undertaking similar activ-
ities are unaffected. Elucidation of risk factors for exposure and disease still
need clarification.

Origin of the organism

PAM was first recognized in Orange County, Florida, USA in 1962 (Butt, 1966)
since when over 250 cases have been reported world-wide.

Clinical features

PAM occurs in otherwise healthly, often active individuals, particularly young
adults and children (Duma etal. 9 1969). Symptoms occur within 7 days of expos-
ure and are indistinguishable from fulminant bacterial meningitis, and include
severe frontal headache, sore throat, high fever, anorexia, vomiting, signs of
meningeal inflammation, altered mental status, occasional olfactory hallucin-
ations, nucal rigidity, somnolence and coma (Lubor, 1981). Signs of brain
stem compression and seizures follow. Disease is characterized by inflamma-
tion of the olfactory bulbs, cerebral hemispheres, brain stem, posterior fossa
and spinal cord. Death typically occurs within 10 days of onset of symptoms,
typically on days 5 or 6.

Pathogenicity, virulence and causation

Infection occurs when the nasal mucosa is exposed to trophozoites following
swimming in contaminated warm waters. Trophozoites adhere to and pene-
trate the nasopharyngeal mucosa and migrate along the olfactory nerve.
N. fowleri exhibits rapid locomotion at 37°C, whereas non-pathogenic
species are more active at 28°C (Thong and Ferrante, 1986). Trophozoites
invade the brain through the cribiform plate and can reside in the subarach-
noid and perivascular spaces. Tissue invasion is achieved by the production of
cytopathic enzymes (Ferrante and Bates, 1988) and pathogenicity and virulence
are assisted by phagocytosis of host neutrophils, possibly by sucker-like struc-
tures on the parasite surface (John et al. 9 1984). Despite the abundance of

320

Naegleria fowleri

N. fowleri in the environment (see above), relatively few cases of PAM occur
and it is likely that a multitude of host defence mechanisms are effective.

Treatment

N. fowleri is sensitive to amphotericin B (Fungizone) but just a handful of sur-
vivors of PAM have been documented (Schmidt and Roberts, 1995). In these
cases there was early diagnosis and administration of intravenous and intrathe-
cal or intraventricular amphotericin B with intensive supportive care. One sur-
vivor received miconazole intravenously and intrathecally and rifampicin orally.

Survival in water and the environment

Although cysts have been identified in soil and dust, N. fowleri appears to pre-
fer warm moist environments, including warm water and soil (John, 1982). In
water, most trophozoites transform into free-living biflagellates. N. fowleri tol-
erates temperatures of 40-45 °C and has been isolated from bodies of temper-
ate and warm fresh water, either natural (such as lakes, rivers and hot springs) or
man-made bodies of water such as swimming pools, thermal effluent, sewage
sludge and drinking water (Martinez, 1993; Fewtrell etal., 1994). The organism
has also been detected in water-cooling circuits from power stations (Hurzianga
etaL, 1990).

Surveys of lakes in Florida have shown that 12/26 (46%) harbour patho-
genic Naegleria spp. (Wellings et aL, 1979) and that they are common in lake
bottom sediment and at the sediment/lake water interface (Wellings etal., 1977).
The source of the organism is probably soil, with heavy rain causing runoff
into bodies of water (Marciano-Cabral, 1988). Occurrence is also linked to
food and nutrient sources and large numbers have been detected where coli-
forms, filamentous cyanobacteria, eubacteria, increased iron and manganese
concentrations are present and there is possibly an interaction between Naegleria
spp. and Legionella spp. (Ma et al. 9 1990). While chlorination of pools and other
artificial waters inactivates Naegleria cysts and trophozoites, control measures
have yet to be established for natural waters (De Jonckheere et al., 1976).

Methods of detection

PAM is usually diagnosed at autopsy, but in suspected cases diagnosis is made
by examination of CSF which shows predominantly polymorphonuclear
leucocytosis, increased protein and decreased glucose concentrations, mimicking

321

Protozoa

bacterial meningitis (Bottone, 1993). Occasionally amoebae may be seen on
Gram-stained smears. For diagnosis during life, there must be a clinical suspi-
cion based on exposure history. If a previously healthy patient has swum in
warm, fresh water within 7 days of onset of symptoms and displays symptoms
of bacterial meningitis with predominantly basilar distribution of exudates
by head CT, N. fowleri infection should be suspected. Examination of
un-refrigerated CSF should be undertaken and care taken to examine further
any atypical mononuclear cells, which may actually be amoebae. Trans-
formation from the amoeboid to the biflagellate form can be induced within
1-20 hours to aid identification and is undertaken by dilution of 1 drop CSF
with 1 ml distilled water.

N. fowleri is readily cultured from environmental samples on non-nutrient
agar plates seeded with Escherichia coli using prior concentration by mem-
brane filtration (Anon, 1989). However, differentiation from other closely
related but non-pathogenic organisms is difficult, but essential for complete
risk assessment. Phenotypic characterization, including morphology, thermo-
tolerance, pathogenicity, detection of antigens by monoclonal antibodies and
isoenzyme electrophoretic profiles, and genotypic characterization by PCR/PFLP
have been used to differentiate the pathogenic N. fowleri from non-pathogenic
thermophilic species (Ma et al. 9 1990; Sparagano et al. 9 1993; Kilvington and
Beeching, 1995).

Epidemiology

N. fowleri is acquired through exposure of the nasal passages to contam-
inated water. Most cases of PAM have occurred in children or in young adults
who have been undertaking water-related activities, particularly during the
summer. Clustering of cases has occurred pointing to a common or single source
of exposure. For example, 16 cases were associated with a public swimming
pool in Czechoslovakia (Cerva and Novak, 1968). The pool was supplied by
river water. A similarly large outbreak of 16 cases occurred in Virginia, USA
among people who had swam in three lakes within a 5-mile radius (Duma et al. 9
1971). Although N. fowleri had been isolated from many of the numerous
lakes in Virginia, cases of illness were not associated with other lakes and the
reason for the clustering around these three lakes is unclear. An outbreak of
five fatal cases of PAM occurred in Mexico in August and September 1990
(Lares-Villa et al. 9 1993). All had swum in a canal from which N. fowleri was
isolated, showing identical isoenzyme patterns to all the cases. Single cases
also occur. For example, in the UK, a girl died of PAM after swimming in one
of the thermal pools at Bath Spa (Cain et al. 9 1981).

In Western Australia, cases have been associated with the reticulated water
supply. Here remote localities are supplied with water via over-ground steel

322

Naegleria fowleri

pipes in which solar heating can be substantial allowing proliferation of
N. fowleri and resulting in about 20 cases of PAM (Robinson et aL 9 1996).
N. fowleri was detected in the household water supply of one case, demon-
strating the link between drinking water and PAM and broadening the risk
factors from recreational activities such as swimming to include drinking water
(Marciano-Cabral, 1988).

Risk assessment

Health effects: occurrence of illness, degree of morbidity and mortality, prob-
ability of illness based on infection:

• Following inhalation of N. fowleri, trophozoites invade the brain through
the nasal passages, typically causing primary amoebic meningoencephalitis
(PAM), which is almost always fatal.

• Antibodies in blood suggest that more people are exposed to Naegleria
than become ill, but the reasons are unknown. Naegleria cysts are plentiful
in the environment, yet few cases occur annually.

• Symptoms occur within 7 days of exposure and are indistinguishable from
fulminant bacterial meningitis and include severe frontal headache, sore
throat, high fever, anorexia, vomiting, altered mental status, occasional olfac-
tory hallucinations, nucal rigidity, somnolence and coma. Death almost
always follows.

Exposure asssessment: routes of exposure and transmission, occurrence in
source water, environmental fate:

• Infection in humans occurs following inhalational exposure to biflagellates
in recreational waters or cysts in soil or dust. Most cases have been linked
to swimming in contaminated water, though two cases in the USA in 2002
might have been caused through ingestion or inhalation of unchlorinated
drinking water.

• N. fowleri tolerates temperatures of 40-45°C, and has been isolated from
bodies of temperate and warm fresh water, either natural (such as lakes,
rivers, hot springs) or man-made such as swimming pools, thermal effluent,
sewage sludge and drinking water.

• Surveys of lakes in Florida have shown that almost half contain pathogenic
Naegleria spp., which are common in lake bottom sediment.

Risk mitigation: drinking-water treatment, medical treatment:

• Chlorination inactivates Naegleria organisms.

• N. fowleri is sensitive to amphotericin B (Fungizone), but just a handful of
survivors of PAM have been documented, usually because diagnosis comes
too late.

323

Protozoa

References

Anon. (1989). Isolation and identification of Giardia cysts, Cryptosporidium oocysts and

freeliving pathogenic amoebae in water etc. In Methods for the Examination of Waters

and Associated Materials. London: HMSO.
Bottone, E.J. (1993). Free-living amoebas of the genera Acanthamoeba and Naegleria:

an overview and basic microbiologic correlates. Mount Sinai J Med, 60: 260-270.
Butt, C.G. (1966). Primary amoebic meningoencephalitis. New Engl J Med, 274:

1473-1476.
Cain, A.R.R., Wiley, P.F., Brownell, B. et al. (1981). Primary amoebic meningencephalitis.

Arch Dis Childh, 56: 140-143.
Cerva, L. and Novak, K. (1968). Amoebic meningoencephalitis: sixteen fatalities. Science,

160: 92.
Cursons, R.T.M., Brown, T.J., Keyes, E.A. et al. (1980). Immunity to pathogenic free-living

amoeba: role of humoral immunity. Infect Immun, 29: 401-407.
De Jonckheere, J. et al. (1976). Differences in the destruction of cysts of pathogenic and

non-pathogenic Naegleria and Acanthamoeba by chlorine. Appl Environ Microbiol, 31:

294-297.
Duma, R.J. et al. (1969). Primary amebic meningoencephalitis. New Engl J Med, 281:

1315-1323.
Duma, R.J., Shumaker, J.B. and Callicott, J.H. (1971). Primary amebic meningo-encephali-

tis: a survey in Virginia. Arch Environ Hlth, 23: 43-47.
Ferrante, A. and Bates, E.J. (1988). Elastase in pathogenic free-living amoebae Naegleria

and Acanthamoeba species. Infect Immun, 56: 3320-3321.
Fewtrell, L., Godfree, A.F., Jones, F. et al. (1994). Pathogenic Microorganisms in

Temperate Environmental Waters. Cardigan, Dyfed: Samara Publishing.
Hurzianga, H.W. and McLaughlin, G.L. (1990). Thermal ecology of Naegleria fowleri

from a power plant cooling reservoir. Appl Environ Microbiol, 56: 2200-2205.
John, D.T. (1982). Primary amebic encephalitis and the biology of Naegleria fowleri.

Ann Rev Microbiol, 36: 101-123.
John, D.T., Cole, T.B. and Merciano-Cabral, F.M. (1984). Sucker-like structures on the

pathogenic amoeba Naegleria fowleri. Appl Environ Microbiol, 47: 12-14.
Kilvington, S. and Beeching, J. (1995). Development of PCR for identification of Naegleria

fowleri from the environment. Appl Environ Microbiol, 61: 3764-3767.
Lares-Villa, F., de Jonckhere, J.F., de Moura, H. et al. (1993). Five cases of primary amebic

meningoencephalitis in Mexicali, Mexico: study of isolates. / Clin Microbiol, 31:

685-688.
Lubor, C. (1981). Amebic meningoencephalitis. In Medical Microbiology of Infectious

Diseases, Brause, A.I., Davies, C.E. and Fierer, J. (eds). Philadelphia, PA: WB Saunders Co,

pp. 1281-1284.
Ma, P., Visvesvara, G.S., Martinez, A.J. et al. (1990). Naegleria and Acanthamoeba infec-
tions. Rev Infect Dis, 12: 490-513.
Marciano-Cabral, F. (1988). Biology of Naegleria spp. Microbiol Rev, 52: 114-133.
Martinez, A.J. (1993). Free-living amebas: infection of the central nervous system. Mount

Sinai J Med, 60: 271-278.
Schmidt, G.D. and Roberts, L.S. (1995). Foundations of Parasitology. Chicago:

William C Brown.
Sparagano, O., Drouet, E., Brebant, R. et al. (1993). Use of monoclonal antibodies to dis-
tinguish pathogenic Naegleria fowleri (cysts, trophozoites, or flagellate forms) from

other Naegleria species./ Clin Microbiol, 31: 2758-2763.
Thong, Y.H. and Ferrante, A. (1986). Migration patterns of pathogenic and non-
pathogenic Naegleria species. Infect Immun, 51: 177-180.
Wellings, F.M., Amuso, P.T. and Chang, S.L. et al. (1977). Isolation and identification of

pathogenic Naegleria from Florida lakes. Appl Environl Microbiol, 34: 661-667.
Wellings, F.M. et al. (1980). Pathogenic Naegleria: Distribution in nature. EPA research

and development bulletin No. 600/1-79-018.

324

Toxoplasma gondii

Basic microbiology

Toxoplasma gondii is the only species in the genus (Phylum Apicomplexa,
Order Eucoccidiorida, Family Eimeriidae). This tissue-cyst forming coccidium
has a heterogeneous life cycle comprising an asexual phase in a variety of warm-
blooded intermediate hosts and a sexual phase in the intestines of carnivorous
definitive hosts (Frenkel, 1973). Felids are the only known definitive hosts and
are the main reservoirs of infection. Cats become infected by eating meat or
offal containing tissue cysts, which are lysed by digestive enzymes following
ingestion, releasing bradyzoites which invade the epithelial cells of the small
intestine (Munday, 1972). Asexual multiplication occurs, followed by the sexual
cycle resulting in the formation of immature oocysts which are shed in the
faeces. Cats can shed oocysts for 1-2 weeks in extremely large numbers, peak-
ing at over a million a day, even after the consumption of just one tissue
cyst (Dubey and Beattie, 1988). The oocysts mature and sporulate over 1-5
days under ambient conditions: sporulation does not usually occur below 4°C
or above 37°C. Mature oocysts measure about 12 X 11 |xm and are infective for
a wide range of warm-blooded intermediate hosts, including humans. When
oocysts are ingested, an asexual cycle is initiated forming crescent or bow-
shaped tachyzoites (2 |xm by 6 |xm) which localize to form tissue cysts (ranging

Protozoa

from<12|xm to over 100 |xm) in muscle and neural tissue, including the
myocardium and brain. The cysts may persist throughout the life of intermedi-
ate hosts.

Intermediate hosts are therefore infected by ingestion of oocysts in cat faeces or
by the consumption of tissue cysts (for example in meat) containing bradyzoites.
Human infection is widespread: in the USA and UK estimates vary between 16
and 40% of the population being infected, while in continental Europe, Central
and South America the figure ranges from 50 to 80% (Dubey and Beattie, 1988).
However, disease occurs more rarely and the sequelae depend on the clinical
status of the patient. If infection is acquired during pregnancy in the absence
of prior immunity, tachyzoites can infect the fetus via the placenta and cause
abortion, or congenital disease resulting in mental retardation or blindness in
the infant (Remington et aL 9 2001). Previously acquired latent infection can be
reactivated in immunocompromised patients resulting in severe disease, includ-
ing encephalitis (Luft and Remington, 1992). Many of the biological features of
T. gondii affect its transmission and epidemiology: robust, highly infectious
oocysts can be shed in large numbers by felids; they are infectious for many hosts
including man; humans can also acquire infection by the ingestion of tissue
cysts in raw or undercooked meat or tachyzoites in milk, via blood transfusion
or organ transplant, or transplacentally from an acutely infected mother.

Origin of the organism

The genus Toxoplasma was first proposed in 1909 by Nicolle and Manceaux fol-
lowing the identification of asexual stages of similar parasites in the tissues of
birds and mammals, and merozoites in the blood of north African rodents called
Gondi. Although several species were named, during the 1930s it was shown that
these were identical to the type species T. gondii. Identification of sexual stages by
electron microscopy during the 1960s provided evidence for the coccidian nature
of the parasite (Levine, 1997). During the 1960s and 1970s the heterogeneous
life cycle was elucidated by the discovery of sexual stages in the small intestine of
cats, which followed the induction of infection in intermediate hosts by inocula-
tion with cat faeces (Dubey and Beattie, 1988).

The first recorded case of human toxoplasmosis was recognized retrospectively
in an 11-month old infant with congential hydrocephalus and microphthalmia.
The clinical importance of X gondii infection was established during the 1930s
with the recognition of T. gondii as the aetiolgical agent of encephalomyelitis in
neonates, and the description of the classic triad of human congenital toxoplas-
mosis: retinochoroiditis, hydrocephalus and encephalitis with cerebral calcifica-
tion (Wolf et aL, 1939). Acutely acquired human disease and vertical transmission
in humans were both recognized in the 1940s when the classic tetrad of symptoms
of congential toxoplasmosis was described: retinochoroiditis, hydrocephalus or
microcephalus, cerebral calcification and psychomotor disturbances (Sabin et al.,

326

Toxoplasma gondii

1952). T. gondii was recognized as a causative agent of lymphadenopathy in the
early 1950s (Sinn, 1952). The major sequelae of congential toxoplasmosis and
recognition of risks to patients with malignancies were described in the 1960s
(Eichenwald, 1960; Vietzke et cd. 9 1968), with recognition of T. gondii as an
opportunistic pathogen in AIDS patients in the 1980s (Luft et al. 9 1984).

Clinical features

T. gondii infection is relatively common but is generally asymptomatic in approxi-
mately 85% of immunocompetent individuals, for whom there is no significant
health risk; 10-20% of acute infections may develop flu-like illness or cervical
lymphadenopathy, with symptoms resolving within a year. However, in immuno-
deficient patients, including HIV/AIDS patients, organ transplant recipients and
patients undergoing chemotherapy, central nervous system disease is common
and chorioretinitis or pneumonitis may develop. This is often due to reactivation
of chronic or latent infection during immunodeficit, the most significant being
toxoplasmic encephalitis due to reactiviation of bradyzoites in brain tissue
(Renold et al. 9 1992). Thus, T. gondii is a significant cause of morbidity and mor-
tality in the immunodeficient. Additionally, organ transplant poses a risk, particu-
larly from bradyzoites in infected heart, which may reactivate causing disease
with 50% mortality rate.

Infection can also have significant health effects if acquired during pregnancy.
Acute primary infection at this time can result in congenital toxoplasmosis and
can cause spontaneous abortion, particularly during the first trimester, and fetal
abnormalities of the brain, eyes and internal organs (Dunn, 1999). The spec-
trum of congenital infection is broad, ranging from slightly impaired vision to
retinochoroiditis, hydrocephalus, convulsions and intracerebral calcification,
ocular disease being most common (Desmonts and Couvreur, 1974; Remington
et al. 9 1995). Accurate diagnosis is essential since treatment of the mother can
reduce fetal infection (Desmonts and Courvreur, 1974). Infants with subclinical
infection at birth will develop signs or symptoms, frequently chorioretinitis, later
in life unless treated.

Pathogenicity and virulence

Infection with T. gondii is common in humans, but clinical disease is largely
restricted to risk groups. Although mild symptoms may occur in immunocom-
petent individuals, the most significant being lymphadenopathy, most infections
are asymptomatic and severe manifestations are rare. Mothers infected during
pregnancy have a temporary parasitaemia and congenital infections acquired

327

Protozoa

during the first trimester are usually more severe than those acquired later
(Desmonts and Courvreur, 1974; Remington et al. 9 1995). General infection of
the fetus occurs following the development of focal lesions in the placenta and
subsequently infection is cleared from visceral tissues to localize in the central
nervous system. Ocular disease is the most common manifestation of congeni-
tal toxoplasmosis. Some controversy has surrounded whether ocular toxo-
plasmosis developing in later life is a manifestation of prenatal infection or
recently acquired primary infection (Holland, 1999). However, multiple cases
and outbreaks have been documented with compelling evidence of recent acqui-
sition from a variety of sources.

Infection of immunosuppressed patients may occur in any organ, but enceph-
alitis is the most clinically important manifestation, the predominant lesion in
the brain being necrosis, usually of the thalamus (Renold et al. 9 1992). Initially
bilateral severe, persistent headaches develop, responding poorly to analgesics,
followed by confusion, lethargy and coma.

There is evidence of at least two clonal lineages within T. gondii, supported
by genetic analyses. Strains within one lineage are virulent for mice and show
vertical transmission while strains in the other lineage are avirulent in mice
(Johnson, 1997).

Causation

T. gondii is a widespread parasitic infection, usually causing no symptoms.
Infection with T. gondii results in life-long immunity, and if acquired before
conception presents no substantial risk of transmission to the fetus. The
exception is in women with systemic lupus erythematosus or AIDS. Risk of
congenital disease appears to be related to gestational age at infection: while
risk of transmission is highest in the third trimester, disease is most severe
if acquired in the first trimester (Hohlfeld et al. 9 1989; Dunn et al. 9 1999). Pre-
viously acquired latent infection can be reactivated in immunocompromised
individuals and is life-threatening, usually manifesting as encephalitits, although
any organ can become infected. It was estimated in the 1980s that 10% AIDS
patients in the USA and 30% in Europe died of toxoplasmosis (Dubey and
Beattie, 1988).

Treatment

Treatment of healthy immunocompetent individuals is not generally indi-
cated: neither is treatment of immunocompromised individuals unless symp-
toms are severe or persistent, when treatment is generally with pyrimethamine

328

Toxoplasma gondii

and sulphadiazine (Kasper, 1998). Treatment of pregnant women is problem-
atic. While treatment of acute primary infection of the mother during preg-
nancy can reduce the incidence of congenital infection by approximately half
(Desmonts and Couvreur, 1974), treatment depends on gestational age and
whether the fetus is known to be infected.

Survival in the environment

Tachyzoites are fragile outside a host and are sensitive to heat, while brady-
zoites can survive refrigeration and some may survive freezing, maintaining
infectivity after storage at — 5°C. However, they do succumb to heat and are
killed by heating to 67°C. Oocysts are shed unsporulated in felid faeces and
take 1-5 days to sporulate and thus become infective. They have been
detected in soil naturally contaminated with cat faeces (Ruiz et al. 9 1973) and
in soil from gardens (Coutinho et al. 9 1982) and can survive in soil for
18 months (Frenkel et al. 9 1975). Maternal contact with soil has been
epidemiologically linked to increased risk of congenital toxoplasmosis
(Decavalas et al. 9 1990). Oocysts can become distributed in the environment
by mechanical spread by invertebrates.

Survival in water

Survival of T. gondii oocysts is readily assessed by the mouse infectivity model
and has shown long-term maintenance of oocyst viability in survival water.
Cell culture models, while currently not as sensitive as the mouse model, offer
an alternative assay, particularly since there is evidence for at least two clonal
lineages within T. gondii, one of which is avirulent for mice (Johnson, 1997).
Few studies have been published on the survival of oocysts in water, but labora-
tory tests have shown they can survive for up to 4.5 years at 4°C (Dubey,
1998). Survival for several months has been demonstrated in water, as has
resistance to many disinfectants, freezing at — 10°C, and drying, but they are
killed by heating to 55-60°C (Kuticic and Wikerhauser, 1996).

Infection has been detected in aquatic mammals and may imply survival of
oocysts contaminating seawater (Cole et al. 9 2000).

Methods of detection

Diagnosis of toxoplasmosis in man is primarily by serology. The methylene
blue dye test for the detection of antibodies, introduced in 1948 (Sabin and

329

Protozoa

Feldman), is maintained as a gold standard for serology tests by reference
laboratories, but is labour-intensive and requires a continual supply of live
organisms. Since IgG can persist for decades, IgM, which typically persists for
6-9 months, is used as a marker of recent infection, although IgM antibodies
have been detected for up to 18 months (Wilson and McAuley, 1999). Other
immunological methods include complement fixation tests, direct agglutin-
ation tests, ELISA, enzyme-linked fluorescent assay, indirect haemagglutination
tests, immunosorbent agglutination test and latex agglutination test. Diag-
nosis in critical clinical cases (pregnant women, HIV/AIDS patients, neonates
etc.) requires specialist testing including enhanced IgA/IgM detection, meas-
urement of IgG avidity and direct detection by PCR, undertaken by reference
laboratories.

Although cats may shed high numbers of oocysts for a limited period, concen-
tration methods using high-density sucrose solution may be required and oocysts
should be definitively recognized following sporulation and bioassay in mice
(Dubey and Beattie, 1988). For epidemiological surveys, however, oocyst detec-
tion is impractical and serological prevalence is a better marker of exposure to
T. gondii. Similarly, the detection of tissue cysts in meat animals is difficult since
the numbers present are low and may be as few as 1/100 g meat. Digestion of the
sample to rupture the cyst wall and release hundreds of bradyzoites prior to
bioassay in mice or application of PCR to detect DNA has been used to assess
T. gondii in meat samples (Dubey, 1988; Jauregue et al., 2001).

Direct detection of oocysts in environmental samples is possible but ham-
pered by the probable low density of oocysts within the sample. A method for
the detection of T. gondii in filtered water retentate using a rodent model has
been validated experimentally by Isaac-Ren ton and colleagues (1998), although
failed to detect T. gondii in water samples following an outbreak in Canada.

Epidemiology

Horizontal transmission via tissue cysts in pork to man was hypothesized in the
1950s (Weinman and Chandler, 1956) and traditionally the consumption of
raw or undercooked meat has been regarded as the main horizontal route of
acquisition by humans, although the importance could vary with dietary habits.
However, the prevalence of infection in meat-producing animals has decreased
and other routes of infection should be considered, particularly environmental
routes, given the robust nature of the oocyst stage. Although the epidemiological
role of cats in human infection was recognized in the late 1960s, the epidemio-
logical importance of different routes of infection has been little evaluated and
it is currently not possible to discriminate between infection caused by oocyst
ingestion and that caused by tissue cyst ingestion.

Potential routes of transmission of oocysts to humans include direct contact
with infected cat faeces, indirect contact following faecal contamination by cats

330

Toxoplasma gondii

of water sources, vegetables, from unwashed hands, contact with contaminated
soil, for example by children during play or by adults during gardening, cat lit-
ter trays or sand pits used as cat latrines. Seroprevalence in wild scavengers,
such as racoons which in the USA was 60% (Dubey and Odening, 2001), indi-
cates widespread occurrence of oocysts in the environment. Vehicles of tissue
cyst infection include the consumption of raw or undercooked meat. T. gondii
is common in many food animals, including sheep, pigs and rabbits. Outbreaks
of acute toxoplasmosis in humans have been associated with the consumption
of undercooked pork. In one outbreak in Korea, three adults ate a meal of
uncooked wild pig spleen and liver and in a second Korean outbreak five sol-
diers were infected following a meal of raw pig liver (Choi et al. 9 1997).

Human seroprevalence studies have shown large geographical variation
both between and within countries and among different ethnic groups within
the same geographic area. However, methods have not always been standard-
ized and may account for some variation. Broad estimates of seroprevalence in
women of child-bearing age with no obstetric history have shown a range of
37-58% in Central Europe, Eastern Europe, Australia and Northern Africa;
51-77% in Latin-American and West African countries; 4-39% in Southeast
Asia, China and Korea and 11-28% in Scandinavian countries (Tenter et al. 9
2000). Although mass antenatal screening is undertaken in a limited number of
countries, including Canada, Austria and France, elsewhere such screening is
not undertaken due to issues of cost-benefit analysis, and reduction in congeni-
tal toxoplasmosis and surveillance may rely on reports of congenital disease or
acute symptoms (Peyron et al. 9 2002).

Community outbreaks of toxoplasmosis are infrequently recognized and very
few waterborne outbreaks have been reported. In those that have been docu-
mented, the association has been made epidemiologically, since accepted micro-
biological methods for the detection of Toxoplasma oocysts in water are lacking.
In addition, it is likely that the time periods involved in the detection of clinical
cases and the probable episodic nature of the occurrence of oocysts in water
would render water testing unreliable for this purpose. The first waterborne
outbreak, defined by descriptive epidemiology, was among US army personnel
who drank untreated water from a stream in Panama, suspected to have been
contaminated with jaguar faeces (Beneson et al. 9 1982). The first outbreak to be
associated with an unfiltered surface municipal water supply occurred in British
Colombia, Canada over a 9-month period in 1995, during which 110 acute
cases were identified, including 42 pregnant women and 11 neonates identified
through an antenatal screening programme (Bowie et al. 9 1997). Statistical asso-
ciations were made between acute infection and residence in one water supply
distribution zone, with no other conventional source of Toxoplasma implicated.
The suspected cause of the outbreak was the contamination by domestic, feral
or wild cats (including cougar) of the feeder streams to the drinking water reser-
voir (Isaac-Renton et al. 9 1998; Aramini et al. 9 1999). Serological studies of the
riparian environment demonstrated that T. gondii cycles endemically among the
wild and domestic animals of the watershed (Aramini et al. 9 1999).

331

Protozoa

Other evidence for a waterborne route has been suggested by Hall et al.
(1999) following a case of congenital toxoplasmosis in a Jain Hindu. Serological
screening of 251 women in India showed no significant difference in seropreva-
lence among Jians, vegetarian Hindus and non-vegetarian Hindus. Although
the numbers were small, the sample suggests that Jain Hindus are exposed to
Toxoplasma. However, Jains strictly follow dietary rules where fruit and leafy
vegetables are thoroughly washed to remove insects, root vegetables are rarely
eaten to prevent introduction of soil to the kitchen, milk is boiled and farming
or keeping cats is not undertaken and it was suggested that exposure might be
from drinking water.

In a toxoplasmosis-endemic area of Brazil, seropositivity was related to
socioeconomic group with 84%, 62% and 23% of subjects in the lower, middle
and upper socioeconomic groups seropositive respectively (Bahia-Oliveira et al.,
2003). In multivariate analyses, drinking unfiltered water increased risk of
seropositivity in the lower and middle socioeconomic groups.

Risk assessment

Advice on the prevention of congenital toxoplasmosis issued to pregnant women
includes handwashing after gardening, contact with soil, uncooked meat and
cat faeces or litter trays. Interestingly, epidemiological studies have shown a lack
of association between seropositivity for T. gondii and cat ownership (Stray-
Pedersen and Lorentzen-Styr, 1980; Wallace et al., 1993; Kapperud et al., 1996;
Cook et al., 2000; Bahia-Oliveira et al., 2003), possibly because cats often defe-
cate away from home. One difficulty, however, is that IgG antibodies to T. gondii
are long-lived and may reflect past rather than the more recent exposures recalled
in questionnaires. IgM persists for 6-9 months and is generally used as an indi-
cator of recent or acute infection (Wilson and McAuley, 1999).

The detection of waterborne outbreaks of toxoplasmosis is more difficult
than, say, Cryptosporidium, since the symptoms of acute cryptosporidiosis are
more likely to be reported. Where antenatal screening is in place, or special stud-
ies have been undertaken, links have been made with drinking water and either
seropositivity or symptoms (Bowie et al., 1997; Bahia-Oliveira et al., 2003).
However, the risk from waterborne infection is generally low where the num-
bers of cats, particularly wild cats, are not large, since only cats shed environ-
mentally robust oocysts. Thus in some countries the potential for felid faeces to
enter the water systems is lower than in others. Tissue cysts may enter water
through degrading or rotting animal carcasses but their survival is unknown.
Waterpipes should be covered during repair and renovation to prevent host ani-
mals entering and flushed prior to entering service.

Survival of oocysts and thus risk of environmental transmission will depend
on climatic conditions since oocysts are susceptible to freezing and desiccation
(Kuticic and Wikerhauser, 1996) and survival is enhanced by a mild, damp

332

Toxoplasma gondii

climate. It is likely that full water treatment (coagulation and filtration) will
remove T. gondii oocysts, although oocysts are probably resistant to chlorine as
used to treat potable water supplies.

Overall risk assessment

Health effects: occurrence of illness, degree of morbidity and mortality, prob-
ability of illness based on infection:

• Prevalence data show that toxoplasmosis is one of the most common human
infections world-wide. In the USA and UK, estimates of infection vary
between 16% and 40% of the population, while in continental Europe and
Central and South America, the figure ranges from 50% to 80%. Infection
is more common in warm climates and at lower altitudes than in cold, high-
altitude regions.

• Toxoplasma infection in immunocompetent persons is generally asymptom-
atic. However, 10-20% of patients may develop cervical lymphadenopathy
or a flu-like illness. Infections are self-limited, and symptoms usually resolve
within a few months to a year.

• Previously acquired latent infection can be reactivated in immunocomprom-
ised patients resulting in severe central nervous system disease including
encephalitis, although any organ can become infected. Immunocompromised
patients are also at greater risk of disease from newly acquired infections.

• If infection is acquired during pregnancy, the fetus can become infected caus-
ing abortion or congenital disease resulting in severe effects such as mental
retardation or blindness in the infant. Most infants who are infected while in
the womb have no symptoms at birth but may develop symptoms later in life
if left untreated. Only a small percentage of infected newborns have serious
eye or brain damage at birth. Though a fetus is more likely to become infected
during the third trimester, the outcome is usually more severe in infections
acquired during the first trimester. An estimated one-half of untreated mater-
nal infections are transmitted to the fetus and an estimated 400-4000 cases of
congenital toxoplasmosis occur each year in the USA.

Exposure assessment: routes of exposure and transmission, occurrence in
source water, environmental fate:

• Cats are the only known definitive hosts. Intermediate hosts, including humans,
are therefore infected by ingesting oocysts from faecally-contaminated food,
hands, or water; from organ transplantation or blood transfusion (rarely);
or from transplacental transmission.

• Very few waterborne outbreaks have been reported. The risk from water-
borne infection is generally low where the numbers of cats, particularly wild
cats, are not large, since only cats shed environmentally robust oocysts. Thus,
in some countries the potential for felid faeces to enter the water systems is

333

Protozoa

lower than in others. There is the potential for T. gondii oocysts to enter sur-
face water through wastewater effluent discharge if people flush cat faeces
into the sewage system.

• Oocysts can survive in the environment for several months and are remark-
ably resistant to disinfectants, freezing and drying, but are killed by heating
to 70°C for 10 minutes. Laboratory tests have shown they can survive for up
to 4.5 years at 4°C.

Risk mitigation: drinking-water treatment, medical treatment:

• T. gondii oocysts are resistant to chlorine, though the process of coagula-
tion and flocculation will probably remove oocysts.

• Medical treatment is not needed for a healthy person who is not pregnant.
If symptoms exist, they will usually resolve within a few weeks. Treatment
with pyrimethamine plus sulphadiazine may be recommended for pregnant
women or persons who have weakened immune systems.

Future implications

It is likely that further studies will identify waterborne risks in areas of high
endemnicity of Toxoplasma infection. Other routes of contamination may be
identified, such as the practice of emptying cat litter trays into domestic toilets
and thus introducing oocysts to the sewage system. Work on Giardia and
Cryptosporidium has shown that variable removal of oocysts occurs during
sewage treatment (Robertson et al., 2000) and thus T. gondii oocysts could
enter surface water through effluent discharge.

References

Aramini, J.J., Stephen, C, Dubey, J.P. et al. (1999). Potential contamination of drinking
water with Toxoplasma gondii oocysts. Epidemiol Infect, 122: 305-315.

Bahia-Oliveira, L.M.G., Jones, J.L., Azevedo-Sila, J. et al. (2003). Highly endemic, water-
borne toxoplasmosis in North Rio de Janeiro State, Brazil. Emerging Infect Dis,
in press.

Beneson, M.W., Takafuji, E.T., Lemon, S.M. et al. (1982). Oocyst-transmitted toxoplas-
mosis associated with ingestion of contaminated water. New Engl J Med, 300:
694-699.

Bowie, W.R., King, S.A., Werker, D.H. et al. (1997). Outbreak of toxoplasmosis associated
with municipal drinking water. Lancet, 350: 173-177.

Choi, W.Y., Nam, H.W., Kwak, N.H. et al. (1997). Foodborne outbreaks of human toxo-
plasmosis. / Infect Dis, 175: 1280-1282.

Cole, R.A., Lindsay, D.S., Howe, D.K. et al. (2000). Biological and molecular character-
isations of Toxoplasma gondii strains obtained from southern sea otters (Enhydra lutris
nereis).] Parasitol, 86: 526-530.

334

Toxoplasma gondii

Cook, A.J., Gilbert, R.E., Buffolano, W. et al. (2000). Source of Toxoplasma infection in

pregnant women: European multicentre case-control study. European Research net-
work on congenital toxoplasmosis. Br Med J, 321: 142-147.
Coutinho, S.G., Lobo, R. and Dutra, G. (1982). Isolation of Toxoplasma from the

soil during an outbreak of toxoplasmosis in a rural area of Brazil. / Parasitol, 68:

866-868.
Decavalas, G., Papapetropoulou, M., Giannoulaki, E. et al. (1990). Prevalence of

Toxoplasma gondii antibodies in gravidas and recently aborted women and study of

risk factors. Eur J Epidemiol, 6: 223-226.
Desmonts, G. and Couvreur, J. (1974). Congenital toxoplasmosis. A prospective study

of 378 pregnancies. New Engl J Med, 290: 1110-1116.
Dubey, J.P. (1988). Refinement of pepsin digestion method for isolation of Toxoplasma

gondii from infected tissues. Vet Parasitol, 74: 75-77.
Dubey, J.P. (1998). Toxoplasma gondii oocyst survival under defined temperatures.

/ Parasitol, 84: 862-865.
Dubey, J.P. and Beattie, C.P. (1988). Toxoplasmosis of Animals and Man. Boca Raton:

CRC Press.
Dubey, J.P. and Odening, K. (2001). Toxoplasmosis and related infections. In Parasitic

Diseases of Wild Animals, Samuel, W.M., Pybus, M.J. and Kocan, A.A. (eds). Ames:

Iowa State University Press, pp. 478-5190.
Dunn, D., Wallon, M., Peyron, F. et al. (1999). Mother to child transmission of toxoplas-
mosis: risk estimates for clinical counselling. Lancet, 353: 1829-1833.
Eichenwald, H.A. (1960). A study of congenital toxoplasmosis. In Human Toxoplasmosis,

Siim, J.C. (ed.). Copenhagen: Ejnar Munksgaard Forlag, pp. 41-49.
Frenkel, J.K. (1973). Toxoplasmosis: parasite life cycle, pathology and immunology. In The

coccidian. Eimeria, Isospora, Toxoplasma and related genera, Hammond, D. and Lond,

P.L. (eds). Baltimore: University Park Press, pp. 343-410.
Frenkel, J.K., Ruiz, A. and Chinchilla, M. (1975). Soil survival of Toxoplasma oocysts in

Kansas and Costa Rica. Am J Trop Med Hyg, 24: 439-443.
Hohlfeld, P., Daffos, F., Thulliez, P. et al. (1989). Fetal toxoplasmosis: outcome of preg-
nancy and infant follow-up after in utero treatment./ Pediatr, 115: 765-769.
Holland, G.N. (1999). Reconsidering the pathogenesis of ocular toxoplasmosis. Am J

Ophthalmol, 128: 502-505.
Isaac-Renton, J., Bowie, W.R., King, A. et al. (1998). Detection of Toxoplasma gondii

oocysts in drinking water. Appl Environ Microbiol, 64: 2278-2280.
Jauregue, L.H., Higgins, J.A., Zarlenga, D.S. et al. (2001). Development of a real-time PCR

assay for the detection of Toxoplasma gondii in pig and mouse tissues./ Clin Microbiol,

39:2065-2071.
Johnson, A.M. (1997). Speculation on possible life cycles for the clonal lineages in the

genus Toxoplasma. Parasitol Today, 13: 393-397.
Kapperud, G., Jenum, P.A., Stray-Pedersen, B. et al. (1996). Risk factors for Toxoplasma

gondii infection in pregnancy. Results of a prospective case-control study in Norway.

Am J Epidemiol, 144: 405-412.
Kasper, L.H. (1998). Toxoplasma infection. In Harrison's Principles of Internal Medicine,

14th edn, Fauci, A.S., Isselbacher, K.J. and Wilson, J. (eds). New York: McGraw-Hill,

Health Professions Division, pp. 1535-1539.
Kuticic, V. and Wikerhauser, T. (1996). Studies of the effect of various treatments on the

viability of Toxoplasma gondii tissue cysts and oocysts. In Toxoplasma gondii,

Gross, U. (ed.). Berlin: Springer- Verlag, pp. 261-265.
Levine, N.D. (1977). Taxonomy of Toxoplasma. J Parasitol, 24: 36-41.
Luft, B.J., Brooks, R.G., Conley, F.K. et al. (1984). Toxoplasmic encephalitis in patients

with acquired immune deficiency syndrome. JAMA, 252: 913-917.
Luft, B.J. and Remington, J.S. (1992). Toxoplasmic encephalitits in AIDS. Clin Infect Dis,

15:211-222.
Nicolle, C. and Manceaux, L. (1909). Sur un protozoaire nouveau du gondii. CR Hebd

Seanes Acad Sci, 148: 369-372.

335

Protozoa

Remington, J.S., McLeod, R. and Desmonts, G. (2001). Toxoplasmosis. In Infectious
Diseases of the Fetus and Newborn, 5th edn, Remington, J.S. and Klein, J.O. (eds).
Philadelphia: WB Saunders, pp. 205-346.

Renold, C, Sugar, A., Chave, J.P. et al. (1992). Toxoplasma encephalitis in patients with
the acquired immunodeficiency syndrome. Medicine, 71: 224-239.

Robertson, L.J., Paton, C.A., Campbell, A.T. etal. (2000). Giardia cysts and Cryptosporidium
oocysts at sewage treatment works in Scotland, UK. Water Res, 34: 2310-2322.

Ruiz, A., Frenkel, J.K. and Cerdas, L. (1973). Isolation of Toxoplasma from soil.
/ Parasitol, 59: 204-206.

Sabin, A.B. and Feldman, H.A. (1948). Dyes as microchemical indicators of a new immun-
ity phenomenon affecting a protozoan parasite (Toxoplasma). Science, 108: 660-663.

Sabin, A.B., Eichenwald, H., Feldman, H.A. et al. (1952). Present status of clinical mani-
festations of toxoplasmosis in man: indications and provisions for routine serologic
diagnosis. JAMA, 150: 1063-1069.

Siin, J.C. (1952). Studies on acquired toxoplasmosis. II. Report of a case with pathologio-
cal changes in lymph node removed at biopsy. Acta Pathol Microbiol Scand, 30:
104-108.

Stray-Pedersen, B. and Lorentzen-Styr, A.M. (1980). Epidemiological aspects of Toxoplasma
infections among women in Norway. Acta Obstet Gynecol Scand, 59: 323-326.

Tenter, A.M., Heckeroth, A.R. and Weiss, L.M. (2000). Toxoplasma gondii: from animals
to humans. Int J Parasitol, 30: 1217-1258.

Vietzke, W.M., Gelderman, A.H., Grimley, P.M. et al. (1968). Toxoplasmosis complicating
malignancy: experience at the National Cancer Institute. Cancer, 21: 816-827.

Wallace, M.R., Rossetti, R.J. and Olson, P.E. (1993). Cats and toxoplasmosis risk in HIV-
infected adults. JAMA, 269: 76-77.

Weinman, D. and Chandler, A.H. (1956). Toxoplasmosis and man and swine - an investi-
gation of the possible relationship. JAM A, 161: 229-232.

Wilson, M. and McAuley, J.B. (1999). Toxoplasma. In Manual of Clinical Microbiology,
7th edn, Murray, P. (ed.). Washington, DC: ASM Press, pp. 1374-1451.

Wolf, A., Cowen, D. and Paige, B.H. (1939). Toxoplasmic encephalomyelitis. III. A new
case of granulomatous encephalitis due to a protozoan. Am J Pathol, 15: 657-694.

336

Part 4

Viruses

Common themes

Introduction

The challenge of proving that outbreaks or sporadic instances of disease are
attributable to waterborne viruses is much greater than that associated with
bacterial or protozoan disease. Viruses are more difficult to detect in all aquatic
matrices, they are often associated with non-specific infections and epidemi-
ology is usually difficult because the triviality of the majority of virus infections
means that very few are reported to medical authorities and therefore tracing
causes of outbreaks is a difficult task.

Nevertheless, evidence has accumulated over the last 30 years which links
the ingestion of water contaminated by faecally-derived viruses with disease
and the numbers of viruses for which this evidence is forthcoming is increas-
ing as technology for their detection improves. Very early studies done in the
1940s as polioviruses were beginning to be propagated in cell culture indicated
that the virus could survive in sewage and be transmitted to receiving waters,
from where it could be recovered in an infectious state (Melnick, 1947). The
notion, since largely refuted, that swimming in domestic pools during the sum-
mer led to outbreaks of poliomyelitis was held strongly in the USA for some
years since the virus was on occasion recovered from domestic swimming
pools where chlorination was defective. The enormous outbreak of infectious

Viruses

hepatitis, now recognized as caused by hepatitis E virus (HEV) in Delhi in
1955/56 demonstrated the potential for infection of large numbers of people
when sanitary conditions are poor.

