g’ my '

259%:9’, ,
at“:

g,

, 1,

a:
'3
5&1
- A

.fii

fin

a ‘-:';.-.7.', 3.5.5.. :1. . A-

i

.g .
~ ‘ as ‘ -
fi!¢§;=r.‘,,;,3
2 ,3ng
3 .x , 3’!

rv'

.
~ ‘ -.~:‘:.»:.‘.1.

«w
:‘3

Ema-v.7...

gr

V» .
$349,?

1.
3

7t .r
41“

“fit:
I “ <

«on-w- \

a
ra’fi
5
:1:

:y
#5

t}'

.'

‘ ,0
v ‘i
v

a).
“a.“
, ‘
p.-
v
401

 

-:-.9
53... . . ,3

T191338

2390 E”

 

LIBRARY
Michigan State
University

 

 

 

This is to certify that the
dissertation entitled

EXAMINING THE PRESENCE AND PREVALENCE OF KEY
HUMAN ENTERIC VIRUSES IN ENVIRONMENTAL SAMPLES
USING CULTIVATION, MOLECULAR AND ARRAY-BASED TOOLS
FOR DETECTION

presented by

Mark Vee-Meng Wong

has been accepted towards fulfillment
of the requirements for the

PhD. degree in Crops and Soil Sciences

 

 

MMWK I96“

agorProfessor’SSig‘hature

MEN/m?

Date

MSU is an affirmative-action. equal-opportunity employer

A4 —.-.-A- -.-._.-A-.-._,-.-.-.-.- —.-.—----— -.-.-,_.- —.-.- —.-

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K:IProj/Acc&Pres/CIRC/DateDue.indd

EXAMINING THE PRESENCE AND PREVALENCE OF KEY
HUMAN ENTERIC VIRUSES IN ENVIRONMENTAL
SAMPLES USING CULTIVATION, MOLECULAR AND
ARRAY-BASED TOOLS FOR DETECTION

By

Mark Vee-Meng Wong

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY

Crops and Soil Sciences

2008

Abstract

EXAMINING THE PRESENCE AND PREVALENCE OF KEY HUMAN ENTERIC
VIRUSES IN ENVIRONMENTAL SAMPLES USING CULTIVATION,
MOLECULAR ANDARRAY-BASED TOOLS FOR DETECTION

By
Mark Vee-Meng Wong

The current recreational water standards are based on the measurement of fecal
indicator bacteria and do not always provide an accurate indication of viral pollution.
Unfortunately, enteric viruses are difficult to detect using conventional cell culture
methods as it requires several days to complete, making it unsuitable as a routine
monitoring tool. Molecular detection methods like integrated cell culture polymerase
chain reaction (PCR) have been successfully used to rapidly detect the presence of virus
nucleic acid from environmental samples but with over a 100 different types of viruses
capable of causing infections in humans, routine testing using PCR is not feasible. In this
dissertation, I describe the development of a virus microarray that can be integrated with
cell culture to rapidly perform high throughput screening of environmental samples.

The microarray holds a total of 780 unique probes broadly targeting 27 different
groups of enteric viruses known or suspected to cause enteric disease and was used to test
the hypothesis that the range of virus types present in the environment is under-estimated
by conventional cell culture and cell culture-PCR. To test this hypothesis, the types of
viruses present at two typical recreational beaches in the Great Lakes were characterized I
using conventional methods to serve as a baseline for the types of viruses present in the
environment. In addition, raw sewage samples which are the major contributing source of

human enteric viruses present in the environment, were analyzed using both PCR

methods and the virus microarray over a period of thirteen months. Using PCR, 12
species of viruses from 7 major viral groups were detected as opposed to 14 different
groups of viruses detected using microarrays. A detailed analysis of the results showed
the following: 16 cases in which both the PCR and the microarray results were positive
and in agreement. 17 cases in which the microarray was positive but could not be
supported by PCR results, 15 cases in which PCR results were positive for an enteric
viruses but were not shown to hybridize significantly on the microarray and 303 instances
in which neither PCR nor the microarray produced a positive signal. This resulted in a p
value of less than 0.0001 for a Fisher’s exact test indicating strong statistical correlation
between the PCR and the microarray. It was observed that while PCR was more sensitive,
able to detect viruses during months where no virus signals were detected on the
microarray, the group specific primers used often biased the results in favor of just one
species of virus when the array was able to indicate the presence of several members
from the same family or related families of viruses.