Despite waterborne viruses being enteric, that is to say they establish their
initial infection in the gastrointestinal tract, conclusive links between polluted
water and viruses causing specific instances of enteric disease have been diffi-
cult to demonstrate, mainly due to the problems in detecting viruses in aquatic
matrices. Many groups of viruses inhabit the gastrointestinal tract of humans
and animals, only some of which cause sufficient local damage for gastroen-
teritis or other illnesses to result. Primarily the faecal-oral route transmits viruses
adapted to replicate in the gut, but epidemiological patterns vary. Although they
may be present in the same water, enteric viruses have a range of important
attributes that distinguish them from other enteric microorganisms. The infec-
tious dose is low; it has been accepted for many years that a single clump, or
plaque forming unit (pfu), of virus may be capable of initiating a viral infec-
tion. Indeed, it was on that premise that Berg (1967) stipulated that any virus
in water would constitute a hazard. It is probable that this influential state-
ment was the basis for the enterovirus parameter of the EU Bathing Water
Directive, 1976. In common with the majority of virus families under normal
circumstances, the enteric viruses are host specific, so animal enteric viruses
do not usually infect humans. Viruses are the smallest of the microorganisms
and the most likely to be able to travel through the pore spaces within rocks
(Wellings e* */., 1975).

The principal areas for consideration of waterborne viral disease are the
nature of outbreaks, the aetiology of the different viruses causing waterborne
infection, the methods for detecting waterborne viruses and the survival of the
agents in the aquatic environment. Predictions of the risk of contracting viral
disease through ingestion of polluted water is becoming a reality as confidence
in detection techniques improves and future trends in water virology will
undoubtedly focus on acquisition of data for risk assessment models. Readily
identified viral outbreaks are those of gastroenteritis as the incubation period is
a matter of days and the symptoms well defined; noroviruses are currently the
most commonly recognized cause. Viral hepatitis outbreaks are recognizable due
to the severity of the symptoms, although the incubation period of 1-2 months
makes investigation difficult; hepatitis A (HAV) and HEV outbreaks are well
known. Outbreaks due to specific viruses are discussed in the section on each
virus but general issues are addressed below.

Viral waterborne disease may be transmitted by consumption of drinking
water, by immersion in recreational water or by contact through skin or inhal-
ation if contamination of the water with human sewage has occurred. Pollution
by animal faecal material is unlikely to pose a risk of transmission of animal
viruses. The health effects caused by waterborne pathogens generally have been
reviewed extensively, e.g. by Galbraith (1987) and by Stanwell-Smith (1994)
who documented drinking water-associated disease outbreaks in the UK, and
Hunter (1997), Dadswell (1993) and Pruss (1998) who considered disease
associated with recreational water contact. A system for attaching strong or

340

Common themes

possible association between disease and exposure to water was developed by
Tillett etal. (1998) and is based on the epidemiological picture, laboratory diag-
nosis and water quality.

Drinking water-associated outbreaks of disease are usually the result of one
of four events:

• inadequate removal of organisms during treatment

• failure in the treatment process

• failure in the chlorination or other disinfection system

• breaks in the integrity of the distribution infrastructure or of sewage removal.

These events may occur alone or in concert, for example in Bramham, Yorkshire
in 1980 (Short, 1988) a large outbreak of gastroenteritis, involving over 3000
people, occurred when sewage from a broken pipe contaminated drinking water
supply pipes. At the time of the breakage the chlorination plant was faulty, so
drinking water contaminated with sewage was distributed without adequate
disinfection.

Viruses have been identified as responsible for a number of drinking water-
associated disease outbreaks world-wide, though much less frequently than
bacteria or protozoa. Though most transmission is by person to person,
noroviruses (members of the family Caliciviridae) are the most widely impli-
cated in the context of waterborne viral disease, though the evidence has often
been circumstantial. Greater confidence in the association between water expos-
ure and disease is generated when the same organism is detected in patients
and in the water to which they have been exposed and there is a temporal con-
sistency in the detection.

It has been even more problematic to link the use of recreational water with
a specific virological cause of illness than drinking water. The most readily iden-
tifiable risks are those associated with swimming pools, spa-pools in leisure
centres and domestic whirlpool baths. Outbreaks of adenovirus conjunctivitis
have been associated with swimming pool exposure (Martone et aL 9 1980;
Turner et al., 1987). Swimming pools and the spread of poliovirus was of
great concern in the USA during the 1940s and 1950s. It is not clear whether
transmission did in fact occur via the water, although the potential for spread
is illustrated by an outbreak of echovirus 30 in Northern Ireland, where an
outdoor pool was contaminated with vomit on its day of opening, followed by
an outbreak of headache, diarrhoea and vomiting in swimmers (Kee etal., 1994).
Chlorine levels were said to be satisfactory and no other pathogens were detected
but it was not stated if faeces were tested for the presence of noroviruses.

Illness after recreational contact in sea water is particularly difficult to confirm
and quantify. Cohort studies on bathers have involved recruitment of individu-
als or by using individuals already on a section of beach. It was shown in the UK
prospective studies (Pike, 1991a,b, 1994) that an increase in mild gastro-
enteritis, eye and ear symptoms occurred after swimming in microbiologically
poor seawater. No virus or other microorganism was shown to be associated
with any of these symptoms. These broad conclusions have been confirmed in
studies all over the world. The quality of the water is most closely linked with

341

Viruses

health effects if enterococci are used as indicators of faecal pollution (Cabelli
et al 9 1983; Fleisher et al 9 1993; Kay et al. 9 1994). Although sewage is the most
likely source of faecal contamination the bathers are also potential polluters
(Fattal e* */., 1991).

Freshwater recreation has been more closely linked with waterborne illness
than contact with seawater. Canoeists using the River Trent developed gastro-
enteritis caused by noroviruses (Gray et al. 9 1997) and undiagnosed gastroen-
teritis was found to be more common in canoeists using the River Trent than
those using unpolluted lakes in North Wales (Fewtrell et al. 9 1992). Studies in
The Netherlands indicate that illness and norovirus infection are linked to
swimming in river water (van Olphen et al. 9 1991; Medema et al. 9 1995; de Roda
Husman, personal communication).

River water used for irrigation has contaminated crops and transmitted dis-
ease. In the UK, the most likely viral pathogen would be the noroviruses.
Regulations to limit the use of such water to particular crops and certain periods
should reduce the small risk of transmission of these pathogens. Use of digested
sewage sludge as a fertilizer in agriculture is currently an important issue.
Retailers, such as the large supermarkets, are concerned that crops may be at
risk of contamination, although no evidence is currently available that condi-
tions in the UK have resulted in consumer disease. Research is now underway
to improve the detection methods for enteric viruses, Salmonella, E. coli 0157
and protozoa. The digestion processes used in sewage treatment are being more
precisely characterized and the regulations governing the use of sewage sludge
have been reviewed. The ADAS matrix (1999) has recently been drawn up that
more closely defines the use of sewage sludge.

Bivalve shellfish, such as oysters, cockles and mussels are grown commercially
in estuaries and in shallow seawater all around the coast of Britain. They are
filter feeders of particulate matter that may include faecal material and conse-
quently viruses. Norovirus has been identified in shellfish tissue (Lees et al. 9
1995), has been detected in stools of individuals with gastroenteritis who have
eaten shellfish and has been strongly linked in epidemiological studies to out-
breaks of gastroenteritis (Advisory Committee on the Microbiological Safety
of Food, 1998). Oysters are the most frequent cause of outbreaks as they are
often eaten raw or poorly cooked. Between 1992 and 1996, 81 HAV infec-
tions transmitted by shellfish were reported to CDSC. An astrovirus outbreak
was reported at a workshop on food-borne viral infections in 1987 (Kurtz and
Lee, 1987) to be associated with shellfish consumption.

References

Advisory Committee on the Microbiological Safety of Food. (1998). Report on foodborne

viral infections. London: HMSO.
ADAS (1999). The safe sludge matrix. Guidelines for the application of sewage sludge to

agricultural land. London: Agricultural Development and Advisory Service (Ministry

Agriculture Food and Fisheries).

342

Common themes

Berg, G. (1967). Transmission of Viruses by the Water Route. New York: Wiley InterScience

Publishers, pp. 1-484.
Cabelli, V.J., Dufour, A.P., McCabe, L.J. et al. (1982). Swimming associated gastro -enteritis

and water quality. Am J Epidemiol, 115: 606-616.
Dadswell, J.V. (1993). Microbiological quality of coastal waters and its health effects. Int

J Environ Hltb Res, 3: 32-46.
Fattal, B., Peleg-Olevsky, E. and Cabelli, V.J. (1991). Bathers as a possible source of con-
tamination for swimming-associated illness at marine bathing beaches. Int J Environ

Hltb Res, 1:204-214.
Fewtrell, L., Godfree, A.F., Jones, F. et al. (1992). Health effects of white-water canoeing.

Lancet, 339: 1587-1589.
Fleisher, J.M., Jones, F., Kay, D. et al. (1993). Water and non-water related risk factors for

gastro -enteritis among bathers exposed to sewage contaminated marine waters. Int J

Epidemiol, 22: 698-708.
Galbraith, N.S. (1987). Water and disease after Croydon: a review of waterborne and

water associated disease in the UK 1937-86./ Instit Water Environ Manag, 1: 7-21.
Gray, J.J., Green, J., Gallimore, C. et al. (1997). Mixed genotype SRSV infections among a

party of canoeists exposed to contaminated recreational water./ Med Virol, 52: 425-429.
Hunter, P.R. (1997). Waterborne Disease: Epidemiology and Ecology. Chichester: John

Wiley &C Sons.
Kay, D., Fleisher, J.M., Salmon, R.L. et al. (1994). Predicting likelihood of gastroenteritis

from sea bathing: results from randomised exposure. Lancet, 344: 905-909.
Kee, F., McElroy, G., Sewart, D. et al. (1994). A community outbreak of echovirus infec-
tion associated with an outdoor swimming pool. / Public Hltb, 16: 145-148.
Kurtz, J.B. and Lee, T.W. (1987). Astroviruses: human and animal. In Novel Diarrhoea

Viruses. Ciba foundation symposium 128, Bock, G. and Whelan, J. (eds). London:

J. and A. Churchill, pp. 92-107.
Lees, D.N., Hensilwood, K., Green, J. et al. (1995). Detection of small round structured

viruses in shellfish by reverse transcription-PCR. Appl Environ Microbiol, 61: 4418-4424.
Martone, W.J., Hierholzer, J.C., Keenlyside, R.A. et al. (1980). An outbreak of adenovirus

type 3 disease at a private recreation center swimming pool. Am J Epidemiol, 111: 229-237.
Medema, G.J., van Asperen, I.A., Kokman-Houweling, J.M. et al. (1995). The relationship

between health effects in triathletes and microbiological quality of freshwater. Water Sci

Technol, 31: 19-26.
Melnick, J.L. (1947). Poliomyelitis virus in urban sewage in epidemic and in non-epidemic

times. Am J Hyg, 45: 240-253.
Pike, E.B. (1991a). Health effects of sea bathing Phase I - Pilot studies at Langland Bay

1989. Medmenham: WRc.
Pike, E.B. (1991b). Health effects of sea bathing Phase II - Studies at Ramsgate and

Moreton. Medmenham: WRc.
Pike, E.B. (1994). Health effects of sea bathing (WM 9021) Phase HI - Final report to the

Department of the Environment. Medmenham: WRc.
Pruss, A. (1998). Review of epidemiological studies on health effects from exposure to

recreational water. Int Epidemiol Assoc, 27: 1-9.
Short, C.S. (1988). The Bramham incident, 1980: an outbreak of waterborne infection.

/ Inst Water Environ Manag, 2: 383-390.
Stanwell-Smith, R. (1994). Recent trends in the epidemiology of waterborne disease. In

Water & Public Health, Golding, A.M.B. (ed.). London: Smith-Gordon, pp. 39-56.
Tillett, H.E., de Louvois, J. and Wall, P. (1998). Surveillance of outbreaks of waterborne

infectious disease: categorizing levels of evidence. Epidem Infect, 120: 37-42.
Turner, M., Istre, G.R., Beauchamp, H. et al. (1987). Community outbreak of adenovirus

type 7a infections associated with a swimming pool. South Med J, 80: 712-715.
van Olphen, M., de Bruin, H.A.M., Havelaar, A.H. et al. (1991). The virological quality of

recreational waters in the Netherlands. Water Sci Technol, 24: 209-212.
Wellings, F.M., Lewis, A.L., Mountain, C.W. et al. (1975). Demonstration of virus in

groundwater after effluent discharge onto soil. Appl Environ Microbiol, 29: 751-757.

343

The survival and
persistence of viruses
in water

The term 'survival' in this context means the ability of an infectious unit of
virus to remain infectious in a body of water over a defined time. 'Persistence'
refers to the continued presence of a particular virus type in a body of water
over a period of time. The rates of decay for viruses vary in respect of a var-
iety of factors, including the methods by which samples are taken, processed
and assayed. There continues to be a need to standardize methodology as far
as possible in order to accommodate this aspect of the variation. Cautionary
notes have been sounded by Morris and Irving (1998).

Since viruses are obligate intracellular parasites and unable to multiply out-
side the host, the number of infectious enteric virus particles can only decline
after being shed into the environment. However, enteric virions are robust and
may survive for long periods dependent on the environmental conditions.
Data on the survival characteristics of the individual virus types are generally
scanty. Virus particles, comprising a protein coat enclosing the genome of nucleic
acid, are protected from degradation by faecal organic material. Particles are
destroyed in the environment by desiccation, by UV light, heat above 56°C,

Viruses

digestion by microorganisms and by predation. Particles shed in faecal material
in soil at low temperature or in sediment under water will survive the longest
and may be detectable for months or years.

Sewage may contain any of the human enteric virus groups that circulate
within a population. Those present will vary from season to season, from year
to year and between different geographical locations. During sewage treat-
ment the heavier solid material settles to form sludge and, as virus particles
clump together and are often attached to organic debris, the virions are most
likely to be present in sludge. Treatment by biological filtration or the acti-
vated sludge process reduces the number of virus particles as a result of micro-
biological predation by an estimated 50% and 90% respectively (Berg, 1973)
and discharge of effluent into fresh or marine waters will reduce the number
of viruses further by microbial activity, the action of light and dispersal.

Treatment processes to reduce the numbers of microorganisms in sludge
may reduce the numbers of infectious viruses present but the mechanisms are
poorly understood since the variable and toxic nature of sludge has made it
very difficult to undertake reproducible studies. Berg et al. (1988) showed that
enteroviruses may survive up to 38 days in aeration sludges at 5°C at pH 6-8, but
also that they sometimes survive at pH 3.5. Consequently the numbers and types
of viruses present in fully treated sludge are unknown and so the risk from
viruses is not yet possible to quantify. Mesophilic anaerobic digestion is cur-
rently the most widespread method of sludge treatment in the UK. This takes
place at about 42° C for approximately a month, during which time virus par-
ticles may be degraded or consumed by other microorganisms. Anaerobic diges-
tion or composting are processes more likely to produce sludge that contains
residual infectious particles than the more extreme liming at pH 9. Drying the
sludge or the production of dry pellets are treatments increasing in use and
should eliminate all infectious virus particles.

Among the enteric viruses the enteroviruses and hepatitis A virus (HAV) have
been the most intensively studied with respect to their survival and persistence
because, although not usually associated with intestinal disease enteroviruses
they are relatively easily recovered from aquatic matrices and can be grown in
cell culture and HAV is a well-established waterborne pathogen. Other types of
enteric virus such as rotaviruses and noroviruses, which do cause intestinal dis-
ease, are more difficult to detect and infectivity of these agents cannot easily, if
at all, be determined in the laboratory. Hence the enteroviruses have been used
as a model for survival of other enteric virus types.

Viruses are generally less numerous in sewage than bacteria; whereas faeces
may consistently contain 10 6 -10 7 coliforms/g, viruses may only be present
occasionally and it is generally accepted that raw (inlet) sewage contains approxi-
mately 10 8 coliforms/lOOml and 10 3 -10 4 pfu enteroviruses per litre (though
other individual studies report widely ranging figures).

Attempts have been made for more than 25 years to determine virus survival
in different matrices, though relatively few have been done in recent years
(e.g. Enriquez et aL 9 1995; Wait and Sobsey, 2001). Since bacterial indicator
organisms have been used to determine microbial water pollution, comparisons

346

The survival and persistence of viruses in water

of bacterial and viral survival are common. Comments on experimental con-
ditions and apparatus for survival studies, given elsewhere in the literature
pertaining to bacterial or protozoan studies, apply generally to virus survival
studies too. The general consensus is that viruses are more robust than bacteria;
Vasl et al. (1981) in a field study of sewage-polluted marine coastal waters
showed that while total and faecal coliforms and enterococci were correlated
with enterovirus levels, enteroviruses were more resistant than coliforms and
of resistance equal to enterococci. Parameters in this, as in other studies, var-
ied; volumes of water analysed for enteroviruses ranged from 35 to 85 litres
per sample, water temperature varied between 15 °C and 29°C and the con-
ductivity also varied. Nasser et al. (1993) compared the survival of poliovirus,
hepatitis A virus (HAV) and F + phage in wastewater. Reduction of poliovirus
and HAV was influenced by temperature. At 10°C the titres of these agents was
reduced by 1-2 log 1( ) after 90 days and the phage titre remained unaffected at the
same temperature. At higher temperatures reduction in titres were seen. Kadoi
and Kadoi (2001) showed that feline calicivirus, sometimes used as a cell-
culturable surrogate for norovirus, remained infectious in seawater at up to
10°C for up to 30 days. Survival comparisons in respect of viruses in seawater
and on sediments were done by Tsai et al. (1993), who found that, in sea-
water at 25°C, F + phage was inactivated faster than poliovirus, HAV and
rotaviruses; at lower temperatures and in the presence of sediments there was
less difference in inactivation rates, though it protected poliovirus and HAV at
5°C and 25 °C, whereas it accelerated the inactivation of rotavirus. Abad et al.
(1997) found 2 logK) units reduction in astrovirus infectivity titre after 60 days at
4°C and 3.2 logK) units at 20°C. In all tests the T 99 99 parameter (4 log 1( ) reduc-
tion) was determined and for the enteric viruses was found to range from 2 weeks
to years. Callahan et al. (1995) looked at virus survival in different types of sea-
water and concluded that F + phages survived about the same as poliovirus,
undergoing a 4 logK) reduction in about one week and Johnson et al. (1997) com-
pared the survival of poliovirus under different environmental conditions with
that of Cryptosporidium, Giardia and Salm. typhimurium in marine waters, with
reference to enterococci as the determinand of microbial water quality. The order
of survival in sunlight was Cryptosporidium (greatest), poliovirus, Giardia and
Salm. typhimurium, though poliovirus survived less well in the dark. The T 9 q (the
time required to reduce the number of viruses by 90%) values for poliovirus
ranged from 10 hours to 26 hours, much less than previous studies, though some
of this difference was probably due to different experimental conditions.

That the T 90 values vary under different circumstances is evident, however,
the extent of this variation needs to be determined. It is also clear that viruses
are more resistant to environmental degradation than other microorganisms,
especially bacteria. Morris and Irving (1998) cited a series of seven papers
which reported T 90 values for enteroviruses ranging, under different conditions,
from 14 to over 288 hours.

The issue of virus adsorption to sediments needs to be considered, both in the
context of the natural state (e.g. above) and in planning experimental protocols
for virus survival studies. Turbidity dramatically attenuates the penetration of

347

Viruses

light into the water column; the nature of the material causing the turbidity may
also markedly affect the survival of microorganisms, particularly the viruses as
these can be readily absorbed to some materials such as clay and this could be
more important than light for these organisms (Gerba and Schaiberger, 1975).

References

Abad, EX., Pinto, R.M., Villena, C. et al. (1997). Astrovirus survival in drinking water.

Appl Environ Microbiol, 63: 3119-3122.
Berg, G. (1973). Removal of viruses from sewage, effluents and waters. Bull Wld Hltb Org,

49: 451-460.
Berg, G., Sullivan, G. and Venosa, A.D. (1988). Optimum pH levels for eluting enteroviruses

from sludges solids with beef extract. Appl Environ Microbiol, 54: 83-841.
Callahan, K.M., Taylor, D.J. and Sobsey, M. D. (1995). Comparative survival of hepatitis A

virus, poliovirus and indicator viruses in geographically diverse seawaters. Water Sci

Tecbnol, 31: 189-193.
Enriquez, C.E. and Gerba, C.P. (1995). Concentration of enteric adenovirus 40 from tap, sea

and waste water. Water Res, 29: 2554-2560.
Gerba, C.P. and Schaiberger, G.E. (1975). Effect of particulates on virus survival in seawater.

/ Water Pollution Control Fed, 27: 125-129.
Johnson, D.C., Enriquez, C.E., Pepper, I.L. et al. (1997). Survival of Giardia,

Cryptosporidium, poliovirus, and Salmonella in marine waters. Water Sci Tecbnol, 35:

261-268.
Kadoi, K. and Kadoi, B.K. (2001). Stability of feline caliciviruses in marine water main-
tained at different temperatures. New Microbiol, 24: 17-21.
Morris, R. and Irving, T. (1998). Review of tbe Fate of Enteroviruses in tbe Environment.

Report WW-11D: United Kingdom Water Industry Research Ltd Research and

Development - Bathing Water Policy.
Nasser, A. M., Tchorch, Y. and Fattal, B. (1993). Comparative survival of E. coli, F +

bacteriophages, HAV and poliovirus 1 in wastewater and groundwater. Water Sci

Tecbnol, 27: 401-407.
Tsai, Y.L., Sobsey, M.D., Sangermano, L.R., Palmer, C.J. (1993). Simple method of con-
centrating enteroviruses and hepatitis A virus from sewage and ocean water for rapid

detection by reverse transcriptase-polymerase chain reaction. Appl Environ Microbiol,

58:3488-3491.
Vasl, R., Fattal, B., Katzenelson, E. et al. (1981). Survival of enteroviruses and bacterial

indicator organisms in the sea. In Viruses and Wastewater Treatment, Goddard, M. and

Butler, M. (eds). Oxford: Pergamon Press, pp. 113-116.
Wait, D.A. and Sobsey, M.D. (2001). Comparative survival of enteric viruses and bacteria

in Atlantic Ocean seawater. Water Sci Tecbnol, 43: 139-142.

348

Methods for the detection
of waterborne viruses

Detection of waterborne viruses is more complex than detection of other
microorganisms because of difficulties in concentrating the sample and then in
detecting the virus by cell culture or molecular biological means. With the
exception of sewage, where the quantity of virus present may be sufficient to
permit detection without further concentration, viruses in water are too dilute
to be detected by direct analysis. Groundwater and drinking water will con-
tain very few viruses and 1001 or more will need to be processed, while recre-
ational fresh or marine waters may contain many more viruses so processing
101 samples will be sufficient. Water samples are therefore taken through a
two-stage process, first to concentrate the virus into a smaller volume (usually
5-10 ml), then detection of virus in the concentrate. The concentrate may be
inoculated into cell cultures to detect infectious virus; if this is done in a quan-
titative fashion any virus present can be enumerated, the count being reported
as plaque-forming units (pfu), tissue culture infectious doses (TCD50), or
MPN units. Where necessary virus can be isolated and identified from the cell
cultures. Viruses which do not produce an identifiable cytopathic effect in cul-
ture may nevertheless be detected by immunoperoxidase or immunofluores-
cence staining. Where there is no viral replication in culture, as is the case with

Viruses

(a)

Figure 26.1 Membrane filtration.

(b)

most gastroenteritis viruses, then molecular biological detection procedures
may be used.

Concentration methods

Viruses exhibit polarity and can adsorb to a wide variety of charged matrices,
which may be immobilized (such as membranes) or fluid (such as glass powder)
(Figures 26.1 and 26.2). They may thus be concentrated by adsorption to such
matrices. Considered as protein, virus particles have a high relative molecular
mass (M r > 10 6 ) and lend themselves to sedimentation from suspension by
ultracentrifugation and to concentration by ultrafiltration. Based on these
general properties, numerous methods have been devised for the concentration
of viruses from water (Table 26.1). These have been reviewed extensively by
Wyn-Jones and Sellwood (1998) in respect of enteroviruses and by Wyn-Jones
and Sellwood (2001) for other virus groups.

To be of optimum practical use any method must fulfil the following criteria
(after Block and Schwartzbrod, 1989):

• be technically easy to accomplish in a short time

• have a high virus recovery rate

• concentrate a range of viruses

• provide a small volume of concentrate

• not be costly

• be capable of processing large volumes of water

• be repeatable (within a laboratory) and be reproducible (between laboratories).

350

Methods for the detection of waterborne viruses

(a)

(b)

Figure 26.2 Filtration through glass wool, (a) column, (b) cartridge.

There is no single method that fulfils all these requirements.

There are four principal techniques, each based on a different property of the
virus particle. Each technique has numerous variations. Most procedures can
be used to concentrate viruses in sample volumes of 1-1001. Adsorption/elution
and some entrapment techniques comprise the first stage (primary concentra-
tion) in a two-stage concentration process which reduces the initial volume to
between 100 and 400 ml. A secondary concentration stage concentrates the
virus further by acid flocculation and low-speed centrifugation to deposit the
virus-containing floe. This is dissolved in 5-10 ml neutral buffer.

Adsorption/elution

The development of virus adsorption/elution (Viradel) methods suitable for
the recovery of viruses from waters stems from the work of Melnick and his
colleagues in Houston, Texas (e.g. Wallis and Melnick, 1967a,b,c; Wallis etaL,
1970). In general terms, the virus-containing sample is brought into contact
with a solid matrix to which virus will adsorb under specific conditions of pH
and ionic strength. Once virus is adsorbed, the water in which it was originally
suspended is discarded. Virus is then eluted from the matrix into a smaller
volume, though this is still too large to be inoculated directly on to cell cul-
tures. Choice of adsorbing matrix, eluting fluid and processing conditions

351

Table 26.1 Summary of concentration techniques for viruses in water and water-related materials

Relative

2a ry

Water

Initial

virus

Capital

Revenue

concn

Technique

Method

quality

volume

content

Recovery

cost

required?

Comments

Adsorption/

Gauze pads

Sewage or

Large

High

Low to

Nil

Very low

Not quantitative

elution

effluent

medium

Electronegative

All waters

1-1000

Low to

50-60% with

Medium

Yes

High vols require

membranes

litres

medium

practice

dosing pumps

Electropositive

All waters

1-1000

Low to

50-60% with

Medium

High

Yes

No preconditioning

membranes

litres

medium

practice

required

Electronegative

Any low

1-50

Low to

Variable:

Low

Yes

Clogs more

cartridges

turbidity

litres

medium

higher with
clean waters

quickly than
membranes

Electropositive

All waters

1-1000

Low to

Variable

Medium

High

Yes

Wide range

cartridges

litres

medium

of viruses

Glass wool

All waters

1-1000
litres

Low to
medium

Variable

Low

Very low

Yes

No preconditioning
required

Glass powder

All waters

<100
litres

Any

20-60%

Medium

Low

If vol >100

litres

Special apparatus

Entrapment/

Alginate

Clean only

Low

High

Good

Low

Very slow. Clogs

ultrafiltration

membranes

rapidly if turbid

Single

Clean

Low

Any

Variable

Medium

Low

Slow

membranes

Tangential

Treated

High

Low

Variable

High

Medium

Sometime

Prefilter for

(= cross) flow

effluents

turbid waters

and hollow

or better

fibres

Vortex flow

Treated
effluents
or better

High

Low

Unknown

High

Medium

Unknown

Undeveloped yet

Hydroextraction

PEG or
sucrose

Any

Low

High

Variable
(toxicity)

Negligible

Very low

High virus loss
in waste waters

Ultra-

Clean

Low

High

Medium

High

Medium

Wide range, but

centrifugation

usu impractical

Other

Fe oxide floe

All

Low

Any

Variable

Low

techniques

Biph partition.

All

<7 litres

Any

Variable

Low

Toxic to cells

Immunoaffinity

Unknown

Low

High

Low

New method

and mag beads

Methods for the detection of waterborne viruses

will be influenced by the nature of the sample and by experience, but is com-
monly done using a solution containing beef extract or skimmed milk, both at
high pH, which displaces the virus from the adsorbing matrix into the eluant.
Glycine/NaOH solution may also be used. Following elution with protein-
aceous fluids the virus is still too dilute to be inoculated directly into cell cul-
tures and is therefore concentrated further; this is termed secondary
concentration. Several methods are available, but that most commonly used is
that of Katzenelson et al. (1976). The pH of the eluate is reduced to 3.5-4.5
which causes isoelectric coagulation (flocculation) of the protein. The virus is
adsorbed to the floe which is deposited by centrifugation and dissolved in
5-10 ml neutral phosphate buffer. It may then be frozen or used for further
analysis. Gilgen et al. (1997) developed a protocol for analysis of bathing waters
and drinking water which used filtration through positively charged membranes
followed by ultrafiltration as a secondary concentration step and Huang et al.
(2000) used positively charged membranes followed by beef extract elution
and PEG precipitation for the concentration of caliciviruses in water.

a. Adsorption to electronegative membranes and cartridges
Concentration of viruses in water using negatively-charged microporous filters
(e.g. Millipore, Sartorius) has been practised for many years and there are
many variations of the technique, though little change in the basic process.
The popularity of membranes, made of cellulose acetate or nitrate, is due to
their availability in various pore sizes, configurations and compositions and,
by judicious choice of pre-filters and adsorbing filters, it is possible to get good
recoveries of virus accompanied by good flow rates and a minimum of filter
clogging, even from turbid waters. In addition, many solids-associated virus
can be recovered. Virus is bound to the filter by opposing electrostatic attrac-
tive forces and not by entrapment. In its simplest form, a virus-containing sam-
ple is passed under positive pressure or vacuum through a cellulose nitrate
membrane 142 mm or 293 mm in diameter and of mean pore diameter 0.45 |xm,
1.2 |xm or 5 fjim. For waters containing particulate material a pre-filter is used
ahead of the membrane to prevent clogging. Virus is adsorbed and is eluted
using beef extract or skimmed milk solution.

Since viruses and the filter materials are both negatively charged at neutral
pH, the water sample is conditioned to allow electrostatic binding of virus to
filter matrix. The water sample is adjusted to pH 3.5 and Al 3+ ions added in the
form of A1C1 3 to a final concentration of 5 X 10~ 4 M. Magnesium salts may
also be used but most reports suggest better recoveries are obtained by using
aluminium salts (e.g. Homma et al., 1973; Metcalf et al., 1974), though opinion
is divided as to whether metal ions are needed at all when using cellulose nitrate
membranes.

The choice of filter will depend partly on the nature of the sample; for sea-
waters filters with pore diameters of 0.45 jxm and 1.2 |xm are commonly used
and pre-filters are employed. Filters may be used singly or in series. This
method is currently the preferred way of recovering viruses from effluent,

353

Viruses

diluted raw sewage and activated sludge samples, as well as recreational and
surface waters in most water virology laboratories in the UK and is the rec-
ommended method (Standing Committee of Analysts, 1995). It is also a tentative
standard method for the recovery of viruses from waters and wastewaters, as
published by the American Public Health Association (APHA) (1980).

Negatively-charged filters may also be used in tube form. Balston filters are
epoxy resin-bound glass fibre filters with an 8 (xm nominal pore diameter. They
were originally used for concentration of viruses from tap water (Jakubowski
et aL, 1974) and have since been employed for concentration of viruses from river
water (e.g. Morris and Waite, 1980) and other waters (e.g. Guttman-Bass and
Nasser, 1984; United States Environmental Protection Agency, 1984). Their
recoveries appear at least as good as membrane filters, they are less expensive
and can be obtained complete in sterile cartridges in disposable form. Guttman-
Bass and Nasser (1984) reported mean recoveries from sea water of 93, 75, 92
and 109% for poliovirus 1, echo 7, coxsackievirus Bl and coxsackievirus A9
respectively. They are, however, prone to clogging and cannot be used with
even moderately turbid water and according to Gerba (1987) cannot be used
at high flow rates. Because of problems of clogging of membrane or tube filters,
the processing of seawater samples in this way is limited to a maximum of 20 1
before filters have to be changed (Block and Schwartzbrod, 1989).

One way of overcoming the problem of clogging without having to change
membranes or tubes frequently is to increase the surface area of filtration by
the use of larger cartridge filters, where sheets of negatively-charged pleated
filter material approximately 25 cm wide are rolled and used in 30 cm cart-
ridge holders. These were evaluated by Farrah et aL (1976) who used fibreglass
membrane material in a pleated format (Filterite Duo-Fine). The increased
surface area allowed for higher flow rates and recoveries were better than given
by membranes. Pre-filters were used in series to prevent further clogging where
0.45 \im pore diameter filters were used in processing marine waters. Seeded
poliovirus was recovered from 3781 volumes of seawater with 53% efficiency.
These filters are more expensive, but these authors reported that they could be
regenerated up to five times by soaking for 5 minutes in 0.1 M NaOH.

Many modifications to the filtration process have been made but have had
little effect on practicability, recovery, or other aspects. For example, Farrah
et al. (1988) reported a higher adsorption of enteroviruses to Filterite (epoxy
resin-bound glass fibre) filters after treatment with polyethyleneimine.
Concentration by filtration through negatively-charged media has also been
shown to be suitable for virus detection by molecular biological methods (e.g.
Wyn -Jones et al., 1995), though there are differences in the use of eluting fluids.

Generally, recovery rates are as variable with negatively-charged filter media
as with any other kind. Block and Schwartzbrod (1989), citing Beytout et al.
(1975), consider cellulose nitrate membranes relatively efficient insofar as
they give 60% recovery of virus; the same authors recorded glass fibre filters
giving a poor average yield on wastewater but 70% recovery with river
water. Payment et aL (1979), using glass fibre filters, reported 38-58% recovery
of 10 2 -10 6 pfu seeded in 100ml to 10001 volumes. Few studies have been

354

Methods for the detection of waterborne viruses

done on recovery efficiencies from marine waters in a controlled way; how-
ever, controlled studies have been done to evaluate the recovery efficiency of
the method using drinking water. Melnick et al. (1984) reported considerable
variation in the quantity of virus recovered following processing of 1001 tap
water samples containing poliovirus. Though the average recovery was 66%
(of 350-860 pfu virus), values ranged from 8 to 20% in two laboratories, 49
to 63% in three laboratories and 198% in one laboratory. Recovery levels were
higher and less variable where a higher input level of virus was used, but it
must be noted that even the 'low' level of 350-680 pfu is more than is routinely
found in bathing waters.

It is not usually possible to conduct studies where virus is deliberately added
to water systems, however, Hovi et al. (2001), in assessing the feasibility of
environmental poliovirus surveillance added poliovirus type 1 into the Helsinki
sewers and recovered it over a period of 4 days by taking samples at down-
stream locations and concentrating 100-fold simply by polymer two-phase
separation. If controlled studies are done then they should use virus input levels
which would normally be expected in the waters to be monitored under actual
conditions. In the UK External Quality Assessment Scheme for Water Virology
the majority of participating laboratories use cellulose nitrate membranes for
concentration; one laboratory uses Balston tube filters. The normal range of
recovery for seeded clean water is up to 60%.

b. Adsorption to electropositive membranes and cartridges
Positively-charged filters adsorb virus from water and other materials without
the need for prior conditioning of the sample with acid or cations. They will
adsorb virus in the pH range 3-6; at pH values above 7 the adsorption falls
off rapidly, so the pH still needs to be careful monitored. These properties
make the use of positively-charged filters attractive, not only for the convenience
of not having to condition the sample, but also because it makes possible the
concentration of other viruses such as rotavirus and coliphages, which are
sensitive to the low pH conditions needed for adsorption to negatively-charged
media. Keswick et al. (1983) reported that type 1 poliovirus and rotavirus
SA11 survived at least 5 weeks on electropositive filters at 4°C, so this may
make them useful for extended surveys or transmission through postal systems.
Other than not needing to condition the water sample, electropositive filters
are used in the same way as electronegative materials. Virus is eluted from the
filter and secondary concentration is carried out as for the electronegative types.
Recoveries from positively-charged filters are similar to those from negatively-
charged filters; Sobsey and Jones (1979) reported 22.5% recovery using a
two-stage procedure in the concentration of poliovirus from drinking water.
The original positively-charged material, Zeta-plus Series S, is made of a cellulose/
diatomaceous earth/ion-exchange resin mixture. It is commercially available
as Virozorb 1MDS cartridges (AMF-Cuno). Sobsey and Glass (1980) com-
pared these filters with Filterite (fibre glass) pleated cartridge filters for recov-
ery of poliovirus from 10001 tap water and obtained recoveries of about

355

Viruses

30% with both types. The advantages of these filters lie in the large volumes
they can handle without the need for conditioning the sample. Elution from
the filter still needs to be carried out at pH 9 or above, which limits their use
to viruses stable below that pH, though Bosch et al. (1988) successfully con-
centrated rotavirus in this way. Organic materials in the sample, especially ful-
vic acid, were reported to interfere more with virus recovery from Virozorb
cartridges than from glass-fibre materials (Sobsey and Hickey, 1985; Guttman-
Bass and Catalano-Sherman, 1986). A different electropositive material (MK)
is cheaper but its recoveries were reported to be not as good as 1MDS in com-
parative tests (Ma et al., 1994).

Advances in membrane technology resulted in charge-modified nylon mem-
branes being available for concentration of viruses from water. Gilgen et al.
(1995, 1997) described the use of positively-charged nylon membranes (Zetapor,
AMF-Cuno) coupled with ultrafiltration for the concentration of a variety of
enteric viruses prior to detection by RT-PCR. Other nylon membranes are also
available; 'Biodyne B' and c N66-Posidyne' (Pall), which are different grades of
the same material, are made in pore sizes which would permit passage of virus
(0.45 fjim, 1.2 |xm and 3 (xm) and have a positive surface charge over the pH
range 3 to 10, which would promote strong binding of negatively-charged
particles. Posidyne filter material is also available in re-sterilizable cartridge
form, which would increase the convenience of use. No studies have been done
on this material in respect of recovering viruses from bathing waters, but its
low cost and ease of use suggest that further evaluative research should be done.
Triple-layered PVDF membranes and cartridges (Ultipor VF, Pall) have been
used in industry for the removal of polio and influenza viruses from pharma-
ceutical products (AranhaCreado et al., 1997), though whether the virus can
be recovered from the filter is not known.

The need to determine the presence of Cryptosporidium and Giardia as well
as viruses in water samples has led some workers to attempt the simultaneous
concentration of both types of microorganism. Watt et al. (2002) compared
the efficiencies of polypropylene spun fibre cartridges with Filterite or 1MDS
microporous cartridges in concentrating Giardia cysts and Cryptosporidium-
oocysts and poliovirus from 4001 volumes of drinking water and treated
wastewater. Filterite filters performed better than the 1MDS and were easier
to process and the protozoa were trapped by both types of filter; the authors
suggested that it is thus possible to concentrate virus and protozoa at the same
time on a composite filter.

c. Adsorption to glass wool

Glass wool is an economic alternative to microporous filters. It is used in a
column and, provided it is evenly packed to an adequate density, adsorption
of enterovirus appears at least as efficient as with other filter types (see Figure
26.1). An advantage of the method is that virus will adsorb to the filter matrix
at or near neutral pH, which makes it suitable for viruses sensitive to acid and
without the addition of cations; elution still has to be done at high pH.

356

Methods for the detection of waterborne viruses

The technique has been pioneered in France principally by Vilagines and
co-workers (e.g. Vilagines et al. 9 1988), who applied it to the concentration
of a range of viruses from surface, drinking and wastewaters. Glass wool
packed into holders at a density of 0.5 g/cm 3 is washed through with HCl, water
and NaOH and water to neutral pH before the sample is passed through the
filter. Different sizes of filter can be prepared according to the type of water
and flow rate.

In the French studies, sample sizes ranged from 100 to 10001 for drinking
waters, 30 1 for surface waters and 10 1 for wastewaters. The only pretreatment
necessary was dechlorination of drinking waters. Surface water samples were
filtered at 50 1/h in a 42 mm diameter filter holder. Virus was eluted from the
filter with 0.5% beef extract solution and secondary concentration done by
organic flocculation.

Recovery efficiency of approximately 10 2 pfu poliovirus seeded into 4001
drinking water averaged 74% (s.d. 18.9%). For surface waters the recovery
rate was 63% and 57% respectively. Lowering the flow rate to 50 1/h reduced
clogging of the filters.

Since virus concentration on glass wool does not need the sample to be con-
ditioned the technique lends itself to large sample monitoring (for surface waters)
and to continuous monitoring (for drinking waters). Other viruses were also
concentrated during field evaluation of the method; adenoviruses and
reoviruses were also recovered, though as expected enteroviruses predomi-
nated. Vilagines et al. (1993) also reported a survey of two rivers over a 44-
month period and concluded that the technique was robust enough in
physical and experimental terms to be used for routine monitoring of surface
waters. Glass wool is also very cheap and thus the method is economic.

Glass wool has been used in other laboratories; Hugues et al. (1991) found
it more sensitive than the glass powder method in terms of number of positives
detected and in the level of virus when analysing biologically treated waste-
waters; Wolfaardt et al. (1995) used glass wool to concentrate small round-
structured viruses (SRSVs) from spiked sewage and polluted water samples
prior to detection by RT-PCR.

d. Adsorption to glass powder

Powdered borosilicate glass with a bead size of 100-200 fjim is a good adsorb-
ent for viruses under conditions similar to those used for glass fibre micro-
porous filters. However, glass beads constitute a fluidized bed and so have the
advantage that the filter matrix cannot become clogged as with glass-fibre sys-
tems. Sarrette et al. (1977) first developed this technique, which was extended
by Schwartzbrod and Lucena-Gutierrez (1978).

For low sample volumes (<100l) the method gives a low eluate volume
which may not need secondary concentration prior to further analysis.
However, the recovery varies widely with the type of sample, from 60% with
potable water to 20% with urban wastewater (Joret et al., 1980). For sample
volumes of less than 1001 containing relatively little organic material the

357

Viruses

powdered glass technique would appear useful in principle. A drawback is the
complexity of the apparatus; specially constructed two-part mixing and elution
chambers are required and this may render the method impracticable unless
there were other strong reasons for its use.

Entrapment

Entrapment refers to those techniques in which the virus in a sample is bound
to a filter matrix principally by virtue of its molecular size rather than by any
charges on the particle, though in practice electrostatic effects can influence
binding to an ultrafilter.

a. Ultrafiltration

Early ultrafiltration methods involved the filtering of the water sample under
pressure through aluminium/lanthanum alginate filters (Poynter et al. 9 1975).
While these had the unique advantage that they were soluble in sodium cit-
rate, the flux obtained was too low for all but the cleanest waters unless they
had been pre-filtered which precluded analysis of surface waters in volumes
over 1 1.

More recent techniques involve passing the sample through capillaries (e.g.
Rotem et al. 9 1979), membranes (e.g. Divizia et al. 9 1989a,b), hollow fibres
(Belfort et al. 9 1982) with pore sizes that permit passage of water and low
molecular mass solutes but exclude viruses and macromolecules, which are then
concentrated on the membrane or fibre. Most laboratories now use mem-
branes or fibre systems with cut-off levels of 30-100 kDa. In systems in which
the fluid passes directly through the filter, non-filterable components quickly
clog the filter or precipitate at the membrane surface, thus this type of filter is
only useful for small volumes (1000 ml or less) of sample.

Recently developed ultrafilters employ tangential flow. The minimum 'dead'
volume (e.g. 10-15 ml, Divizia et al. 9 1989a), which depends on the apparatus
in use, is that beyond which it is impossible to reduce the retentate volume
further; this is the final volume of concentrate. If this is small enough then it
may be inoculated on to cell cultures or it may have to be further processed by
secondary concentration.

Some workers have experienced binding of virus to the membrane rather than
just prevention of its passage through it. In these cases the virus was eluted by
backwashing with glycine buffer or beef extract and the eluate reconcentrated
by organic flocculation. Some authors have even reported differences in binding
between viruses that are related. Divizia et al. (1989b) for example noted that
hepatitis A virus was recovered with 100% efficiency; poliovirus on the other
hand was recovered very poorly under standard conditions but this improved
if the membranes were pretreated with different buffers. Further, recovery was
best if the virus was eluted with beef extract at neutral (not high) pH.

358

Methods for the detection of waterborne viruses

Ultrafiltration has also been used to reconcentrate viruses recovered from
treated wastewater by adsorption/elution.

A variation in ultrafiltration is vortex flow filtration (VFF). The technique
appears to offer a further step in the reduction of clogging while still retaining
the ability of filters to process large volumes of sample, but there are few sig-
nificant reports in the literature of its use in the field. One is that of Tsai et al.
(1993), who used it for inshore waters in Southern California. Fifteen litres of
each sample were concentrated to 100 ml using a 100 kDa cut-off membrane
and the samples were further concentrated to 100 |Jiml using Centriprep and
Centricon units at lOOOXg.

The advantages of ultrafiltration are principally that the sample requires
no preconditioning and that a wide range of viruses can therefore be recov-
ered, including bacteriophages (e.g. Urase et al., 1993; Nupen et al., 1981).
Efficiency of recovery is usually good, though as with all methods it is vari-
able. The main constraints upon its use are the high initial cost of the equip-
ment and that, despite the advantages of tangential flow, turbid samples still
tend to clog the membrane. Surface waters may take a long time to process
if they are turbid; Nupen et al. (1981) were able to filter 501 volumes but
this took from 40 hours to 72 hours depending on the sample. In other stud-
ies workers have used different parameters (though generally all use
30-100 kDa membranes or hollow fibre cartridges) with differing results.
The technique is generally seen as an advance on adsorption/elution (e.g.
Grabow et al., 1984; Muscillo et al., 1997) and recovery efficiencies, though
variable, appear to be higher than those obtained using adsorption/elution
techniques.

b. Ultracentrifugation

Ultracentrifugation is a catch-all method since it is capable of concentrating
all viruses in a sample provided sufficient g-force and time are used. Differential
ultracentrifugation allows separation of different virus types. A number of
studies have been reported using ultracentrifugation, including one in which
virus from a polluted well was recovered (Mack et al., 1972) and one where
viral numbers in natural waters were as high as 2.5 X 10 8 /ml, 10 3 to 10 7 times
as high as had been found by plaque assay (Bergh et al., 1989). However, the
limited volumes that can be processed, even using continuous flow
systems, together with the high capital costs and lack of portability of the
equipment, limit its usefulness in concentrating viruses directly from natural
waters. It does find a use as a secondary concentration method, however;
Murphy et al. (1983) in an investigation of a gastroenteritis outbreak associ-
ated with polluted drinking water, concentrated 5 1 samples of borehole water
using an ultrafiltration hollow fibre device to 50 ml and followed this by
ultracentrifugation to pellet the virus for electron microscopical examination.
They were thus able rapidly to detect rotaviruses, adenoviruses and small
round-structured viruses, as well as enteroviruses, which were confirmed by
cell culture.

359

Viruses

Other methods

Hydro extraction

Hydroextraction is the concentration of virus in a sample by the removal of
water using a hygroscopic solid. Two solids are commonly used; polyethylene
glycol (PEG, as the polymer in the range 6000-20 000) and sucrose. The tech-
nique involves filling a dialysis bag with the sample, immersing it in the solid
and leaving it at 4°C for several hours. Water is drawn out of the sample
which thus reduces in volume. Further dialysis against phosphate buffer is
then required to remove the PEG/sucrose which has entered the bag.

Hydroextraction has been employed with some success, but its use is limited.
Clearly, volume size is the principal constraint and the maximum volume that
can be handled is about 1 1; sewage and wastewater were successfully concen-
trated in this way nearly 30 years ago (Wellings et aL 9 1976). Further, dialysis
membranes have a M r cut-off of approximately 12 000. Thus some organic
compounds will be concentrated along with the virus. Many of these are cyto-
toxic and the virus cannot be assayed in cell culture. Hydroextraction has
been used frequently as a secondary concentration step following micro-
porous filtration or ultrafiltration (e.g. Ramia and Sattar, 1979).

Iron oxide flocculation

Several reports in the literature describe the adsorption of enteroviruses to
magnetic and non-magnetic iron oxides, either Fe 2 03 or Fe3C>4. Rao et al.
(1968) described adsorption of a range of viruses to magnetic iron oxide and
Bitton etal. (1976) carried out a detailed study of the adsorption of poliovirus
to magnetite (Fe304). Adsorption occurred at pH 5-8 and elution could be
effected with beef extract at pH 8-9. In the case of magnetic oxides the virus
may be concentrated by removing the oxide with adsorbed virus from the
sample water with a magnet, then eluting.

Talc-celite adsorption

Talc (magnesium silicate) mixed with celite (diatomaceous earth) forms a
good combined adsorbent which can be used as a fluid bed or sandwiched
between two layers of filter paper. A range of viruses can be concentrated by
this means and different waters, including tap water and wastewater can be
processed (e.g. Sattar and Westwood, 1978; Sattar and Ramia, 1979; Ramia
and Sattar, 1979).

Adsorption to bituminous coal

Dahling et al. (1985) used powdered coal as an adsorbent with a view to
transferring virus concentration technology to developing countries. Filters
made this way were effective over the pH range 3 to 7 and recoveries did not
differ significantly from those obtained with membranes filters when used

360

Methods for the detection of waterborne viruses

with 1001 tap water samples. They could also be used for wastewater sam-
ples. Lakhe and Parhad (1988) described a similar system, and Chaudhiri and
Sattar (1986) reported a system which could be used for the removal of
enterovirus from water with a view to improving its quality. The same kind of
matrix in a more refined state was used as granular activated carbon by
Jothikumar et al. (1995) for the first stage concentration of enteroviruses, HEV
and rotaviruses. Using RT-PCR as a detection method these authors claimed a
74% recovery of poliovirus 1.

Two-phase separation

Viruses and macromolecules can be partitioned between the two immiscible
phases produced when two different organic polymers are dissolved in water.
Lund and Hedstrom (1966) used sodium dextran sulphate and polyethylene
glycol 6000 mixture for enterovirus recovery from sewage. By controlling the
phases viruses can be partitioned into one of them. If the virus-containing phase
is made small relative to the original volume of sample then concentration is
achieved. Though effective, this method is limited by the occasional toxicity
of the polymer for cell culture; it is also limited to about 71 maximum volume.

Immuno affinity columns and magnetic beads

These are relatively new techniques which have been used in a biochemical or
molecular biological context. They are useful for small volumes but their
application to virus concentration from larger volumes has yet to be demon-
strated. Schwab et al. (1996) developed a broad-based antibody capture tech-
nique for a variety of viruses, in conjunction with primary concentration
through 1MDS positively-charged filters and detection by RT-PCR. Myrmel
etaL (2000) also described the separation of noroviruses in this way. Water sam-
ples seeded with genogroup I norovirus were brought in contact with mag-
netic beads coated with polyclonal antibodies raised against a recombinant
capsid protein of the same virus. Virus was captured and detected by RT-PCR.
An important attribute of this method is that it acts as a clean-up stage in that
RT-PCR inhibitors are removed as well as the virus being concentrated.

D ryingl freeze- drying

Forced removal of the water in samples has been used by several authors. The
commonest approach seems to be the vacuum-drying of the sample; Bosch
etaL (1988) concentrated rotavirus and astrovirus in this way and Kittigul et al.
(2001) reported significantly higher rotavirus antigen recovery following
SpeedVac concentration than when polyethylene glycol (PEG) precipitation
was used as a secondary step, both after membrane filtration as a first stage.
It will be clear from the foregoing that there is a wide range of methods for
concentration of viruses from different types of water. The UK preferred
method is to concentrate the sample using negative polarity cellulose nitrate

361

Viruses

membranes in a 142 mm or 293 mm diameter filter holder and to elute the
virus with beef extract solution. This is also the method recommended by the
Standing Committee of Analysts (1995). Table 26.1 is intended as a quick ref-
erence to the methods for enterovirus concentration.

Detection and enumeration of waterborne viruses

After sampling and concentration, the third major aspect of virological analysis
of waters is the detection and enumeration of the viruses in the concentrate.
Detection and enumeration are conveniently considered together since, for
many viruses they are performed simultaneously, particularly where the virus
multiplies in culture and infectivity assays are done. Broadly speaking, detection
may be done by infectivity-based methods where the virus undergoes at least
partial multiplication in cell culture, or it may be done by techniques based on
properties other than infectivity. Most important in this latter category are the
molecular biological techniques, especially the polymerase chain reaction which
has found wide application in water virology as it has in other biological dis-
ciplines. Enumeration by molecular means may be semi-quantitative, such as
end point dilution assays using the disappearance of a PCR band as the end
point in a titration, or, increasingly, real-time PCR is becoming a realistic pos-
sibility for enumerating genome copies of a target virus, though the relation-
ship between numbers of infectious units and genome copies depends on many
variables and may not be achievable for all viruses.

Detection of virus infectivity is done by inoculating cell cultures with part
or all of the concentrate and allowing the virus to multiply in the cells so that they
are killed. Virus-specific cell killing (the cytopathic effect, CPE) of enteroviruses
and some other types is visible to the naked eye. If a range of cell cultures is
inoculated under liquid assay it should be possible to detect polio, coxsackie,
echoviruses, as well as some adenoviruses and reoviruses. Hepatitis A virus may
also be detected in this way but only after prolonged incubation of cultures
and it is therefore not an approach used in routine waterborne HAV detection.

There are two approaches to the enumeration of virus infectivity. The
plaque assay, where virus-mediated cell destruction is confined to a small area
(the plaque) by incorporation of agar in the maintenance medium, may be done
in cultures of cells growing in single layers (monolayers) or in cultures where
the cells are suspended in the maintenance medium. In both cases, plaques
develop following incubation and may be counted as they become visible, in
the case of enteroviruses usually after about 3 days. The suspended cell assay,
since it offers more adsorption sites to viruses, is three to four times more sen-
sitive than the monolayer assay (UK Public Health Laboratory Service Water
Virology External Quality Assurance (EQA) Scheme unpublished results),
though the latter requires only a fifth of the cells. One plaque is taken as being
the progeny of one infectious unit of virus; this may be the same as one virus

362

Methods for the detection of waterborne viruses

particle, but this is unlikely given the association of virions with particulate
matter, both organic and inorganic.

Virus infectivity may also be assayed in liquid culture, where virus concen-
trate is added to replicate cell cultures which are then observed for specific CPE.
Computation of the positives allows a titre to be determined in terms of Most
Probable Number units or Tissue Culture Dose 50 (TCD 50 ) units.

Virus infectivity may also be determined by immunofluorescence or
immunoperoxidase techniques, which are particularly useful where limited
replication occurs and a distinct CPE is not produced.

Cell culture

Because viruses are obligate intracellular parasites they will only grow in living
cells. They are also quite species specific, so that, for example, human viruses
generally only grow in human cells and bovine viruses will usually only grow in
bovine cells. As cultures and viruses become more adapted this division becomes
blurred but, in general, human viruses will multiply only in cells of primate origin.

Cell cultures may be divided into three kinds; continuous cell lines, primary
cell cultures and semi-continuous cell strains. The first are generally used in
water virology because of their availability and susceptibility to virus strains
normally encountered in water samples.

Continuous cell lines arise through spontaneous transformation of primary
cultures or in vivo transformation as tumour tissue. They are usually hetero-
ploid or even aneuploid. They tend to be sensitive to fewer viruses than primary
cells but, being transformed, they will undergo extensive or indefinite serial
passaging, though their properties change gradually as they do so. Provided
the correct choice is made initially this type of cell culture is the most useful for
a routine water virology laboratory.

There are many cell lines suitable for growing enteroviruses, including HEp-2,
HeLa and VERO cells. The line most favoured for enumeration of water-
borne enteroviruses is the Buffalo Green Monkey (BGM) line first described by
Barron et al. (1970) and in a water context by Dahling et al. (1974). This was
reported to give higher plaque assay titres of poliovirus, coxsackieviruses B,
some echovirus and reoviruses than obtained in rhesus or grivet monkey kidney
cells. It is interesting to note, however, that this apparent better sensitivity is not
continued with isolation of enteroviruses from clinical specimens; Schmidt et al.
(1976) found that BGM cells were less sensitive than primary RMK or human
fetal kidney cells for the isolation of some echo and adenoviruses from clinical
specimens and Pietri and Hugues (1985) found there was no difference between
BGM and KB cells for quantification of poliovirus. Nevertheless, BGM cells
are used almost exclusively for the detection and enumeration of waterborne
enteroviruses. Morris (1985) examined ten cell lines for their ability to grow
enteroviruses isolated from wastewater effluent. Eighty-two per cent of isol-
ates were positive in BGM cells, 73% in RD cells and 64% in chimpanzee
liver cells. BGM was also the most sensitive in the number of plaques counted.

363

Viruses

Dahling and Wright (1986) carried out an extensive set of experiments to
optimize the BGM line in respect of a number of assays for waterborne viruses
and made recommendations in respect of many cell culture and assay param-
eters, as well as doing a comparative virus isolation study involving BGM cells
and nine other cell lines. This work has become the accepted basis for many
standard methods on detection of water-associated viruses.

Other cell lines, derived from intestinal tissue, have been investigated for
their ability to support the growth of enteric viruses. Most of these studies have
been directed at growing the more fastidious agents like rota- and astroviruses,
but Patel et al. (1985) carried out a large survey on the susceptibility of a range
of lines to different enteroviruses, including all 31 serotypes of echovirus; they
found that two lines, HT-29 and SKCO-1, had a markedly wider sensitivity,
with comparable or wider sensitivity for enteroviruses than PMK or rhabdo-
myosarcoma cell cultures. HT-29 and SKCO-1 (both of which supported the
growth of all echoviruses) are derived from human colonic carcinoma tissue
and so it is not surprising that they respond better to infection with enteric
viruses. They require a high seed density and do not grow quickly, however
and perhaps this is why they have not found greater favour, along with
CaC0 2 cells (Fogh, 1977) which are of similar origin, in the detection of
waterborne enteric viruses generally.

The plaque assay

The plaque assay is the method most used for the estimation of waterborne
enteroviruses; it is in plaque-forming units (pfu) that the levels of virus per-
mitted per 101 sample under the EU Bathing Water Directive are expressed.
Although other methods are available, and have some advantages, this is the
principal method in use. All the concentrate should be tested, but many labora-
tories test only a proportion of the concentrate and multiply the resulting
count accordingly. This is unwise where there are likely to be small numbers
of virus particles present and will not necessarily be randomly distributed. A
subsample will therefore not be representative of the whole and an erroneous
titre will be recorded. If all the sample is not assayed in one test adequate internal
quality controls need to be included to verify the performance of the assay.

Monolayer plaque assay

The virus concentrate is inoculated on to preformed monolayers in Petri dishes
or flasks and the cells are reincubated under an agar overlay until a CPE is
seen. Infection is confined to local areas of cell death (plaques) since the agar
prevents the virus spreading over the whole of the monolayer and it can only
infect adjacent cells. It has been recognized for many years that plaque morph-
ology varies between enterovirus types (Hsiung and Melnick, 1957) and
careful experienced examination of plaques in appropriate cell cultures may
provide an indication of the enteroviruses present. Plaques are counted and
the titre recorded as the number of plaque-forming units (pfu) in the sample. It
is assumed that one plaque is the progeny of one virus particle or one infectious

364

Methods for the detection of waterborne viruses

unit (but see below), hence 10 plaques counted in an assay means there were
10 infectious particles in the original sample, assuming the whole of the sam-
ple was assayed.

However, because viruses aggregate to different degrees under different
conditions the one plaque = one particle assumption may be false and the
actual numbers of particles may be greater than the plaque count would suggest.
Whether this matters in practice is debatable; the decision in setting levels for
compliance rests on the plaque count, not on the number of virus particles in
the sample.

Plaques are counted daily starting at day 2. Since viruses multiply at different
rates counting is continued after the first appearance of plaques. Echoviruses,
for example, take longer to form plaques, if they do at all. The SCA (1995)
method recommends counting plaques for 2-5 days; the US EPA (1984, revised
1987) method suggests counting should continue for 12 days or until no new
plaques appear between counts; Block and Schwartzbrod (1989) recommend
6-14 days. The duration of the reading period will depend partly on the state
of the negative control cultures; these are counted for up to 16 days, after which
the test may be discarded. Visualization is enhanced either by incorporating
neutral red into the agar overlay medium as the cells are inoculated, or by
removing the agar and staining the monolayers with Giemsa or methylene blue
(Figure 26.3). Neutral red stains only live cells, so plaques appear as clear areas
against a pink background.

The monolayer plaque assay method is favoured in many European labora-
tories, often using Giemsa as a stain to reveal plaques.

Suspended cell plaque assay

The suspended cell assay (Cooper, 1967) increases the sensitivity of the ordinary
plaque assay. More cells are used per vessel, suspended in the agar instead of
being in a layer underneath it and thus offer many more adsorption sites to
any virus present. It is also quicker in that it involves no prior establishment
of monolayers or fluid changes since cells and concentrate are added to the
culture vessels at the same time. The method is more sensitive than the mono-
layer assay, though more expensive in cells (by a factor of five to ten); Dahling
and Wright (1988) reported a five to eight times greater sensitivity using the
suspended cell assay compared with the monolayer assay and the UK Water
Virology EQA Scheme records that the suspended cell assay has three times
the sensitivity of the monolayer plaque assay. Plaques are easier to pick for
subsequent identification and the system is independent of the surface area of
the plaque assay vessel. However, the cultures only last a week (US EPA, 1984,
revised 1987) to 10 days, so slow-plaquing viruses may not be detected. Plaques
may also be more difficult to count compared with plaques in monolayers.
Despite the constraints, it is a method well suited to the enumeration of
enteroviruses in low numbers, such as are found in environmental samples.
The USEPA method recommends that the suspended cell assay should be used
where the level of indigenous virus is likely to be less than 5 pfu/ml.

365

Viruses

Figure 26.3 Plaques in BGM cell monolayer caused by polioviruses. One plaque is
the result of multiplication of one virus (or, more correctly, one virus aggregate) and so
the technique can be used to enumerate plaque-producing viruses in concentrates.

Sometimes 'plaques' are produced by toxic components in the medium, or by
an undissolved particle of agar. Recognition of these is clearly important and
is a matter of experience. Where there is doubt as to the nature of a plaque the
area is 'picked' and the medium plus cells in the vicinity is withdrawn and
inoculated into a fresh tube culture to confirm its viral origin.

Calculation of the titre of a preparation assayed by the plaque method makes
a number of assumptions. Further, the number that is calculated for the viruses
in a sample will only at best be proportional to the number really present. The
proportionality factors (usually unknown) depend on the virus type and the
calculation method. The assumptions made are (a) that the presence of one
virus is sufficient to cause cellular destruction and that the absence of cellular
destruction signifies absence of virus; and (b) that viruses are randomly dis-
tributed in the sample.

Since it is impossible to predict the number of virus clumps that will produce
a single plaque the titres are invariably expressed as plaque-forming units, or

366

Methods for the detection of waterborne viruses

pfu. Four different ways of calculating this have been proposed by Block and
Schwartzbrod (1989) in an attempt to allow for different ranges of counts in
a plaque assay; calculation based on (a) fewer than 15 pfu counted at a level
of dilution; (b) 15-100 pfu counted at one level of dilution; (c) more than 100 pfu
counted at a level of dilution; and (d) counts from several successive dilutions
are used. Different equations, all using 95% confidence intervals, are used in
each case. Comparison of the results seems to indicate that the amplitude of the
confidence intervals increases as the number of counted plaques decreases and
the best results are obtained with a large number of plaques. However, the
risks of confluence of plaques is then magnified and a truncation method of
calculation (Beytout et al. 9 1975) is recommended.

Most UK laboratories employ the suspended plaque assay method, though
few EU laboratories appear to use it.

Liquid assays

Cells under liquid media may support the growth of more enteric virus
serotypes than cells growing in or under agar. Many enteroviruses, especially
some echoviruses, do not form plaques and so will not be detected under agar;
some viruses take a long time to produce a CPE and agar cultures may have
deteriorated too far to be useful. In these cases cells growing under liquid
medium are used. Virus multiplication produces cell degeneration and often a
CPE characteristic of the infecting virus (though the CPE produced by all
enteroviruses is consistent), so some idea may be gained of the agent at hand.

Most probable number assay

In the most probable number of cytopathic units (MPNCU) assay the concen-
trate is divided into several fractions, each of which is inoculated into a separate
cell culture. Dilutions of the concentrate may be used. The number of fractions
and the size of cultures vary between laboratories but may range from five frac-
tions each being inoculated into a 75 cm 2 flask of cells to over 40 fractions each
being inoculated into a well in a microplate. Cultures are incubated for up to 21
days. Since plaques are not formed, the titre must be calculated in a different
way. The MPNCU is calculated from a collection of positive results obtained
for a series of dilutions, where a positive culture is one showing typical CPE.
Probability tables (Chang et aL 9 1958) are used for small numbers of replicates
(e.g. three or five per dilution) and there are computer programs available for
calculation of the MPNCU for larger numbers (e.g. Hurley and Roscoe, 1983).
This method is favoured by French workers and, until recently, by those in
Austria.

End point dilution assay (TCD^o)

Serial dilutions of the concentrate are inoculated into cell cultures and each

culture is scored positive or negative after incubation. The titre is calculated

367

Viruses

(e.g. by the method of Reed and Muench, 1938) as the logarithm of the dilu-
tion of virus producing a CPE in 50% of the cultures. Though the method is
simple and economic, its precision is difficult to evaluate. It is the least
favoured of the three methods described.

Choice of assay method

Comparison of assay methods in agar and under liquid media provides the

pointers indicated in Table 26.2.

It will be seen that there is no clear-cut best method. However, the plaque
assay has a greater advantage of individualizing the plaque-forming units and
providing entities (plaques) which are countable and directly related to the
number of virus particles or to 'real' viruses. For many users this is an easier
concept to grasp than the more abstract MPN or TCD 50 . Despite its limita-
tions, the plaque assay remains widely used as the unofficial standard method
of assay. When enumeration methods are compared, a mathematical relation-
ship can be deduced. Hugues (1981) showed that if 96 cultures per dilution
were inoculated in an end point dilution assay, that 1 TCD 50 = 0.69 pfu, and
that lpfu = 1.44 TCD 50 . The MPN assay has advantages only when larger
numbers of cultures are inoculated per dilution, since it is on this that the pre-
cision of the test depends. The three calculation techniques give very similar
results if used properly and the 95% confidence limits overlap. The MPN is
more reliable than the others provided the number of cultures inoculated per
dilution exceeds 30 (Block and Schwartzbrod, 1989).

Several comparative studies have been done on methods for the detection of
enteroviruses in water. Morris and Waite (1980), for example concluded that
monolayers were the least sensitive system, tube cultures were of intermediate
sensitivity (for MPN determination, though only four tubes were set up per
dilution) and the suspended cell assay was the most sensitive. BGM cells gave
the best recoveries and RD cells were variable. It is interesting to note that the

Table 26.2 Comparison of assay methods

Attribute

Liquid

Agar

Range of viruses
detected
Blind passage

Sensitivity
Subculture
Virus separation
Statistical precision

Wide range possible

Blind passage possible to
increase titres to detectable
levels

Greater sensitivity
(especially than monolayers)

Subculturing easy

Impossible to separate

virus types
Bad precision, large bias

where few replicates

used (as is usual)

Non-plaquing viruses not

detected
Faster-growing viruses in a

mixture overgrow slower

ones, which are not isolated
Sensitivity improved using

suspended cell assay
Subculturing difficult

(impossible without CPE)
Separation of viruses

possible by plaque picking
Good, especially where all

concentrate tested in one

assay

368

Methods for the detection of waterborne viruses

RD cells have been reported susceptible to coxsackievirus A strains (Block and
Schwartzbrod, 1989) though they are less sensitive than suckling mice, which
is the only other system that supports growth of this group of viruses; other
cell cultures are refractory to these viruses which have to be assayed in suckling
mice and are not therefore looked for routinely in waterborne virus detection.
Toxicity problems can occur with both systems; experience from the UK EQA
scheme suggests that toxicity occurs frequently in liquid cultures and less so in the
plaque assay. Block and Schwartzbrod (1989) however suggest the reverse is true.

Isolation and identification

The EU Bathing Water Directive requires that waters are analysed for
enteroviruses, where circumstances demand; though this is a mixed group no
requirement is made in respect of a specific identification, since the presence
of any enterovirus in a sample will indicate the presence of human faecal con-
tamination of the water and the potential presence of enter opathogenic viruses.
However, it is essential to confirm that 'plaques' detected in an assay are indeed
virus-specific and this is done by isolating the virus. Further, it may on occasion
be necessary to do additional investigations on a sample, for example where
repeated virological failures make it useful to see if the same organism is causing
the failures.

Isolation

Isolation of one specific virus type is done by picking plaques from the enu-
meration test. Dead cells constituting the plaque and the medium surrounding
them are withdrawn and inoculated into a fresh cell culture. In 2-5 days any
virus will have grown up sufficiently to be identified.

Identification

Most laboratories identify isolated enteroviruses by the serum neutralization
test (SNT). Aliquots of the isolate are incubated with different sera and each
mixture inoculated into a cell culture. Absence of CPE indicates neutralization
of the virus by the serum and thus indicates the identity of the agent.

The SNT, while flexible in that it can be used to identify groups or specific
serotypes, is a time consuming and laborious test. It is difficult to standardize
and sometimes gives equivocal results, necessitating a repeat of the procedure.
Other serological procedures have been used to identify enteroviruses isolated
from environmental sources, with varying degrees of success. Payment et al.
(1982) described an immunoassay method for typing poliovirus isolates, and

369

Viruses

Payment and Trudel (1985, 1987) described methods for detection and iden-
tification of enteric viruses, including enteroviruses, using immunoperoxidase
techniques; they also suggested using this as an enumeration technique, based
on scoring the number of peroxidase-positive cultures and calculating the titre
by the MPN method. Pandya et al. (1988) used an immunosorbent assay sys-
tem to identify coxsackievirus A isolated from sewage. Generally such methods
show advantages over the SNT, though they do not indicate whether the virus
is infectious, which is an important consideration. Further, cross-reactions
between sera against different enterovirus types has been known, limiting the
specificity of the tests. Polyclonal antibodies of sufficient specificity for enzyme
immunoassays have not been available and refinement of serological tests has
had to await detailed investigation of enterovirus structure.

A new approach to the identification of enteroviruses is the use of fluorescent
monoclonal antibodies (S.J. Mabs, personal communication). These new
reagents are available as blends or against individual serotypes, though not all
are currently available. Infected cells, grown under liquid medium and show-
ing a cytopathic effect, are pelleted, fixed and stained in wells on PTFE-coated
slides and examined under a fluorescence microscope. The process is much
quicker than the SNT, less ambiguous in its results and scanning of the test is
also quicker. Testing in parallel with the SNT indicates a comparable level of
sensitivity and specificity. Flow cytometry has been used by Abad et al. (1998)
and Baradi et al. (1998) to sort automatically rotavirus infected CaCC>2 and
MA-104 cells.

Detection of viruses by molecular biology

Enteroviruses are the group that has been investigated most in the context of
waterborne virus disease and is the viral determinant chosen for the EU Bathing
Water Directive because they can be concentrated successfully and grown eas-
ily in cell culture, not because of their gastrointestinal pathogenesis. Detection
of enteroviruses in a water sample implies that there may be other non-
detectable agents present, which either do not survive the pH shifts in the con-
centration procedure or do not multiply in the available cell cultures (or both).
Enteric viruses other than polio, coxsackie and echo types are associated
with gastrointestinal disease and therefore it is essential that methods are
developed for their detection. Serological tests have been developed but these
have not found favour generally, because they detect only viral protein and
this as such does not represent a health hazard. The use of molecular biological
tests offers a partial solution to the problem of the detection of enteropatho-
genic viruses in water. Based on knowledge of part or all of the sequence of a
viral genome, complementary sequences can be synthesized and probes or
PCR primers prepared which can be used to detect the relevant virus in con-
centrates of the sample.

370

Methods for the detection of waterborne viruses

Although the ultimate goal of such research is the direct (and rapid) detection
of pathogenic viruses such as rotavirus and SRSVs in water, techniques first
have to be validated against proven methods. This has led to the development
of molecular biology-based detection methods for enteroviruses in environ-
mental concentrates; these methods can be validated against traditional cell
culture techniques and then taken forward in the development of methods for
the detection of enteric pathogens. Richardson etal. (1991) reviewed the water
industry application of gene probes.

Gene probes were the first approach made in the molecular biological detec-
tion of enteric viruses and have been widely used (Enriquez et al. 9 1993;
Dubrou et aL 9 1991; Moore and Margolin, 1993; Margolin et al. 9 1993).
They are easy to prepare and, if used in conjunction with Southern blotting,
they produce satisfactory results. However, they lack sensitivity and, even though
some now incorporate digoxigenin instead of using radioisotopes, they have
largely been superseded.

The PCR reaction (Saiki et al. 9 1988) overcomes these shortcomings. The
reaction has been used extensively in all branches of biology and has found a
particular use in analysis of environmental materials. Detection of enteroviruses
is a practical proposition since the picornavirus group contains well-conserved
nucleotide sequences at the 5 ' end of the genome which are used to prepare
pan-enterovirus primers which are the starting reagents in the PCR. Since they
contain RNA, the nucleic acid of enteroviruses must first be reverse transcribed
and cDNA prepared before the PCR proper can be done. The whole reaction
is termed RT-PCR.

Methods and applications for detection of enteroviruses generally have been
extensively described (e.g. Rotbart, 1990, 1991a,b; Chapman etal. 9 1990; Tracy
et al. 9 1992). Numerous investigations have been done using RT-PCR to detect
enteroviruses in different environmental samples, including river and marine
recreational waters (e.g. Kopecka et al. 9 1993; Wyn-Jones et al. 9 1995; Gilgen
etal. 9 1995), ground waters (Abbaszadegan etal., 1993; Regan and Margolin,
1997) and sludge-amended field soils (Straub et al. 9 1995). The technique has
been extended to cover other virus groups present in water including adeno-
viruses (Puig et al. 9 1994), hepatitis A (Graff et al. 9 1993), astrovirus (Marx
etal. 9 1995) and rotavirus (Gajardo et al. 9 1995).

Though the technique is straightforward in principle it is not without prac-
tical problems. Chief among these is the presence of fulvic and humic acids in
the concentrates which inhibit the RT and/or polymerase reactions. Different
solutions have been found to remove these but most rely on adsorption of the
extracted RNA to silica (e.g. Shieh et al. 9 1995). Pallin et al. (1997) devised a
method for recovering all the virus in a concentrate into a single PCR tube,
which allowed direct comparisons of sensitivity with cell culture methods where
the whole of the concentrate is tested at one time.

Refinement of the RT-PCR and restriction enzyme analysis of amplicons has
permitted the differentiation of virus types within the enterovirus group.
Hughes et al. (1993) compared the nucleotide sequences of six coxsackievirus
B4 isolates from the aquatic environment with those of four CB4 isolates from

371

Viruses

clinical specimens and found that the isolates fell into two distinct groups not
related to their origin. Wyn -Jones, Pallin and Lee (unpublished results) devised
a method for identifying enterovirus concentrated from bathing waters based
on restriction enzyme analysis which assigned any isolate to at least group
level and many to serotype, using just four enzymes, the groupings correlating
with those determined by immune electron microscopy and Sellwood et al.
(1995) reported a system using RFLP analysis to discriminate between wild
and vaccine-like strains of poliovirus. Egger et al. (1995) devised a multiplex
PCR for the differentiation of polioviruses from non-polioviruses, which made
an important step in the accumulation of public health information. None of
these investigations would have been possible without RT-PCR.

RT-PCR detects viral RNA, not infectious virus and there is thus a theoret-
ical objection to its use in a public health context. However, if virus-specific
RNA is detected, even if it came from a non-infectious particle, that signifies
there are likely to be infectious particles present also and that the water in
which the RNA was detected is indeed polluted with enteroviruses. Further, it
is well known that RNA does not survive in the environment so if a positive
result is obtained by RT-PCR it indicates that it can only have come from
intact, i.e. infectious virus particles. There is an important use for RT-PCR in
the screening of samples for enteroviruses; negative ones can be discarded and
positives investigated further for presence of infectious virus. The technique
has been further employed by the combination of cell culture with RT-PCR.
Reynolds et al. (1996) and Murrin and Slade (1997) inoculated BGM cultures
with concentrates and tested the supernatants at intervals up to 10 days. Virus
was detectable by RT-PCR as early as one day post-inoculation, instead of
more than 3 days by normal visualization of CPE. This thus allows a more
rapid analysis of waters to be carried out.

It will be obvious that RT-PCR detection of enteroviruses is a useful adjunct
to the technology for analysing bathing waters for enteroviruses and its use
should be considered when formulating strategies and monitoring schemes.

References

Abad, EX., Pinto, R.M. and Bosch, A. (1998). Flow cytometry detection of infectious

rotaviruses in environment and clinical samples. Appl Environ Microbiol, 64:

2392-2396.
Abbaszadegan, M., Huber, M.S., Gerba, C.P. et al. (1993). Detection of enteroviruses in

ground water by PCR. Appl Environ Microbiol, 65: 444-449.
American Public Health Association. (1980). Standard methods for examination of water,

15th edn.
AranhaCreado, H., Oshima, K., Jafari, S. et al. (1997). Virus retention by a hydrophilic

triple-layer PVDF microporous membrane filter./ Pbarm Sci Technol, 51: 119-124.
Baradi, C.R.M., Emslie, K.R., Vesey, G. et al. (1998). Development of a rapid and sensitve

quantitative assay for rotavirus based on flow cytometry./ Virol Metb, 74: 31-38.
Barron, A.L., Olshevsky, C. and Cohen, M.M. (1970). Characteristics of the BGM line of

cells from African green monkey kidney. Arcbiv Gesamte Virus forscbung, 32: 389-392.

372

Methods for the detection of waterborne viruses

Belfort, G., Paluszek, A. and Sturman, L.S. (1982). Enterovirus concentration using auto-
mated hollow fiber ultrafiltration. Water Sci Technol, 14: 257-272.

Bergh, O., Borsheim, K.Y., Bratbak, G. et al. (1989). High abundance of viruses found in
aquatic environment. Appl Environ Microbiol, 340: 467-468.

Beytout, D., Laveran, H. and Reynaud, M.P. (1975). Methode practique d'evaluation
numerique applicable aux techniques miniaturisees de titrage en plages. Ann Biol Clin,
33: 379-384.

Bitton, G., Pancorbo, O. and Gifford, G.E. (1976). Factors affecting the adsorption of
poliovirus to magnetite in water and wastewater. Water Res, 10: 973-980.

Block, J.C. and Schwartzbrod, L. (1989). Viruses in Water Systems. Detection and Identi-
fication. New York: VCH Publishers Inc.

Bosch, A., Pinto, R.M., Blanch, A.R. and Jofre, J.T. (1988). Detection of human rotavirus
in sewage through two concentration procedures. Water Res, 22: 343-348.

Chang, S.L., Berg, K.A., Busch, K. et al. (1958). Application of the 'most probable number'
method for estimating concentrations of animal viruses by the tissue culture technique.
Virology, 6: 27-31.

Chapman, N.M., Tracy, S., Gauntt, C.J. et al. (1990). Molecular detection and identifica-
tion of enteroviruses using enzymatic amplification and nucleic acid hybridization.
/ Clin Microbiol, 28: 843-850.

Chaudhiri, M. and Sattar, S.A. (1986). Enteric virus removal from water by coal-based sor-
bents: development of low -cost water filters. Water Sci Technol, 18: 77-82.

Cooper, P.D. (1967). The plaque assay of animal viruses. Adv Virus Res, 8: 319-378.

Dahling, D.R. and Wright, B.A. (1986). Optimization of the BGM cell line and viral assay
procedures for monitoring viruses in the environment. Appl Environ Microbiol, 51:
790-812.

Dahling, D.R. and Wright, B.A. (1988). Optimisation of suspended cell method and com-
parison with cell monolayer technique for virus assays./ Virol Meth, 20: 169-179.

Dahling, D.R., Berg, G. and Berman, D. (1974). BGM, a continuous cell line more sensitive
than primary rhesus and African green kidney cells for the recovery of viruses from
water. Hltb Lab Sci, 11: 275-282.

Dahling, D.R., Phirke, P.M., Wright, B.A. et al. (1985). Use of bituminous coal as an alter-
native technique for the field concentration of waterborne viruses. Appl Environ
Microbiol, 49: 1222-1225.

Divizia, M., de Filippis, P., di Napoli, A. et al. (1989a). Isolation of wild-type hepatitis A
virus from the environment. Water Res, 23: 1155-1160.

Divizia, M., Santi, A.L. and Pana, A. (1989b). Ultrafiltration: an efficient second step for
hepatitis A and poliovirus concentration./ Virol Meth, 23: 55-62.

Dubrou, S., Kopecka, H., Lopez-Pila, J.M. et al. (1991). Detection of hepatitis A virus and
other enteroviruses in wastewater and surface water samples by gene probe assay. Water
Sci Technol, 24: 267-272.

Egger, D., Pasamontes, L., Ostermayer, M. et al. (1995). Reverse transcription multiplex
PCR for differentiation between polio- and enteroviruses from clinical and environ-
mental samples./ Clin Microbiol, 33: 1442-1447.

Enriquez, C.E., Abbaszadegan, M., Pepper, I.L. et al. (1993). Poliovirus detection in water
by cell culture and nucleic acid hybridization. Water Res, 27: 1113-1118.

Farrah, S.R., Gerba, C.P., Wallis, C. et al. (1976). Concentration of viruses from large vol-
umes of tap water using pleated membrane filters. Appl Environ Microbiol, 31: 221-226.

Farrah, S.R., Girard, M.A., Toranzos, G.A. et al. (1988). Adsorption of viruses to diatom-
aceous earth modified by in situ precipitation of metallic salts. TLent Ges Hyg, 34:
520-521.

Fogh, J. (1977). Absence of Hela cell contamination in 169 cell lines derived from human
tumours./ Natl Cancer Inst, 58: 209-220.

Gajardo, R., Bouchrit, N., Pinto, R.M. et al. (1995). Genotyping of rotaviruses isolated
from sewage. Appl Environ Microbiol, 61: 3460-3462.

Gerba, C.P. (1987). Recovering viruses from sewage, effluents and water. In Methods for
Recovering Viruses from the Environment, Berg, G. (ed.). Boca Raton, FL: CRC Press,
pp. 1-23.

373

Viruses

Gilgen, M., Germann, D., Luethy, J. et al. (1997). Three-step isolation method for sensitive
detection of enterovirus, rotavirus, hepatitis A virus and small round-structured viruses
in water samples. Int J Food Microbiol, 37: 189-199.

Gilgen, M., Wegmuller, B., Burkhalter, P. et al. (1995). Reverse transcription PCR to detect
enteroviruses in surface water. Appl Environ Microbiol, 61: 1226-1231.

Grabow, W.O.K., Nupen, E.M. and Bateman, B.W. (1984). South African research on
enteric viruses in drinking water. Monogr Virol, 15: 146-155.

Graff, J., Ticehurst, J. and Flehmig, B. (1993). Detection of hepatitis A virus in sewage by
antigen capture polymerase chain reaction. Appl Environ Microbiol, 59: 3165-3170.

Guttman-Bass, N. and Nasser, A. (1984). Simultaneous concentration of four enteroviruses
from tap, waste and natural water. Appl Environ Microbiol, 47: 1311-1315.

Guttman-Bass, N. and Catalano-Sherman, J. (1986). Humic acid interference with virus
recovery by electropositive microporous filters. Appl Environ Microbiol, 52: 556-561.

Homma, A., Sobsey, M.D., Wallis, C. et al. (1973). Virus concentration from sewage.
Water Res, 7:945-952.

Hovi, T., Stenvik, M., Partanen, H. et al. (2001). Polio virus surveillance by examining
sewage specimens. Quantitative recovery of virus after introduction into sewerage at
remote upstream location. Epidemiol Infect, 127: 101-106.

Hsiung, G.D. and Melnick, J.L. (1957). Morphologic characteristics of plaques produced
on monkey kidney monolayer culture by enteric viruses (poliomyelitis, coxsackie and
ECHO groups). Immunology, 78: 128-131.

Huang, P.W., Laborde, D., Land, V.R. et al. (2000). Concentration and detection of cali-
civiruses in water samples by reverse transcription-PCR. Appl Environ Microbiol, 66:
4383-4388.

Hughes, M.S., Hoey, E.M. and Coyle, P.V. (1993). A nucleotide sequence comparison of
Coxsackievirus B4 isolates from aquatic samples and clinical specimens. Epidemiol
Infect, 110: 389-398.

Hugues, B. (1981). Cited from Block, J.C. and Schwartzbrod, L. (19 89). Viruses in Water
Systems; Detection and Identification. New York: VCH Publishers Inc.

Hugues, B., Andre, M. and Champsaur, H. (1991). Virus concentration from waste water:
glass wool versus glass powder. Biomed Lett, 46: 103-107.

Hurley, M.A. and Roscoe, M.E. (1983). Automated statistical analysis of microbial enu-
meration by dilution series. / Appl Bacteriol, 55: 159-164.

Jakubowski, W., Hoff, J.C, Anthony, N.C. et al. (1974). Epoxy-fiberglass adsorbent for
concentrating viruses from large volumes of potable water. Appl Microbiol, 28: 501.

Joret, J.C, Block, J.C, Lucena-Gutierrez, F. et al. (1980). Virus concentration from sec-
ondary wastewater: Comparative study between epoxy fibreglass and glass powder
adsorbents. Europ J Appl Microbiol Biotech, 10: 245-252.

Jothikumar, N., Khanna, P., Paulmurugan, R. et al. (1995). A simple device for the con-
centration and detection of enterovirus, hepatitis E virus and rotavirus from water sam-
ples by reverse transcription-polymerase chain reaction./ Virol Metb, 55: 410-415.

Katzenelson, E., Fattal, B. and Hostovesky, T. (1976). Organic flocculation: an efficient sec-
ond step concentration method for the detection of viruses in tap water. Appl Environ
Microbiol, 32: 638-639.

Keswick, B.H., Pickering, L.K., DuPont, H.L. et al. (1983). Organic flocculation: an effi-
cient second step concentration method for the detection of viruses in tap water. Appl
Environ Microbiol, 46: 813-816.

Kittigul, L., Khamoun, P., Sujirarat, D. et al. (2001). An improved method for concentrat-
ing rotavirus from water samples. Mem Inst Oswaldo Cruz, 96: 815-821.

Kopecka, H., Dubrou, S., Prevot, J. et al. (1993). Detection of naturally-occurring
enteroviruses in waters by reverse transcription, polymerase chain reaction, and
hybridization. Appl Environ Microbiol, 59: 1213-1219.

Lakhe, S.B. and Parhad, N.M. (1988). Concentration of viruses from water on bituminous
coal. Water Res, 22: 635-640.

Lund, E. and Hedstrom, CE. (1966). The use of an aqueous polymer phase system for
enterovirus isolations from sewage. Am J Epidemiol, 84: 287-291.

374

Methods for the detection of waterborne viruses

Ma, J.-F., Naranjo, J. and Gerba, C.P. (1994). Evaluation of MK filters for recovery of

enteroviruses from tap water. Appl Environ Microbiol, 60: 1974-1977.
Mack, W.N., Yue-Shoung, L. and Coohon, D.B. (1972). Isolation of poliomyelitis virus

from a contaminated well. Public Hltb Rep, 87: 271-274.
Margolin, A.B., Gerba, C.P., Richardson, K.J. et al. (1993). Comparison of cell culture and

a poliovirus gene probe assay for the detection of enteroviruses in environmental water

samples. Water Sci Tecbnol, 27: 311-314.
Marx, F.E., Taylor, M.B. and Grabow, W.O.K. (1995). Optimization of a PCR method for

the detection of astro virus type 1 in environmental samples. Water Sci Tecbnol, 31:

359-362.
Melnick, J.L., Safferman, R., Rao, V.C. et al. (1984). Round robin investigations of methods

for the recovery of poliovirus from drinking water. Appl Environ Microbiol, 47: 144-150.
Metcalf, T.G., Wallis, C. and Melnick, J.L. (1974). Environmental factors influencing isol-
ation of enteroviruses from polluted surface waters. Appl Environ Microbiol, 27: 920.
Moore, N. and Margolin, A.B. (1993). Evaluation of radioactive and non-radioactive gene

probes and cell culture for detection of poliovirus in water samples. Appl Environ

Microbiol, 59: 3145-3146.
Morris, R. (1985). Detection of enteroviruses: an assessment of ten cell lines. Water Sci

Tecbnol, 17: 81-88.
Morris, R. and Waite, W.M. (1980). Evaluation of procedures for the recovery of viruses

from water - I Concentration systems. Water Res, 14: 791-793.
Murphy, A.M., Grohmann, G.S. and Sexton, M.F.H. (1983). Infectious gastroenteritis in

Norfolk Island and recovery of viruses from drinking water. / Hyg, 91: 139-146.
Murrin, K. and Slade, J. (1997). Rapid detection of viable enteroviruses in water by tissue

culture and semi-nested polymerase chain reaction. Water Sci Tecbnol, 35: 429-432.
Muscillo, M., Carducci, A., la Rosa, G. et al. (1997). Enteric virus detection in Adriatic sea-
water by cell culture, polymerase chain reaction and polyacrylamide gel electrophoresis.

Water Res, 31: 1980-1984.
Myrmel, M., Rimstad, E. and Wasteson, Y. (2000). Immunomagnetic separation of a

Norwalk-like virus (genogroup I) in artificially contaminated environmental water sam-
ples. Int J Food Microbiol, 62: 17-26.
Nupen, E.M., Basson, N.C. and Grabow, W.O.K. (1981). Efficiency of ultrafiltration for

the isolation of enteric viruses and coliphages from large volumes of water in studies on

wastewater reclamation. Water Pollution Res, 13: 851-863.
Pallin, R., Place, B.M., Lightfoot, N.F. et al. (1997). The detection of enteroviruses in large

volume concentrates of recreational waters by the polymerase chain reaction. / Virol

Metb, 57: 67-77.
Pandya, G., Jana, A.M., Tuteja, U. et al. (1988). Identification of Group A Coxsackie-
viruses from sewage samples by indirect enzyme-linked immunosorbent assay. Water

Res, 22: 1055-1057.
Patel, J.R., Daniel, J. and Mathan, V.I. (1985). A comparison of the susceptibility of three

human gut tumour-derived differentiated epithelial cell line, primary monkey kidney

cells and human rhabdomyosarcoma cell line to 66-prototype strains of human

enteroviruses./ Virol Metb, 12: 209-216.
Payment, P. and Trudel, M. (1979). Efficiency of several micro -fiber glass filters for

recovery of poliovirus from tap water. Appl Environ Microbiol, 38: 365-368.
Payment, P. and Trudel, M. (1985). Immunoperoxidase method with human immune

serum globulin for broad spectrum detection of cultivable viruses in environmental

samples. Appl Environ Microbiol, 50: 1308-1310.
Payment, P. and Trudel, M. (1987). Detection and quantitation of human enteric viruses in

wastewaters: increased sensitivity using a human immune serum immunoglobulin

immunoperoxidase assay on MA-104 cells. Can J Microbiol, 33: 568-570.
Payment, P., Tremblay, C. and Trudel, M. (1982). Rapid identification and serotyping of

poliovirus isolates by an immunoassay./ Virol Metb, 5: 301-308.
Pietri, C. and Hugues, B. (1985). The effect of the cell system on the quantification of

viruses present in sewage eluates. Microbios Lett, 30: 67-72 (French).