We conclude that microarrays are capable of screening for a broad number of
pathogenic viruses which may be circulating in the population and excreted in the
community wastewater but that PCR remained a more sensitive detection method. This is
the first demonstration of an environmental microarray for detection of viruses in water

and could be used to improve public health surveillance.

Table of Contents

 

 

 

 

 

 

LIST OF TABLES vii
LIST OF FIGURES viii
ABBREVIATIONS ix
CHAPTER 1. INTRODUCTION TO HUMAN VIRUSES AND THE WATER
ENVIRONMENT l
1.1 Viral Gastroenteritis, Disease Surveillance and Outbreak Information ................ 1
1.2 Productivity Losses, Economic Costs of Viral Infections ..................................... 2
1.3 Knowledge Gaps, Limitations of Current Tests .................................................... 3
1.4 Microarray Detection of Pathogens ...................................................................... 4
CHAPTER 2. LITERATURE REVIEW 6
2.1 Recreational Water Quality Standards .................................................................. 6
2.2. The Enteric Viruses .............................................................................................. 7
2.2.1 The Enteroviruses ....................................................................................... 11
2.2.2 The Adenoviruses ....................................................................................... 12
2.2.3 The Rotaviruses .......................................................................................... 13
2.2.4 The Noroviruses .......................................................................................... 15
2.2.5 The Astroviruses ......................................................................................... 16
2.2.6 The Picobimaviruses ................................................................................... 17
2.2.7 The Parvoviruses ......................................................................................... 18
2.2.8 The Toroviruses .......................................................................................... 19
2.2.9 The Coronaviruses ...................................................................................... 19
2.3. Virus Sampling and concentration methods for water ........................................ 20
2.3.1 Concentration of Viruses from the Water Environment ............................. 20
2.3.2 Virus Adsorption Elution ............................................................................ 21
2.3.3 Cation coated filter method ......................................................................... 23
2.3.4 Microfiltration and Ultrafiltration .............................................................. 23
2.3.5 Organic Flocculation ................................................................................... 24
2.3.6 Polyethylene Glycol Precipitation .............................................................. 24
2.3.7 SpeedVac Concentration ............................................................................. 25
2.4 Methods of Viral Detection ................................................................................. 25
2.4.1 Cell Culture for Environmental samples ..................................................... 25
2.4.2 Polymerase chain reaction (PCR) ............................................................... 26
2.4.3 Hybridization probes and microarrays ........................................................ 27
CHAPTER 3. RESEARCH QUESTIONS, HYPOTHESES. GOALS AND
OBJECTIVES 15
3.1 Research Questions ............................................................................................. 35
3.2 Research Hypotheses and Methodology ............................................................. 35
3.3 Research Objectives ............................................................................................ 36

iv

CHAPTER 4. EVALUATION OF PUBLIC HEALTH RISKS AT
RECREATIONAL BEACHES IN LAKE MICHIGAN VIA DETECTION OF
ENTERIC VIRUSES AND A HUMAN SPECIFIC BACTERIOLOGICAL

 

 

 

MARKER 38
4.1 Abstract ............................................................................................................... 38
4.2 Introduction ......................................................................................................... 39
4.3 Materials and Methods ........................................................................................ 44

4.3.1 Area of study ............................................................................................... 44
4.3.2 Sample Collection ....................................................................................... 45
4.3.3 Cell culture .................................................................................................. 48
4.3.4 PCR and RT PCR for viruses ...................................................................... 48
4.3.5 Extraction of Enterococci ........................................................................... 51
4.3.6 PCR Conditions for esp Marker .................................................................. 51
4.3.7 Statistical Analysis ...................................................................................... 52
4.4 Results ................................................................................................................. 53
4.4.1 Viral results ................................................................................................. 53
4.4.2 Sequencing Confirmation ........................................................................... 54
4.4.3 Enterococcus esp gene PCR results ............................................................ 56
4.4.4 Virus PCR and Cell Culture results correlation .......................................... 56
4.4.5 Physical Indicators ...................................................................................... 57
4.4.6 Modeling Virus Contamination at the Beaches .......................................... 58
4.4.7 Risk Estimation ........................................................................................... 66
4.5 Discussion ........................................................................................................... 66
4.5.1 Overall Discussion of Results ..................................................................... 66
4.5.2 Modeling of Pathogens and Indicators ....................................................... 68
4.5.3 Significance of Detection Methods ............................................................. 68
4.5.4 Adenovirus Presence ................................................................................... 69
4.6 Conclusions ......................................................................................................... 70