375

Viruses

Poynter, S.F.B., Jones, H.H. and Slade, J.S. (1975). Virus concentration by means of soluble
ultrafilters. In Methods for Microbiological Assay, Board, R.G. and Lovelock, D.W.
(eds). London: Academic Press, pp. 65-74.

Puig, M., Jofre, J., Lucena, F. (1994). Detection of adenoviruses and enteroviruses in pol-
luted water by nested PCR amplification. Appl Environ Microbiol, 60: 2963-2970.

Ramia, S. and Sattar, S.A. (1979). Second-step concentration of viruses in drinking and sur-
face waters using polyethylene glycol extraction. Can] Microbiol, 25: 587.

Rao, C, Sullivan, R., Read, R.B. et al. (1968). A simple method for concentrating and
detecting viruses. / Am Water Works Assoc, 60: 1288-1294.

Reed, L.J. and Muench, H. (1938). A simple method of estimating fifty percent endpoints.
AmerJ Hyg, 27: 493-495.

Regan, P.M. and Margolin, A.B. (1997). Development of a nucleic acid capture probe with
reverse transcriptase-polymerase chain reaction to detect poliovirus in groundwater.
J Virol Meth, 64: 65-72.

Reynolds, C.A., Gerba, C.P. and Pepper, I.L. (1996). Detection of infectious enterovirus by
an integrated cell culture-PCR procedure. Appl Environ Microbiol, 62: 1424-1427.

Richardson, K.J., Stewart, M.H. and Wolfe, R.L. (1991). Application of gene probe tech-
nology to the water industry. JAW WA, 83: 71-81.

Rotbart, H.A. (1990). Enzymatic RNA amplification of the enteroviruses. / Clin
Microbiol, 28: 438-442.

Rotbart, H.A. (1991a). New methods for rapid enteroviral diagnosis. Prog Med Virol, 38:
96-108.

Rotbart, H.A. (1991b). Nucleic acid detection systems for enteroviruses. Clin Microbiol
Rev, 4: 156-168.

Rotem, Y., Katzenelson, E. and Belfort, G. (1979). Virus concentration by capillary ultra-
filtration./ Environ Eng-ASCE, 5: 401-407.

Saiki, R.K., Gelfand, D.H., Stoffel, S. et al. (1988). Primer-directed enzymatic amplification
of DNA with a thermostable DNA polymerase. Science, 239: 487-491.

Sarrette, B., Danglot, B. and Vilagines, R. (1977). A new and simple method for the recu-
peration of enteroviruses from water. Water Res, 11: 355-358.

Sattar, S.A. and Ramia, S. (1979). Use of talc-selite layers in the concentration of
enteroviruses from large volumes of potable waters. Water Res, 13: 1351-1353.

Sattar, S.A. and Westwood, J.C.N. (1978). Viral pollution of surface waters due to chlorin-
ated primary effluents. Appl Environ Microbiol, 36: 427-431.

Schimdt, N.J., Ho, H.H. and Lennette, E.H. (1976). Comparative sensitivity of the BGM
cell line for isolation of enteric viruses, tilth Sci, 13: 115-117.

Schwab, K.J., Leon, R. and Sobsey, M.D. (1996). Immunoaffinity concentration and
purification of waterborne enteric viruses for detection by reverse transcriptase PCR.
Appl Environ Microbiol, 62: 2086-2094.

Schwartzbrod, L. and Lucena-Gutierrez, F. (1978). Concentration des enterovirus dans les
eaux par adsorption sur poudre de verre: proposition d'un appareillage simplifie.
Micro bia, 4: 55-58.

Sellwood, J., Litton, P.A., McDermott, J. et al. (1995). Studies on wild and vaccine strains
of poliovirus isolated from water and sewage. Water Sci Technol, 31: 317-321.

Shieh,Y.-S.C, Wait, D., Tai, L. et al. (1995). Methods to remove inhibitors in sewage and
other faecal wastes for enterovirus detection by the polymerase chain reaction. / Virol
Meth, 54: 51-66.

Sobsey, M.D. and Glass, J.S. (1980). Poliovirus concentration from tap water with elec-
tropositive adsorbent filters. Appl Environ Microbiol, 40: 201-210.

Sobsey, M.D. and Hickey, A.R. (1985). Effects of humic and fulvic acids on poliovirus
concentration from water by microporous filtration. Appl Environ Microbiol, 49:
259-264.

Sobsey, M.D. and Jones, B.L. (1979). Concentration of poliovirus from tapwater using
positively-charged microporous filters. Appl Environ Microbiol, 37: 588-595.

Standing Committee of Analysts (SCA). (1995). Methods for the isolation and identifica-
tion of human enteric viruses from waters and associated materials. In Methods for the
examination of waters and associated materials. London: HMSO.

376

Methods for the detection of waterborne viruses

Straub, T.M., Pepper, I.L. and Gerba, C.P. (1995). Comparison of PCR and cell culture for
detection of enteroviruses in sludge-amended field soils and determination of their
transport. Appl Environ Microbiol, 61: 2066-2068.

Tracy, S., Chapman, N.M. and Pistillo, J.M. (1992). Detection of human enteroviruses
using the polymerase chain reaction. In Frontiers of Virology, 1: Diagnosis of Human
Viruses by Polymerase Chain Reaction Technology, Becker, Y. and Darai, G. (eds).
New York: Springer- Verlag, pp. 331-344.

Tsai, Y.L., Sobsey, M.D., Sangermano, L.R. et al. (1993). Simple method of concentrating
enteroviruses and hepatitis A virus from sewage and ocean water for rapid detection by
reverse transcriptase-polymerase chain reaction. Appl Environ Microbiol, 59: 3488-3491.

United States Environmental Protection Agency. (1984). USEPA/APHA Standard Methods
for the examination of water and wastewater.

Urase, T., Yamamoto, K. and Ohgaki, S. (1993). Evaluation of virus removal in membrane
separation processes using coliphage Q beta. In Development and Water Pollution
Control in Asia, Bhamidimarri, R. et al. (eds), Oxford: Pergamon, pp. 9-15.

Vilagines, P.H., Sarrette, B. and Vilagines, R. (1988). Detection en continu du poliovirus
dans des eaux de distribution publique. CR Acad Sci Paris, 307, serie III: 171-176.

Vilagines, P., Sarrette, B., Husson, G. et al. (1993). Glass wool for virus concentration at
ambient water pH level. Water Sci Technol, 27: 299-306.

Wallis, C. and Melnick, J.L. (1967a). Concentration of viruses from sewage by adsorption
on to Millipore membranes. Bull Wld Hlth Org, 36: 219-225.

Wallis, C. and Melnick, J.L. (1967b). Concentration of viruses on aluminium and calcium
salts. Am J Epidemiol, 85: 459-468.

Wallis, C. and Melnick, J.L. (1967c). Concentration of viruses on membrane filters.
/ Virol, 1: 472-477.

Wallis, C.j Melnick, J.L. and Fields, J.E. (1970). Detection of viruses in large volumes of
natural waters by concentration on insoluble polyelectrolytes. Water Res, 4: 787-796.

Watt, P.M., Johnson, D.C. and Gerba, C.P. (2002). Improved method for concentration of
Giardia, Cryptosporidium and poliovirus from water./ Environ Sci Hlth, 37: 321-330.

Wellings, F.M., Lewis, A.L. and Mountain, C.W. (1976). Demonstration of solids-
associated virus in wastewater and sludge. Appl Environ Microbiol, 31: 354-360.

Wolfaardt, M., Moe, C.L. and Grabow, W.O.K. (1995). Detection of small round-
structured viruses in clinical and environmental samples by polymerase chain reaction.
Water Sci Technol, 31: 375-382.

Wyn-Jones, A.P. and Sellwood, J. (1998). Review of methods for the isolation, concentra-
tion, identification and enumeration of enteroviruses. Project WW-11B Research and
Development - Bathing Water Policy, UK Water Industry Research Ltd.

Wyn-Jones, A.P. and Sellwood, J. (2001). Enteric viruses in the aquatic environment,
invited review. / Appl Microbiol, 91: 945-962.

Wyn-Jones, A. P., Pallin, R., Sellwood, J. et al. (1995). Use of the polymerase chain reaction
for the detection of enteroviruses in river and marine recreational waters. Water Sci
Technol, 31: 337-344.

377

Adenovirus

Basic microbiology

One of the earliest reports of the isolation of adenoviruses was when secretions
from patients with respiratory infections were inoculated on to human cell cul-
tures resulting in cell death (Hilleman and Werner, 1954). The term 'adeno' was
derived from the adenoid tissue which was found to be associated with persist-
ent adenovirus infection (Rowe etaL, 1953). The human strains of adenoviruses
are classified in the Mastadenovirus genus of the Adenovirus family, which also
includes many other mammalian adenovirus strains. Forty-nine serotypes of
human adenoviruses are recognized; these are based on neutralization studies
using antisera raised against epitopes of the hexon protein and the knob portion
of the fibre protein of the virion capsid. These serotypes form six subgroups,
designated A to F, on the basis of their ability to agglutinate red blood cells.

The icosahedral structure of the virus particle is very distinct when visual-
ized by electron microscopy (Figure 27.1). The virions are 70-100 nm in
diameter with no membrane. The protein coat or capsid comprises 252 capso-
meres of which 240 are hexons (surrounded by six other capsomeres) and 12
are pen tons (surrounded by five capsomeres). The pentons form the angles of
the icosaherdron and each has a gylcoprotein fibre extending from it. Four
proteins and a genome of double-stranded DNA, approximately 36kb in

Viruses

Figure 27.1 Transmission electron micrograph of adenovirus. (Courtesy of
Dr G. William Gary, Jr, CDC, USA.)

length, form the core of each virion. These symmetrical particles can form
extensive arrays or inclusions within the nucleolus of infected cells.

Adenoviruses are stable at pH 3, resistant to intestinal enzymes and replicate
in the epithelial cells of the intestine. As adenoviruses do not possess a mem-
brane, ether and chloroform do not inactivate the virus. The limited data that
are available on the inactivation of adenoviruses by chlorine demonstrates that
the virus is killed but it is not clear if the virus is more or less resistant than the
enteroviruses. Adenoviruses are more resistant to the action of UV light than
enteroviruses (Gerba et aL 9 2002). Sewage seeded with adenovirus 40 and 41
serotypes had detectable virus after 60 days held at 15°C and 4°C which was
longer than for enterovirus (Enriquez and Gerba, 1995; Enriquez et aL 9 1995).

Origin of the organism

380

The origin of human adenovirus strains in the environment will be human
sewage. In common with all viruses the adenoviruses are obligate intracellular
parasites, therefore no replication can occur in the environment.

Adenovirus

Clinical features and virulence

Adenoviruses infect the respiratory tract, the eye and the gastrointestinal tract.
Asymptomatic infections are common and shedding of virus from pharynx and
gut may be prolonged. The respiratory illness caused by adenoviruses is often
associated with pharyngitis, fever, cough and cold-like symptoms. Tonsillitis
may be involved as well as conjunctivitis (Wadell, 2000). As well as these
common symptoms in children, an acute respiratory disease (ARD) has been
recognized in young adults. It occurs particularly in young military recruits
undergoing training and can be severe. Adenovirus infection has been associ-
ated with Bordetella pertussis infection and the symptoms of whooping cough.

Follicular conjunctivitis can occur as part of the general respiratory illness
but also as an eye infection without generalized symptoms. It is usually a mild,
short infection from which there is a full recovery. Epidemic keratoconjunc-
tivitis is a more aggressive disease involving lymphadenopathy, pain and
swelling around the eye. The residual effects may include corneal opacity and
last for some years. It was easily spread in specific areas of work and became
known as 'shipyard eye'.

Gastroenteritis in young children is caused by Group F adenoviruses, com-
prising serotypes 40 and 41. Several hundred reports are sent to the Health
Protection Agency Communicable Disease Surveillance Centre throughout
each year. Although many adenovirus serotypes may replicate in the gut and
are detectable in faeces, it is now recognized that only serotypes 40/41 cause
gastroenteritis (Krajden et aL 9 1990; Lew et al., 1991) unless part of a wider
systemic illness. Several serotypes that replicate in the intestine have been
linked to a syndrome termed 'intussusception'. Part of the lower bowel tele-
scopes into itself thereby causing a blockage and may be caused by adenovirus
replication in the associated mesenteric glands.

It is not understood why the severity of disease varies between different
serotypes of adenovirus. Nor has the mechanism behind latency and long-
term shedding of virus been elucidated.

Pathogenicity

Intranuclear inclusions consisting of an array of virions may be found in
some host cells such as alveolar cells when pneumonia has developed. The
infection shuts off the expression of host cell messenger RNA which, in
conjunction with the production of excess virus protein, leads to cell death.
The different subgenera (A-F) may have a range of cell tropisms, e.g. adeno-
virus 8 usually infects the eye, while 40 and 41 infect the gastrointestinal tract.
However, most strains gain entry to the host via the mouth or eye; initial
infection is in non-ciliated epithelium cells and glandular tissue (adenoids and
tonsils) in the nasopharynx. Incubation periods vary according to the

381

Viruses

serotypes and the site of infection; respiratory illness and gastrointestinal dis-
ease can result after a few days, while eye infections may have an incubation
period of 2-6 days.

Immune responses include the production of 'complement-fixing anti-
bodies' or immunofluorescent antibody which are group-specific (raised
against the soluble hexon antigen) and of neutralizing antibodies which are
type specific (raised against a mixture of hexon and fibre antigens) and are
measured by a neutralizing or haemagglutination test. All the antibody levels
wane in time but will rise after further infection. Immunity to group F viruses
appears to be life-long as the majority of reported infections occur in very
young children.

Transmission and epidemiology

The transmission route for group F, serotypes 40/41 is faecal-oral spread as
the virus is shed in faecal material in large numbers. Other serotypes are
spread by aerosols of respiratory droplets and mechanically into the conjunc-
tiva of the eyes. Swimming pool waters have been suggested to be the vehicle
of transmission for serotypes 3 or 7 (Foy et al. 9 1968; Martone et al. 9 1980;
Turner et al. 9 1987; Papapetropolou and Vantarakis, 1995; Harley et al. 9
2001) resulting in eye infections. Problems with the chlorination of these
pools have usually been identified. Outbreaks of eye infections, often serotype 8,
have also occurred in hospital eye departments after the use of contaminated
equipment on a series of patients (Jernigan et al. 9 1993).

Infection with the adenoviruses that cause respiratory infections is wide-
spread and common with and without obvious symptoms. Group F adeno-
virus infection causes diarrhoea in babies and immunity to these serotypes can
reach 50% or more by the age of 4 years.

Treatment

There is no specific treatment for enteric adenovirus infection and symptoms
are rarely sufficiently severe to require rehydration.

Distribution in the environment

Adenoviruses do not form plaques and thus have not been detected in the
widely used BGM cell plaque assays but do grow well in human cell culture

382

Adenovirus

and monkey kidney cell culture under liquid medium. They form an easily
recognizable cytopathic effect which can be confirmed as adenovirus by IF or
ELISA. Many of the culturable serotypes were isolated from sewage by
Sellwood et al. (1981) and Irving and Smith (1981), however, this method of
detection is laborious and few reports have been published.

Molecular methods such as PCR have allowed all enteric adenoviruses to be
detected with greater ease from sewage, other water matrices and shellfish.
Primers based on the hexon gene detect both culturable and group F serotypes
of human origin and other mammals (Allard et al., 1990; Puig et al., 1994),
whereas primers based on the fibre gene can be directed at group F specifically
(Tiemessen and Nel, 1996). An alternative method is to use RFLP to distin-
guish the different sized PCR products of the different subgenuses (Kidd et al.,
1996). Genthe et al. (1995) used not PCR but gene probes to identify adeno-
viruses in a high proportion of sewage and other waters in South Africa.

The Spanish group of Puig et al. (1994) and Pina et al. (1998) have detected
adenoviruses in sewage, polluted water and shellfish throughout the year in
most samples investigated. Sewage from the south of England also contains
adenovirus in most samples all year round (Sellwood unpublished results).
This frequency of detection has prompted several calls for the virus group
to be used as a marker of human virus contamination in water. A European
collaborative project comprising Spain, UK, Sweden and Greece identified
adenoviruses in shellfish from each of the countries using hexon-based primers
and a quantitative PCR (Formiga-Cruz et al., 2002). Adenoviruses including
group F serotypes were detected in sewage from San Paulo in Brazil (Santos
et al., 2002) and in the UK (unpublished results Sellwood and Wyn-Jones).
Drinking water was shown to contain adenoviruses including some 'closely
related' to adenovirus 40/41 in nearly 40% of Korean tap water samples by
Lee et al. (2002). The disinfection system was found not to be effective and
therefore level of chlorination of the water was low.

Integrated cell culture-PCR may be used to identify infectious virus by an
initial incubation period in cell culture and then, to increase the sensitivity
using PCR as the detection method. This was used by Chapron et al. (2000)
to detect adenoviruses 40 and 41 (among others) in 29 surface water samples;
14 were adenovirus 40/41 positive, of which 11 contained infectious virus.
Grabow et al. (2001) has detected adenovirus in 4% of drinking water sup-
plies in South Africa using a similar approach.

Adenoviruses were detected in the air filters of a building's ventilation sys-
tem where an outbreak of adenovirus respiratory illness had occurred
(Echavarria et al., 2000).

Waterborne outbreaks

Kukkula et al. (1997) identified adenoviruses in the sewage contaminated water
that had caused an outbreak of norovirus infection as well as the norovirus itself.

383

Viruses

Most other reports have been swimming pool associated, as mentioned above,
particularly when the chlorination system had failed (Harley et al. 9 2001).

Risk assessment

Health effects: occurrence of illness, degree of morbidity and mortality, prob-
ability of illness based on infection:

• Adenoviruses mostly cause respiratory illness; however, depending on the
infecting serotype, they may also cause gastroenteritis (from group F), con-
junctivitis, cystitis and rash illness. In healthy people, illness from adeno-
virus is generally mild.

• In a summary of studies of sources of diarrhoeal illness, enteric adenovirus
was identified as a sole cause around 5-10% of the time, representing
between and 40% of all serotypes in adenovirus cases; generally, 0-20%
prevalence of faecal shedding in healthy hosts.

• Asymptomatic infections are common and shedding of virus from pharynx
and gut may be prolonged. About 50% of childhood enteric adenovirus
infections result in illness; the probability of illness increases when the
infection is respiratory. Immunity to group F (enteric) viruses appears to be
life-long as the majority of reported infections occur in very young children.

Exposure assessment: routes of exposure and transmission, occurrence in
source water, environmental fate:

• Adenovirus has not been positively associated with any outbreaks from
drinking water. A number of outbreaks of conjunctivitis have occurred as a
result of exposure to recreational water.

• Routes of exposure include direct (touching), aerosolization and ingestion.
The main route of exposure for enteric adenovirus infection is most likely
faecal-oral.

• There is limited evidence of secondary spread except among very young
(preschool age) children.

• Water is contaminated by human faecal matter/sewage, so the occurrence in
source water is dependent on level of human faecal contamination. In pub-
lished reports, concentrations in river water have ranged from to 25 pfu/1
and from 70 to 3200 cpu/1.

Risk mitigation: drinking-water treatment, medical treatment:

• The limited data that are available suggest that adenoviruses are inactivated
by free chlorine, but its level of resistance compared to other viruses is not
clear. Adenoviruses are more resistant to the action of UV light than
enteroviruses.

• The illness is self-limiting. Rehydration therapy may be necessary in some
children with diarrhoea.

384

Adenovirus

References

Allard, A., Girones, R., Juto, P. et al. (1990). PCR for detection of adenovirus in stool
samples./ Clin Microbiol, 28: 2659-2667.

Chapron, CD., Ballester, N.A., Fontaine, J.H. et al. (2000). Detection of astroviruses,
enteroviruses and adenovirus types 40 and 41 in surface waters collected and evaluated
by the Information Collection Rule and an Integrated Cell Culture-Nested PCR proced-
ure. Appl Environ Microbiol, 66: 2520-2525.

Echavarria, M., Kola vie, S.A., Cersovsky, S. et al. (2000). Detection of adenoviruses in cul-
ture negative environmental samples by PCR during an adenovirus associated respira-
tory disease outbreak./ Clin Microbiol, 38: 2982-2984.

Enriquez, C.E. and Gerba, C.P. (1995). Concentration of enteric adenovirus 40 from tap,
sea and waste water. Water Res, 29: 2554-2560.

Enriquez, C.E., Hurst, C.J. and Gerba, C.P. (1995). Survival of the enteric adenoviruses
40 and 41 in tap, sea and waste water. Water Res, 29: 2548-2553.

Formiga-Cruz, M., Toflno-Quesada, G., Bofill-Mas, S. etal. (2002). Distribution of human
virus contamination in shellfish from different growing areas in Greece, Spain, Sweden
and the UK. Appl Environ Microbiol, 68: 5990-5998.

Foy, H.M., Cooney, M.K. and Hatlen, J.G. (1968). Adenovirus type 3 epidemic associated
with intermittent chlorination of a swimming pool. Arch Environ Hltb, 17: 795-802.

Genthe, B., Gericke, M., Bateman, B. et al. (1995). Detection of enteric adenoviruses in
South African waters using gene probes. Water Sci Tecbnol, 31: 345-350.

Gerba, C.P., Ramos, D.M. and Nwachuka, N. (2002). Comparative inactivation of
enteroviruses and adenovirus 2 by UV light. Appl Environ Microbiol, 68: 5167-5169.

Grabow, W.O., Taylor, M.B. and Villiers, J.C. (2001). New methods for the detection of
viruses: call for review of drinking water quality guidelines. Water Sci Tecbnol, 43: 1-8.

Harley, D., Harrower, B., Lyon, M. et al. (2001). A primary school outbreak of pharyngo-
conjunctival fever caused by adenovirus type 3. Commun Dis Intell, 25: 9-12.

Hilleman, M.R. and Werner, J.H. (1954). Recovery of new agents from patients with acute
respiratory illness. Proc Soc Exp Biol Med, 85: 183-188.

Irving, L.G. and Smith, F.A. (1981). One-year survey of enteroviruses, adenoviruses and
reoviruses isolated from effluent at an activated sludge purification plant. Appl Environ
Microbiol, 41: 51-59.

Jernigan, J.A., Lowry, B.S. and Hayden, F.G. (1993). Adenovirus type 8 epidemic
keratoconjunctivitis in an eye clinic: risk factors and control. / Infect Dis, 167:
1307-1313.

Kidd, A.H., Jonsson, M., Garwicz, D. et al. (1996). Rapid subgenus identification of
human adenovirus isolates by a general RT-PCR./ Clin Microbiol, 34: 622-627.

Krajden, M., Brown, M. and Petrasek, A. (1990). Clinical features of adenovirus enteritis:
a review of 127 cases. Pediatr Infect Dis J, 9: 636-641.

Kukkula, M., Arstila, P., Klossner, M.L. et al. (1997). Waterborne outbreak of viral gastro-
enteritis. Scand J Infect Dis, 29: 415-418.

Lee, S.H. and Kim, S.J. (2002). Detection of infectious enteroviruses and adenoviruses in
tap water in urban areas in Korea. Water Res, 36: 248-256.

Lew, J.F., Moe, C.L. and Monroe, S.S. (1991). Astro virus and adenovirus associated with
diarrhoea in children in day care settings. / Infect Dis, 164: 673-678.

Martone, W.J., Hierholzer, J.C, Keenlyside, R.A. et al. (1980). An outbreak of adenovirus
type 3 disease at a private recreation center swimming pool. Am J Epidemiol, 111:
229-237.

Papapetropolou, M. and Vantarakis, A.C (1995). Detection of adenovirus outbreak at a
municipal swimming pool by nested PCR amplification./ Infect, 36: 101-103.

Pina, S., Puig, M., Lucena, F. et al. (1998). Viral pollution in the environment and in shell-
fish: human adenovirus detection by PCR as an index of human viruses. Appl Environ
Microbiol, 64: 3376-3382.

Puig, M., Jofre, J., Lucena, F. et al. (1994). Detection of adenoviruses and enteroviruses in
polluted water by nested PCR amplification. Appl Environ Microbiol, 60: 2963-2970.

385

Viruses

Rowe, W.P., Huebner, R.J., Gillmore, L.K. et al. (1953). Isolation of a cytopathogenic agent

from human adenoids undergoing spontaneous degeneration in tissue culture. Proc Soc

Exp Biol Med, 84: 570-573.
Santos, F.M., Vieira, M.J., Monezi, T.A. et al. (2002). Discrimination of adenovirus types

circulating in urban sewage and surface waters in San Paulo City, Brazil. Poster

International Water Association. Health Related Microbiology Symposium. Cascaise

Portugal.
Sellwood, J., Dadswell, J.V. and Slade, J.S. (1981). Viruses in sewage as an indicator of

their presence in the community. / Hyg Camb, 86: 217-225.
Tiemessen, C.T. and Nel, M.J. (1996). Detection and typing of subgroup F adenoviruses

using PCR. / Virol Metb, 59: 73-82.
Turner, M., Istre, G.R., Beauchamp, H. et al. (1987). Community outbreak of adenovirus

type 7a infections associated with a swimming pool. South Med J, 80: 712-715.
Wadell, G. (2000). Adenoviruses. In Principles and Practice of Clinical Virology, 4th edn,

Zuckerman, A.J., Banatalava, J.E. and Pattison, J.R. (eds). Chichester: John Wiley and

Sons, pp. 307-327.

386

Astrovirus

Introduction

Among the viruses recognized as agents of viral gastroenteritis, the role of astro-
viruses is perhaps the least well understood. First described in stool specimens
from babies with gastroenteritis by Appleton and Higgins (1975) and
since observed regularly, their significance in enteric disease has been less well
defined than that of noroviruses, rotaviruses or the enteric adenoviruses.
Generally, in healthy patients, astroviruses cause only a mild or symptomless
infection, but several authors have reported incidences second only to rota-
viruses as causing infantile gastroenteritis (e.g. Marx et al. 9 1998a) and they are
occasionally associated with more severe disease, especially in the immunocom-
promised individual. In addition to numerous reports of their involvement in
diarrhoea in babies and young children (e.g. Madeley, 1979), astroviruses have
also been associated with disease in the elderly (Wilson and Cubitt, 1988) and
in a wide variety of young animals including lambs (Snodgrass and Gray, 1977),
calves (Woode and Bridger, 1978), deer (Tzipori et al. 9 1981), piglets (Bridger,
1980), kittens (Hoshino et aL, 1981), mice (Kjeldsberg and Hem, 1985)
and puppies (Marshall et al. 9 1984). In most animals they cause mild diarrhoeal
disease, though in avian species they are more virulent (e.g. Gough et al. 9 1984;
Koci et aL 9 2000; Imada et al. 9 2000). Astrovirus infections appear to be species

Viruses

specific and there is currently no animal model for infection in humans. Given
that the role of viruses in general in waterborne disease has yet to be fully under-
stood, it will be clear that the less well defined role of human astrovirus infec-
tion will compound the difficulties in ascribing outbreaks of waterborne disease
to this agent.

Basic microbiology

Astroviruses were discovered in the UK by electron microscopical examination
of faecal specimens (Appleton and Higgins, 1975; Madeley and Cosgrove,
1975). While most particles appear as round structures approximately 28 nm in
diameter, about 15% of particles (in a fresh specimen) show a characteristic
five- or six-pointed star-shaped motif on the particle surface. This structure is
clearly different from that of other viruses and serves as a diagnostic feature.
Buoyant densities range from 1.32 g/cm 3 to 1.35 g/cm 3 for human strains to
1.39 g/cm 3 for ovine strains (Herring et al. 9 1981). Astroviruses are moderately
resistant to chemical and physical agents; they are stable at pH 3, are resistant
to chloroform, a variety of detergents (non-ionic, anionic and zwitterionic) and
to other lipid solvents. Human astroviruses retain their infectivity after 5, but
not 10, minutes at 60°C (Kurtz and Lee, 1987). They are stable to storage at low
temperatures ( — 70°C) but, in common with most viruses and other micro-
organisms, may be disrupted by repeated freezing and thawing. There is no
evidence to suggest that astrovirus is not sensitive to chlorine. Ovine astrovirus
capsids were shown by Herring et al. (1981) to contain at least two structural
proteins of M r 30 and 32kDa, an observation which initiated the classification
of the group separate from the picornaviruses and caliciviruses.

Astroviruses contain a positive-sense single-stranded polyadenylated RNA
genome of 6.8-7.2 kb encompassing three open reading frames (ORFla,
lb and 2). Complete nucleotide sequences are available for two (HastV-1 and
HastV-2) of the eight recognized astrovirus serotypes (Jiang et aL, 1993;
Willcocks et al., 1994). The subgenomic fragment of about 2500 nucleotides
encodes a single open reading frame (ORF2) and hence the capsid precursor
polypeptide. A database search revealed the amino acid sequence of this polypep-
tide to be unique and it was thus proposed that astroviruses belong to a separate
family, the Astroviridae (Monroe et aL, 1993).

In 1981, Lee and Kurtz (1981) reported the adaptation of astroviruses to
grow in cell culture, first in human embryo kidney (HEK) monolayers and, after
about six passages, to grow in the LLCMK 2 monkey kidney cell line. This is
achieved by growing the virus in the presence of trypsin sufficient to cleave the
major capsid protein but not high enough to damage the host cells. Viral infec-
tivity may be monitored by immunofluorescence because the trypsin causes
detachment of the cells and a cytopathic effect is difficult to determine, though
with experience it is possible to see a deterioration in the quality of the cells

388

Astrovirus

compared with uninfected controls (Wyn-Jones, unpublished observations).
The need for initial passage in HEK cells was circumvented by the observation
that the virus could be grown directly from clinical specimens by inoculation
into monolayers of the human colonic carcinoma cell line CaCo-2 (Willcocks
et al. 9 1990; Wyn-Jones and Herring, 1991). Following growth in cell culture
further studies on the viral proteins were possible. Willcocks et aL (1990) deter-
mined that there are three virus specific polypeptides in human astroviruses
of 33.5, 31.5 and 24 kDa, the latter being loosely associated with the particle.
For astroviruses generally, it is agreed that the capsid consists of three major
proteins, two in the range 29-33 kDa and one of more variable composition
in the range 13-26.5 kDa (Willcocks et aL, 1992). The polypeptides are the
specific cleavage products of a 90 kDa precursor, which is cleaved only in the
presence of trypsin.

Human astroviruses are currently divided into eight serotypes on the basis of
immune EM, immunofluorescence and enzyme immunoassays (Kurtz and Lee,
1984; Lee and Kurtz, 1994). High divergence in the capsid protein sequences
of astroviruses infecting hosts of different species reflects human strains being
serologically distinct from the animal astroviruses, though limited similarities
between capsid sequences of human, feline and porcine astroviruses have sug-
gested that zoonoses involving these species could occur (Jonassen et aL, 2001).
There is no epidemiological evidence for this however. Nucleotide sequence
analysis of a limited region of the ORF2 of human astrovirus strains by Noel
et aL (1995) showed there is good correlation between antigenic and genomic
types. Relatively few isolates of HAstV type 8 have been reported and there is
little information available with regard to the antigenic and genetic relation-
ships between HAstV type 8 (HAstV-8) and the other serotypes. Taylor et aL
(2001) reported on the analysis of a human astrovirus isolated from a South
African paediatric patient with diarrhoea. Immune electron microscopy and
EIA, and genetic analysis of selected regions of the ORFla, ORFlb and ORF2
characterized the virus as HAstV-8 and confirmed that HAstV-8 represents a
distinct antigenic type and genotype.

The principal technique for detection of astroviruses in clinical specimens
remains electron microscopy (Figure 28.1), done on faecal specimens from
children of five years and under in sporadic cases and adults in outbreaks. This
technique has the advantage of being a catch-all method, in that it will reveal
the presence in a specimen of a range of enteric viruses provided they have a
distinct morphology and are there in sufficient quantity, about 10 6 particles/g.
However, it suffers from the very real problem of missing astrovirus-positive
specimens owing to fewer than 15% of particles usually showing the surface
star structure in a fresh specimen. This will be less (5-10%) if the specimen is
old or has been stored. Immune EM is not a good technique as the antibody
tends to obscure the surface morphology and leads to false identification. Solid
phase immune electron microscopy (SPIEM) has, however, been used with
good effect and contributed to the separation of serotypes (Kurtz, 1994).
Astrovirus EIA kits of high sensitivity, based on monoclonal antibodies, are
available but are not widely used, mainly due to cost considerations.

389

Viruses

Figure 28.1 Transmission electron microscope micrograph of astrovirus.

Other diagnostic techniques involve inoculation of CaCo-2 cell cultures and
performing immunofluorescence on the culture 18-24 hours later; this has the
advantage of detecting infectious virus and is a sensitive technique. Detection
can also be carried out using the reverse-transcriptase polymerase chain reaction
(RT-PCR); reagents are available which will detect all the serotypes of astrovirus
in one reaction (Jonassen etaL 9 1995). These authors reported on 38 astrovirus
EM-positive stool specimens. Using RT-PCR, 36 were positive by RT-PCR
and the remaining two were considered to have been either EM false positives
or had lost the RNA. Tests for other enteric viruses were negative. These
authors and others have also described reagents for the detection of separate
astrovirus serotypes by this technique. Authors have also described integrated
cell culture RT-PCR for astrovirus detection in environmental sample concen-
trates; here, concentrate is inoculated into permissive or partially permissive cell
cultures such as PLC/PRF/5 or CaCo-2 and allowed to develop for between
24 and 60 hours; the culture is then analysed by RT-PCR. This technique has
been used with good effect by several groups, e.g. Pinto et al. (1996), Chapron
et al. (2000) and Taylor et al. (2001) as well as for enteroviruses by Reynolds
et al (1998).

Plaque assays in cell culture have also been described (Hudson et al., 1989),
but the methods have not been generally adopted, either because they have not

390

Astrovirus

been found reproducible or because they depend on the use of primary human
cell culture. Attempts to develop a plaque assay in CaCo-2 cells have been
unsuccessful (Wyn-Jones, unpublished data; M. Carter, personal communica-
tion), though Tree (1997) reported a plaque assay in HT29 cells, a colonic
carcinoma cell line different from CaCo-2.

Typing of astroviruses is carried out by immunofluorescence, immune elec-
tron microscopy (Lee and Kurtz, 1994), enzyme immunoassay and nucleotide
sequencing (Noel et al. 9 1995) and immunogold staining electron microscopy
(Kjeldsberg, 1994).

Origin of the organism

The source of astroviruses in the environment may be defined as the host animal
(including man) from which they are derived. In common with all viruses multi-
plying in the gut, the source of human astrovirus in the environment is faecal
material from infected individuals, whether they are symptomatic or not. No
multiplication takes place outside the host. The input to the environment
is the material which contains the virus before it enters a controlled water.
Inputs include those materials, such as faeces, sewage, sludge, sediments and sep-
tic tanks which may contain human faeces and subsequently enter controlled
waters.

Pathogenesis and clinical features

Phillips et al. (1982) showed that astroviruses infect the mucosal epithelial cells
of the lower parts of the duodenal villi and Snodgrass et al. (1979) reported
transient villous atrophy in the small intestine of infected lambs. The virus is
assembled in pseudo-crystalline arrays and is released from cells into the gut
lumen; a characteristic of astrovirus infection is the occurrence of these arrays
on EM grids. Excretion of virus may be followed by diarrhoea, though this is
rarely as severe as rotavirus diarrhoea and asymptomatic excretion is common.
Diarrhoea, where present, generally lasts 2-3 days but may continue for a
week and is paralleled by virus shedding. During the acute stage, virus levels of
10 10 particles/g have been observed. A good immunity follows infection so that
repeated episodes of disease are rare in a healthy individual.

The incubation period of the disease, determined by human volunteer studies,
is 3-4 days (Kurtz and Lee, 1987) though the virus is excreted for a day prior
to symptoms appearing and Konno et al. (1982) calculated a 24-36 hours incu-
bation period based on secondary spread characteristics. In addition to diar-
rhoea, which may be watery but is usually just soft stools and sometimes not
even recorded as diarrhoea, clinical features of the infection include headache,

391

Viruses

malaise, nausea and occasionally vomiting. There is also a mild fever. The diar-
rhoea generally lasts 2-3 days but there are reports of it continuing for as long
as 14 days. Generally, astrovirus diarrhoea is milder than that caused by
rotavirus and does not lead to significant dehydration.

The early volunteer studies (Kurtz and Lee, 1987) showed that not all
patients suffer diarrhoea. These studies were done on adults, which may reflect
a different picture from children, but nevertheless showed fewer individuals
affected with diarrhoea; of 17 volunteers only one had overt diarrhoea (2-6
loose motions per day). There is probably a dose-response mechanism operat-
ing; following the Marin County outbreak (see below; Midthun et al., 1993)
virus was given to 19 volunteers; none of the first 17 became ill so the dose was
increased by 20-fold for the remaining two, of whom one became ill. These
studies confirm the relatively low pathogenicity of astrovirus, at least for
adults, compared with that of noroviruses.

Transmission and epidemiology

Astroviruses are transmitted from person to person by the faecal-oral route.
The PHLS Communicable Disease Surveillance Centre totals included 116
astrovirus laboratory reports for 2001 and 96 for 2002. This compares with
256 enteric adenovirus (serotypes 40/41) for 2001 and 56 for 2002; 28 'classic
calicivirus' (now known as sappovirus) in 2001 and 32 in 2002. These three
enteric viruses share a similar epidemiology. Though person-to-person spread
is the most common mode of enteric virus transmission within closed commu-
nities (hospital wards, schools, families etc.) modifications of this route exist
including transmission via water or sewage. However, unequivocal waterborne
transmission of astroviruses has not been reported (see below).

The epidemiology of astroviruses was reviewed by Glass et al. (1996) and a
3-year study by Guix et al. (2002) described a molecular epidemiological
approach. Person-to-person spread occurs in families, nurseries or paediatric
wards, where endemic infections may occur (Caul, 1996). Astrovirus infections
have been found world-wide, mainly in young children with diarrhoea.
Outbreaks have also been recorded among the elderly (Gray et al., 1987) and
in military recruits (Belliot et al., 1997). As with any virus, infection is more
pronounced in immunocompromised individuals. Many infected individuals
shed virus in the absence of marked symptoms, hence they may still partake in
daily activities; in the case of children, they will still go to school and will mix
with other children. The agent is thus likely to be transmitted efficiently
through the community, though numbers of individuals reporting illness may
be quite low and many cases will go unreported.

In Marin County, California, 51% of the residents in a home for the
elderly developed gastroenteritis due to astrovirus type 5 (Herrmann et al.,
1990). Outbreaks have been reported where staff in day-care centres or on
paediatric wards have become infected, showing that susceptibility into adult

392

Astrovirus

life does occur (Caul, 1996). Gray et al. (1987) reported an outbreak of gas-
troenteritis in a home for the elderly where 80% of the residents and 44% of
the staff were affected. Calicivirus was involved as well as astrovirus.

Nevertheless a defined role for astroviruses in gastrointestinal disease is dif-
ficult to determine; reports indicate that their incidence and significance may be
underestimated. On the one hand, they seem to be a minor cause of gastro-
enteritis in children (see above) and are only rarely associated with adult disease;
surveys have shown that astroviruses are endemic in most communities and so
it would seem that there is a reservoir of virus ready to be taken up by suscep-
tible individuals. Conversely, studies in Guatemala, Thailand (Hermann et al.,
1991) and Australia (Palombo and Bishop, 1996) found that astroviruses were
the second most frequently detected viral pathogen in children with diarrhoea,
exceeding adenoviruses in frequency of detection and only rotaviruses being
more prevalent. The high frequency of detection is therefore not confined to
developing countries. Astrovirus outbreaks in adults are reported infrequently;
Konno et al. (1982) described an outbreak in Japan involving four adults and
84 children aged 5-6 years and Oishi et al. (1994) reported a large outbreak
in Osaka, Japan affecting both students and teachers; over 4700 people were
affected. Astrovirus was detected by EM and confirmed by serology and molec-
ular biology tests. Utagawa et al. (1994) reported on samples collected between
1982 and 1992 in Japan from sporadic and outbreak cases and used different
tests to identify the agent.

Distribution in the environment

Not all cases of astrovirus diarrhoea arise from direct person-to-person trans-
mission. Food, water and fomites have been implicated in astrovirus disease,
which suggests that astroviruses persist in the environment. The 1991 outbreak
described by Oishi et al. (1994) was food-borne, canteens in 14 schools having
received food prepared in three central kitchens which, in turn, had been sup-
plied from a single source. Astroviruses have also been implicated in shellfish-
related outbreaks, though not as commonly as noroviruses; Kurtz and Lee
(1987) cited a report where a third of 39 naval officers attending a party suf-
fered gastroenteritis 5 days after consuming oysters. Astrovirus was seen in
specimens from five patients and two showed astro-specific IgM antibodies.

Astroviruses have been detected in most aquatic matrices. Generally speak-
ing, however, there are no unequivocal accounts of the waterborne transmis-
sion of astrovirus disease and little circumstantial or anecdotal evidence to
suggest that astroviruses constitute a significant waterborne public health risk.
Cubitt (1991) reported one instance where children became ill 24-36 hours
after drinking stream water which was the presumed source of the infection.
Astrovirus particles were detected in stool samples from the children and high
titres of IgG suggestive of recent infection were detected. In another incident in
1994, soldiers returning from a training exercise in the country drank treated

393

Viruses

effluent at a treatment works under the misconception that it was drinking
water. Four developed gastroenteritis and astrovirus particles were seen by
electron microscopy in specimens from two of them (CDR, 1996).

Detection of astrovirus in several types of environmental sample has been
described by Marx et al. (1995, 1998b). These authors tested sewage sludge,
hospital and abattoir effluents, source water (raw water, 40-litre samples),
stream water (20-litre samples) and treated wastewater. The source water and
stream water were concentrated by glass wool adsorption. No information was
given on the sludge and wastewater processing. Concentrates were tested by a
modification of the RT-PCR method of Jonassen etal. (1993) for astrovirus type
1 or using more broadly reactive primers described by Mitchell etal. (1995) and
by Jonassen et al. (1995). Virus detection was achieved either directly follow-
ing concentration or after a virus amplification step involving inoculating the
concentrates in cell cultures. Positive results were obtained for the source water
even without concentration and water from the suburban stream was positive
following cell culture amplification. Effluent and sludge were also positive by
RT-PCR. In contrast, no samples were positive by EM or IEM.

Taylor et al. (2001) analysed river and dam water in South Africa over a 12-
month period for human astrovirus and hepatitis A virus (HAV) and found
astrovirus in 11/51 river samples and 3/51 dam samples. The river was used by
the local rural community for 'domestic purposes' and both sites were used for
recreation; the dam water was also abstracted for drinking water treatment.
The river was continually faecally polluted and the dam, while a large storage
facility, nevertheless, received input from a number of sources of varying water
quality. The authors also highlighted that virus was detected in dam water
samples in the absence of bacterial and phage indicators and that there was no
seasonal peak for astroviruses, although there was for HAV.

Pinto et al. (1996) reported detection of infectious astroviruses in 500-litre
samples from a dam receiving sewage effluent in an area of Spain during a gas-
troenteritis outbreak. Astrovirus detection in one sample out of six was done by
molecular hybridization with an astrovirus-specific cDNA probe following virus
growth in CaCo-2 cell cultures. Stool samples from affected individuals were
shown to contain astrovirus particles by electron microscopy. These authors
determined that the limit of detection by this technique was 10 virus particles.
Pinto et al. concluded that the water concentrates and unconcentrated samples
would have contained 200 and 20 astrovirus particles per litre, respectively. The
samples positive for astrovirus also contained enterovirus, rotavirus and adeno-
virus 40.

The same group later reported astrovirus detection (by RT-PCR) in samples
from three sewage treatment plants in France and two in Barcelona, and found
that peak environmental prevalence occurred in the winter months, coincident
with the peak of clinical activity (Pinto et al., 2001). Samples could be typed by
restriction fragment length polymorphism (RFLP) analysis of a fragment of the
ORFla. By this means different patterns were established, three different pat-
terns of isolates (corresponding to different viral genotypes) together account-
ing for about 87% of isolates.