CHAPTER 5. DETECTION OF PATHOGENIC VIRUSES CIRCULATING IN

COMMUNITY WASTEWATER USING OLIGONUCLEOTIDE

MICROARRAYS 72
5.1 Abstract ............................................................................................................... 72
5.2 Introduction ......................................................................................................... 73
5.3 Materials and Methods ........................................................................................ 75
5.4 Results and Discussion ........................................................................................ 91
5.5 Conclusion ......................................................................................................... 120

CHAPTER 6. TESTING THE SENSITIVITY OF THE MICROARRAY WITH

POLIOVIRUS SPIKED INTO SEWAGE 122
6.1 Introduction ....................................................................................................... 122
6.2 Literature Survey ................................................................................................... 122

6.2.1 Levels of Human Viruses in the Environment ............................................... 122
6.2.2 Sensitivity of Various Detection Methods ..................................................... 124
6.3 Materials and Methods ...................................................................................... 125
6.3.1 Poliovirus Plaque Assay ........................................................................... 125

6.3.2 Wastewater Sample Collection and Cell Culture ...................................... 126

 

6.3.3 Poliovirus spiking .......................................................................................... 127
6.3.4 Preparation of Samples for Hybridization ................................................ 128
6.3.5 Hybridization of Samples ......................................................................... 129
6.3.6 Data Analysis Using DetectiV .................................................................. 130

6.4 Results ............................................................................................................... 130
6.4.1 Viruses Present in January 2008 Control Sample ..................................... 130
6.4.2 Sensitivity Testing using Spiked Poliovirus LS-C-l ................................ 131

6.5 Discussion ......................................................................................................... 134
6.5.1 Viruses present in January 2008 sewage .................................................. 134
6.5.2 Spiking sensitivity ..................................................................................... 135
6.5.3 Conclusions ............................................................................................... 136
CHAPTER 7. SUMMARY AND CONCLUSION - __ _ - 138
REFERENCES .......... - ----- - 147

 

vi

LIST OF TABLES

Table 2.1. Baltimore Classification of viruses by genome type ................................... 10
Table 2.2. Viral Microarrays and Their Applications .................................................. 31
Table 4.1. RT/PCR primers used in this study for the detection of viruses ................. 50
Table 4.2. Viral Water Quality at Lake Michigan Beaches ......................................... 55

Table 4.3. Average pH, water temperature, wind speed, wind direction, turbidity and
number of swimmers recorded for Washington Park and Silver Beach samples. ............ 58

Table 5.1. List of viral targets, type of genomes and Genbank accession numbers ..... 80

Table 5.2. List of primers used for the polymerase chain reaction PCR detection and
sequencing for each viral target ......................................................................................... 88

Table 5.3. t test results for the hybridization of control viruses and sewage extracted
samples to the viral microarray .......................................................................................... 97

Table 5.4. Comparison of positive signal fractions, software analysis, group specific
PCR results and sequencing data for the identification of viruses within a sample ........ 101

Table 6.1. p values and log; average signal for the ten most likely viruses present in
the January 2008 sewage sample ..................................................................................... 132

Table 6.2. Distribution of virus groups among winter month samples ...................... 137

vii

LIST OF FIGURES

Figure 4.1. Overhead satellite image of the Washington Park sampling location. Stars
indicate the sampling points ............................................................................................... 47

Figure 4.2. Overhead satellite image of the Silver Beach sampling location. Stars indicate
the sampling points. ........................................................................................................... 47

Figure 4.3. The effect of wind direction in the near-shore region depends on the
orientation of the beach ...................................................................................................... 60

Figure 4.4. Scatter plot matrix showing the association between the variables for the two
beaches ..................................................................................................................... 63

Figure 4.5. Observed and simulated viral data plotted on a 1:1 line for Silver Beach and
Washington Park. ............................................................................................................... 65

Figure 5.1. Schematic illustrating the steps involved with virus concentration, nucleic
acid extraction, labeling and hybridization on the virus microarray ................................. 85

Figure 5.2. Bar plots showing the logz-normalized hybridization signals for various
control experiments ............................................................................................................ 94

Figure 5.3. Adenovirus type 40 and 41 hybridization results .......................................... 100
Figure 5.4. Bar plot showing logz-normalized hybridization signals for the different target

virus groups when Cy3-labeled DNA and Cy5-label RNA extracted from sewage is
hybridized unto the array ................................................................................................. 105