394

Astrovirus

Abad et al. (1997) studied the survival of astroviruses in different water
types. Study of survival in dechlorinated tap water at 4°C and marine water
at 20°C showed a reduction of around 2 logs^ at 4°C and about 3.6 logs^ at
20°C over a period of 60 days. After 90 days the reduction was 3.3 logs 1( ) and
4.3 logsjo respectively. These figures would indicate that virus persists in
water for extended periods of time but that inactivation does eventually occur.
The virus displayed marked stability in dechlorinated drinking water at 4°C,
with only a 1.2 logio reduction over a 45-day period.

Several groups have studied the presence of astroviruses in shellfish. Le
Guyader et al. (2000) reported the detection, over a 3-year period, of HAV,
noroviruses, enteroviruses, rotavirus and astrovirus by reverse transcription-
PCR and hybridization in shellfish tissue. In respect of astroviruses, noroviruses
and rotaviruses, mussel samples were more highly contaminated than oysters.
Viral contamination was mainly observed during winter months, although there
were some seasonal differences among the viruses.

Chlorine has an inactivating effect on astrovirus, as would be expected. In
the presence of 0.5 mg/1 free chlorine astrovirus showed a log titre reduction of
2.5 after a one-hour contact time. Increasing the free chlorine concentration to
1 mg/1 resulted in a 3 log reduction in titre (Abad et al., 1997). The survival
characteristics found in these studies were comparable to those determined for
rotavirus and enteric adenoviruses, though the astrovirus decay was more pro-
nounced at the higher temperatures. Astroviruses will therefore be inactivated
by chlorine to the same extent as other viral agents.

Quantification of astroviruses in environmental samples is difficult given the
poor growth in cell culture. Enumeration by plaque assay (Hudson et al.,
1989) has not been repeated in other laboratories so, until recently, semiquan-
titative approaches such as end-point dilution methods have been the best
available techniques. In 2002, El-Senousy et al. (2002) reported the detection
and quantification of astroviruses in sewage and river water in the Cairo
(Egypt) area by competitive RT-PCR. An unrelated internal control RNA
which reverse transcribed and co-amplified with the astrovirus target was used
to provide a quantitative standard. In eight raw sewage samples the number of
RNA copies ranged from 3.4 X 10 3 /1 to 2.3 X 10 6 /1, and in two treated efflu-
ent samples the titres were 3.4 X 10 3 /1 and 1.1 X 10 4 /1. Two water samples
(before and after treatment) were positive, both with titres of 2.3 X 10 3 copies
astrovirus RNA per litre. Using an integrated cell culture RT-PCR technique
estimates were also made of the viral infectious units in the samples and these
ranged from zero to 33.3 'cell culture-RT-PCR units' per litre.

Increasing sophistication of techniques has promoted the study of astro-
viruses in the environment. One of the persistent problems associated with
cause and effect in waterborne disease is to demonstrate that the virus detected
in the patient is the same as that found in the water. Nadan et al. (2003) com-
pared astrovirus cDNA sequences from contemporaneous clinical and envir-
onmental sources by sequence analysis and found that there was coincidence of
types. Human astroviruses types 1, 3, 5, 6 and 8 were found in the clinical
specimens and types 1, 2, 3, 4, 5, 7 and 8 were detected in the environmental

395

Viruses

samples. Phylogenetic analysis revealed that the types 1, 3, 5 and 8, present in
stool specimens and environmental isolates, clustered together, thus demon-
strating they were closely related. Whether this was coincidence or true cause
and effect cannot be further determined but it is a significant step forward. The
enigma was further unravelled in a study by Gofti-Laroche et al. (2003) which
showed that the presence of astrovirus RNA in tap water samples was associ-
ated with a significant risk of acute digestive conditions (defined as an episode
of abdominal pain, nausea, vomiting with or without diarrhoea; RR = 1.51
and a P = 0.002). Thus astroviruses may yet have a part to play in waterborne
gastrointestinal disease.

Risk assessment

Health effects: occurrence of illness, degree of morbidity and mortality, prob-
ability of illness based on infection:

• Astroviruses are endemic in most communities world-wide and mostly infect
young children and babies, although they have been associated with disease
in the elderly. They appear to have a very low pathogenicity for adults.

• Generally, astroviruses cause only a mild or symptomless infection, but
reported incidences are second only to rotaviruses as causing infantile gas-
troenteritis.

• In addition to diarrhoea, which may be watery but is usually just soft stools,
clinical features of astrovirus infection include headache, malaise, nausea,
occasionally vomiting and a mild fever. The diarrhoea generally lasts 2-3
days but there are reports of it continuing for as long as 14 days. Diarrhoea
is paralleled by virus shedding.

• A good immunity follows infection, so that repeated episodes of disease are
rare in a healthy individual.

Exposure assessment: routes of exposure and transmission, occurrence in
source water, environmental fate:

• Person-to-person spread by the faecal-oral route is thought to be the most
common route of transmission. Food, water and fomites have been impli-
cated in astrovirus disease, which suggests that astroviruses persist in
the environment; however, there are no confirmed cases of waterborne
transmission.

• Astroviruses have been found in sewage, wastewater and surface water
(over 50% of samples were positive for astrovirus (Chapron et al., 2000)).
Also, 12% of tap water samples in France were positive and the presence of
astrovirus RNA was associated with a significant increased risk of gas-
troenteritis (Gofti-Laroche et al. 9 2003).

• Astroviruses are moderately resistant to chemical and physical agents. Human
astroviruses retain their infectivity after 5, but not 10 minutes at 60°C.

396

Astrovirus

Risk mitigation: drinking-water treatment, medical treatment:

• Chlorine inactivates astrovirus. In the presence of 0.5mg/l free chlorine,
astrovirus showed a log titre reduction of 2.5 after a one-hour contact time.
Increasing the free chlorine concentration to 1 mg/1 resulted in a 3 -log reduc-
tion in titre. Astroviruses should be inactivated by chlorine to the same
extent as other viral agents.

• No medical treatment is necessary as astrovirus diarrhoea is mild and self-
limiting and does not lead to significant dehydration.

References

Abad, EX., Pinto, R.M., Villena, C. et al. (1997). Astrovirus survival in drinking water.

Appl Environ Microbiol, 63: 3119-3122.
Appleton, H. and Higgins, P.G. (1975). Viruses and gastroenteritis in infants. Lancet, 1: 7919.
Belliot, G., Laveran, H. and Monroe, S.S. (1997). Outbreak of gastro -enteritis in military

recruits associated with serotype 3 astrovirus infection./ Med Virol, 51: 101-106.
Bridger, J.C. (1980). Detection by electron microscopy of caliciviruses, astroviruses and

rotavirus-like particles in the faeces of piglets with diarrhoea. Vet Rec, 107: 532-533.
Caul, E.O. (1996). Viral gastroenteritis: small round-structured viruses, caliciviruses and

astroviruses. Part II - the epidemiological perspective./ Clin Pathol, 49: 959-964.
CDR (1996). General outbreaks of infectious intestinal disease in England and Wales

1992-1994. Commun Dis Rep, 6: R57-R63.
Chapron, C.D., Ballester, N.A., Fontaine, J.H. et al. (2000). Detection of astroviruses,

enteroviruses, and adenovirus types 40 and 41 in surface waters collected and evaluated

by the information collection rule and an integrated cell culture-nested PCR procedure.

Appl Environ Microbiol, 66: 2520-2525.
Cubitt, W.D. (1991). A review of the epidemiology and diagnosis of waterborne viral infec-
tions. Water Sci Tecbnol, 24: 197-203.
El-Senousy, W.M., Caballero, S., Pinto, R.M. et al. (2002). Detection and quantification

of human astroviruses in sewage and river water from Cairo (Egypt) by a competitive

RT-PCR. Am Water Works Assoc. International Symposium on Waterborne Pathogens.

Poster.
Glass, R.I., Noel, J., Mitchell, D. et al. (1996). The changing epidemiology of astrovirus-

associated gastroenteritis: a review. Arch Virol, 12(Suppl.): 287-300.
Gofti-Laroche, L., Gratacap-Cavallier, B., Demannse, D. et al. (2003). Are waterborne

astrovirus implicated in acute digestive morbidity (EMIRA study)./ Clin Virol, 27(1):

74-82.
Gough, R.E., Collins, M.S., Borland, E. et al. (1984). Astrovirus-like particles associated

with hepatitis in ducklings. Vet Rec, 114: 279.
Gray, J.J., Wreghitt, T.G., Cubitt, W.D. et al. (1987). An outbreak of gastroenteritis in a

home for the elderly associated with astrovirus type 1 and human calicivirus. / Med

Virol, 23: 377-381.
Guix, S., Caballero, S., Villena, C. et al. (2002). Molecular epidemiology of astrovirus

infection in Barcelona, Spain./ Clin Microbiol, 40: 133-139.
Herring, A.J., Gray, E.W. and Snodgrass, D.R. (1981). Purification and characterization of

ovine astrovirus./ Gen Virol, 53: 47-55.
Herrmann, J.E., Cubitt, D.W., Hudson, R.W., Perron-Henry, D.M. et al. (1990).

Immunological characterisation of the Marin county strain of astrovirus. Arch Virol,

110:213-220.
Herrmnan, J.E., Taylor, D.N., Echevarria, P. et al. (1991). Astroviruses as a cause of gastro-
enteritis in children. New Engl J Med, 324: 1757-1760.

397

Viruses

Hoshino, Y., Zimmer, J.F., Moise, N.S. et al. (1981). Detection of astro viruses in faeces of

a cat with diarrhoea. Arch Virol, 84: 135-140.
Hudson, R.W., Herrmann, J.E. and Blacklow, N.R. (1989). Plaque quantitation and virus

neutralization assays for human astro viruses. Arch Virol, 108: 33-38.
Imada, T., Yamaguchi, S., Masaji, M. et al. (2000). Avian nephritis virus (ANV) as a new

member of the family Astroviridae and construction of infectious ANV cDNA. / Virol,

74: 8487-8493.
Jiang, B., Monroe, S.S., Koonin, E.V. et al. (1993). RNA sequence of astrovirus: distinctive

genome organisation and a putative retrovirus-like ribosomal frameshifting signal that

directs the viral replicase synthesis. Proc Natl Acad Sci USA, 90: 10539-10543.
Jonassen, CM., Jonassen, T.O., Saif, Y.M. et al. (2001). Comparison of capsid sequences

from human and animal astroviruses. / Gen Virol, 82: 1061-1067.
Jonassen, T.O., Kjeldsberg, E. and Grinde, B. (1993). Detection of human astrovirus

serotype 1 by the polymerase chain reaction./ Virol Meth, 44: 83-88.
Jonassen, T.O., Monceyron, C, Lee, T.W. et al. (1995). Detection of all types of human

astrovirus by the polymerase chain reaction. / Virol Meth, 52: 327-334.
Kjeldsberg, E. (1994). Serotyping of human astrovirus strains by immunogold staining

electron microscopy./ Virol Meth, 50: 137-144.
Kjeldsberg, E. and Hem, A. (1985). Detection of astroviruses in gut contents of nude and

normal mice. Arch Virol, 84: 135-140.
Koci, M.D., Seal, B.S. and Schultz-Cerry, S. (2000). Molecular characterization of an avian

astrovirus./ Virol, 74: 6173-6177.
Konno, T., Suzuki, H., Ishida, N. et al. (1982). Astrovirus-associated epidemic gastro-
enteritis in Japan./ Med Virol, 9: 11-17.
Kurtz, J.B. (1994). Astroviruses. In Viral Infections of the Gastrointestinal Tract, 2nd edn,

Kapikian, A.Z. (ed.). New York: Marcel Dekker.
Kurtz, J.B. and Lee, T.W. (1984). Human astrovirus serotypes. Lancet, 2: 1405.
Kurtz, J.B. and Lee, T.W. (1987). Astroviruses: human and animal. In Novel Diarrhoea

Viruses, CIBA Foundation Symposium No. 128. Chichester: Wiley Interscience.
Lee, T.W. and Kurtz, J.B. (1981). Serial propagation of astroviruses in tissue culture with

the aid of trypsin./ Gen Virol, 57: 421-424.
Lee, T.W. and Kurtz, J.B. (1994). Prevalence of human astrovirus serotypes in the

Oxford region 1976-92, with evidence for two new serotypes. Epidem Infect, 112:

187-193.
Le Guyader, F., Haugarreau, L., Miossec, L. et al. (2000). Three-year study to assess human

enteric viruses in shellfish. Appl Environ Microbiol, 66: 3241-3248.
Madeley, C.R. (1979). Viruses in the stools./ Clin Pathol, 32: 1-10.

Madeley, C.R. and Cosgrove, B.P. (1975). 28 nm particles in faeces in infantile gastro-
enteritis. Lancet, 2: 451-452.
Marshall, J.A., Healey, D.S., Studdert, M.J. et al. (1984). Viruses and virus-like particles in

the faeces of dogs with and without diarrhoea. Aust Vet J, 61: 33-38.
Marx, F.E., Taylor, M.B. and Grabow, W.O.K. (1995). Optimization of a PCR method for

the detection of astrovirus type 1 in environmental samples. Water Sci Technol, 31:

359-362.
Marx, F.E., Taylor, M.B. and Grabow, W.O.K. (1998a). The prevalence of human astro-
virus and enteric adenovirus infection in South African patients with gastroenteritis.

S Afr J Epidem Infect, 13: 5-9.
Marx, F.E., Taylor, M.B. and Grabow, W.O.K. (1998b). The application of a reverse

transcriptase-polymerase chain reaction-oligonucleotide probe assay for the detection

of human astrovirus in environmental water. Water Res, 32: 2147-2153.
Midthun, K., Greenberg, H.B., Kurtz, J. et al. (1993). Characterization and seroepidemiology

of a type 5 astrovirus associated with an outbreak of gastroenteritis in Marin County,

California./ Clin Microbiol, 31: 955-962.
Mitchell, D.K., Monroe, S.S., Jiang, X. et al. (1995). Virologic features of an astrovirus

diarrhea outbreak in a day care center revealed by reverse transcriptase polymerase

chain reaction. / Infect Dis, 172: 1437-1444.

398

Astrovirus

Monroe, S.S., Jiang, B. and Stine, S.E. (1993). Subgenomic RNA sequence of human
astrovirus supports classification of Astroviridae as a new family. / Virol, 65:
641-648.

Nadan, S., Walter, J. E., Grabow, W.O.K. et al. (2003). Molecular characterization of astro-
viruses by reverse transcriptase PCR and sequence analysis: comparison of clinical and
environmental isolates from South Africa. Appl Environ Microbiol, 69: 747-753.

Noel, J.S., Lee, T.W., Kurtz, J.B. etal. (1995). Typing of human astroviruses from clinical isol-
ates by enzyme immunoassay and nucleotide sequencing./ Clin Microbiol, 33: 797-801.

Oishi, I., Yamazaki, K., Kimoto, T. et al. (1994). A large outbreak of acute gastroenteritis
associated with astrovirus among students and teachers at schools in Osaka, Japan.
/ Infect Dis, 170: 430-443.

Palombo, E.A. and Bishop, R.F. (1996). Annual incidence, serotype distribution, and
genetic diversity of human astrovirus isolates from hospitalized children in Melbourne,
Australia./ Clin Microbiol, 43: 1750-1753.

Phillips, A.D., Rice, S.J. and Walker-Smith, J. A. (1982). Astrovirus within human small
intestinal mucosa. Gut, 23: A923-924.

Pinto, R.M., Abad, F.X., Gajardo, R. et al. (1996). Detection of infectious astroviruses in
water. Appl Environ Microbiol, 62: 1811-1813.

Pinto, R.M., Villena, C, Le Guyader, F. et al. (2001). Astrovirus detection in wastewater
samples. Water Sci Tecbnol, 43: 73-76.

Reynolds, C.A., Gerba, C.P. and Pepper, I.L. (1998). Detection of infectious enterovirus by
an integrated cell culture-PCR procedure. Appl Environ Microbiol, 62: 1424-1427.

Snodgrass, D.R. and Gray, E.W. (1977). Detection and transmission of 30 nm particles
(astroviruses) in the faeces of lambs with diarrhoea. Arch Virol, 55: 287-291.

Snodgrass, D.R., Angus, K.W., Gray, E.W. et al. (1979). Pathogenesis of diarrhoea caused
by astrovirus infections in lambs. Arch Virol, 60: 217-226.

Taylor, M.B., Cox, N., Very, M.A. et al. (2001). The occurrence of hepatitis A and astro-
viruses in selected river and dam waters in South Africa. Water Res, 35: 2653-2660.

Tree, J. (1997). PhD Thesis, University of Surrey, UK.

Tzipori, S., Menzies, J.D. and Gray, E.W. (1981). Detection of astroviruses in the faeces of
red deer. Vet Rec, 108:286.

Utagawa, E.T., Nishizawa, S., Sekine, S. et al (1994). Astrovirus as a cause of gastro-
enteritis in Japan./ Clin Microbiol, 32: 1841-1845.

Willcocks, M.M., Brown, T.D.K., Madeley, C.R. et al. (1994). The complete sequence of a
human astrovirus./ Gen Virol, 75: 1785-1788.

Willcocks, M.M., Carter, M.J. and Madeley, C.R. (1992). Astroviruses. Rev Med Virol, 2:
97-106.

Willcocks, M.M., Carter, M.J., Laidler, F.R. et al. (1990). Growth and characterisation of
human faecal astrovirus in a continuous cell line. Arch Virol, 113: 73-81.

Wilson, S.A. and Cubitt, W.D. (1988). The development and evaluation of radioimmune
assays for the detection of immunoglobulins M and G against astrovirus. / Virol Meth,
19: 151-160.

Woode, G.N. and Bridger, J.C. (1978). Isolation of small viruses resembling astroviruses
and caliciviruses from acute enteritis of calves./ Med Microbiol, 11: 441-452.

Wyn -Jones, A.P. and Herring, A.J. (1991). Growth of clinical isolates of astrovirus in a cell
line and preparation of viral RNA. Water Sci Technol, 24: 285-290.

399

Enterovirus

(poliovirus, coxsackievirus, echovirus)

Basic microbiology

Enteroviruses are small, icosahedral particles approximately 27 nm in diam-
eter with no obvious surface structure, as seen by EM, and no envelope. The
virus consists of a protein capsid enclosing a single-strand genome of positive
sense RNA. The RNA is approximately 7.5 kb long and contains a single open
reading frame (ORF) from which the various proteins are derived (Racaniello,
2001).

The Enterovirus genus is part of the Picornaviridae and is further classified
by biomolecular, biochemical characterization and serological typing which
also relates to the species in which the viruses are found. Many animal
enteroviruses have been identified including bovine and porcine species,
which have distinct genomic organizations from the human species. In com-
mon with all viruses they are obligate intracellular parasites and, as for most
virus groups, under normal conditions they are species specific.

Viruses

The capsid of the enteroviruses is an icosahedron with 20 faces and 12
apices, composed of 60 protomers, each containing a single copy of each of
the proteins VP1, VP2, VP3 and VP4. The capsid outer surface undulates
markedly with deep canyons between star-shaped protrusions. The canyons
have been shown to be the receptor binding sites of poliovirus (Mendelsohn
et al. 9 1989). The virus attaches to the cell membrane receptor, a protein in the
case of poliovirus, after which the virus gains entry into the cell cytoplasm by
endocytosis. After the protein coat is lost the single strand of viral RNA is
translated directly into virion proteins which are cleaved and processed by
proteases into structural and non-structural units. Viral RNA synthesis takes
place on small sections of cell membranes. Replication continues with the
assembly of capsids and genome into infectious particles but also of empty
capsids in the cellular cytoplasm. Cell lysis and death follows.

The enterovirus virions are stable within the pH range 3-10, resistant
against quaternary ammonium compounds, 70% ethanol and stable in many
detergents. As they have no lipid envelope, they are resistant to ether and
chloroform. The virions are sensitive to chlorine, sodium hypochlorite,
formaldehyde, gluteraldehyde and UV radiation (Porterfield, 1989).

Human enteroviruses were initially classified by their ability to cause dis-
ease in suckling mice and later by growth characteristics in monkey and
human cell culture (Melnick, 1976). The strains have now been shown to
have different genomic sequences by molecular typing including the use of RT-
PCR and RFLP (Muir et al. 9 1998). The International Committee for the
Taxonomy of Viruses Picornavirus Study Group (Van Regenmortel et al.,
2000) has decided that the genus Enterovirus contains the human species of
poliovirus, human enterovirus A, B, C, D (HEV A-D) as well as bovine and
porcine species. In the traditional groupings shown below coxsackievirus B
and many echoviruses are classed as HEV-B while coxsackievirus A are
classed as mostly HEV-A and HEV-C:

• polioviruses (serotypes 1-3)

• coxsackievirus A (serotypes 1-22, A24)

• coxsackievirus B (serotypes 1-6)

• echoviruses (serotypes 1-9, 11-27, 29-33)

• enteroviruses 68-71 (within species HEV-A, B, D).

Coxsackievirus A strains do not grow well in cell culture so less is known of
their occurrence than the other enteroviruses. The other three groups grow
readily in human or monkey cell cultures and have been widely studied.
Poliovirus and coxsackievirus B form plaques in monkey cells (such as BGM
cells) under agar, but echoviruses vary in this respect.

Since 1968, newly identified enteroviruses have been given sequential num-
bers (enterovirus 68-71). Prior to 1991 hepatitis A virus (HAV) was included
as enterovirus 72 but distinct differences in protein size, cell culture growth
characteristics and temperature resistance compared with the majority of
human enteroviruses determined that it should be placed in the separate genus
'Hepatovirus' (Minor, 1991) and is discussed in detail in Chapter 30.

402

Enterovirus

Virology of infectious poliovirus vaccine

Poliovirus vaccine was developed from the serial passage of wild-type virus
strains in monkey kidney cells to produce an attenuated strain which causes
asymptomatic infection (Sabin and Boulger, 1973; Minor, 1998). Attenuated
virus replicates only in the small intestine and does not usually have a viraemic
phase. The local immune response includes IgA as well as circulating IgG,
both of which are protective against future infection. The molecular basis
of this shift in virulence was reviewed by Minor (1992) and was identified as
a small number of mutations. Attenuation of serotype 1 resulted from one
major change at base 480 of the 5' non-coding region and some smaller
changes elsewhere in the genome. Serotype 2 had changes at base 481 in the
5' non-coding region and another at 2903 that affected residue 143 of the cap-
sid protein VP1. The major attenuating mutations for serotype 3 were changes
at base 2034 which produced a change at residue 91 of the VP3 capsid pro-
tein and at base 472 in the 5' non-coding region. The viruses that are shed
after vaccination have mutated and often reverted to virulent forms but the
genomes remain distinguishable from wild-type. Despite these common changes
the vaccine in practice is very safe and does not revert to the virulence of wild-
type (Freidrich, 1998). Symptoms of paralytic polio occur only at approxi-
mately 1 in a million doses of vaccine.

This live vaccine strain of poliovirus will be present in sewage throughout
the world wherever a programme of vaccination is in place. The USA, Finland
and increasingly other countries have introduced inactivated poliovirus vac-
cine as the WHO Poliovirus Eradication Programme progresses. This killed
vaccine, given by intramuscular injection will not be present in sewage. The
elimination of wild and then the vaccine strain of poliovirus from the envir-
onment should be complete in the foreseeable future.

Origin of the organism

In common with all viruses, the enteroviruses are obligate intracellular para-
sites and, as for most virus groups, under normal conditions they are species
specific. The origin of human viruses in sewage will be human faecal material.
No multiplication will occur outside the living host cell.

Clinical features

Enterovirus infections may cause a wide range of symptoms but most com-
monly infection is asymptomatic (Grist et aL 9 1978). Approximately 1% of
those infected show clinical illness. For many serotypes, when symptoms do

403

Viruses

occur they present as flu-like including fever, malaise, respiratory disease,
headache and muscle ache. Occasionally the virus is more neurotropic and
meningitis develops. Paralysis is a major feature of poliovirus symptomatic
infections and a temporary feature of some coxsackievirus B infections.
Coxsackievirus A infections may be associated in particular with hand, foot
and mouth disease; coxsackievirus B infections with Bornholme disease, peri-
carditis and myocarditis; echoviruses with encephalitis and Guillain-Barre
syndrome; enterovirus 70 with haemorrhagic conjunctivitis. None of the
enterovirus symptomatic infections are associated with diarrhoea and vomiting
other than as part of the wider disease spectrum.

The virulence of vaccine poliovirus was discussed previously. For other
enteroviruses the virulence is dependent on genetic makeup but also on the
initial virus dose and the host age and immune response.

Virulence and pathogenicity

The alimentary canal and the respiratory system are the main sites of multi-
plication for the enteroviruses with entry through the mouth (Loria, 1988);
replication may also take place in the lymphoid tissue of the pharynx. The
incubation period from exposure to onset of symptoms may be between 2 and
30 days for different strains but is most commonly 5-14 days (Pallansch and
Roos, 2001). Virus is shed, in asymptomatic and symptomatic infections, in
faeces and in oral secretions usually from 7 days after infection and may be
prolonged, especially in the faeces. It is via this route that enteroviruses reach
the environment. A viraemia may occur during which specific target organs
may be infected, determined by the serotype involved.

Of all the enteroviruses, wild poliovirus causes the most severe neurological
disease. Virus replication in the lower motor neurons causes flaccid paralysis;
when respiratory muscles are affected death will ensue. Viral meningitis, most
commonly caused by enterovirus infection, is the inflammation of the
meninges together with fever and headache. Recovery is usually without last-
ing effects. Serotypes coxsackievirus B5 and echovirus 4, 6, 9, 11, 13 and 30
are known to be more neurotropic than other strains. Encephalitis that
involves the brain tissue and symptoms of confusion and seizures is more
serious in outcome and may produce lasting damage.

Treatment

No specific treatment is available for enterovirus infection but symptomatic
treatment is used for serious infections. Normal human immunoglobulin

404

Enterovirus

comprising high titre antibody has been used to support neonates with
echovirus 11 infection (Nagington, 1982).

Environment

More information is available on the occurrence of enteroviruses in the envir-
onment than any other virus group because their identification in water sam-
ples has been, for many years, the most straightforward. However, even the
type of enterovirus detected in water is in part dependent on the isolation
method used. In general, if a range of human and monkey cell cultures under
liquid medium are used, poliovirus, coxsackievirus B and echovirus serotypes
will be detectable (Sellwood et al. 9 1981; Irving and Smith, 1981; Hovi et al. 9
1996). When an agar-based plaque assay with BGM cells is used then
poliovirus and coxsackievirus B are the most likely serotypes to be found
(Morris, 1985).

Many studies on the presence of enteric viruses in water and associated
materials took place between 1975 and 1985 in the UK and the USA. The
techniques in use at that time were primarily for detection of culturable
enterovirus and were similar in efficiency to those used for current enterovirus
surveys and monitoring projects and may, indeed have had greater resources
available. The studies that included the use of liquid culture and a range of cell
culture types were done during this early period. More recently, molecular
methods, RT-PCR, real-time RT-PCR and integrated cell culture and RT-PCR
(ICC-PCR), have been utilized thus increasing the sensitivity of detection. The
latter assay incorporates an infectivity component which addresses a major
disadvantage of molecular techniques. (See also section on Concentration and
detection.)

Coxsackievirus B serotypes 1-6 have all been detected and reported,
although B3, B4 and B5 are the most frequently identified (Irving and Smith,
1981; Lewis et al. 9 1986; Hughes et al. 9 1992). The predominating serotype
has been shown to change over time and to be associated with the serotype
most frequently found in samples from clinical material. Echoviruses are least
frequently identified (Martins et al. 9 1983; Morris and Sharp, 1984; Krikelis
et al. 9 1986; Hovi et al. 9 1996) except if liquid assay is used when the com-
bined serotypes of echoviruses are the most numerous strain (Sellwood et al. 9
1981).

Enteroviruses are usually detectable all year round with some seasonal vari-
ation in the studies cited above. An increase in the number of isolates has been
reported in summer and autumn for both northern and southern hemispheres
in the above references. Poliovirus isolates also seem to follow this pattern.
Sampling error, sample size, differences in water flow quantity, differences in
techniques and recovery efficiency between laboratories makes a comparison
of numbers difficult per standard volume at any single time.

405

Viruses

Of the polioviruses, serotype 2 is usually predominant, type 3 being the
least numerous across all water types (Sellwood et al. 9 1981; Payment et al. 9
1983; Morris and Sharp, 1984). Wild strains were identified (Sattar and
Westwood, 1977; Payment et al. 9 1983; Lucena et al. 9 1985) in early studies,
although in the UK and more recently in other parts of the world only vaccine
strains have been reported (Sellwood et al 9 1981, 1995; Poyry et al. 9 1988). Wild
strains were found in water associated with a poliovirus outbreak in Finland
(Poyry et al. 9 1988).

Surveillance for poliovirus has been carried out in several countries during
outbreaks of poliomyelitis. Sewage in Japan has been investigated by
Matsuura et al. (2000) and Horie et al. (2002) for the presence of poliovirus.
Only vaccine strains were identified but some of the strains had neurovirulent
characteristics. Manor et al. (1999), in Israel, has developed a selective cell
culture system for processing large numbers of water samples and plaques and
found a small number of wild poliovirus isolates. Poyry et al. (1988) and
Bottinger et al. (1992), in Scandinavia, found poliovirus of vaccine and wild
type during the previous ten years. During an outbreak in the Netherlands
in 1992-93 van der Avoort et al. (1995) studied poliovirus in sewage and
Oostvogel et al. (1994) analysed the isolates from patient samples. Wild type
strains were identified in both and sewage collected early in the epidemic also
contained wild type. An outbreak in Albania was investigated by Divizia et al.
(1999) and wild type, vaccine related and some recombinant isolates were
identified. As part of the WHO Poliovirus Eradication Programme, RD cells
have been recommended for monitoring purposes.

The investigation of water and associated materials for the distribution of
enteroviruses has been undertaken in many countries. The epidemiology of
enteroviruses in communities and the presence of enteroviruses in sewage and
water have been shown to follow a similar pattern in temperate countries
across the world. UK reports include those of Sellwood et al. (1981), Edwards
and Wyn-Jones (1981), Morris and Sharp (1984), Hughes et al. (1992),
Murrin et al. (1997), Pallin et al. (1997).

Other European studies have been reported from Sweden (Bottinger, 1973;
Bottinger and Herrstrom, 1992), Finland (Poyry etal 9 1988; Hovi et aL 9 1996),
Denmark (Lund et al., 1973), Italy (Carducci et al. 9 1995), France (Agbalika
et al., 1983; Kopecka et al. 9 1993), Greece (Kirkelis et al. 9 1986), Romania
(Nestor and Costin, 1976), Hungary (Palfi, 1971) and Russia (Drozdov and
Kazantseva, 1977).

Early work by Melnick et al. (1954) in the USA has continued to the pre-
sent day by researchers such as Gerba et al. (1978), Sobsey et al. (1973),
Abbaszadegan et al. (1999), Griffin et al. (1999) and Reynolds et al. (2001).
Chapron et al. (2000) investigated a range of enteric viruses, including
enterovirus in source waters for drinking water treatment, as part of the moni-
toring required by the US Environmental Protection Agency Information
Collection Rule. Payment et al. (1983, 1988) and Sattar and Ramia (1978)
have continued to investigate enteroviruses over many years in Canada. Studies
in Japan (Tani et al. 9 1992; Katayama et al. 9 2002), South Africa (Grabow and

406

Enterovirus

Nupen, 1981; Geldenhuys and Pretorius, 1989; Vivier et aL, 2001), Australia
(Irving and Smith, 1981; Grohmann et aL, 1993), New Zealand (Lewis et aL,
1986) and Israel (Fattal etaL, 1977; Guttman-Bass et aL, 1981) have all con-
firmed the presence of enteroviruses in a wide range of water types, such as
surface freshwater and marine waters and throughout the seasons. High num-
bers of enteroviruses have been identified in tropical countries such as Puerto
Rico (Dahling et aL, 1989) and Colombia (Tambini et aL, 1993).

A range of human enteric viruses are potentially present in any type of
water contaminated by human faecal material or polluted by untreated or
treated sewage. Viruses are likely to be present associated with debris and will
be clumped together in faecal material. Therefore untreated sewage will con-
tain the most viruses. Sewage that has been processed to break up clumps may
have more evenly dispersed virus particles, but these should then be more
accessible to factors that may destroy them. Any treatment which removes the
solids of sewage before disposal to a receiving water will also remove a pro-
portion of the viruses. The resulting effluent will contain fewer viruses.

Most rivers in the UK have treated sewage discharged into them and therefore
are likely to contain enteroviruses. Sewage is discharged into in-shore marine
waters all along the coasts of Britain. Designated bathing waters, the majority
of which are marine sites in the UK, are more likely to have higher levels of
enteroviruses if sewage is discharged nearby through short sea-outfalls than
if long sea-outfalls are in place. Fresh water, i.e. rivers, streams, brooks and
canals, contain enteroviruses if that water has had treated sewage, storm water
overflows or untreated discharges added to it. The increasing use of UV as a ter-
tiary treatment in the UK will reduce the number of infectious virions in sewage
effluent. The vast majority of enteroviruses in controlled waters originate from
un-disinfected, continuous point source sewage discharges. The particulate
phase of river water was also found to contain enteroviruses by Payment et aL
(1988). Figure 29.1 shows the range of enterovirus serotypes present in a UK
river in 2002. Coxsackieviruses B2 and B4 were the most common serotypes
identified throughout 2002 but numbers waned later in the year. During the
summer coxsackievirus B5 numbers increased. Coxsackievirus Bl isolates
became increasingly numerous after the middle of the year and coxsackievirus
B3 was present in low numbers throughout the year.

Enteroviruses have been detected in sewage sampled at the point of entry to
a treatment works and final effluent in studies by Sell wood et aL (1981), Irving
and Smith (1981), Payment et aL (1983), Martins et aL (1983), Lewis et aL
(1986),P6yry et aL (1988), Dahling et aL (1989),Hovi et aL (1996) and Green
and Lewis (1999). The enteroviruses of effluent were investigated by Carducci
et aL (1995) and Rose et aL (1989); effluent and wastewater were studied in
association with river waters by Morris and Sharp (1984), Lucena etal. (1985),
Payment etal. (1988) and Geldenhuys and Pretorius (1989). Figure 29.2 shows
the range of enterovirus serotypes identified in sewage and effluent from a
sewage treatment works in the UK in 2001. Coxsackievirus B4 was the most
common serotype identified throughout the year. During the autumn months
coxsackieviruses B2 and B5 were found in similar numbers to that of B4.

407

Viruses

■*—•

0)
.Q

Jan Feb

Mar Apr May June July

Month

Aug Sept Oct

Nov

Figure 29.1 Enterovirus serotypes, UK recreational river water isolates, 2002.

Sewage sludge comprises a high proportion of solids to which enteroviruses
are attached (Hamparian et aL 9 1985; Hurst and Goyke, 1986). Anaerobic
digestion of sludge reduces, but does not remove all enteroviruses. Re-suspension
or introduction of this material into the water cycle by dumping at sea,
even in mid-Atlantic (Goyal et aL, 1984) or spread onto pasture or soil
(Straub et al. 9 1992) and subsequent rain may release any viruses present.
Enteroviruses in sludge spread to land will be denatured over time, at a rate
dependent on the weather conditions. Lagoons and oxidation ponds for
sewage sludge may also contain enteroviruses. Leachates from landfill may
contain enteroviruses but it is unlikely to be in high numbers. Aerosols of sewage
have occasionally been reported to contain low numbers of enteroviruses
(Carducci et al., 1995).

408

Enterovirus

Poliovirus 3-i
Poliovirus 2

Poliovirus 1

Coxsackievirus B5

Poliovirus
Unknown

Coxsackievirus B2
Coxsackievirus B3

n = 311

Coxsackievirus B4

Figure 29.2 Enterovirus serotypes, UK sewage isolates, 2001 percentage distributions.

Enteroviruses have been detected in bathing water, especially seawater in
many parts of the world, including the UK (Merrett etaL, 1989; Hughes etaL,
1992; Sellwood, 1994), Spain (Lucena et aL, 1985), the Netherlands (Van
Olphen et aL, 1991), France (Hugues et aL, 1979), Italy (Patti et aL, 1990;
Pianetti et aL, 2000), Australia (Grohmann et aL, 1993), Hawaii (Reynolds
et aL, 1998) and Florida Keys (Donaldson et aL, 2002). Bathers, especially
children, may pollute recreational waters directly (Fattal et aL, 1991) and
affect the virus count in localized areas.

Sediments from fresh and marine waters have been shown to be associated
with enteroviruses (Lewis et aL, 1985, 1986), Rao et aL (1984) and Bosch
et aL (1988). These sediments may be re-suspended by rainfall, strong winds,
strong tides and currents thereby releasing their attached viruses back into the
water column. The enteroviruses in sediments may persist for long periods
and so be capable of affecting counts for some time after discharging of sewage
or sludge may have stopped. Enteroviruses, rotaviruses and adenoviruses
were found in sediments near Bondi beach up to 2 years after disposal of
sewage had moved off-shore (G. Grohmann, personal communication).

Enteroviruses have occasionally been detected in groundwater (Slade, 1981;
Abbaszadegan, 1993, 1999) and in wastewater-recharged groundwater
(Vaughn et aL, 1978) and there has been concern that this may affect the qual-
ity of the source water for drinking water abstraction. An urban aquifer in the
UK was shown to be contaminated with enteroviruses and noroviruses during
different times of the year (Powell et aL, 2003). Drinking water in the UK
after treatment contains little organic material and therefore a minimal
chance of containing microorganisms. Before distribution as potable water it
is also disinfected with chlorine which inactivates any viruses. Morris and
Sharp (1985) found no evidence of viruses after treatment but Tyler (personal
communication) detected occasional enteroviruses in other waters origina-
ting in smaller reservoir treatment works. This was thought to be because of

409

Viruses

failure of chlorination equipment. Using molecular assays Grabow et ah
(2001) and Lee and Kim (2002) have detected enteroviruses in finished tap
water in South Africa and Korea respectively.

Bird and animal waste is unlikely to contain human enteroviruses.

Waterborne outbreaks

Swimming pools and the spread of poliovirus was of great concern in the USA
during the 1940s and 1950s. It is not clear whether transmission did in fact
occur via the water although the potential for spread is illustrated by an out-
break of echovirus 30 in Northern Ireland. An outdoor pool was contam-
inated with vomit on its day of opening followed by an outbreak of headache,
diarrhoea and vomiting in swimmers (Kee et al., 1994). Chlorine levels were
said to be satisfactory and no other pathogens were detected, but it was not
stated if faeces were tested for the presence of norovirus.

Although often investigated, little strong evidence of waterborne transmission
of poliovirus or any enterovirus has been documented in developed countries
(Metcalf et aL, 1995). As enterovirus infections are so common in the commu-
nity and easily spread person to person and within families, any transmission
by drinking water or recreational water contact will be difficult to confirm by
cohort studies. In addition, only 1 in 100 infections results in symptomatic illness
making it difficult to trace contacts. One case control study investigated children
with febrile illness who presented at a hospital paediatric clinic (D'Alessio et al. 9
1981). When an enterovirus infection was confirmed by laboratory isolation,
a history of swimming in a local lake was significantly more common than in
controls. Coxsackievirus infection was found to be more common in divers using
a river in France than in a control population (Garin et aL, 1994). Community
outbreaks of coxsackievirus B infections in Cyprus in 1998 (Papageorgiou,
personal communication) and Belarus (Amvrosieva et aL, 1998) have again
raised questions regarding the significance of enteroviruses in water and sewage
as possible vehicles of transmission, but epidemiological or virological evidence
remains scanty. Enterovirus will be readily detectable in sewage and associated
polluted recreational water if it is circulating in a community, but to confirm
transmission via water is problematic.

The relevance of enterovirus in recreational water as a pathogen is unclear.
The EU Bathing Water Directive, 1976 included the parameter for enterovirus
to be monitored if the quality of the water was suspected to have changed.
The choice of this test was a practical one, no other virus group could at that
time, or can still be concentrated and detected reliably. Although the entero-
viruses should be regarded as pathogens and a hazard potentially transmissible
in water, the degree of risk via water, remains a matter of debate.

In areas of the world where full poliovirus vaccination cover is yet to be
achieved the virulent virus continues to cause serious outbreaks. Gaspar
et al. (2000) reported that between January and June 1999 there were 1100
suspected cases of poliomyelitis in Angola, where poor sanitation and water

410

Enterovirus

supply and low herd immunity contributed to the rapid spread of the virus
and a high incidence of paralytic disease. In Albania in 1996 an outbreak
occurred with a high (12%) mortality rate. Samples of water and patient
stools were analysed by RT-PCR and significant homology found between
sequences of environmental isolates and faecal isolates (Divizia et al. 9 1999).
Further characteristics of the genomes of isolates from the two sources were
found in common. Polluted drinking water supplies may have facilitated the
rapid spread of poliovirus within these communities once the infection was
established and therefore significant amounts of virus were being shed into
water supplies.

Drinking water that is efficiently chlorinated or disinfected should not
contain infectious enteroviruses.

Epidemiology

Transmission of virus is primarily person to person by the faecal-oral route
and by secretory droplets from the nasopharynx. Faecal material contains
large amounts of virus which is subsequently incorporated into sewage. There
is no known animal reservoir for human strains. Spread of the viruses within
families and between close contacts, such as in schools, is rapid and takes
place with ease. Poor hygiene and lack of hand washing facilitate viral spread.
Adults are more likely to demonstrate symptoms. A clear seasonal pattern
shows a peak of reported illness during the summer months of the general and
neurological manifestations of enterovirus infection (Maguire et al. 9 1999).
This pattern has been reported world-wide in temperate climates. During any
one season a single serotype may predominate clinical reports but other
serotypes will also be present. In the summer of 2000, echovirus 13 was a
common cause of meningitis in the UK (PHLS Communicable Disease Report
August 2000) but a range of coxsackievirus B serotypes were found in sewage
(Figure 29.3) reflecting the range of serotypes present in the community.

Several studies have attempted to determine the infectious dose for
enteroviruses using poliovirus (Minor et al. 9 1981) and echovirus 12 (Schiff
et al. 9 1984) but the results are difficult to assess. The lowest dose was 17
plaque forming units (pfu) of echovirus 12, which would infect 1% of those
inoculated. Electron microscopy of the virus preparation suggested this was
equivalent to 700 virus particles. Racaniello (2001) states that poliovirus and
other picornaviruses have a particle to pfu range between 30 and 1000. Such
studies are extremely difficult to standardize and it is generally assumed that
for most viruses one infectious particle is capable of infecting a susceptible
host (Schiff et al. 9 1984). It is difficult to distinguish experimentally between
one virion (virus particle) and one 'infectious unit', which may comprise an
unknown and variable number of virions clumped together. Risk assessment
for water contact assumes that one virus infectious unit is capable of initiat-
ing an infection in a susceptible host (Berg, 1967).

411

Viruses

Unknown

Poliovirus 3
Poliovirus 2

Coxsackievirus B2

Poliovirus 1

Coxsackievirus B5

Coxsackievirus B3

Coxsackievirus B4

n= 166

Figure 29.3 Enterovirus serotypes, UK sewage isolates, 2000 percentage distributions.

Risk assessment

Health effects: occurrence of illness, degree of morbidity and mortality, prob-
ability of illness based on infection:

• Occurrence of enterovirus infection and illness varies by serotype, geog-
raphy, season and the host's age and antibody status.

• For many serotypes, when symptoms occur they present as fever, malaise,
respiratory disease, headache and muscle ache. Coxsackievirus primarily
causes meningitis, unspecified febrile illness, hand-foot-and-mouth disease
and conjunctivitis. Echovirus primarily causes meningitis, unspecified febrile
illness, and respiratory illness. Enterovirus 70 is associated with haemor-
rhagic conjunctivitis. Wild poliovirus causes the most severe neurological
disease. Most enterovirus disease resolves on its own within a week of
infection.

• Susceptibles, such as newborns, can have a fatality rate of up to 10%, but
otherwise, fatality is uncommon.

• The probability of illness based on infection varies widely by serotype and
the host's age and antibody status. Overall, approximately 1% of those
infected show clinical illness.

Exposure assessment: routes of exposure and transmission, occurrence in
source water, environmental fate:

• Transmission is faecal-oral and by secretory droplets from the nasophar-
ynx: primarily person to person directly or via fomites. Theoretically, trans-
mission is possible through water or food ingestion or aerosolization.

• A volunteer study on infectious doses of enteroviruses was performed with
echovirus 12. The study of 149 volunteers concluded that 919 pfu are

412

Enterovirus

necessary to infect 50% of an exposed population and estimated that
17 pfu are necessary to infect 1% of an exposed population. Such studies are
extremely difficult to standardize and it is generally assumed that for most
viruses one infectious particle is capable of infecting a susceptible host.

• Attack rates of infection have been reported variously, depending on the
serotype and host characteristics.

• Secondary spread of enterovirus infections is common. Viral shedding gen-
erally lasts from a week to a month. This relatively long period increases
the probability of secondary spread. Generally, illness in secondary cases
is lessened from that of the index case; frequently secondary infection
is asymptomatic.

• Enteroviruses are found in all types of water that can be polluted by human
sewage: lakes, rivers, sediments, marine environments, groundwater, waste-
water, and drinking water. As enterovirus infections are so common in the
community and easily spread person to person and within families, any
transmission by drinking water or recreational water contact will be diffi-
cult to confirm. Little evidence supports a waterborne route of infection.
Most outbreaks appear to result from person-to-person spread. The pre-
dominating serotype in water sources has been shown to change over time
and to be associated with the serotype most frequently found in samples
from clinical material.

Risk mitigation: drinking-water treatment, medical treatment

• The enterovirus virions are sensitive to chlorine, sodium hypochlorite,
formaldehyde, gluteraldehyde and UV radiation. Standard water treatment
with chlorine disinfection should efficiently chlorinate or control
enteroviruses.

• Until recently, no antiviral drugs effectively prevented or treated echovirus
infections. Gammaglobulin therapy and exchange transfusions have been
used with limited success in the critically ill. However, an antipicornaviral
agent, pleconaril, has shown promise in clinical trials for echovirus menin-
gitis. It is expected to receive approval from the FDA (Rotbart and Webster,
2001).

References

Abbaszadegan, M., Huber, M.S., Gerba, C.P. et al. (1993). Detection of enteroviruses

in ground water with the polymerase chain reaction. Appl Environ Microbiol, 59:

1319-1324.
Abbaszadegan, M., Stewart, P. and LeChevallier, M. (1999). A strategy for detection of

viruses in groundwater by PCR. Appl Environ Microbiol, 65: 444-449.
Agbalika, R, Hartemann, P., Briigaud, M. et al. (1983). Enterovirus circulation in the

wastewaters of the Lorraine region. Rev Epidemiol Sante Publique, 31: 209-221.
Amvrosieva, T.V., Titov, L.P., Mulders, M., Hovi, T. et al. (2001). Viral contamination

as the cause of aseptic meningitis outbreak in Belarus. Cent Eur J Public Hltb,

9: 154-157.

413

Viruses

Bell, E.J. and Grist, N.R. (1967). Viruses in diarrhoeal disease. Br Med J, 582: 741-742.

Berg, G. (1967). Transmission of Viruses by the Water Route. New York: Wiley InterScience
Publishers, pp. 1-484.

Bosch, A., Lucena, R, Girones, R. et al. (1988). Occurrence of enteroviruses in marine
sediment along the coast of Barcelona, Spain. Can J Microbiol, 34: 921-924.

Bottinger, M. (1973). Experiences from investigations of virus isolations from sewage
over a two year period with special regard to poliovirus. Arch Gesamte Virusforsch,
41: 80-85.

Bottinger, M. and Herrstrom, E. (1992). Isolation of poliovirus from sewage and their
characteristics: experience over two decades in Sweden. Scand J Infect Dis, 24:
151-155.

Carducci, A., Arrighi, S. and Ruschi, A. (1995). Detection of coliphages and enteroviruses
in sewage and aerosol from an activated sludge wastewater treatment plant. Lett Appl
Microbiol, 21: 207-209.

Chapron, C.D., Ballester, N.A., Fontaine, J.H. et al. (2000). Detection of astroviruses,
enteroviruses and adenovirus types 40 and 41 in surface waters collected and evaluated
by the Information Collection Rule and an Integrated Cell Culture-Nested PCR proced-
ure. Appl Environ Microbiol, 66: 2520-2525.

D'Alessio, D.J., Minor, T.E., Allen, C.I. et al. (1981). A study of the proportions of swim-
mers among well controls and children with enterovirus-like illness shedding or not
shedding an enterovirus. Am J Epidemiol, 113: 533-541.

Dahling, D.R., Safferman, R.S. and Wright, B.A. (1989). Isolation of enterovirus and
reovirus from sewage and treated effluents in selected Puerto Rican communities. Appl
Environ Microbiol, 55: 503-506.

Divizia, M., Palombi, L., Buonomo, E. et al. (1999). Genomic characterization of human
and environmental polio viruses isolated in Albania. Appl Environ Microbiol, 65:
3534-3539.

Donaldson, K.A., Griffin, D.W. and Paul, J.H. (2002). Detection, quantitation and identi-
fication of enteroviruses from surface waters and sponge tissue from the Florida Keys
using real-time RT-PCR. Water Res, 36: 2505-2514.

Drozdov, S.G. and Kazantseva, V.A. (1977). Importance of the results of a virological
examination of sewage. Vopr Virusol, 5: 597-602.

Edwards, E. and Wyn -Jones, A.P. (1981). Incidence of human enteroviruses in water and
wastewater systems associated with the River Wear. In Viruses and Wastewater Treat-
ment, Goddard, M. and Butler, M. (eds). Oxford: Pergamon Press, pp. 109-112.

Fattal, B. and Nishmi, M. (1977) Enterovirus types in Israel sewage. Water Res, 11:
393-396.

Fattal, B., Peleg-Olevsky, E. and Cabelli, V.J. (1991). Bathers as a possible source of con-
tamination for swimming-associated illness at marine bathing beaches. Int J Environ
Hlth Res, 1: 204-214.

Freidrich, F. (1998). Neurological complications associated with oral poliovirus vaccine
and genomic variability of the vaccine strains after multiplication in humans. Acta Virol,
42: 187-195.

Garin, D., Fuchs, F., Crance, J.M. et al. (1994). Exposure to enteroviruses and hepatitis A
virus among divers in environmental waters in France, first biological and serological
survey of a controlled cohort. Epidemiol Infect, 113: 541-549.

Gaspar, M., Morais, A., Brumana, L. and Stella, A. (2000). Outbreak of poliomyelitis in
Angola. / Infect D is, 181: 1777-1779.

Geldenhuys, J.C. and Pretorius, P.D. (1989). The occurrence of enteric viruses in polluted
water, correlation to indicator organisms and factors influencing their numbers. Water
Sci Technol, 21: 105-109.

Gerba, C.P., Farrah, S.R., Goyal, S.M. et al. (1978). Concentration of enteroviruses from
large volumes of tap water, treated sewage and seawater. Appl Environ Microbiol,
35: 540-548.

Goyal, S.M., Adams, W.N., O'Malley, M.L. et al. (1984). Human pathogenic viruses at
sewage sludge disposal sites in the middle Atlantic region. Appl Environ Microbiol, 48:
758-763.

414

Enterovirus

Grabow, W.O.K. and Nupen, E.M. (1981). Comparison of primary kidney cells with

the BGM cell line for the enumeration of enteric viruses in water by means of a tube

dilution technique. In Viruses and Wastewater Treatment, Goddard, M. and Butler, M.

(eds). Oxford: Pergamon Press, pp. 253-256.
Grabow, W.O., Taylor, M.B. and de Villiers, J.C. (2001). New methods for the detection of

viruses: call for review of drinking water quality guidelines. Water Sci Tecbnol,

43: 1-8.
Green, D.H. and Lewis, G.D. (1999). Comparative detection of enteric viruses in waste-
waters, sediments and oysters by RT-PCR and cell culture. Water Res, 33: 1195-1200.
Griffin, D.W., Gibson, C.J. and Lipp, E.K. (1999). Detection of viral pathogens by RT-PCR

and of microbial indicators by standard methods in the canals of the Florida Keys. Appl

Environ Microbiol, 65: 4118-4125.
Grist, N.R., Bell, E.J. and Assaad, F. (1978). Enteroviruses in Human Disease. Frog Med

Virol, 24: 114-157.
Grohmann, G., Ashbolt, N.J., Genova, M.S. et al. (1993). Detection of viruses in coastal

and river water systems in Sydney, Australia. Water Sci Tecbnol, 27: 457-461.
Guttman-Bass, N., Tchorsh, Y., Nasser, A. et al. (1981). Rapid detection of enteroviruses

in water by a quantitative fluorescent technique. In Viruses and Wastewater Treatment,

Goddard, M. and Butler, M. (eds). Oxford: Pergamon Press, pp. 247-251.
Hamparian, V.V., Ottolenghi, A.C. and Hughes, J.H. (1985). Enteroviruses in sludge: mul-

tiyear experience with four wastewater treatment plants. Appl Environ Microbiol, 50:

280-286.
Horie, H., Yoshida, H., Matsuura, K. et al. (2002). Neurovirulence of type 1 polioviruses

isolated from sewage in Japan. Appl Environ Microbiol, 68: 138-142.
Hovi, T., Stenvik, M. and Rosenlew, M. (1996). Relative abundance of enterovirus

serotypes in sewage differs from that in patients: clinical and epidemiological implica-
tions. Epidemiol Infect, 116: 91-97.
Hughes, M.S., Coyle, P.V. and Connolly, J.H. (1992). Enteroviruses in recreational waters

of Northern Ireland. Epidemiol Infect, 108: 529-536.
Hugues, B., Bougis, M.A., Plissier, M. et al. (1979). Evaluation of the viral contamination

of the sea water after the emission of an effluent into the sea. Zentralbl Bakteriol, 169:

253-264.
Hurst, C.J. and Goyke, T. (1986). Stability of viruses in wastewater sludge eluates. Can J

Microbiol, 32: 649-653.
Irving, L.G. and Smith, F.A. (1981) One-year survey of enteroviruses, adenoviruses and

reo viruses isolated from effluent at an activated sludge purification plant. Appl Environ

Microbiol, 41: 51-59.
Katayama, H., Shimasaki, A. and Ohgaki, S. (2002). Development of a virus concentration

method and its application to detection of enterovirus and Norwalk virus from coastal

seawater. Appl Environ Microbiol, 68: 1003-1039.
Kee, F., McElroy, G., Sewart, D. et al. (1994). A community outbreak of echovirus infec-
tion associated with an outdoor swimming pool. / Public Hlth, 16: 145-148.
Kopecka, H., Dubrou, S., Prevot, J. et al. (1993). Detection of naturally-occurring

enteroviruses in waters by reverse transcription, polymerase chain reaction, and

hybridization. Appl Environ Microbiol, 59: 1213-1219.
Krikelis, V., Markoulatos, P. and Spyrou, N. (1986). Viral pollution of coastal waters

resulting from disposal of untreated sewage effluents. Water Sci Tecbnol, 18: 43-48.
Lee, S.H. and Kim, S.J. (2002). Detection of infectious enteroviruses and adenoviruses in

tap water in urban areas in Korea. Water Res, 36: 248-256.
Lewis, G.D., Austin, M.W. and Loutit, M.W. (1986). Enteroviruses of human origin and

faecal coliforms in river water and sediments down stream from a sewage outfall in the

Taieri River, Otago. NZ J Marine Freshwater Res, 20: 101-105.
Lewis, G.D., Loutit, M.W. and Austin, F.J. (1985). A method for detecting human

enteroviruses in aquatic sediments./ Virol Meth, 10: 153-162.
Loria, R.M. (1988). Host conditions affecting the course of Coxsackievirus infections. In

Coxsackieviruses: A General Update, Beninelli, M. and Friedman, H. (eds). New York:

Plenum Press, pp. 125-150.

415

Viruses

Lucena, R, Bosch, A., Jofre, J. et al. (1985). Identification of viruses isolated from sewage,

river water and coastal seawater in Barcelona. Water Res, 19: 1237-1239.
Lund, E. and Ronne, V. (1973). On the isolation of viruses from sewage treatment plant

sludges. Water Res, 7: 863-866.
Maguire, H.C., Atkinson, P., Sharland, M. et al. (1999). Enterovirus infections in England

and Wales: laboratory surveillance data: 1975 to 1994. Commun Dis Public Hlth, 2:

122-125.
Manor, Y., Handsher, R., Halmut, T. et al. (1999). Detection of poliovirus circulation by

environmental surveillance in the absence of clinical cases in Israel and the Palestinian

authority./ Clin Microbiol, 37: 1670-1675.
Martins, M.T., Soares, L.A., Marques, E. et al. (1983). Human enteric viruses isolated from

influents of sewage treatment plants in S Paulo Brazil. Water Sci Tecbnol, 15: 69-73.
Matsuura, K., Ishikura, M., Yoshida, H. et al. (2000). Assessment of poliovirus eradication

in Japan: genomic analysis of polioviruses isolated from river water and sewage in

Toyama Prefecture. Appl Environ Microbiol, 66: 5087-5091.
Melnick, J.L. (1976). Enteroviruses. In Viral Infections of Humans: Epidemiology and

Control, Evans, A.S. (ed.). New York: J Wiley &c Sons, pp. 163-207.
Melnick, J.L., Emmons, J., Opton, E.M. et al. (1954). Coxsackieviruses from sewage. Am

} Hyg, 59: 183-195.
Mendelsohn, C.L., Wimmer, E., Racaniello, V.R. et al. (1989). Cellular receptor for

poliovirus: molecular cloning, nucleotide sequence, and expression of a new member of

the immunoglobulin superfamily. Cell, 56: 855-865.
Merrett, H., Pattinson, C, Stackhouse, C. et al. (1989). The incidence of enteroviruses

around the Welsh coast - a three year intensive survey. In Watershed 89. The Future of

Water Quality in Europe. Oxford: Pergamon Press, pp. 345-351.
Metcalf, T.G., Melnick, J.L. and Estes, M.K. (1995). Environmental virology: from detec-
tion of virus in sewage and water by isolation to identification by molecular biology -

a trip of over 50 years. Ann Rev Microbiol, 49: 461-487 '.
Minor, P. (1991). Picornaviridae. In Classification and Nomenclature of Viruses (Arch Virol

Suppl 2), Francki, R.I.B., Fauquet, CM., Knudson, D.L. et al. (eds). Wien: Springer-

Verlag, pp. 320-326.
Minor, P. (1992). The molecular biology of poliovaccines./ Gen Virol, 72: 3065-3077.
Minor, P. (1998). Picornaviruses. In Topley and Wilson's Microbiology and Micro-
bial Infections, 9th edn, Mahy, B.W.J, and Collier, L. (eds). London: Arnold, pp.

485-509.
Minor, T.E., Allen, C.I., Tsiatis, A. A. et al. (1981). Human infective dose determination for

oral poliovirus type 1 vaccine in infants./ Clin Microbiol, 13: 388-389.
Morris, R. (1985). Detection of enteroviruses: an assessment of ten cell lines. Water Sci

Technol, 17: 81-88.
Morris, R. and Sharp, D.N. (1984). Enteric viruses levels in wastewater effluents and sur-
face waters in the Severn Trent Water Authority 1979-81. Water Res, 18: 935-939.
Morris, R. and Sharp, D.N. (1985). Failure to detect cytopathic enteroviruses in drinking

water. Water Sci Technol, 17: 105-109.
Muir, P., Kammerer, U., Korn, K. et al. (1998). Molecular typing of enteroviruses: current

status and future requirements. Clin Microbiol Rev, 11: 202-227.
Murrin, K. and Slade, J. (1997). Rapid detection of viable enteroviruses in water by

tissue culture and semi-nested polymerase chain reaction. Water Sci Technol, 35:

429-432.
Nagington, J. (1982). Echo virus 11 infection and prophylactic antiserum. Lancet, 1: 446.
Nestor, I. and Costin, L. (1976). Presence of certain enteroviruses (Coxsackievirus) in

sewage effluents and in river waters of Romania. / Hyg Epidemiol Microbiol Immunol,

20: 137-149.
Oostvogel, P.M., van Wijngaarden, J.K., van der Avoort, H.G.A.M. et al. (1994).

Poliomyelitis outbreak in an un vaccinated community in the Netherlands, 1992-93.

Lancet, 344: 665-669.
Palfi, A.B. (1971). Virus content of sewage in different seasons in Hungary. Acta Microbiol

Acad Sci Hung, 18: 231-237.

416

Enterovirus

Pallansch, M.A. and Roos, R.P. (2001). Enteroviruses: Polio viruses, Coxsackieviruses,
Echoviruses, and newer Enteroviruses. In Fields Virology, 4th edn, Knipe, D.M. and
Howley, P.M. (eds). Philadelphia, PA: Lippincott Williams &c Wilkins, pp. 723-775.

Pallin, R., Place, B.M., Lightfoot, N.F. et al. (1997). The detection of enteroviruses in large
volume concentrates of recreational waters by the polymerase chain reaction. / Virol
Meth, 57: 67-77.

Patti, A.M., Aulicino, F.A., de Filippis, P. et al. (1990). Identification of enteroviruses isol-
ated from seawater: indirect immunofluorescence. Boll Soc Ital Biol Sper, 66: 595-600.

Payment, P., Ayache, R. and Trudel, M. (1983). A survey of enteric viruses in domestic
sewage. Can] Microbiol, 29: 111-119.

Payment, P., Affoyon, F. and Trudel, M. (1988). Detection of animal and human enteric
viruses in water from the Assomption River and its tributaries. Can J Microbiol, 34:
967-973.

Pianetti, A., Baffone, W., Citterio, B. et al. (2000). Presence of enteroviruses in the waters
of the Italian coast of the Adriatic Sea. Epidemiol Infect, 125: 455-462.

Porterfield, J.S. (1989). Picornaviridae. In Andrewes' Viruses of Vertebrates. London:
Bailliere Tindall, pp. 120-145.

Powell, K.L., Taylor, R.G., Cronin, A.A. et al. (2003). Microbial contamination of two
urban sandstone aquifers in the UK. Water Res, 37: 339-352.

Poyry, T. and Stenvik Hovi, T. (1988). Viruses in sewage waters during and after a
poliomyelitis outbreak and subsequent nationwide oral poliovirus vaccination cam-
paign in Finland. Appl Environ Microbiol, 54: 371-374.

Rao, V.C., Seidel, K.M., Goyal, S.M. et al. (1984). Isolation of enteroviruses from water,
suspended solids and sediments from Galveston Bay: survival of poliovirus and
rotavirus adsorbed to sediments. Appl Environ Microbiol, 48: 404-409.

Ravcaniello, V.R. (2001). Ficornaviridae: The viruses and their replication. In Fields
Virology, 4th edn, Knipe, D.M. and Howley, P.M. (eds). Lippincott Williams &C
Wilkins, pp. 685-722.

Reynolds, K.A., Roll, K., Fujioka, R.S. et al. (1998). Incidence of enteroviruses in Mamala
Bay, Hawaii using cell culture and direct PCR methodologies. Can J Microbiol, 44:
598-604.

Reynolds, K.A., Gerba, C.P., Abbaszadegan, M. et al. (2001). ICC/PCR detection of entero-
viruses and hepatitis A virus in environmental samples. Can] Microbiol, 47: 153-157.

Rose, J.B., de Leon, R. and Gerba, C.P. (1989). Giardia and virus monitoring of sewage
effluent in the state of Arizona. Water Sci Technol, 21: 43-47.

Rotbart, H.A. and Webster, A.D. (2001). Treatment of potentially life-threatening
enterovirus infections with pleconaril. Clin Infect Dis, 32: 228-235.

Sabin, A.B. and Boulger, L.R. (1973). History of Sabin attenuated poliovirus oral live
vaccine strains. / Biol Standard, 1: 115-118.

Sattar, S.A and Westwood, J.C. (1977). Isolation of apparently wild strains of poliovirus
type 1 from sewage in the Ottawa area. Can Med Assoc J, 116: 25-27.

Sattar, S.A. and Ramia, S. (1978). Viruses in sewage: effect of phosphate removal with
calcium hydroxide (lime). Can J Microbiol, 24: 1004-1008.

Schiff, G.M., Stefanovic, G.M., Young, E.C. et al. (1984). Studies of echovirus 12 in
volunteers: determination of minimal infectious dose and the effect of previous infection
on infectious dose. / Infect Dis, 150: 858-866.

Sellwood, J. (1994). Viruses and the EC Bathing Water Directive. PHLS Microbiol Digest,
11: 81-82.

Sellwood, J., Dadswell, J.V. and Slade, J.S. (1981). Viruses in sewage as an indicator of
their presence in the community. / Hyg Camb, 86: 217-225.

Sellwood, J., Litton, P.A., McDermott, J. and Clewley, J. P. (1995). Studies on wild and vac-
cine poliovirus isolated from water and sewage. Water Sci Technol, 31: 317-321.

Slade, J.S. (1981). Viruses and bacteria in a chalk well. Water Sci Technol, 17: 111-126.

Sobsey, M.D., Wallis, C, Henderson, M. et al. (1973). Concentration of enteroviruses
from large volumes of water. Appl Environ Microbiol, 26: 529-531.

Straub, T.M., Pepper, I.L. and Gerba, C.P. (1992). Persistence of viruses in desert soils amended
with anaerobically digested sewage sludge. Appl Environ Microbiol, 58: 636-641.

417

Viruses

Tambini, G., Andrus, J.K., Marques, E. et al. (1993). Direct detection of wild poliovirus
circulation by stool surveys of healthy children and analysis of community wastewater.
J Infect Dis, 168: 1510-1514.

Tani, N., Shimamoto, K., Ichimura, K. et al. (1992). Enteric virus levels in river water.
Water Res, 26:45-48.

van der Avoort Reimerink, J.H.J. , Ras, A., Mulders, M.N. et al. (1995). Isolation of epi-
demic poliovirus from sewage during the 1992-93 type 3 outbreak in the Netherlands.
Epidemiol Infect, 114: 481-491.

Van Regenmortel, M.H.V., Faugeuet, CM., Bishop, E.B. et al. (2000). Virus taxonomy:
The classification and nomenclature of viruses. Seventh Report of the International
Committee on Taxonomy of Viruses. San Diego, CA: Academic Press.

Van Olphen, M., de Bruin, H.A.M., Havelaar, A.H. et al. (1991). The virological quality of
recreational waters in the Netherlands. Water Sci Tecbnol, 24: 209-212.

Vaughn, J. M., Landry, E.F., Baranosky, L.J. et al. (1978). Survey of human virus occurrence
in wastewater recharged groundwater on Long Island. Appl Environ Microbiol, 36:
47-51.

Vivier, J.C., Clay, C.G. and Grabow, W.O. (2001). Detection and rapid differentiation
of human enteroviruses in water sources by restriction enzyme analysis. Water Sci
Tecbnol, 43:209-212.

418

Hepatitis A virus (HAV)

Basic microbiology

Mild jaundice as a clinical entity has been recognized for thousands of years and
by the beginning of the last century epidemic hepatitis was associated with a
short incubation period and transmission by the faecal-oral route. By the 1960s,
Krugman differentiated between clinical cases of the blood-borne hepatitis B
virus and 'infectious hepatitis', hepatitis A (Krugman et al. 9 1962).

The virion of HAV is 25-28 nm in diameter without obvious morphology
by EM and has no envelope. The RNA is a single strand of 7478 bases with
three sections including a large open reading frame. This organization is typical
of picornaviruses, but the HAV organization is sufficiently distinct to be clas-
sified in its own genus: the Hepatovirus. A range of strains of the single HAV
serotype are found around the world but the antibodies produced after infection
are cross-protective.

The virus is stable at pH 3, resistant to intestinal enzymes and a temperature
of 60°C for 10 hours, although it is inactivated at 100°C after a few minutes
(Provost et al. 9 1975). As the virions have no membrane ether, freon and chloro-
form have no effect. The virus remains viable for months after storage at room
temperature and in water, sewage and shellfish (Sobsey et al. 9 1988). The virion

Viruses

is inactivated by free residual chlorine concentration at 2 mg/1 after 15 minutes,
by iodine, by UV radiation and by 70% ethanol.

Origin of the organism

The origin of HAV strains in the environment will be human sewage. Other than
man, only primates are known to be susceptible to HAV infection. In common
with all viruses HAV is an obligate intracellular parasite, therefore no replication
can occur in the environment.

Clinical features and virulence

A prodromal stage of malaise and fever is followed in a few days by nausea,
vomiting and abdominal pain, then dark urine and jaundice develop. Jaundice
may persist for 1-2 weeks and malaise for 2-3 months. There is rarely progres-
sion to fulminant liver disease and no chronic or carrier state. Severity of the
symptoms is age related with infection more severe in adulthood. The icteric
stage is short or absent in children and asymptomatic infections are common.

Pathogenicity

The virus replicates in the hepatocytes of the liver which causes local necrosis and
a marked response by lymphocytes (inflammatory cell infiltration). Blockage
of the bile canaliculi may result in cholestasis, although the basic structure of the
liver remains unchanged. Repair of damaged tissue is usually complete in 3-4
months. Subsequent to infection, immunity is life long as neutralizing antibody
will prevent re-infection.

Transmission and epidemiology

Spread is by the faecal-oral route as large numbers of virus particles are shed in
faecal material for approximately a week before and a few days after jaundice
is apparent. There is a short period of low titre viraemia which has resulted
in rare cases of blood-borne transmission in men having sex with men, in
post-transfusion hepatitis and in the past as a result of blood products such as

420

Hepatitis A virus (HAV)

factor VIII. Current inactivation procedures used on blood products now elimi-
nate this latter route.

The transmission of HAV via the faecal-oral route is most common within
families and close contacts (Maguire et al., 1995). If hand hygiene is poor, water
is scarce or sanitation lacking, the virus will be easily spread. The population of
many countries with poor sanitation have asymptomatic infections early in
life resulting in a high level of immunity in adults. HAV infection is not common
in the UK (Gay et aL 9 1994) and many infections are due to imports from trav-
ellers returning from endemic areas. Localized outbreaks have occurred in the
UK involving drug users living in poor housing conditions. The poor standard
of hygiene and the contacts between groups in different cities lead to localized
but linked increased numbers of cases (Crowcroft, 2003). The virus spreads
readily between children within schools and day-care facilities, but infection is
often only recognized when adult family members develop symptomatic disease.

Diagnosis is by detection of circulating antibody HAV IgM which is detectable
in serum for 3 months after the onset of symptoms. Virions are present in faeces
for only a few days after the onset of symptoms. Although the virus may even-
tually grow in a very limited range of cell culture and has been detected using
EM and the more sensitive PCR, they are not appropriate for diagnosis.

Faecal contamination of food and water with HAV has led to many point
source outbreaks and is discussed below. HAV is endemic in many developing
countries of the world where children become infected early in life and a high
level of immunity is present in the adult population.

Treatment

There is no specific treatment for HAV infection and symptoms are rarely severe
enough to warrant hospitalization. Bed rest and adequate diet during the
periods of anorexia provide the required support. There is an effective vaccine
which is recommended for travellers to endemic areas and to close contacts of
cases. High titre immunoglobulin is also available for close contacts of cases and
may prevent the spread of the virus if given within 10 days of contact.

Distribution in the environment

As HAV is very difficult and slow to grow in cell culture, the detection of virus
particles in water and associated materials has relied mainly on molecular means.
Only during the last 10 years has virus been reported in water and shellfish.
HAV antigen capture PCR was used by Graff et al. (1993) in Germany for
sewage samples and by Deng et al. (1994) in the USA on septic tank effluent.

421

Viruses

Shieh et al. (1991) used gene probes to detect HAV in Danube River water, in
well water in Maryland during an outbreak and from a drinking water supply
in Mexico. Gajardo et al. (1991), also using gene probes, found HAV in sewage
samples collected in Barcelona.

RT-PCR was used to detect HAV in sewage and occasionally in treated efflu-
ent from Pune in India, an endemic area throughout the year (Vaidya et al.,
2002). Drinking water tested by Grabow et al. (2001) in South Africa had HAV
in 3% of samples. In contrast, the sewage and effluent from New Zealand was
not shown to contain HAV, although marine sediments from near the sewage
treatment works outfall did have detectable virions. It was not clear if this was
as a result of continuing shedding in the community or the persistence of virus
(Green and Lewis, 1999). Only three domestic wells out of 50, sampled in
Wisconsin four times a year, were found to contain HAV by the use of PCR on
one occasion (Borchardt et al., 2003). HAV, however, was the enteric virus most
frequently found but did not correlate with the presence of any of the bacter-
ial indicator organisms. The wells were of generally good bacterial quality.
Groundwater in the eastern states also contained HAV in nearly 9% of samples
(Abbaszadegan et al., 1999). Jothikumar et al. (1998), also in the USA, inves-
tigated the distribution of HAV in sewage from California using PCR linked to
an immunocapture bead system. Although problems of inhibition were encoun-
tered, samples from May and November were positive. Sewage contaminated
canal water was found to contain HAV in the Florida Keys in 63% of samples
tested (Griffin et al, 1999).

Shellfish have been implicated in the transmission of HAV and some of the
outbreaks are discussed below. Shellfish material is toxic to cell culture and
inhibitory to some molecular systems, consequently detection of virus is not often
reported. Enriquez et al. (1992) were able to detect HAV by cell culture in mus-
sels that had been kept in faecally-contaminated river water. Pietri et al. (1988),
Le Guyader et al. (1994) and Chung et al. (1996) detected HAV, with consider-
able difficulty, from oysters raised in sea water. A survey of shellfish in Europe
(Formiga-Cruz et al., 2002) found HAV in shellfish from all the countries'
samples except Sweden. However, the number of positive samples was small
compared to other enteric viruses found. Romalde et al. (2002a) detected
HAV in 27% of sampled bivalves grown in north-west Spain and remarked on
the inadequacy of EU standards of microbiological quality. The same group
(Romalde et al., 2002b) investigated imported shellfish from South America
after an outbreak of HAV and found four of 17 batches to be contaminated,
including the batch implicated in a Spanish outbreak of hepatitis. Green and
Lewis (1999) did not detect HAV by PCR in New Zealand oysters.

HAV will only be present in sewage if the virus is circulating within the com-
munity and therefore is less likely to be found in developed countries where the
level of endemic disease is low. The virus has not been identified in UK sewage
but has been detected in shellfish that was influenced by sewage associated
with a local outbreak (Lees, personal communication). Thornton et al. (1995) in
Ireland found no evidence of HAV infection in a follow-up epidemiological study
after a large outbreak of gastroenteritis resulting from a sewage contaminated

422

Hepatitis A virus (HAV)

borehole supplying drinking water to a town. Garin etal. (1994) did not find any
evidence that divers using surface water in France had a higher prevalence of
antibody than military personnel in the same age range. These results indicate
that the level of HAV present in sewage in northern Europe should be low.

Waterborne outbreaks

Hepatitis virus A may be spread by drinking polluted water, by contact with
recreational water and by consuming infected shellfish. The detection of virus in
the implicated environmental matrix has been reported in relatively few episodes.
Divizia et al. (1993) linked an outbreak in an Italian school with the pollution
of well water that supplied the drinking water for the school. HAV was
detected from the water after concentration using ultrafiltration and detection
by RT-PCR. Epidemiological evidence linking polluted water with outbreaks
has been successively reinforced by the use of probes (e.g. Shieh et al., 1991)
and RT-PCR (Jothikumar et al., 1998). HAV was detected in well water, a
cesspool and specimens from patients associated with a community outbreak
in Quebec, Canada (De Serres et al., 1999). Prior to the use of molecular
techniques, cell culture was able to detect infectious virus after prolonged
incubation of samples (Bloch et al., 1990). This was also described by Hejkal
et al. (1982) using radioimmunoassay on sewage and well water in Texas.

Conaty et al. (2000) undertook an epidemiological investigation that impli-
cated HAV infection with the consumption of oysters in New South Wales,
Australia. HAV was detected in samples of shellfish grown in the lake linked to
the outbreak. During the large outbreak of hepatitis in Shanghai, China, during
1988, HAV was grown in cell culture and experimentally infected marmosets
after inoculation with a suspension of clam material (Xu et al., 1992).

More commonly it is only epidemiological evidence that links HAV outbreaks
with water or consumption of shellfish. Contaminated spring water was respon-
sible for an outbreak in Kentucky (Bergeisen etal., 1985). An outbreak in India
was associated with use of a drinking water reservoir in a college cafeteria which
was shown to be heavily contaminated with indicator bacteria (Poonawagul
et al., 1995). A recent outbreak associated with tap water in Albania was
described in a ProMED website posting (January 2003). A swimming pool in
a campsite in Louisiana was implicated in an outbreak of HAV (Mahoney etal.,
1992). Direct contact with sewage after storm flooding was linked to a com-
munity outbreak in Florida (Vonstille et al., 1993). Contaminated well water
was responsible for a hardware store outbreak (Bowen and McCarthy, 1983).

A large outbreak of hepatitis was attributed to the consumption of clams in
China (Halliday et al., 1991), to clams in the UK (Anon, 1982) and to oysters
(Desenclos et al., 1991). Many other waterborne outbreaks have been
reported, especially from developing countries (Hunter, 1997).

Transmission may occur by other food routes not directly associated with
water such as raw fruits (Niu et al., 1992; Hutin et al., 1999) and by direct
faecal-oral route if an infected person with poor hygiene is involved in food

423

Viruses

preparation (CDC, 1993). Glasses used to serve drinks by a barman in a pub
were linked by Sundkvist et al. (2000) to customers who developed hepatitis.

Risk assessment

Health effects: occurrence of illness, degree of morbidity and mortality, prob-
ability of illness based on infection:

• Many hepatitis A viral infections are asymptomatic. The severity of the symp-
toms is age related; infection in adults is more severe.

• If symptoms are present, a prodromal stage of malaise and fever is followed
in a few days by nausea, vomiting and abdominal pain, then dark urine and
jaundice. Jaundice may persist for 1-2 weeks and malaise for 2-3 months.
A few patients are ill for as long as 6 months. There is rarely progression to
fulminant liver disease and no chronic or carrier state.

• The population of many countries with poor sanitation have asymptomatic
infections early in life resulting in a high level of immunity in adults.

Exposure assessment: routes of exposure and transmission, occurrence in
source water, environmental fate:

• Since the virus is difficult to detect in the environment using cell culture, infor-
mation on occurrence has been reported in the past 10 years using molecular
means. HAV has been found in sewage, surface water sources, groundwater
sources, and in food, especially shellfish. The occurrence in the environment
is dependent on the amount of infection circulating in a community; therefore,
developing countries will have higher levels of HAV in sewage than developed
countries.

• Hepatitis A is spread through faecal-oral contact. Therefore, it may be spread
by drinking polluted water, by contact with recreational water or by consum-
ing infected shellfish. It is also transmitted by close person-to-person contact.

• Bloodborne cases of transmission have been rarely reported.

• Many waterborne outbreaks have been linked to drinking water and recre-
ational water contact.

Risk mitigation: drinking-water treatment, medical treatment:

• There is no specific treatment for HAV infection, though symptoms are rarely
severe enough to warrant hospitalization.

• There is an effective vaccine which is recommended for travellers to endemic
areas and other high-risk populations, such as intravenous drug users and
people with liver disease.

• Immunoglobulin is a preparation of antibodies that can be given before
exposure for short-term protection against hepatitis A and for persons who
have already been exposed to hepatitis A virus.

424

Hepatitis A virus (HAV)

HAV is susceptible to chlorine disinfection, so adequate residuals in drink-
ing water should be sufficient to inactivate it.

References

Abbaszadegan, M., Stewart, P. and LeChevallier, M. (1999). A strategy for detection of

viruses in groundwater by PCR. Appl Environ Microbiol, 65: 444-449.
Anon. (1982). Hepatitis A: frozen cockles. Commun Dis Rep, 82: 18.
Bergeisen, G.H., Hinds, M.W. and Skaggs, J.W. (1985). A waterborne outbreak of hepatitis A

in Meade County, Kentucky. Am J Public Hlth, 75: 161-164.
Bloch, A.B., Stramer, S.L., Smith, J.D. etal. (1990). Recovery of hepatitis A virus from a water

supply responsible for a common source outbreak of hepatitis A. Am J Public Hltb, 80:

428-430.
Borchardt, M.A., Bertz, P.D., Spencer, S.K. et al. (2003). Incidence of enteric viruses in

groundwater from household wells in Wisconsin. Appl Environ Microbiol, 69:

1172-1180.
Bowen, G.S. and McCarthy, MA. (1983). Hepatitis A associated with a hardware store

water fountain and a contaminated well in Lancaster County, Pennsylvania, 1980.

Am J Epidemiol, 117: 695-705.
CDC (1993). Foodborne Hepatitis A - Missouri, Wisconsin, Alaska, 1990-1992. MMWR,

1993: 526-529.
Chung, H., Jaykus, L. and Sobsey, M.D. (1996). Detection of human viruses in oysters by

in vivo and in vitro amplification of nucleic acids. Appl Environ Microbiol, 62: 3772-3778.
Conaty, S., Bird, P., Bell, G. et al. (2000). Hepatitis A in New South Wales, Australia from

consumption of oysters: the first reported outbreak. Epidemiol Infect, 124: 121-130.
Crowcroft, N.S. (2003). Hepatitis A infections in injecting drug users. CDPH, 6: 82-85.
De Serres, G., Cromeans, T.L., Levesque, B. et al. (1999). Molecular confirmation of

Hepatitis A virus from well water: epidemiology and public health implications./ Infect

Dis, 179: 37-43.
Deng, M.Y., Day, S.P. and Cliver, D.O. (1994). Detection of HAV in environmental samples

by antigen-capture PCR. Appl Environ Microbiol, 60: 1927-1933.
Desenclos, J.-CA., Klontz, K.C., Wilder, M.H. et al. (1991). A multi-state outbreak of

hepatitis A caused by the consumption of oysters. Am J Public Hlth, 81: 1268-1272.
Divizia, M., Gnesivo, C, Amore Bonapasta, R., Morace, G. et al. (1993). Hepatitis A virus

identification in an outbreak by enzymatic amplification. Eur J Epidemiol, 9(2): 203-208.
Enriquez, R., Frosner, G.G., Hochstein-Mintzel, V. et al. (1992). Accumulation and per-
sistence of HAV in mussels./ Med Virol, 37: 174-179.
Formiga-Cruz, M., Tofino-Quesada, G., Bofill-Mas, S. etal. (2002). Distribution of human

virus contamination in shellfish from different growing areas in Greece, Spain, Sweden

and the UK. Appl Environ Microbiol, 68: 5990-5998.
Gajardo, R., Diez, J.M., Jofre, J. et al. (1991). Adsoption-elution with negatively and

positively-charged glass powder for the concentration of hepatitis A virus from water.

] Virol Meth, 31:345-352.
Garin, D., Fuchs, F., Crance, J.M. et al. (1994). Exposure to enteroviruses and HAV among

divers in environmental waters in France, first biological and serological survey of a

controlled cohort. Epidemiol Infect, 113: 541-549.
Gay, N.J., Morgan-Capner, P., Wright, J. et al. (1994). Age-specific antibody prevalence to

hepatitis A in England: implications for disease control. Epidemiol Infect, 113: 113-120.
Grabow, W.O., Taylor, M.B. and de Villiers, J.C. (2001). New methods for the detection of

viruses: call for review of drinking water quality guidelines. Water Sci Technol, 43: 1-8.
Graff, J., Ticehurst, J. and Flehmig, B. (1993). Detection of hepatitis A virus in sewage by

antigen capture polymerase chain reaction. Appl Environ Microbiol, 59: 3165-3170.

425

Viruses

Green, D.H. and Lewis, G.D. (1999). Comparative detection of enteric viruses in waste-
waters, sediments and oysters by RT-PCR and cell culture. Water Res, 33: 1195-1200.

Griffin, D.W., Gibson, C.J., Lipp, E.K. et al. (1999). Detection of viral pathogens by RT-
PCR and of microbial indicators by standard methods in the canals of the Florida Keys.
Appl Environ Microbiol, 65: 4118-4125.

Halliday, M.L., Kang, L.Y., Zhou, T.K. et al. (1991). An epidemic of Hepatitis A attribut-
able to the ingestion of raw clams in Shanghai China. / Infect Dis, 164: 852-859.

Hejkal, T.W., Keswick, B., LaBelle, R.L. et al. (1982). Viruses in a community water supply
associated with an outbreak of gastro -enteritis and infectious hepatitis. JAWA, 74:
318-321.

Hunter, P.R. (1997). Viral hepatitis. In Waterborne Disease Epidemiology and Ecology.
Chichester: John Wiley &c Sons, pp. 206-221.

Hutin, Y.J., Pool, V, Crammer, E.H. et al. (1999). A multistate, foodborne outbreak of
HAV. New Engl J Med, 340: 595-602.

Jothikumar, N., Cliver, D.O. and Mariam, T.W. (1998). Immunomagnetic capture RT-PCR
for rapid concentration and detection of HAV from environmental samples. Appl
Environ Microbiol, 64: 504-508.

Krugman, S., Ward, R. and Giles, J.P. (1962). The natural history of infectious hepatitis.
Am ] Med, 32:717-728.

Le Guyader, R, Dubois, E., Menard, D. et al. (1994). Detection of HAV, rotavirus, and
enterovirus in naturally contaminated shellfish and sediment by RT-PCR. Appl Environ
Microbiol, 60: 3665-3671.

Maguire, H.C., Handford, S., Perry, K.R. et al. (1995). A collaborative case control study
of sporadic hepatitis A in England. Commun Dis Rep CDR Rev, 5: R33-R40.

Mahoney, F.J., Farley, T.A., Kelso, K.Y. et al. (1992). An outbreak of HAV associated with
swimming in a public pool./ Infect Dis, 165: 613-618.

Niu, M.T., Polish, L.B., Robertson, B.H. etal. (1992). Multistate outbreak of HAV associated
with frozen strawberries./ Infect Dis, 166: 518-524.

Pietri, Ch., Huges, B., Crance, J.M. et al. (1988). HAV levels in shellfish exposed in a natural
marine environment to effluent from a treated sewage outfall. Water Sci Tecbnol, 20:
229-234.

Poonawagul, U., Warintrawat, S., Snitbhan, R. et al. (1995). Outbreak of hepatitis A in a
college traced to contaminated water reservoir in cafeteria. SE Asian J Trop Med Public
Hltb, 26: 705-708.

Provost, P.J., Wolanski, W., Miller, W.J. et al. (1975). Physical, chemical and morphological
dimensions of human HAV strain CR326. Proc Soc Exp Biol Med, 148: 532-539.

Romalde, J.L., Area, E., Sanchez, G. et al. (2002a). Prevalence of enterovirus and HAV in
bivalve molluscs from Galicia (NW Spain): inadequacy of the EU standards of microbio-
logical quality. Int J Food Micro, 74: 119-130.

Romalde, J.L., Torrado, I. and Barja, J.L. (2002b). Global market: shellfish imports as a
source of re-emerging food-borne HAV infections in Spain. Int Microbiol, 4: 223-226.

Shieh, Y.S., Baric, R.S., Sobsey, M.D., Ticehurst, J. et al. (1991). Detection of hepatitis A
virus and other enteroviruses in water by ssRNA probes./ Virol Methods, 31(1): 119-136.

Sobsey, M.D., Shields, P.A., Haunchman, F.S. etal. (1988). Survival and persistence of hep-
atitis A virus in environmental samples. In Viral hepatitis and liver disease, Zuckerman,
AJ. (ed.). New York: Alan R Liss, pp. 121-124.

Sundkvist, T., Hamilton, G.R., Hourihan, B.M. et al. (2000). Outbreak of HAV spread by
contaminated glasses in a public house. Commun Dis Public Hlth, 3: 60-62.

Thornton, I., Fogarty, J., Hayes, C. et al. (1995). The risk of HAV from sewage contam-
ination of a water supply. Commun Dis Rep, 5: R1-R4.

Vaidya, S.R., Chitambar, S.D. and Arankalle, V.A. (2002). PCR based prevalence of HAV,
HEV and TT viruses in sewage from an endemic area./ Hepatol, 37: 131-136.

Vonstille, W.T., Stille, W.T. and Sharer, R.C. (1993). Hepatitis A epidemics from utility
sewage in Ocoee, Florida. Arch Environ Hlth, 48: 120-124.

Xu, Z.Y., Li, Z.H., Wang, J.X. et al. (1992). Ecology and prevention of a shellfish associ-
ated hepatitis A epidemic in Shanghai, China. Vaccine, lO(SuppL): 67-68.

426

Hepatitis E virus (HEV)

Basic microbiology

The most recently recognized viral cause of hepatitis is now known as hepa-
titis E virus (Wong et aL 9 1980), but its classification is currently unspecified.
Although initially classified in the Caliciviridae family, the similarities between
HEV, beet necrotic yellow vein virus (a furovirus) and rubella virus (a toga virus)
suggest that they should be linked in related families. The genome is single-
stranded, positive sense RNA approximately 7.5 kb long. It is a non-enveloped
particle about 30 nm in diameter with little surface structure visible by EM.
Several strains have been identified, but all seem to belong to a single serotype.
A virus closely resembling human HEV has been shown to infect pigs in Europe
and in the USA, but it is unclear what the exact relationship is (Meng et aL, 1999).
As the virus successfully passes through the stomach, the virions must be acid
stable and similarly as no envelope is present, virions should be resistant to ether
and chloroform. Further studies on sensitivity ranges to chlorine and other chem-
icals have yet to be done.

Origin of the organism

The origin of HEV strains in the environment will be human sewage, although
the precise distribution of the pig HEV-like virus remains to be investigated.

Viruses

In common with all viruses HEV is an obligate intracellular parasite, therefore
no replication can occur in the environment.

Clinical features and virulence

The acute clinical features are indistinguishable from HAV and similar to a
mild case of HBV Anorexia, abdominal pain and jaundice are the main features
but are usually mild and the illness is self-limiting; there is no chronic or carrier
state. The infection can, however, be severe in pregnant women with a signifi-
ant mortality of up to 20% (Khuroo, 1991). It is not known what causes this
increase in virulence. Symptoms are most apparent in young adults. Prevalence
studies in Vietnam demonstrated an immunity rate of only 9%, indicating a sig-
nificant potential for outbreaks to occur (Hau et aL, 2000).

Pathogenicity

It is likely that after replication in the intestine the virus reaches the liver through
the portal vein and then replicates in the hepatocytes, as is the case for HAV.
Progeny virus particles are likely to be shed back into the intestine in bile.
Cell-mediated factors may contribute to liver damage and symptoms.

Treatment

There is no specific treatment for HEV infection. Neither specific immunoglobu-
lin nor vaccine is available.

Transmission and epidemiology

Spread of this virus is by the faecal-oral route but water and also food seem to
be the major routes rather than direct person-to-person contact (Aggarwal and
Naik, 1994). Major epidemics of waterborne disease have been linked with
HEV and are discussed below. The main diagnostic tool is circulating antibody
HEV IgM which is detectable in serum for 3 months after onset of symptoms.
Virions are present in faeces for a week or more after the onset of symptoms,

428

Hepatitis E virus (HEV)

which is longer than is the case for HAV. The virus has been detected using
EM and PCR which is more sensitive. Cell culture is not appropriate.

Outbreaks of HEV hepatitis have been recognized in India and most of Asia,
the Middle East, Africa and Mexico. Antibody studies suggest some circulation
of the virus in the USA and in Europe, but again the relationship to the pig virus
is currently unclear.

Distribution in the environment

Methods have been developed to detect HEV in water matrices (Grimm and
Fout, 2002) but there are few reports on natural environmental samples.
Jothikumar et al. (1993) were the first to publish a study on the detection of
HEV in sewage in Asia. Vaidya et al. (2002) found 10% of sewage samples
collected in India over a period of one year assayed using PCR were positive
for HEV compared to 24% positive for HAV.

Waterborne outbreaks

A large outbreak occurred in Somalia during 1988 and 1989 which was linked
to consumption of river water, particularly after heavy rain. Communities that
drank well water were less likely to suffer disease (Bile et al., 1994). As seen in
other outbreaks the mortality rate among pregnant women was significant at
13%. This was similar to the outbreak reported by Naik et al. (1992) in India.
Singh et al. (1995) reported an outbreak in a small, educated Indian community
during a time of water scarcity and subsequent contamination of a piped water
supply. The attack rate was nearly 2% in both children and adults. The first rec-
ognized waterborne outbreak of HEV in Indo China was reported by Corwin
et al. (1996) and linked to consumption of river water for drinking. The same
group undertook epidemiological studies which indicated that boiling river water
used for drinking was a significant protective measure (Corwin et al., 1999).
A mixed outbreak of HAV and HEV was identified in Djibouti which was
reported as waterborne (Coursaget et al., 1998). French soldiers were infected
with HAV and the local residents with HEV. A large waterborne outbreak of
HEV occurred in Islamabad when the drinking water treatment plant malfunc-
tioned (Rab et al., 1997). HEV was identified as the causative agent of an out-
break in Kashmir in 1978 (Skidmore et al., 1992).

The potential for food-borne HEV infection spreading from endemic areas
to non-endemic areas has been reviewed by Smith (2001). As evidence of person-
to-person spread of HEV is not strong and waterborne transmission well
documented in endemic areas, it seems likely that the drinking water treatment
infrastructure in developed countries should safeguard the population from HEV
infection if the virus is introduced.

429

Viruses

Risk assessment

Health effects: occurrence of illness, degree of morbidity and mortality, prob-
ability of illness based on infection:

• Many hepatitis E viral infections are asymptomatic. The severity of the symp-
toms is age related; infection in young to middle-aged adults is more severe.

• Typical clinical signs and symptoms of acute hepatitis E are similar to those
of other types of viral hepatitis and include abdominal pain, anorexia, dark
urine, fever, hepatomegaly, jaundice, malaise, nausea and vomiting.

• No evidence of chronic infection has been detected in long-term follow up of
patients with hepatitis E.

• The case fatality rate for hepatitis E is 1-3%; however, the fatality rate is up
to 20% in pregnant women.

• Death of the mother and fetus, abortion, premature delivery, or death of a
live-born baby soon after birth are common complications of hepatitis E infec-
tion during pregnancy.

Exposure assessment: routes of exposure and transmission, occurrence in source
water, environmental fate:

• HEV is transmitted primarily by the faecal-oral route and faecally-
contaminated drinking water is the most commonly documented vehicle of
transmission.

• Unlike hepatitis A infection, person-to-person transmission of HEV appears
to be uncommon.

• Hepatitis E may be a zoonotic disease, with pigs possibly serving as reservoirs
for human infection.

• In developed countries, most cases that occur are in people who have travelled
to areas where hepatitis E is endemic.

• There are few studies looking at the occurrence of hepatitis E in source water;
however, it is probably common in endemic areas, because of the number of
waterborne infections that occur.

Risk mitigation: drinking-water treatment, medical treatment:

• There is no specific treatment for HEV.

• No products are available to prevent hepatitis E, though a vaccine is currently
in development.

• Hepatitis E is susceptible to chlorination. Water that has been adequately dis-
infected should be free of infectious hepatitis E.

References

Aggarwal, R. and Naik, S.R. (1994). Hepatitis E: Intrafamilial transmission versus waterborne
spread./ Hepatol, 21: 718-723.

430

Hepatitis E virus (HEV)

Bile, K., Isse, A., Mohamud, O., Allebeck, P. et al. (1994). Contrasting roles of rivers and
wells as sources of drinking water on attack and fatality rates in a hepatitis E epidemic in
Somalia. Am J Trop Med Hyg, 51: 466-474.

Corwin, A.L., Khiem, H.B., Clayson, E.T. et al. (1996). A waterborne outbreak of HEV trans-
mission in southwestern Vietnam. Am J Trop Med Hyg, 54: 559-562.

Corwin, A.L., Tien, N.T., Bounlu, K. et al. (1999). The unique riverine ecology of HEV trans-
mission in South-East Asia. Trans Roy Soc Trop Med Hyg, 93: 255-260.

Coursaget, P., Buisson, Y., Enogat, N. et al. (1998). Outbreak of enterically-transmitted hepa-
titis due to HAV and HEV. / Hepatol, 28: 745-750.

Grimm, A.C. and Fout, G.S. (2002). Development of a molecular method to identify hepatitis
E virus in water. / Virol Meth, 101: 175-188.

Hau, C.H., Hien, T.T., Khiem, H.B. et al. (2000). Prevalence of enteric hepatitis A and E
viruses in the Mekong River delta region of Vietnam. Am J Trop Med Hyg, 60: 277-280.

Jothikumar, N., Aparna, K., Kamatchiammal, S. et al. (1993). Detection of hepatitis E
virus in raw and treated wastewater with the polymerase chain reaction. Appl Environ
Microbiol, 59: 2558-2562.

Khuroo, M.S. (1991). Hepatitis E: the enterically transmitted non-A, non-B hepatitis. Indian
J Gastroenterol, 10: 96-100.

Meng, X.J., Dea, S., Engle, R.E. et al. (1999). Prevalence of antibodies to the hepatitis E virus
(HEV) in pigs from countries where hepatitis E is common or rare in the human popula-
tion./ Med Virol, 59: 297-302.

Naik, S.R., Aggarwal, R., Salunke, P.N. et al. (1992). A large waterborne viral hepatitis E
epidemic in Kanpur, India. Bull WHO, 70: 597-604.

Rab, M.A., Bile, M.K., Mubarik, M.M. et al. (1997). Waterborne HEV epidemic in
Islamabad, Pakistan: a common source outbreak traced to the malfunction of a modern
water treatment plant. Am J Trop Med, 57: 151-157.

Singh, J., Aggarwal, N.R., Bhattacharjee, J. et al. (1995). An outbreak of viral hepatitis E:
role of community practices. / Commun Dis, 27: 92-96.

Skidmore, S.J., Yarbough, P.O., Gabor, K.A. et al. (1992). Hepatitis E virus: The cause of a
waterborne hepatitis outbreak./ Med Virol, 37: 58-60.

Smith, J.L. (2001). A review of HEV/ Food Protect, 64: 572-586.

Vaidya, S.R., Chitambar, S.D. and Arankalle, V.A. (2002). PCR based prevalence of HAV,
HEV and TT viruses in sewage from an endemic area./ Hepatol, 37: 131-136.

Wong, D.C., Purcell, R.H. and Sreenivasan, M.A. (1980). Epidemic and endemic hepatitis in
India: evidence for nonA/nonB hepatitis virus aetiology. Lancet, 2: 876-878.

431

Norovirus and sapovirus

Basic microbiology

Norwalk virus was identified in faecal material by immune electron microscopy
during an outbreak of winter vomiting disease in Ohio (Kapikian et al., 1972)
after which many similar viruses causing similar outbreaks were identified
around the world. New strains have been named after the place in which they
were found. A name for this collection of viruses has been more problematic,
but recently, the International Committee for the Taxonomy of Viruses has des-
ignated 'norovirus' for the name of the genus, to replace the terms Norwalk
virus, Norwalk-like virus, NLV and the UK term 'small round-structured virus'
(SRSV) (Caul and Appleton, 1982). Comparisons of the RNA polymerase and
capsid sequences of the genome have shown that noroviruses are a genus within
the Caliciviridae and have at least two distinct genogroups. Genogroup I viruses
include Norwalk virus and Southampton virus, genogroup II includes Hawaii,
Bristol and Lordsdale viruses (Green et al., 2000b, 2001). Genomic relation-
ships of newly identified viruses are determined by sequence analysis of the cap-
sid proteins (Green et aL 9 1995, 2000a). Clusters (genotypes) of strains with
approximately 80% or more similarity in amino acid sequence are grouped
together and linked, with the first reported strain as reference virus. Bovine and
porcine noroviruses have also been identified.

Viruses

Three other genera within the Caliciviridae are currently recognized. The
International Committee for the Taxonomy of Viruses has designated the term
'sapovirus' as the name of the genus, to replace Saporo virus, classic calicivirus,
human calicivirus or Saporo-like virus (SLV). The genus includes Saporo virus
and Manchester virus. Virus particles have a very distinct cup-shape pattern on
the surface and ragged outline visible by EM and genetically are significantly
different to the noroviruses. The porcine enteric calicivirus is a sapovirus. The
other two genera do not contain human strains of virus, but of note is the feline
calicivirus which, although a respiratory virus, has been used as a surrogate
for norovirus in some experimental studies due to its ability to replicate in cell
culture.

Calicivirus particles are 27-30 nm in diameter and have a ragged outline. The
capsid surface is a series of arches and cup-like depressions which, using elec-
tron microscopy, is seen as an amorphous surface with no discernible features in
noroviruses. In contrast, sapovirus virions have well-defined cup shapes on the
surface. The capsid is made of a single protein arranged in icosahedral symmetry
and the genome is single-stranded RNA (Clarke et aL 9 1998) approximately
7.5 kb in length with three open reading frames (ORFs). As the virus cannot yet
be grown in cell culture and there are few animal models, the characterization
of the virus components has been difficult and little is known of the replication
mechanisms in vivo.

Studies on Norwalk virus indicate that the virus is not inactivated by pH 3,
or ether or chloroform or by 30 minutes at 60°C and is only partially inacti-
vated by 70% alcohol (Clarke et al. 9 1998). However, it is inactivated by
chlorine at >10mg/l, although not by free residual chlorine at 0.5-1.0 mg/1
(Keswick et aL, 1985). Studies on the disinfection of noroviruses are ham-
pered by the lack of an infectivity assay, but current limited data suggest that
they may be more resistant than enteroviruses. Further studies are needed to
confirm this.

Origin of the organism

In common with all viruses the noroviruses and sapoviruses are obligate
intracellular parasites and, as for most virus groups, under normal conditions
they are species specific. The origin of human viruses in sewage will be human
faecal material. No multiplication will occur outside the living host cell.

Clinical features and virulence

Diarrhoea and vomiting are the predominant symptoms of norovirus infection
with accompanying headache and myalgia. Projectile vomiting is often present

434

Norovirus and sapovirus

in more than 50% of adult cases and is one of the features that indicate an out-
break may be due to norovirus (Kaplan etal. 9 1982a). Asymptomatic infection
does occur although may not be common. Symptoms can be incapacitating but
usually resolve within 2-3 days with no complications.

Sapovirus causes gastroenteritis in which diarrhoea is the predominate
feature and vomiting is less likely to occur.

Nothing is known of the factors that influence virus virulence.

Pathogenicity

Virus particles enter the host through the mouth; the virions are unaffected by
the acidic pH as they pass through the stomach. Virus replication probably
takes place in the mucosal epithelium of the small intestine resulting in damage
to the epithelium (enterocytes) and flattening of the villi. The incubation period
is commonly 24 hours followed by the sudden onset of nausea and vomiting.
The projectile vomiting that is such a marker feature of norovirus symptoms
may be due to delayed emptying of the stomach caused by abnormal gastric
motor function (Meeroff et al. 9 1980). Virus shedding in faecal material con-
tinues for a week or sometimes two but rarely longer.

Immunity appears to be short lived and different strains of noroviruses are
not cross-protective. Children become infected early in life but symptomatic
re-infections occur throughout adult life. Results from volunteer infection stud-
ies are complex but suggest that immunity to any one strain may only last 9
months in some individuals (Green et al. 9 2001). Factors other than circulating
antibody may be important in determining the host response to infection.

Sapovirus immunity has been less studied but as infection is most common
in babies and young children, it may be more long lasting than norovirus.

Epidemiology

Transmission of noroviruses is mainly direct person to person by the faecal-oral
route or by the ingestion of particles of vomitus. Projectile vomiting has been
shown to contaminate the environs from which re-suspended, aerosolized parti-
cles may be swallowed and infection transmitted (Chadwick and McCann, 1994;
Cheesbrough et al 9 2000). Marks et al. (2000) demonstrated transmission of
the virus most frequently in persons situated near to an index case of vomiting
and less frequently when further away. In a report by Evans et al. (2002) trans-
mission occurred over time when an index case contaminated an area of theatre
seating and successive clients using the seating became ill. Transmission by food
and water is discussed in detail below. Norovirus gastroenteritis, with vomiting
a particular feature, occurs in all ages and occurs in sporadic and epidemic

435

Viruses

disease patterns (Caul, 1996a,b). Secondary cases are common among close
contacts in the family, in schools, in residential settings, in hospitals and on
cruise ships.

Initially called 'winter vomiting disease' the infection is now recognized
throughout the year in temperate climates. The Infectious Intestinal Disease Study
in the UK 1993-1996 investigated gastrointestinal disease in the community and
found that norovirus gastroenteritis was significantly more frequent than previ-
ously recognized (Wheeler et ai 9 1999). Noroviruses were the most common
cause of gastroenteritis in adults and have an epidemic or outbreak pattern over-
lying the endemic disease. The PHLS Communicable Disease Report of December
19, 2002 stated that 3029 cases of noroviruses in the UK had been reported in
the first 10 months of that year. This was the highest level yet recognized and
substantially more than any other gastroenteritis-causing pathogen. There had
not been the usual decline in the summer months. Patients aged over 65 years of
age had accounted for 68% of cases. As noroviruses are readily transmitted, out-
breaks are common within enclosed and residential areas such as schools, nursing
homes, hospital wards, hotels and cruise ships (Caul et ai 9 1979; Cubitt et aL 9
1981; Dolin et al. 9 1982; Ho et al 9 1989).

More outbreaks of gastroenteritis are caused by noroviruses than any other
infectious agent in the UK. In 2000, over 250 outbreaks were reported as caused
by noroviruses compared to approximately 50 due to Salmonella (Public Health
Laboratory Service unpublished data). Similar levels of infection have been
reported world-wide but as laboratory diagnosis is difficult much is based on
clinical presentation and outbreak features (Kaplan et ai 9 1982a).

Oliver et al. (2003) investigated the risk of infection to humans by bovine
noroviruses and concluded that there was no evidence of transmission.

Detection of human noroviruses in faecal material may be done by electron
microscopy (EM), but this requires specialist equipment and staff. It is also rela-
tively insensitive as more than 10 6 particles/ml must be present for the virus to
be identified. However, much of the initial investigation of this virus group was
based on EM (Caul and Appleton, 1982). It remains the only method of detect-
ing any virus that may be present in faecal material. More recent advances have
been possible using molecular techniques directed specifically at noroviruses.
Polymerase chain reaction (PCR) is discussed fully under 'Detection'; it is much
more sensitive than EM which has allowed virus identification for a wider time
range after the onset of symptoms. This has in turn lead to a more accurate pic-
ture of norovirus infections. ELISA kits are now available for clinical specimens
which will mean that diagnosis will not be limited to specialist centres.

Reported clinical infections of noroviruses are usually linked to genogroup II
viruses but the predominant genotype or genetic cluster can vary from year to
year (Hale et al., 2000). Paradoxically the same strain can cause large out-
breaks year after year.

The transmission and epidemiology of sapoviruses are less well under-
stood. Direct person-to-person spread by the faecal-oral route is assumed to
be the most common method of transmission. Symptomatic illness in the UK

436

Norovirus and sapovirus

occurs mainly in babies throughout the year (Public Health Laboratory Service
Communicable Disease Report, 2002). Outbreaks have been identified mainly
in nurseries but occasionally in schools.

Treatment

There is no specific treatment and the symptoms are rarely sufficiently severe
to require rehydration.

Distribution in the environment

Noroviruses have only recently been identified in aquatic samples. As with
enterovirus investigation, sewage, river and marine samples must usually be
processed to concentrate the virus. Processing methods that have been used for
enterovirus concentration have been effective for noroviruses. Techniques,
particularly adaptation of molecular methods used to detect virus in clinical
samples, have now been used successfully to identify the norovirus genome in
a range of water matrices. It is not possible to grow the virus in animal models
or in cell culture so it cannot be proven that the molecular material originated
in an infectious particle.

In the UK, noroviruses have been identified in sewage, effluent, river and sea-
water samples (Wyn-Jones et al., 2000; Sellwood, unpublished results) (Table
32.1). Both genogroups and a range of strains of virus have been found through-
out the year using the primer set that is also used in the UK for many clinical
investigations. Lodder et al. (1999) detected both genogroups of virus in sewage
in the Netherlands. Griffin et al. (1999) identified noroviruses in 10% of canal
water sites in the Florida Keys using one of two primer sets. Interestingly, only
a primer set that detected Norwalk strain gave positive results. Waters most
influenced by sewage discharge were the most often found to contain virus.
Katayama et al. (2002) detected noroviruses in Tokyo Bay coastal sea water
during the winter.

Less polluted water has been shown to contain noroviruses; in the UK, urban
groundwater samples were positive during the winter months (Powell et al.,
2003). Mineral waters were investigated by Beuret et al. (2002) in Switzerland
and several brands were positive by their tests in litre size samples; this study
remains to be repeated by others. Borchardt et al. (2003) found evidence of
transient norovirus contamination in a small number of private household
wells in the USA.

Environmental contamination in hospitals, hotels and restaurants has been
reported as the transmission route for noroviruses during outbreaks (Green et al.,
1998b; Cheeseborough et al., 1997; Marks et al., 2000). Contamination of

437

Viruses

Table 32.1 Norovirus detected in UK sewage, 2000

Crude

sewage 3

Crude

sewage b

Final effluent

Month

unprocessed

100 ml

processed

1 litre processed

January

February

March

April

May

June

July

August

September

October

November

December

a 140|xl aliquot was used for all RT-PCR (Wyn-Jones etal., 2000).

b 100 ml sewage sample concentrated by beef extract protein precipitation to 10 ml final volume.

c 1 I final sewage effluent sample concentrated by beef extract protein precipitation to 10 ml final

volume.

D: detected.

Nd: not detected.

carpets was detected in a hotel where a prolonged outbreak of norovirus gas-
troenteritis occurred (Cheesbrough etal.^ 2000). Swabs from areas where vomit-
ing had contaminated carpets and also from curtain material at high level were
positive, thus supporting the suggestion that such contamination is a significant
factor in the transmission of virus. Virus has also been detected from surfaces
used in food preparation (Taku etal., 2002).

Bivalve shellfish such as oysters and mussels have been found to contain
noroviruses by groups in the UK (Lees et ai 9 1995; Henshilwood et al. 9 1998),
other parts of Europe (Le Guyader et al. 9 2000; Hernroth et al. 9 2002; Formiga-
Cruz et al. 9 2002) and the USA (Le Guyader et al. 9 1996; Kingsley et al. 9 2002).

Waterborne outbreaks

Outbreaks on record with a greater or lesser degree of association with
noroviruses include one at a tourist resort hotel, where sewage had seeped
through faults in the rock strata into the borehole water used to supply the hotel
(Lawson, 1991) and another at a mobile home park, which was the result of
sewage gaining access to the well (McNulty, 1993). Sewage also gained access
to a borehole supply in Naas, Ireland in 1991, which resulted in a reported
6000 cases of gastroenteritis (Fogarty, 1995). Nearly 50 cases of gastroenteritis
occurred in 1992 after water from the River Thames, being used for irrigation
purposes, entered the drinking-water distribution system as a result of a back-
flow from a farm installation (Gutteridge et ai 9 1994) and a similar incident

438

Norovirus and sapovirus

occurred in Riding Mill, Northumberland, in 1992 (Stan well-Smith, 1994). In all
of these early outbreaks it was epidemiological assessment and clinical symp-
toms that suggested noroviruses as the infective cause.

Epidemiological evidence combined with the detection of norovirus in stool
specimens supported the assertion that contaminated ice produced on a cruise
ship caused the outbreak of gastroenteritis reported by Khan et al. (1994). An
outbreak, in which 45 out of 70 hikers who drank water from a general store
on the Appalachian Trail in the USA became ill (Peipins et al., 2002), was con-
firmed by epidemiology, clinical symptoms and the detection of norovirus in
stool specimens and somewhat strengthened by the finding of faecal coliform
organisms in the water. Contaminated water that was used in the production of
cakes in a bakery is likely to have been the cause of an outbreak of noroviruses
among employees and customers in South Wales (Brugha et al., 1999); the
strength of association here was the detection and typing of the virus in stool
samples and again poor bacterial quality of a drinking water supply.

A well in Finland, polluted by river water during spring floods, transmitted
several groups of pathogens to the inhabitants of a village who then developed
gastroenteritis (Kukkula et al., 1997). The drinking water of an Italian resort
(Boccia et al., 2002) during the summer of 2000 and drinking water in Andorra
during the skiing season (Pedalino et al., 2003) were strongly associated with
widespread outbreaks of noroviruses and poor quality water. Kukkula et al.
(1999) reported another outbreak of norovirus gastroenteritis in Heinavesi,
Finland where by epidemiological estimate 1700-3000 cases occurred. Here, not
only was the virus detected in the stool samples of those affected, but it was also
detected in the municipal water supply and in consumers' taps. The water supply
to the village had poor or non-existent chlorination. The virus was typed and
genogroups I and II found in the stools and genogroup II virus found in the water,
so the cause and effect were well reconciled. Maurer and Sturchler (2000)
reported an outbreak of gastroenteritis of mixed aetiology in La Neuveville,
Switzerland involving over 1600 individuals. Campylobacter jejuni, Shigella
sonnei and noroviruses were detected in stool samples and a norovirus isolate
showed identical sequence homology with one isolated from the water supply.
Another instance of the value of molecular techniques was demonstrated by
Brown et al. (2001) in linking the water supply to an outbreak of gastroenteri-
tis involving 448 individuals in a resort hotel in Bermuda where 18 out of 19
stool specimens were positive for genogroup II noroviruses which were also
detected in a 3 -litre water sample. Epidemiology and clinical features of a
norovirus outbreak combined with noroviruses identified in stool samples and
noroviruses in water with identical nucleotide sequences in an outbreak was
reported by Anderson et al. (2003) in snowmobilers at a winter resort.

Many of these outbreaks are associated with small drinking-water systems
with limited supply that were in challenging weather conditions: dry and hot or
frozen/thawing or flood. In these circumstances the integrity of the infrastruc-
ture can be compromised and drinking-water supply becomes contaminated
with sewage. If disinfection is not available or is not adequate, transmission may
occur.

439

Viruses

Freshwater recreation has been more closely linked with waterborne illness
than contact with sea water. Canoeists using the River Trent developed gastro-
enteritis caused by noroviruses (Gray et al. 9 1997) and undiagnosed gastro-
enteritis was found to be more common in canoeists using the River Trent
than those using unpolluted lakes in North Wales (Fewtrell, 1992). Studies in
the Netherlands indicate that illness and norovirus infection are linked to
swimming in river water (Medema etal., 1995; van Olphen et al. 9 1991; de Roda
Husman personal communication).

Bivalve shellfish, such as oysters and mussels, are a major transmission route
of norovirus gastroenteritis and have been reviewed by Lees (2000). The large
outbreak in the USA in 1993 included at least 25 clusters of gastroenteritis
spread over seven states and infected over 100 people (Kohn et al. 9 1995).
Linking outbreaks in restaurants to batches of shellfish and obtaining stool
specimens is difficult in practice so much of the evidence for shellfish transmis-
sion is epidemiological.

A common feature of waterborne and shellfish-associated norovirus out-
breaks is the finding that mixed genogroups of virus are present in the stool.
Many of the outbreaks described above noted this finding. A single genogroup is
usually found in faecal specimens from patients infected by the major route of
norovirus transmission, i.e. person-to-person transmission as seen in sporadic
cases, hospitals, residential and commercial settings. Transmission by aerosolized
vomitus and environmental contamination is being increasingly recognized
and has been reported in cruise ships, theatre seating, hospital wards and
restaurants.

Surface contamination of food by food-handlers is another well-recognized
transmission route of norovirus gastroenteritis (Advisory Committee on the
Microbiological Safety of Food, 1998). Food which is served uncooked or
handled after cooking is a potential hazard. Many types of food have been asso-
ciated, on epidemiological evidence, with transmission: sandwiches, salad, rasp-
berry gateaux, watercress, frosting mix, carrots, lobster tail, melon cocktail
and even hamburger and fries. As yet detection of the virus from the food is
difficult and under development. Good hygiene and information for food
handlers is crucial for prevention of transmission.

Risk assessment

Health effects: occurrence of illness, degree of morbidity and mortality, prob-
ability of illness based on infection:

• Seroprevalence of identified strains is usually high - approaching
100% - by adulthood. Seropositivity does not affect the susceptibility of
re-infection, except perhaps in sapoviruses. Noroviruses are estimated to
be the number one cause of viral gastroenteritis outbreaks in the USA
and the UK.

440

Norovirus and sapovirus

• Clinical symptoms of norovirus are generally mild and self-limiting: nau-
sea, vomiting, diarrhoea and fever that last for 1-3 days. Mortality is rare and
has occurred mainly in people with pre-existing conditions and the elderly.
Sapovirus causes gastroenteritis in which diarrhoea is the more predominat-
ing feature.

• Attack rates in drinking water outbreaks have ranged from 31 to 87%.
Three studies have shown 46%, 68% and 85% of people developing symp-
toms after becoming infected. The probability of developing illness may
relate to genetic susceptibility in the host.

Exposure assessment: routes of exposure and transmission, occurrence in
source water, environmental fate:

• Transmission is faecal-oral, either person to person or through a common
source such as water or food. Aerosolization from vomitus is a source of
infection as well.

• Secondary spread can be high in noroviruses but apparently uncommon in
sapoviruses.

• An infectious dose of 10 PCR-detectable units has been shown through oral
ingestion.

• Waterborne outbreaks of norovirus have been identified. Researchers esti-
mated that 23% of waterborne outbreaks in the USA between 1975 and 1981
were due to Norwalk-like viruses. Freshwater recreation has been more closely
linked with waterborne illness than contact with seawater. Food has been
identified as a common outbreak source - especially shellfish.

• Noroviruses have only recently been identified in aquatic samples. In the UK
and other countries, noroviruses have been identified in sewage, effluent, river
and seawater. Noroviruses have been found in groundwater and bottled min-
eral water samples. Concentrations at intake are unknown, but dependent on
level of human faecal contamination in the source water.

Risk mitigation: drinking-water treatment, medical treatment:

• Studies on the disinfection of noroviruses are hampered by the lack of
an infectivity assay, but norovirus is inactivated by chlorine at >10mg/l,
although not by free residual chlorine at 0.5-1.0 mg/1.

• The illness is self-limiting. Rehydration therapy may be necessary in rare
cases.

References

Advisory Committee on the Microbiological Safety of Food. (1998). Report on foodborne

viral infections. London: HMSO.
Anderson, A.D., Heryford, A.G., Sarisky, J.P. et al. (2003). A waterborne outbreak of

Norwalk-like virus among snowmobilers - Wyoming, 2001. / Infect Dis, 15:

303-306.

441

Viruses

Beuret, C, Kohler, D., Baumgartner, A. etal. (2002). Norwalk-like virus sequences in mineral
waters: one-year monitoring of three brands. App I Environ Microbiol, 68: 1925-1931.

Boccia, D., Tozzi, A.E., Cotter, B. et al. (2002). Waterborne outbreak of Norwalk-like virus
gastro-enteritis at a tourist resort, Italy. Emerg Infect Dis, 8: 563-568.

Borchardt, M.A., Bertz, D., Spencer, S.K. et al. (2003). Incidence of enteric viruses in ground-
water from household wells in Wisconsin. Appl Environ Microbiol, 69: 1172-1180.

Brown, CM., Cann, J.W., Simons, G. et al. (2001). Outbreak of Norwalk virus in a
Caribbean island resort: application of molecular diagnostics to ascertain the vehicle of
infection. Epidemiol Infect, 126: 425-432.

Brugha, R., Vipond, I.B., Evans, M.R. et al. (1999). A community outbreak of food-borne
SRSV gastro-enteritis caused by a contaminated water supply. Epidemiol Infect, 122:
145-154.

Caul, E.O. (1996a). Viral gastroenteritis: small round structured viruses, caliciviruses and
astroviruses. Part I. The clinical and diagnostic perspective./ Clin Pathol, 49: 874-880.

Caul, E.O. (1996b). Viral gastroenteritis: small round structured viruses, caliciviruses and
astroviruses. Part II. The epidemiological perspective./ Clin Pathol, 49: 959-964.

Caul, E.O. and Appleton, H. (1982). The electron microscopical and physical characteris-
tics of SRSVs: an interim scheme for classification. / Med Virol, 9: 257-265.

Caul, E.O., Ashley, C.R. and Pether, J.V.S. (1979). 'Norwalk'-like particles in epidemic
gastro-enteritis in the UK. Lancet, ii: 1292.

Chadwick, P.R. and McCann, R. (1994). Transmission of SRSV by vomiting during a
hospital outbreak of gastro-enteritis./ Hosp Infect, 26: 251-259.

Cheesbrough, J.S., Barkess-Jones, L. and Brown, D.W.G. (1997). Possible prolonged envi-
ronmental survival of SRSV. / Hosp Infect, 35: 325-326.

Cheesbrough, J.S., Green, J. and Gallimore, C.I. (2000). Widespread environmental contam-
ination with Norwalk-like viruses (Norovirus) detected in a prolonged hotel outbreak
of gastro-enteritis. Epidemiol Infect, 125: 93-98.

Clarke, I.N., Lambden, P.R. and Caul, E.O. (1998). Human enteric RNA viruses: Caliciviruses
and astroviruses. In Topley and Wilsons' Microbiology and Microbial Infections, 9th edn,
Collier, L., Ballows, A. and Sussman, M. (eds). London: Edward Arnold, pp. 511-535.

Cubitt, W.D., Paed, P.J. and Saeed, A. A. (1981). A new serotype of calicivirus associated
with an outbreak of gastro-enteritis in a residential home for the elderly./ Clin Pathol,
34: 924-926.

Dolin, R., Reichmann, R.C., Roessner, K.D. et al. (1982). Detection by immune electron
microscopy of the Snow Mountain agent of acute viral gastro-enteritis. / Infect Dis,
146: 184-189.

Evans, M.R., Meldrum, R., Lane, W., Gardner, D. et al. (2002). An outbreak of viral gas-
troenteritis following environmental contamination at a concert hall. Epidemiol Infect,
Oct, 129(2): 355-360.

Fewtrell, L., Godfree, A.F., Jones, F., Kay, D. et al. (1992). Health effects of white-water
canooeing. Lancet, 339: 1587-1589.

Fogarty, J., Thornton, L. and Hayes, C. (1995). Illness in a community associated with an
episode of water contamination with sewage. Epidemiol Infect, 114: 289-295.

Formiga-Cruz, M., Tofino-Quesada, G., Bofill-Mas, S. etal. (2002). Distribution of human
virus contamination in shellfish from different growing areas in Greece, Spain, Sweden
and the UK. Appl Environ Microbiol, 68: 5990-5998.

Gray, J.J., Green, J., Gallimore, C. et al. (1997). Mixed genotype SRSV infections among a
party of canoeists exposed to contaminated recreational water./ Med Virol, 52: 425-429.

Green, J., Hale, A.D. and Brown, D.W.G. (1995). Recent developments in the detection and
characterisation of small round structured viruses. PHLS Microbiol Digest, 12: 219-222.

Green, J., Wright, P.A., Gallimore, C.I. et al. (1998). The role of environmental contami-
nation with small round structured viruses in a hospital outbreak investigated by
reverse-transcriptase polymerase chain reaction assay./ Hosp Infect, 39: 39-45.

Green, J., Vinje, J., Gallimore, C.I. et al. (2000a). Capsid diversity among Norwalk-like
viruses. Virus Genes, 20: 227-236.

Green, K.Y., Ando, T., Balayan, M.S. et al. (2000b). Taxonomy of the caliciviruses./ Infect
Dis, 181: S322-S330.

442

Norovirus and sapovirus

Green, K.Y., Chanock, R.M. and Kapikian, A.Z. (2001). Human Caliciviruses. In Fields

Virology, Knipe, D.M. and Howley, P.M. (eds). Philadelphia, PA: Lippincott Williams &C

Wilkins, pp. 841-874.
Griffin, D.W., Gibson, C.J., Lipp, E.K. and Riley, K. (1999). Detection of viral pathogens by

reverse transcriptase PCR and of microbial indicators by standard methods in the canals of

the Florida Keys. Appl Environ Microbiol, 65: 4118-4425. Erratum in: Appl Environ

Microbiol, 2000 Feb, 66(2): 876.
Gutteridge, W. and Haworth, E.A. (1994). An outbreak of gastrointestinal illness associ-
ated with contamination of the mains supply by river water. PHLS Commun Dis Rep

CDRRev,4:R50-R51.
Hale, A., Mattick, K., Lewis, D. et al. (2000). Distinct epidemiological patterns of

Norwalk-like virus infections./ Med Virol, 62: 99-103.
Henshilwood, K., Green, J., Gallimore, C.I. et al. (1998). The development of PCR assays

for detection of SRSV and other human enteric viruses in molluscan shellfish./ Shellfish

Res, 17: 1675-1678.
Hernroth, B.E., Conden-Hansson, A.-C, Rehnstam-Holm, A.-S. et al. (2002). Environmental

factors influencing human viral pathogens and their potential indicator organisms in the

Blue Mussel, Mytilus edulis: the first Scandinavian report. Appl Environ Microbiol,

68: 4523-4533.
Ho, M.S., Glass, R.I.M. and Monroe, S.S. (1989). Viral gastro -enteritis aboard a cruise

ship. Lancet, 2: 961-965.
Kapikian, A.Z., Wyatt, R.G., Dolin, R. et al. (1972). Visualisation by immune electron

microscopy of a 27nm particle associated with infectious non-bacterial gastro-enteritis.

/ Virol, 10: 1075-1081.
Kaplan, J.E., Feldman, R., Campbell, D.S. et al. (1982a). The frequency of a Norwalk-
like pattern of illness in outbreaks of acute gastro-enteritis. Am J Public Hlth, 72:

1329-1332.
Katayama, H., Shimasaki, A. and Ohgaki, S. (2002). Development of a virus concentration

method and its application to detection of enterovirus and Norwalk virus from coastal

seawater. Appl Environ Microbiol, 68: 1033-1039.
Keswick, B.H., Satterwhite, T.K., Johnson, P.C. et al. (1985). Inactivation of Norwalk virus

in drinking water by chlorine. Appl Environ Microbiol, 50: 261-264.
Khan, A.S., Moe, C.L., Glass, R.I. et al. (1994). Norwalk virus associated gastro-enteritis

traced to ice consumption aboard a cruise ship in Hawaii: comparison and application

of molecular method-based assays. Appl Environ Microbiol, 32: 318-322.
Kingsley, D.H., Meade, G.K. and Richards, G.P. (2002). Detection of both HAV and

Norwalk-like viruses in imported clams associated with food-borne illness. Appl

Environ Microbiol, 68: 3914-3918.
Kohn, M., Farley, T.A., Ando, M. et al. (1995). An outbreak of Norwalk virus gastro-
enteritis associated with eating raw oysters. JAMA, 273: 466-471.
Kukkula, M., Arstila, P., Klossner, M.-L. et al. (1997). Waterborne outbreak of viral

gastro-enteritis. Scand J Infect Dis, 29: 415-418.
Kukkula, M., Maunula, L., Silvennoinen, E. et al. (1999). Outbreak of viral gastroenteritis

due to drinking water contaminated by Norwalk-like viruses. / Infect Dis, 180:

1771-1776.
Lawson, H.W., Braun, M.M., Glass, R.I.M. et al. (1991). Waterborne outbreak of

Norwalk virus gastro-enteritis at a southwest US resort: role of geological formations in

contamination of well water. Lancet, 337: 1200-1204.
Le Guyader, F., Haugarreau, L., Miossec, L. et al. (2000). Three year study to assess human

enteric viruses in shellfish. Appl Environ Microbiol, 66: 3241-3248.
Le Guyader, E, Neill, F.H., Estes, M.K. et al. (1996). Detection and analysis of a SRSV

strain in oysters implicated in an outbreak of acute gastro-enteritis. Appl Environ

Microbiol, 62: 4268-4272.
Lees, D.N. (2000). Viruses and bivalve shellfish. Int J Food Microbiol, 59: 81-116.
Lees, D.N., Hensilwood, K., Green, J. et al. (1995). Detection of small round structured

viruses in shellfish by reverse transcription-PCR. Appl Environ Microbiol, 61:

4418-4424.

443

Viruses

Lodder, W.J., Vinje, J., van de Heide de Roda Husman, A.M. et al. (1999). Molecular

detection of Norwalk-like caliciviruses in sewage. Appl Environ Microbiol, 65:

5624-5627.
Maurer, A.M. and Sturchler, D. (2000). A waterborne outbreak of small round structured

virus, Campylobacter and Shigella co-infection in La Neuveville, Switzerland, 1998.

Epidemiol Infect, 125: 325-332.
Marks, P.J., Vipond, I.B., Carlisle, D. et al. (2000). Evidence for Norwalk-like virus (NLV)

in a hotel restaurant. Epidemiol Infect, 124: 481-487.
McAnulty, J.M., Rubin, G.L., Carvan, C.T., Huntley, E.J., Grohmann, G. and Hunter, R.

(1993). An outbreak of Norwalk-like gastroenteritis associated with contaminated

drinking water at a caravan park. Aust J Public Hltb, 17: 36-41.
Medema, G.J., van Asperen, I. A., Kokman-Houweling, J.M., Nooitgedagt, A. et al. (1995).

The relationship between health effects in triathletes and microbiological quality of

freshwater. Water Sci Tecbnol, 31: 19-26.
Meeroff, J.C., Schreiber, D.S., Trier, J.S. et al. (1980). Abnormal gastric motor function in

viral gastro-enteritis. Ann Intern Med, 92: 370-373.
Oliver, S.L., Dastjerdi, A.M., Wong, S. et al. (2003). Molecular characterisation of bovine

enteric caliciviruses: a distinct third genogroup of noroviruses (NLV) unlikely to be of

risk to humans./ Virol, 77: 2789-2798.
Pedalino, B., Feely, E., McKeown, B. et al. (2003). An outbreak of Norwalk-like viral

gastro-enteritis in holidaymakers travelling to Andorra, January-February 2002.

Eur Comm Dis Bull, 8: 1-8.
Peipins, L.A., Highflll, K.A., Barrett, E. et al. (2002). A Norwalk-like virus outbreak on the

Appalachian Trail./ Environ Hltb, 64: 18-23.
Powell, K.L., Taylor, R.G., Cronin, A.A. et al. (2003). Microbial contamination of two

urban sandstone aquifers in the UK. Water Res, 37: 330-352.
Public Health Laboratory Service. (2002). Communicable Disease Report 2002. Annual

figures.
Stanwell-Smith, R. (1994). Recent trends in the epidemiology of waterborne disease. In

Water & Public Health, Golding, A.M.B. (ed.). London: Smith-Gordon, pp. 39-56.
Taku, A., Gulati, B.P., Allwood, P.B. et al. (2002). Concentration of caliciviruses from food

contact surfaces./ Food Protect, 65: 999-1004.
Van Olphen, M., de Bruin, H.A.M., Havelaar, A.H. and Schijven, J.F. (1991). The viro-

logical quality of recreational waters in the Netherlands. Water Sci Tecbnol, 24: 209-212.
Wheeler, J.G., Sethi, D., Cowden, J.M. et al. (1999). Study of infectious intestinal disease

in England: rates in the community, presenting to general practice, and reported to

national surveillance. Br Med J, 318: 1046-1050.
Wyn-Jones, A.P., Pallin, R., Dedoussis, C. et al. (2000). The detection of small round-
structured viruses in water and environmental materials./ Virol Meth, 87: 99-107.

444

Rotavirus

Basic microbiology

Rotaviruses are classified on the basis of serogroups, serotypes and most
recently, genogroups. Six serogroups, termed groups A-F, are recognized and
between each of these groups no cross-reaction occurs using assays based on
the inner capsid antigen, VP6. Group A is the most common human rotavirus
infection; members of groups B and C are the only other groups known to infect
humans. Serotypes are based on neutralization assays for VP7 and to a lesser
extent VP4. Genogroups are based on the structure of specific genes; G types
(G1-G14) refer to the VP7 protein (gene 9) and P types (P1A[8]-P1B[4]) refer
to VP4 protein (gene 4) and are differentiated by RT-PCR. Rotaviruses can
therefore be typed and strains circulating in the community can be monitored
(Iturriza et aL 9 2000). The strains change in frequency of circulation over time
and occasionally new reassortment strains are introduced to a community.

Rotaviruses, at 75 nm in diameter, are relatively large icosahedral virus par-
ticles with a triple layer of protein (Estes, 2001). The virion consists of a core
and a shell of intermediate and outer layers from which 60 spikes protrude. The

Viruses

Figure 33.1 Transmission electron micrograph of rotavirus. (Courtesy of Dr Erskine
Palmer, CDC, USA.)

genome consists of 11 segments of double-stranded RNA (dsRNA) that are cap-
able of genetic reassortment. The complete, infectious particle has a smooth
appearance by EM which becomes a rough outline when the outer shell is lost
(Figure 33.1). It was the wheel -like appearance of the complete particle that was
the basis for the name 'rotavirus' (Flewett et al., 1974). The core comprises
nucleic acid and three structural proteins (VP1, VP2 and VP3). The inter-
mediate layer of the rough, non-infectious particle is made up of protein VP6 and
the outer shell of VP4 (the spikes) and VP7. One hundred and thirty two chan-
nels link the outer capsid with the inner core (Desselberger, 1998).

Virus replication in vitro may be enhanced by pre-treatment with a proteo-
lytic enzyme such as trypsin to cleave the outer capsid spike protein VP4 into
two products, designated VP5* and VP8* (Beards, 1992). Infectivity is lost when
the outer shell is disrupted by the action of a chelating agent such as EDTA.
Infectivity and particle integrity are usually resistant to ether and chloroform.
Infectivity is stable from pH 3-9 and at 4°C especially when stabilized in
CaCl2- Phenol, formalin, chlorine and ethanol inactivate the virus.

Replication takes place in the cytoplasm of infected cells, involves the endo-
plasmic reticulum and release is by cell lysis after a 12-hour replication cycle.

446

Rotavirus

Binding of the virion to the cell receptor site is by the VP4 spike and may depend
on sialic acid presence in the cell membrane. The method by which the virus
penetrates the membrane is not yet fully understood (Estes, 2001). Two modes
of entry seem to be available in vitro depending on whether the virus has been
treated with a proteolytic enzyme such as trypsin or not (Suzuki et aL 9 1985).
Trypsin-treated particles may enter directly through the cell membrane within
moments of attachment, whereas non-trypsin-treated particles are internalized
more slowly by phagocytosis and then released. Accumulations of proteins
that assemble into the incomplete particles occur in cytoplasmic inclusions called
viroplasms. Virus replication in vivo results in the production of large numbers
of incomplete, non-infectious particles as well as infectious particles. Rotavirus is
fastidious in vitro and will grow only in a limited range of cell culture, often only
in an abortive replication cycle producing only incomplete particles.

The rotavirus was initially identified by Bishop in biopsy material (Bishop
et aL, 1973) and then was seen in faeces (Flewett et aL 9 1973) and was quickly
associated with diarrhoea in children.

Origin of the organism

In common with all viruses rotavirus is an obligate intracellular parasite and,
as for most other groups, under normal conditions they are species specific.
The origin of human viruses in sewage will be human faecal material. No
replication will occur outside the living host cell.

Clinical features and virulence

Diarrhoea is the predominant symptom of rotavirus infection. This can become
a life-threatening illness if severe dehydration develops. It has been estimated
that 1 in 8 rotavirus infections in developing countries is severe and that 1 in 160
infected children will die (Kapikian et aL, 2001) resulting in an annual death toll
of nearly 900 000. Although not as severe an illness in developed countries, in
the under 5-year age group admitted to hospital with gastroenteritis, rotavirus is
the most common pathogen found. Immunity studies have shown that by age
3 years 90% of children have had at least one rotavirus infection.

Pathogenicity

Rotavirus replication takes place in the small intestine, in the mature entero-
cytes lining the villi leading to shortening and atrophy. Diarrhoea is the main

447

Viruses

symptom of primary rotavirus infection, although re-infections may be less
severe or asymptomatic. The virus is eliminated and cellular recovery begins
within 48 hours. The incubation period is also about 48 hours and transmission
is by the faecal-oral route. First infections with rotavirus are usually at less than
1 year old resulting in virtually all children over 5 years being immune (Kapikian
et al., 2001). Infection confers long-lasting, cross-strain protection, to some
degree based on antibody to VP7, although adults can be re-infected though usu-
ally without symptoms (Hrdy, 1987). The elderly in residential care may become
more susceptible as their immunity wanes and the opportunity for transmission
occurs. Symptoms in babies may include vomiting as well as diarrhoea and can
be severe with resulting dehydration.

Transmission and epidemiology

The rotaviruses are transmitted directly person to person by the faecal-oral route.
The transfer of virus from surfaces in nurseries to susceptible children may
occur as rotavirus have been detected from this environment (Sattar et aL 9 1986;
Wilde and Pickering, 1992). It has been suggested that the respiratory route
may be involved but there is little evidence to support the theory. It would,
however, help to explain the rapid transmission between young children in all
settings. The usual age of the first episode of infection and hence symptoms is
between 6 months and 2 years old.

Rotaviruses have been detected world-wide and throughout the year in non-
temperate climates. In temperate countries a marked peak of incidence occurs in
the cooler, often winter months. No satisfactory explanation of this seasonality
has been found.

Very large numbers of particles, many of which are the rough non-infectious
virions, are shed into the faeces. For diagnostic purposes this means that EM
is a suitable tool for identification. An ELISA has been available for many years
and is used widely for diagnosis. It is based on detection of the VP6 antigen
of the capsid of the rough particle which is present at more than 10 6 /g in faeces.
Rotavirus only grows in certain cell culture types and this is not suitable for
diagnostic purposes. Other methods available are latex agglutination, reverse
passive haemagglutination assay or immunoblot assay. None of these methods
is useful or practical for investigation of environmental samples. Investigation
by PCR of the genes for the outer capsid proteins (VP4 and VP7) is used to
genogroup strains and hence describe the molecular epidemiology.

Treatment

No specific antiviral treatment is available so the main aim is rehydration by
the replacement of lost fluid and electrolytes. This may be by intravenous

448

Rotavirus

administration or by oral fluids. The latter, containing a mixture of sodium,
chloride and potassium salts plus glucose, has been used very successfully in
many developing countries resulting in a decrease in infant mortality.

Distribution in the environment

There are only limited numbers of reports on the presence of rotavirus in the
environment. The first report was by Steinmann (1981) of virus in sewage in
Germany followed by Rao et al. (1986), Smith and Gerba (1982) in the USA,
Oragui et al. (1995) and Mehnert and Stewien (1993) in Brazil. Hejkal
et al. (1984) reported on the occurrence of rotavirus in sewage in Texas. The
presence of rotavirus in seawater and river water was reported in South Africa
by Genthe et al. (1991), in the UK by Merrett et al. (1991), in Canada by
Raphael et al. (1985) and in Spain by Bosch et al. (1998). Abad et al.

(1998) reported a semi-automated system of rotavirus detection from river
water using MA 104 cells and flow cytometry. The latter stage counted the
cells labelled with rotavirus specific antibody conjugated to a fluorescein dye.
Failure to detect rotavirus was reported by Guttman-Bass et al. (1987) and Tsai
etal. (1994).

Dubois et al. (1997) and Le Guyader et al. (1994) reported the detection of
rotavirus using PCR on treated effluent samples and shellfish and sediments
from western France. Gajardo et al. (1995) detected and genotyped rotavirus
using RT-PCR. A single well water sample was positive for the presence of
rotavirus in a study based in Wisconsin (Borchardt etal., 2003). Green and Lewis

(1999) detected rotavirus from sewage, effluent and oysters in New Zealand
using RT-PCR.

Difficulties with concentration and virus detection are the reason for the
limited number of reported studies. Unlike most other viruses investigated,
rotavirus cannot be concentrated effectively using standard membrane filtra-
tion methods. Methods of detection of infectious virus based on IF or
immunoperoxidase are extremely tedious and very difficult to undertake for
large numbers of samples. The stability of complete particles in the environ-
ment is uncertain as the outer shell, which is essential for infection, is readily
lost but the resulting incomplete particle is robust and stable, consequently the
detection of infectious particles is difficult. Of the rotavirus that reaches water
it is not known how much is infectious and therefore a potential hazard.
Methods based on RT-PCR will detect genome from infectious and non-
infectious particles and have allowed greater numbers of samples to be assayed
using a sensitive technique.

As the majority of symptomatic rotavirus infections in developed countries
occur in babies in disposable nappies this may affect the quantity of virus reach-
ing sewage. Information on the seasonal variation in sewage or its persistence
in the environment is limited.

449

Viruses

Waterborne outbreaks

Rotavirus waterborne outbreaks are uncommon as childhood infection results
in life-long immunity. Infections may recur throughout life but they are mainly
asymptomatic. In a review for the American Water Works Association Research
Foundation, Gerba et al. (1996) reported eight outbreaks, some of which were
on Pacific islands with known water supply problems.

Risk assessment

Health effects: occurrence of illness, degree of morbidity and mortality, prob-
ability of illness based on infection:

• Rotavirus is the most common cause of severe diarrhoea among children.
This can become a life-threatening illness if severe dehydration develops.
Other symptoms can include vomiting and fever, especially in young children.

• Immunization from infection is incomplete, but re-infections are less severe
or asymptomatic.

• Immunity studies have shown that by the age of 3 years, 90% of children
have had at least one rotavirus infection and almost all children over 5 years
are immune.

• The elderly in residential care may become more susceptible as their immunity
wanes and the opportunity for transmission occurs.

• It is estimated that 900 000 children die each year around the world from
rotavirus infection - mostly in developing countries; however, in developed
countries, it is still the primary cause of hospitalization in children under 5
years.

Exposure assessment: routes of exposure and transmission, occurrence in
source water, environmental fate:

• Transmission is by the faecal-oral route either directly from person to person
or via fomites. There is some evidence of a respiratory route, though that
has not been proven.

• Rotavirus has been detected in sewage as well as in seawater and river
water as well as shellfish and sediments, but because of difficulties in test-
ing, there are few data on rotavirus occurrence in source waters.

• Rotavirus waterborne outbreaks are uncommon since childhood infection
results in lifelong immunity.

Risk mitigation: drinking-water treatment, medical treatment:

• Rotaviruses are susceptible to chlorine, ozone and UV light disinfection,
though they are more resistant to UV than enteroviruses.

• For persons with healthy immune systems, rotavirus gastroenteritis is a
self-limited illness, lasting for only a few days. Rehydration to replace

450

Rotavirus

fluid and electrolytes may be required; no specific antiviral treatment is
available.

References

Abad, EX., Pinto, R.M. and Bosch, A. (1998). Flow cytometry detection of infectious
rotaviruses in environment and clinical samples. Appl Environ Microbiol, 64:
2392-2396.

Beards, G.M. (1982). Polymorphism of genomic RNAs within rotavirus serotypes and sub-
groups. Arch Virol, 74: 65-70.

Bishop, R.F., Davidson, C.P., Holmes, J.H. et al. (1973). Virus particles in epithelial cells of
duodenal mucosa from children with acute non-bacterial gastro-enteritis. Lancet, 2:
1281-1283.

Borchardt, M.A., Bertz, P.D., Spencer, S.K. et al. (2003). Incidence of enteric viruses in ground-
water from household wells in Wisconsin. Appl Environ Microbiol, 69: 1172-1180.

Bosch, A., Pinto, R.M., Blanch, A.R. et al. (1998). Detection of human rotavirus in sewage
through two concentration procedures. Water Res, 22: 343-348.

Desselberger, U. (1998). Viruses associated with acute diarrhoeal disease. In Principles and
Practice of Clinical Virology, 4th edn, Zuckerman, A.J., Banatvala, J.E. and Pattison, J.R.
(eds). London: J Wiley and Sons.

Dubois, E., Le Guyader, F., Haugarreau, L. et al. (1997). Molecular epidemiological survey
of rotaviruses in sewage by reverse transcriptase seminested PCR and restriction frag-
ment length polymorphism assay. Appl Environ Microbiol, 63: 1794-1800.

Estes, M. (2001). Rotavirus and their replication. In Fields Virology, 4th edn, Knipe, D.M. and
Howley, P.M. (eds). Philadelphia, PA: Lippincott Williams and Wilkins, pp. 1747-1785.

Flewett, T.H., Bryden, A.S. and Davies, H. (1974). Virus particles in gastro-enteritis.
Lancet, 2: 1497.

Gajardo, R., Bouchriti, N., Pinto, R.M. et al. (1995). Genotyping of rotaviruses isolated
from sewage. Appl Environ Microbiol, 61: 3460-3462.

Genthe, B., Idema, G.K., Krif, R. et al. (1991). Detection of rotavirus in South African
waters: a comparison of a cytoimmunolabelling technique with commercially available
immunoassays. Water Sci Tecbnol, 24: 241-244.

Gerba, C.P., Rose, J.B., Haas, C.N. et al. (1996). Waterborne rotavirus: a risk assessment.
Water Res, 30: 2929-2940.

Green, D.H. and Lewis, G.D. (1999). Comparative detection of enteric viruses in waste-
waters, sediments and oysters by RT-PCR and cell culture. Water Res, 33: 1195-1200.

Guttman-Bass, N., Tchorsh, Y. and Marva, E. (1987). Comparison of methods for
rotavirus detection in water and results of a survey of Jerusalem wastewater. Appl
Environ Microbiol, 53: 761-767.

Hejkal, T.W., Smith, E.M. and Gerba, C.P. (1984). Seasonal occurrence of rotavirus in
sewage. Appl Environ Microbiol, 47: 588-590.

Hrdy, D.B. (1987). Epidemiology of rotaviral infection in adults. Rev Infect Dis, 9: 461-469.

Iturriza Gomara, M., Green, J., Brown, D.W.G. et al. (2000). Seroepidemiological and
molecular surveillance of human rotavirus infections in the UK. London: Public Health
Laboratory Service.

Kapikian, A.Z., Hoshino, Y. and Chanock, R.M. (2001). Rotavirus. In Fields Virology,
4th edn, Knipe, D.M. and Howley, P.M. (eds). Philadelphia, PA: Lippincott Williams &C
Wilkins, pp. 1787-1833.

Le Guyader, F., Dubois, E., Menard, D. et al. (1994). Detection of hepatitis A virus,
rotavirus and enterovirus in naturally contaminated shellfish and sediment by RT-PCR.
Appl Environ Microbiol, 60: 3665-3671.

Mehnert, D.U. and Stewien, K.E. (1993). Detection and distribution of rotavirus in raw
sewage and creeks in San Paulo, Brazil. Appl Environ Microbiol, 59: 140-143.

451

Viruses

Merrett, H., Stackhouse, C. and Cameron, S. (1991). The incidence of rotavirus in the
marine environment: a two-year study. In Proceedings of the First UK Symposium on
Health-Related Water Microbiology, Glasgow, 3-5 Sept., Morris, R., Alexander, L.,
Wyn-Jones, A.P. et al. (eds), pp. 148-157.

Oragui, J.I., Arridge, H., Mara, D., Pearson, H.W. et al. (1995). Rotavirus removal in
experimental waste stabilisation pond systems with different geometries and configur-
ations. Water Sci Technol, 31: 285-290.

Rao, V.C., Metcalf, T.G. and Melnick, J.L. (1986). Development of a method for concen-
tration of rotavirus and its application to recovery of rotavirus from estuarine waters.
Appl Environ Microbiol, 52: 484-488.

Raphael, R.A., Sattar, S.A. and Springthorpe, V. (1985). Rotavirus concentration from raw
water using positively charged filters./ Virol Meth, 11: 131-140.

Sattar, S.A., Lloyd-Evans, N. and Springthorpe, V.S. (1986). Institutional outbreaks of
rotavirus diarrhoea: potential role of fomites and environment surfaces as vehicles for
virus transmission./ Hyg (Camb), 96: 277-289.

Smith, E.M. and Gerba, C.P. (1982). Development of a method for detection of human
rotavirus in water and sewage. Appl Environ Microbiol, 43: 1440-1450.

Steinmann, J. (1981). Detection of rotavirus in sewage. Appl Environ Microbiol, 41:
1043-1045.

Suzuki, H., Kitoaka, S., Sato, T. et al. (1985). Further investigation on the mode of entry of
human rotavirus into cells. Arch Virol, 91: 135-144.

Tsai, Y.L., Tran, B., Sangermano, L.R. et al. (1994). Detection of poliovirus, hepatitis A
virus, and rotavirus from sewage and ocean water by triplex reverse transcriptase PCR.
Appl Environ Microbiol, 60: 2400-2407.

Wilde, J. and Pickering, L. (1992). Detection of rotaviruses in the day care environment by
RT-PCR. / Infect Dis, 166: 507-511.

452

Part 5

Helminths

Dracunculiasis

Dracunculiasis is the only helminth that has unequivocally been shown to be
spread through drinking infected water. The disease is one of those that are
targeted by the World Health Organization for global eradication and it had
been hoped that by the year 2000, this campaign would have achieved its object-
ive. However, the disease remains a problem in certain parts of the world. In
1986 there were an estimated 3.5 million cases of dracunculiasis world-wide,
though by 1995 this had fallen to 129 825 cases. In the years since then the
decline in reported cases has been much less marked with 63 717 cases reported
in 2001 (Figure 34.1) (Anon, 2002). These cases were reported from 17 coun-
tries, though 49471 (77.6%) were reported from Sudan. Other countries
reporting more than 1000 cases were Burkino Faso (1032), Ghana (4739),
Nigeria (5355) and Togo (1354).

Basic parasitology

The agent of dracunculiasis, Dracunculus medinensis, is the only nematode
worm to have been shown unequivocally to be transmitted through drinking
water. The adult female worm measures up to 1 m long and 2 mm in diameter.

Helminths

1000 000

800 000-

600 000-

g- 400 000

200 000-

□ Sudan
I Other countries

■ ■ ■- ■-

N # N # N <£> N c? N c£> ^ N c? ^ ^ N c£> ^p ^p

Year

Figure 34.1 Cases of dracunculiasis reported to the World Health Organization for
the years 1990-2001. (Taken from Weekly Epidemiological Record, Anon, 2002.)

The head end of the worm is rounded with a triangular mouth. Most of the body
of the worm is taken up with a double uterus. The adult males remain small,
about 4 cm in length.

The worm lives in connective tissue where it does no harm until it migrates
down (usually to the legs and feet). Here lytic secretions from glands in the head
probably combined with the irritant effect of the larvae cause a blister that
eventually ruptures to expose the head of the worm. When the head is doused
in water the uterus is extruded through the mouth of the worm and larvae are
expelled into the water.

Origin of the organism

Dracunculiasis is a disease that has been known since antiquity and accurate
descriptions can be found in texts dating to 1350 BC (Muller, 2001; Cox,
2002). The name dates back to Lineus in 1758. The worm is a nematode or
roundworm belonging to the phylum Nemathelminthes, the class Nematoda
and the superfamily Dracunculoidea.

Life cycle

Once expelled from the uterus, the larvae can survive in water for about 6 days.
In muddy water or moist earth, survival can be as long as 3 weeks. Larvae are

456

Dracunculiasis

then ingested by a species of microcrustaceans previously known as Cyclops,
now divided into three genera, Tropocyclops, Mesocyclops and Metacyclops.
Once ingested the larvae penetrate the gut wall into the body cavity where they
develop further, requiring 14 days at above 21 °C. The next stage of the life
cycle happens when the water is drunk by the next human host. The infected
Cyclops are dissolved in the stomach acid and the larvae released. The larvae
then penetrate the gut wall and enter the abdominal cavity. About 3-4 months
later, the males and females mate in the subcutaneous tissues of the thoracic
region. After mating the males die, though the females continue to grow and
eventually migrate downward, usually to the feet where the blister forms until
the foot is placed in water. The blister then ruptures, the larvae are released
and the life cycle starts again. The whole cycle of infection takes about 1 year.

Clinical features

In the majority of cases the first clinical feature is the appearance of the blis-
ter, which grows over a few days to about 3 cm diameter (Cairncross et al. 9
2002). The site of the blister is usually preceded by burning, intense itching and
urticaria. The blister then ruptures and the worm becomes extruded, about
1 cm per day. Left to itself and assuming the site does not become secondarily
infected, the worm track will resolve within about 6 weeks. Unfortunately,
secondary infection is frequent, affecting more than 50% of cases and this can
cause severe pain and disability and rarely death (Adeyeba, 1985). Tetanus, as
a result of secondary infection, has also been described. If the worm breaks
before complete removal, the remnant can cause severe inflammation and scar-
ring. Dracunculiasis is rarely fatal, though it can be severely disabling. Given that
agricultural labourers are among the people most frequently affected, this can
have severe consequences for the family due to the reduced harvest.

There are no laboratory diagnostic tests for dracunculiasis and diagnosis is
made clinically by a history of residence in an endemic area and examination
of the ulcer.

Treatment

No specific antiparasitic agents are effective against the disease and the trad-
itional method of slow removal of the worm by winding around a small stick
is still the method of choice. There is some evidence that niridazole reduces
inflammation around the worm and aids removal. Surgical removal of the worm
before emergence will reduce associated disability.

Care must be taken to avoid the risk of secondary infection; clean dressings
and the prophylactic use of antibiotics are important. Any secondary infection

457

Helminths

should be treated aggressively with antimicrobial agents. Tetanus immuniza-
tion is also recommended.

Epidemiology

The epidemiology follows what can be predicted from a knowledge of the life
cycle of this helminth. Cairncross et al. (2002) review the published studies on
the epidemiology of dracunculiasis. They point out that there is a strong rela-
tionship in a number of studies that villages where cases occur take their drink-
ing water from ponds or shallow step wells. Villages taking their water from
deep wells or from running water tend not to be affected.

There are a number of different interventions that have been shown to reduce
the risk of infection: provision of a safe water supply, filtering drinking water,
active case ascertainment and adequate treatment, preventing cases from hav-
ing contact with water supplies and killing cyclops in ponds (Cairncross et al.,
2002). Filtration is particularly easy as the cyclops are quite large and many
simple fabrics are effective.

Risk assessment

Health effects: occurrence of illness, degree of morbidity and mortality, prob-
ability of illness based on infection:

• There were 63 717 cases of Dracunculus reported in 2001, mostly in Africa.

• The worm lives harmlessly in connective tissue until it migrates down to the
legs and feet. Secretions from glands in the worm's head probably combined
with the irritant effect of the larvae cause a blister that grows to about 3 cm
diameter and that eventually ruptures to expose the head of the worm. The
blister ruptures and the worm extrudes about 1 cm per day. Left alone, the
worm track will resolve within about 6 weeks. Secondary infection is frequent
and affects more than 50% of cases. This can cause severe pain and disability
and, rarely, death.

• If the worm breaks before complete removal, the remnant can cause severe
inflammation and scarring.

Exposure assessment: routes of exposure and transmission, occurrence in
source water, environmental fate:

• Dracunculus is the only helminth that has been shown to be spread through
drinking infected water.

• Larvae can survive in water for about 6 days. In muddy water or moist
earth, survival can be as long as 3 weeks.

458

Dracunculiasis

Risk mitigation: drinking-water treatment, medical treatment:

• Interventions to reduce the risk of infection include the provision of a safe
water supply, filtering drinking water and preventing infected people from
having contact with water supplies. Filtration is particularly easy as the organ-
ism is large and simple fabrics are effective.

• No specific antiparasitic agents are effective against the disease. The trad-
itional method of slow removal of the worm by winding around a small
stick is still used. Any secondary infection should be treated aggressively
with antimicrobial agents.

References

Adeyeba, O.A. (1985). Secondary infection in dracunculiasis: bacteria and morbidity.

IntJ Zoonoses, 12: 147-149.
Anon. (2002). Dracunculiasis eradication. Global surveillance summary, 2001. Wkly

Epidemiol Rec, 77: 143-152.
Cairncross, S., Muller, R. and Zagaria, N. (2002). Dracunculiasis (Guinea Worm Disease)

and the eradication initiative. Clin Microbiol Rev, 15: 223-246.
Cox, F.E.G. (2002). History of human parasitology. Clin Microbiol Rev, 15: 595-612.
Muller, R. (2001). Dracunculiasis. In Principles and Practice of Clinical Parasitology,

Gillespie, S.H. and Pearson, R.D. (eds). Wiley: Chichester, pp. 553-559.

459

Part 6

Future

Emerging waterborne
infectious diseases

It could be argued that one of the more important philosophical developments
in our understanding of the epidemiology of infectious diseases was the intro-
duction of the concept of the emerging infectious disease. Some of the first
authors to use the term were Morse and Schluederberg (1990). It was only
some 5 years later in 1995 that the new journal Emerging Infectious Diseases
was first published. This journal is now one of the most highly cited infectious
disease journals.

Emerging infections can be defined as those infections that have newly
appeared in the population, or have existed but are rapidly increasing in inci-
dence or geographic range (Morse, 1995). Sometimes emerging infections are
distinguished from re-emerging infections. The latter being those diseases that
were once common in a community but then declined in incidence only to
increase again a number of years later.

Many of the diseases that are covered in this book can be defined as emer-
ging infections. Cryptosporidiosis, enterohaemorrhagic £. co//, noroviruses, as
well as many other pathogens not known even 30 years ago. Others such as
cholera and typhoid are better described as re-emerging.

In this chapter we shall consider the factors that are responsible for the emer-
gence and re-emergence of infections, especially as they apply to waterborne

Future

Table 35.1 Factors responsible for the emergence or re-emergence of water-
borne pathogens

Factor Examples

Microbial evolution E. coli 0157:H7, Vibrio cholerae 0139

Improved diagnostic technology Cryptosporidium, hepatitis E virus

New technology Legionnaires' disease and air conditioning systems

Ecological change Schistosomiasis following building of dams

Demographic change Increased pressure on water supplies

International travel and trade Movement of Vibrio cholerae

Breakdown in public health Re-emergence of cholera and typhoid after the
systems collapse of the Soviet Union

pathogens. Morse (1995) described a range of factors (Table 35.1). These fac-
tors will be discussed in turn.

Microbial evolution

Some infections are emerging because they are due to new pathogens that simply
did not exist a few decades ago. The two water borne pathogens that have recently
evolved are Escherichia coli 0157:H7 and Vibrio cholerae 0139 (Rubin et aL,
1998). Escherichia coli 0157:H7 was first recognized as a human pathogen in
1982. E. coli 0157:H7 appears to have developed by a series of steps whereby the
organism evolved from a related strain by the acquisition of genes for the expres-
sion of additional virulence factors (Whittam et aL, 1998). Vibrio cholerae 0139
is also thought to have evolved in a similar way (Rubin et al., 1998).

Clearly for any newly evolved microbial pathogen to be thought of as emer-
ging, the evolution must give the organism some advantage, either to transmit
itself in the environment or to increase its capacity to cause disease. In the two
examples quoted above, the likely explanation is the acquisition of additional
virulence factors that gave organisms that were already present in the envir-
onment the enhanced ability to cause disease. In the case of E. coli 0157:H7
the increased virulence probably also increased the ability to spread in the
environment by allowing increased excretion into the environment from
infected cattle and other mammals.

Improved diagnostic technology

It is probably obvious when thinking about it that a disease will become emer-
gent only after a diagnostic test has been developed or introduced. Perhaps the
two most obvious emerging pathogens in this regard are Cryptosporidium
and hepatitis E virus.

464

Emerging waterborne infectious diseases

Cryptosporidiosis is a useful example in that the technology was not par-
ticularly complex. Indeed the veterinary pathologists had known about
Cryptosporidium since 1907 when it was described in mice. It was not until
1976 that the pathogen was first described in humans (Meisel et al. 9 1976;
Nime et al. 9 1976). Since then of course, microbiologists increasingly realized
that Cryptosporidium is a cause of diarrhoea and more and more laboratories
started to look for the organism in stool samples with the result that reports
increased substantially over the following years. The technology was around
to diagnose Cryptosporidium for decades but nobody thought to look. Hepa-
titis E, on the other hand, is a disease that had to await the development of
molecular biological tools before it could become emergent. Although non-A,
non-B hepatitis was known about for many years, hepatitis E could not be
readily distinguished from other non-A, non-B hepatitis infections as no
serological test was available. It was not until it was possible to synthesize syn-
thetic peptides and recombinant antigens from the viral genome for use in ELISA
or Western blot technologies that it was possible to diagnose the condition. Once
commercial diagnostic tests became available, reports of infections increased and
the disease became emergent.

New technology

The classic example of an emergent disease developing because of new technol-
ogy is Legionnaires' disease. The first outbreak to be diagnosed was in
Philadelphia in 1976 (Fraser et al. 9 1977). Although it was a little time before the
causative agent could be identified and a diagnostic test became available, it soon
thereafter became clear that a major risk factor for this infection was cooling sys-
tems that used water to cool the air. We now know, of course, that Legionella
bacteria are widely disseminated in the water environment. However, the special
conditions in cooling towers allow the multiplication of these organisms to the
great numbers necessary to cause disease. Although not all cases of Legionnaires'
disease are acquired from cooling towers, it is interesting to speculate whether
the effort would have been put into identifying this important pathogen without
the large outbreaks associated with these towers.

Ecological change

Ecological change can affect infectious disease in several different ways. Several
authors have discussed the potential impact of climate change on waterborne
disease (Rose et al. 9 2001; Hunter, 2003). Perhaps one of the most important
effects may be due to changes in rainfall. For example, extreme rainfall events
have been shown to be a significant risk factor for outbreaks of waterborne

465

Future

disease (Curriero et al. 9 2001). Some ecological changes are directly related to
human intervention. An example of this is the local emergence of schistosomia-
sis following the construction of dams (N'Goran et al. 9 1997; Sow et aL 9 2002).
The appearance of a large body of still water provides opportunities for vectors
to breed and the parasite to increase in the local human population.

Demographic change

Demographic change will also have a number of effects on the emergence and
re-emergence of infectious disease. The world's population is increasing and
this is putting increased stress on available water resources, even in several
industrialized nations such as the USA (Hunter, 1997). As water becomes
scarcer lower quality water sources will be used and this can provide oppor-
tunities for increased pathogen transmission. The rapid increase in the
acquired immune deficiency syndrome (AIDS) has been one of the main drivers
affecting the emergence of new diseases. Much of our early interest in crypto-
sporidiosis has been because of its severe impact in people living with AIDS
(Hunter and Nichols, 2002).

International travel and trade

Clearly people who travel abroad, especially to tropical countries, are at risk
of disease that they would not expect to experience in the home country.
Many of these travel-associated illnesses are potentially waterborne (Payment
and Hunter, 2001). Another route for international dissemination of water-
borne pathogens has been carriage in ships' ballast waters (McCarthy and
Khambaty, 1994).

Breakdown in public health systems

Finally, one of the lessons that was associated with the collapse of the Soviet
Union has been the risk to public health that comes with severe economic
collapse. Such economic collapse can in turn threaten public health systems.
This impacts on waterborne disease when such economic collapse impairs
the effectiveness of water treatment and distribution systems. When water
systems decay diseases once controlled, such as cholera and typhoid, can
re-emerge (Semenza et al. 9 1998). It is very sobering to consider what may be the
impact of such economic collapse in some of the world's largest urban centres.

466

Emerging waterborne infectious diseases

Conclusions

As has been discussed, there are many factors that can lead to the emergence
or re-emergence of new pathogens. Although some of these factors are outside
the control of humanity, many are directly the result of human activity or
human pressure on the environment. Emerging infectious diseases pose a
particular problem to the water utility. It is often impossible to be certain of
the eventual impact of the disease and the contribution to disease burden of
drinking-water transmission. How many people realized the full importance
of cryptosporidiosis in the years after the pathogen was first described in
humans? On the other hand, we know many of the human factors that impact
on risk of emergence and we can at least try to manage these factors effectively.

References

Curriero, F.C., Patz, J. A., Rose, J.B. et al. (2001). The association between extreme preci-
pitation and waterborne disease outbreaks in the United States, 1948-1994. Am J
Public Hlth, 91: 1194-1199.

Fraser, D.W., Tsai, T.R., Orenstein, W. et al. (1977). Legionnaires disease: description of an
epidemic of pneumonia. New Engl J Med, 297: 1189-1197.

Hunter, P.R. (1997). Waterborne Disease: Epidemiology and Ecology. Chichester: Wiley.

Hunter, P.R. (2003). Climate change and waterborne and vector-borne disease. / Appl
Microbiol (in press).

Hunter, P.R. and Nichols, G. (2002). The epidemiology and clinical features of
Cryptosporidium infection in immune-compromised patients. Clin Microbiol Rev, 15:
145-154.

McCarthy, S.A. and Khambaty, F.M. (1994). International dissemination of epidemic
Vibrio cbolerae by cargo ship ballast and other non-potable waters. Appl Environ
Microbiol, 60: 2597-2601.

Meisel, J.L., Perera, D.R., Meligro, C. et al. (1976). Overwhelming watery diarrhoea asso-
ciated with a Cryptosporidium in an immunosuppressed patient. Gastroenterology, 70:
1156-1160.

Morse, S.S. (1995). Factors in the emergence of infectious diseases. Emerg Infect Dis, 1: 7-15.

Morse, S.S. and Schluederberg, A. (1990). Emerging viruses: the evolution of viruses and
viral diseases. / Infect Dis, 162: 1-7.

N'Goran, E.K., Diabate, S., Utzinger, J. et al. (1997). Changes in human schistosomiasis
levels after the construction of two large hydroelectric dams in central Cote dTvoire.
Bull World Health Org, 75: 541-545.

Nime, F.A., Burek, J.D., Page, D.L. et al. (1976). Acute enterocolitis in a human being
infected with the protozoan Cryptosporidium. Gastroenterology, 70: 592-598.

Payment, P.R. and Hunter, P.R. (2001). Endemic and epidemic infectious intestinal disease
and its relation to drinking water. In Water Quality: Guidelines, Standards and Health.
Risk Assessment and Management for Water-related Infectious Disease, Fewtrell, L. and
Bartram, J. (eds). London: IWA Publishing, pp. 61-88.

Rose, J.B., Epstein, P.R., Lipp, E.K. et al. (2001). Climate variability and change in the
United States: Potential impacts on water- and foodborne diseases caused by microbio-
logic agents. Environ Hlth Perspec, 109(Suppl. 2): 211-221.

Rubin, E.J., Waldor, M.K. and Mekalanos, J.J. (1998). Mobile genetic elements and
the evolution of new epidemic strains of Vibrio cholerae. In Emerging Infections,
Krause, R.M. (ed.). New York: Academic Press, pp. 147-461.

467

Future

Semenza, J.C., Roberts, L., Henderson, A. et al. (1998). Water distribution system and

diarrheal disease transmission: A case study in Uzbekistan. Am J Trop Med Hyg, 59:

941-946.
Sow, S., de Vlas, S.J., Engels, D. et al. (2002). Water-related disease patterns before and

after the construction of the Diama dam in northern Senegal. Ann Trop Med Parasitol,

96: 575-586.
Whittam, T.S., McGraw, E.A. and Reid, S.D. (1998). Pathogenic Escherichia coli 0157:H7:

a model for emerging infectious diseases. In Emerging Infections, Krause, R.M. (ed.).

New York: Academic Press, pp. 163-183.

468

Index

Acanthamoeba spp., 145, 148, 221-29

basic microbiology, 221-2

clinical features, 223

detection of, 225-6

epidemiology, 226

future implications, 227-8

origins of, 222

pathogenicity and virulence, 223

risk assessment, 226-7
exposure assessment, 227
health effects, 226-7
risk mitigation, 227

survival in environment, 224-5

treatment, 224
Acanthamoeba astronyxis, 221
Acanthamoeba castellanii, 163, 221

clinical features, 223

pathogenicity and virulence, 223
Acanthamoeba culbertsoni, 221
Acanthamoeba griffini, 221
Acanthamoeba hatch etti, 221
Acanthamoeba keratitis, 222, 223
Acanthamoeba lugdenensis, 221
Acanthamoeba polyphaga, 221

clinical features, 223

pathogenicity and virulence, 223
Acanthamoeba rhysodes, 221
Achromobacter spp., 22, 30
Achromobacter punctatum, 30
Acinetobacter spp., 21-8, 131

basic microbiology, 21

clinical features, 23

detection of, 25-6

epidemiology, 26

inactivation rate, 25

metabolism and physiology, 22-3

origin and taxonomy, 21-2

pathogenicity and virulence, 23-4

risk assessment, 26-7
exposure assessment, 27
health effects, 26-7
risk mitigation, 27

survival in the environment, 24-5

treatment, 24

Acinetobacter baumannii, 22

clinical features, 23
Acinetobacter calcoaceticus, 22

metabolism and physiology, 23
Acinetobacter haemolyticus, 22
Acinetobacter johnsonii, 22

clinical features, 23
Acinetobacter Iwoffii, 22

clinical features, 23
Acinetobacter radioresistens, 22
Adenovirus, 379-86

basic microbiology, 379-80

clinical features and virulence, 381

distribution in environment, 382-4

origins of, 380

pathogenicity, 381-2

risk assessment, 384-5

structure, 379-80

transmission and epidemiology, 382

treatment, 382

waterborne outbreaks, 383-4
Adhesins, 34
Aerobacter spp., 30
Aerobacter aero genes, 128
Aerobacter liquefaciens, 30
Aeromonas spp., 25, 29-41

basic microbiology, 29

clinical features, 32-3

epidemiology, 37

isolation and detection, 36-7

metabolism and physiology, 31-2

origin and taxonomy, 30-1

pathogenicity and virulence, 33-4

presence in environment, 34-5

risk assessment, 37-8
exposure assessment, 38
health effects, 37-8
risk mitigation, 38

treatment, 34
Aeromonas allosaccharophila, 31
Aeromonas caviae, 29

clinical features, 32

presence in environment, 35
Aeromonas eucrenophila, 31

Index

Aeromonas bydrophila, 29, 31

clinical features, 32

enterotoxin, 33

haemolysin, 34

presence in environment, 35
Aeromonas jandaei, 31
Aeromonas salmonicida, 31

haemolysin, 34
Aeromonas scbubertii, 31
Aeromonas sobria, 29, 32

presence in environment, 35
Aeromonas trota, 31
Aeromonas veronii, enterotoxin, 33
Air conditioning systems, 149-50
Albendazole, 244, 304
Alcaligenes spp., 22, 25
Amarantbus viridans, 293
Amikacin, 161

Aminoglycosides, 34, 81, 99, 134, 212
Amoebic dysentery, 287
Amoxicillin, 100
Amphotericin B, 321
Ampicillin, 81, 135, 177, 188, 233
Anabaena cylindrical 62
Anabaena flosaquae, 61
Anatoxins, 63

Apbanizomenon flosaquae, 61-2
Arcobacter spp., 43-8, 93

basic microbiology, 43

clinical features, 44-5

environment, 45

epidemiology, 46

isolation and detection, 45-6

metabolism and biochemistry, 44

origins, 43-4

risk assessment, 47

exposure assessment, 47
health effects, 47
risk mitigation, 47

treatment, 45
Arcobacter butzleri, 44, 104

clinical features, 44

in environment, 45

epidemiology, 46

exposure assessment, 47

health effects, 47

metabolism and biochemistry, 44
Arcobacter cryaeropbila, 44

clinical features, 44

in environment, 45

exposure assessment, 47

health effects, 47

Arcobacter nitrofigilis, 44

in environment, 45

metabolism and biochemistry, 44
Arcobacter skirrowii, 44

clinical features, 44-5

exposure assessment, 47

health effects, 47

metabolism and biochemistry, 44
Ascaris lumbricoides, 234
Astroviruses, 387-99

basic microbiology, 388-91

distribution in environment, 393-6

origins of, 391

pathogenesis and clinical features,
391-2

risk assessment, 396-7
exposure assessment, 396
health effects, 396
risk mitigation, 397

serotypes of, 389

structure, 388

transmission electron microscopy,
389,390

transmission and epidemiology, 392-3
Atypical mycobacteria, 156
Azithromycin, 148, 161, 244

Bacillus bydropbilus fuscus, 30
Bacillus bydropbilus Sanarelli, 30
Bacillus icbtbyosmius, 30
Bacillus punctatus, 30
Bacillus ranicida, 30
Bacillus spbaericus, 92
Bacitracin, 235

BACTEC 460 system, 164, 165
BACTEC 9000 system, 164
Bacterium anitratum, 22
bacterium coli commune, 72
Bacterium enter ocoliticum, 210
Bacterium punctatum, 30
Balantidiasis, 232
Balantidium coli, 231-6

basic microbiology, 231-2

clinical features, 232

detection of, 233

epidemiology, 233-4

future implications, 235

pathogenicity, virulence and
causation, 232

risk assessment, 234-5
exposure assessment, 235

470

Index

health effects, 234
risk mitigation, 235

survival in environment and water, 233

treatment, 233
Beet necrotic yellow vein virus, 427
Benzylpenicillin, 99
Bismuth, 100

Bloody diarrhoea, 74, 86-7
Bordetella pertussis, 381
Brevundimonas spp., 126
Burkbolderia spp., 126
Burkbolderia cepacia, 126
Burkbolderia pseudomallei, 126

Campylobacter spp., 49-60, 93

basic microbiology, 49-50

clinical features, 51

detection of, 54

epidemiology and waterborne
outbreaks, 55-6

metabolism and physiology, 50-1

origin and taxonomy, 50

pathogenicity and virulence, 51-3

risk assessment, 56-7
exposure assessment, 57
health effects, 56-7
risk mitigation, 57

survival in environment and water, 53-4

treatment, 53
Campylobacter coli, 50

oxidative stress defence, 52

pathogenicity and virulence, 51—2
Campylobacter cryaeropbila, 44
Campylobacter gracilis, 49

metabolism and physiology, 51
Campylobacter jejuni, 50, 104

clinical features, 51

metabolism and physiology, 50-1

oxidative stress defence, 52

pathogenicity and virulence, 51-2
Campylobacter pyloridis, 93
Campylobacter upsaliensis, 50
Carbapenems, 34, 135
Cell culture of viruses, 363-7

monolayer plaque assay, 364-5

plaque assay, 362, 364

suspended cell plaque assay, 365-7
Centers for Disease Control and

Prevention (CDC), 9
Cephalosporins, 34, 81, 99, 134
Cercomonas intestinalis, 301

Cbilmastix mesnili, 285

Chloramphenicol, 34, 177, 188, 212

Cholera see Vibrio cbolerae

Cholera toxin, 199,200

Cbryseobacterium meningosepticum, 131

Ciprofloxacin, 53, 134, 148, 177, 188, 271

Cirpofloxacillin, 177

Clams, contamination of, 311

Clarithromycin, 100, 148, 161

Clindamycin, 135

Clofazimine, 161

Clostridium jejuni, 109-10

Clotrimazole, 226

Co-amoxiclav, 134

Co-trimoxazole, 148, 177, 188

Codex Committee on Food Hygiene, 10

Comamonas spp., 126

Complement-fixing antibodies, 382

Conjunctivitis, 381

Contact lenses see Acantbamoeba spp.

Corynebacterium spp., 156

Coxsackievirus

growth of, 402

serotypes, 405, 408, 409, 412

see also Enterovirus
Crohn's disease, 160
Cryptosporidium spp., 7, 237-64, 356

basic microbiology, 237-9

clinical features, 240-2

detection of, 247-9

diagnosis of, 464-5

epidemiology, 249-53

future implications, 258

origins of, 239-40

pathogenicity, 242

risk assessment, 254-8

exposure assessment, 256-7
health effects, 256
risk mitigation, 257

survival in environment, 245-7

survival in water, 246-7

treatment, 244

virulence, 243-4
Cryptosporidium andersoni, 238
Cryptosporidium baileyi, 238
Cryptosporidium canis, 238
Cryptosporidium felis, 238
Cryptosporidium bominis, 237, 238
Cryptosporidium meleagridis, 238,

240,250
Cryptosporidium muris, 238
Cryptosporidium nasorum, 238

471

Index

Cryptosporidium parvum, 237, 238, 249
Cryptosporidium sauropbilum, 238
Cryptosporidium serpentis, 238
Cryptosporidium wrairi, 238
Cyanobacteria, 61-70

basic microbiology, 61

clinical features, 62-3

epidemiology of waterborne outbreaks,
65-7

methods of detection, 65

origin and taxonomy, 61-2

pathogenicity and virulence, 63-4

risk assessment, 67—8

exposure assessment, 67-8
health effects, 67
risk mitigation, 68

survival in environment and water, 64-5
Cyclidium spp., 148
Cyclospora cayetanensis, 267-84

basic microbiology, 267-8

causation, 270-1

clinical features, 269-70

detection of, 273-5

epidemiology, 275-8

future implications, 280-1

origins of, 268-9

pathogenicity and virulence, 270

risk assessment, 278-80
exposure assessment, 280
health effects, 279-80
risk mitigation, 280

survival in environment, 271-3

treatment, 271

waterborne outbreaks, 272-3
Cyclosporiasis, 269-70

Demographic change, 465-6
Diagnostic technology, 464-5
Diffusely adherent E. coli, 75, 81
Diloxanide, 289
Dose-response assessment, 5, 9
Doxycycline, 148
Dracunculiasis, 455-9

basic parasitology, 455-6

clinical features, 457

epidemiology, 458

life cycle, 456-7

origins of, 456

risk assessment, 458-9
exposure assessment, 458
health effects, 458
risk mitigation, 459

treatment, 457-8
Dracunculus medinensis, 455
Drinking water, risk assessment, 3-17

Echinamoeba spp., 145, 148
Echovirus:

serotypes, 405, 408, 409, 412

see also Enterovirus
Ecological change, 465-6
Eimeria spp., 268, 274-5
Embden-Meyerhof-Parnas pathway, 130
End point dilution assay, 367-8
Endolimax nana, 285
Entamoeba coli, 285
Entamoeba dispar, 285, 287, 291
Entamoeba gingivalis, 285
Entamoeba hartmanni, 285, 287
Entamoeba histolytica, 236, 285-97

basic microbiology, 285-6

causation, 288-9

clinical features, 287-8

detection of, 290-2

epidemiology, 292-3

future implications, 295

origins of, 287

pathogenicity and virulence, 288

risk assessment, 293-4
exposure assessment, 294
health effects, 294
risk mitigation, 294

survival in environment and water,
289-90

treatment, 289
Entamoeba polecki, 285, 287
Enteroaggregative E. coli, 75, 78, 80, 83, 87
Enterobacter spp., 7
Enter obius vermicularis, 234
Enterohaemorrhagic E. coli, 77, 80, 86-7
Enteroinvasive E. coli, 75, 78, 80, 84, 87
Enteropathogenic E. coli, 75, 77, 79,

83,85
Enterovirus, 401-18

basic microbiology, 401-3

clinical features, 403-4

environmental distribution, 405-10

epidemiology, 411-12

origins of, 403

risk assessment, 412-13

exposure assessment, 412-13
health effects, 412
risk mitigation, 413

472

Index

serotypes, 408, 409, 412

structure, 402

treatment, 404-5

virulence and pathogenicity, 404

waterborne outbreaks, 410-11
Environmental Protection Agency

microbial risk assessment, 10

microbiological standards for E. coli, 179

Surface Water Treatment Rule, 6, 254
Erythromycin, 53, 99, 148
Escherichia spp., 30
Escherichia blattae, 72
Escherichia coli, 71-90, 185, 225

basic microbiology, 71

clinical features, 72-8

detection of, 82-5

diffusely adherent, 75, 81

enteroaggregative, 75, 78, 80, 83, 87

enterohaemorrhagic, 77, 80, 86-7

enteroinvasive, 75, 78, 80, 84, 87

enteropathogenic, 75, 77, 79, 83, 85

enterotoxigenic, 74, 75, 76-7, 78-9,
83-4, 85-6

epidemiology, 85-7

evolution of, 464

metabolism and physiology, 72

origins, 72

risk assessment, 87-9
exposure assessment, 88
health effects, 87-8
risk mitigation, 88-9

survival in environment, 81-2

treatment, 81

vero cytotoxigenic, 75, 77, 84-5

virulence and pathogenicity, 78-81

in water, 82
Escherichia fergusoni, 72
Escherichia hermannii, 72
Escherichia vulneris, 72
Ethambutol, 161
EU Bathing Water Directive, 369
European Drinking Water Inspectorate

Directive, 82
Exposure assessment, 5, 9

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "كتب في علم الميكروبيولوجي"

See other formats