Figure 6.1. logz-transformed average poliovirus signal versus number of poliovirus
plaque-forming units (PFID. ........................................................................................... 133

viii

ABBREVIATIONS

CL — confidence limit

DNA — deoxyribonucleic acid

dNTP — deoxy—nucleotide triphosphate
L — liters

M — molars

mM — millimolar

MPN — Most Probable Number estimate
PCR — polymerase chain reaction

PFU — plaque forming units

RNA ~ ribonucleic acid

ix

CHAPTER 1 INTRODUCTION TO HUMAN VIRUSES AND

THE WATER ENVIRONMENT

1.] Viral Gastroenteritis, Disease Surveillance and Outbreak Information
Viruses which cause gastroenteritis, which is an inflammation in the
gastrointestinal tract, are known as enteric viruses and are found in contaminating
sewage, surface and groundwater. They are primarily associated with drinking
recreational and food-borne disease though some enteric viruses are capable of causing
more severe diseases like meningitis and cardiomyopathy. In 2002, the World Health
Organization (WHO) estimated that there were 1.7 million deaths related to unsafe water,
sanitation and hygiene, mainly through infectious diarrhea (WHO, 2002). The estimated
4 billion cases of diarrhea annually account for over 82 million disability adjusted life
years (DALYs) and represents 5.7 percent of the global morbidity (Pruss and Havelaar,
2001). In the United States, there were 764 documented waterborne outbreaks attributed
to drinking water between 1971 and 2002, resulting in 575,457 cases of illness and 79
deaths (Blackburn et al., 2004). The most recent report from the US CDC reported that
within the 2003-2004 time period, there were 98 waterborne disease outbreaks for both
drinking and recreational waterborne outbreaks (5458 illness, 58 hospitalization and 5
deaths) , 50 of which were caused by infectious agents leading to gastroenteritis. Viruses
were known or suspected to have been responsible for 8 out of 50 of the infectious

gastroenteritis cases reported (Dziuban et al., 2006; Liang et al., 2006).

These outbreak figures however do not represent the true numbers of waterborne
disease, the possible etiological agents and the cost. An estimate of the waterborne
infection and illness rate in the US by Reynolds et al postulated that 10.7 million
infections per year and 5.4 million illnesses per year occur in populations served by
community groundwater systems; 2.2 million infections per year and 1.1 million illnesses
per year occur in non—community groundwater systems; and 26.0 million infections per
year and 13.0 million illnesses per year occur in municipal surface water systems. The
total estimated number of waterborne illnesses per year in the US. was thus estimated to

be 19.5 million per year (Reynolds et al., 2008).

1.2 Productivity Losses, Economic Costs of Viral Infections

While the symptoms of the diarrhea are generally mild in adults and the majority
of sufferers recover within a few days, the annual productivity loss due to rotavirus
infections alone has been estimated to be in excess of 6.3 million pounds in the United
Kingdom and 352 million dollars in the United States (Barnes et al., 1998). Within the
state of Michigan, tourism, fishing, boating and other recreational activities involving
contact with water contributes over 17.5 billion dollars to the State’s economy and
generates over 192,700 jobs. In 2006, 4 percent of the 207 regularly monitored tier 1
beaches in Michigan were found to exceed the State’s daily maximum bacterial standard
of 300 colony forming units per 100 milliliters of water. Nationwide, the number of
closings and advisory days at Ocean, bays, and Great Lakes beaches for 2006 increased

by 28 percent compared to the previous year resulting in a total of 25,643

closing/advisory days. The majority of these closings (over 67 percent) were in response

to known pollution events or actual monitoring results (Dorfrnan and Stoner, 2007).

1.3 Knowledge Gaps, Limitations of Current Tests

Despite the large economic and productivity costs attributed to enteric viral
disease. Limited data are available on the prevalence of viruses in the environment.
Previous work on surveying the environment for the presence of enteric viruses have
been restricted by the cost and manpower needed to analyze such matrices. Although
bacterial phages have been proposed as an alternative indicator of viral pollution,
research has shown that they do not adequately represent the extent and degree of human

and animal viral pollution in the environment (J iang and Chu, 2004).

In 2000, Congress authorized the “Beaches Environmental Assessment and
Coastal Health Act of 2000” which required that the US Environmental Protection
Agency develop within 5 years “new or revised water quality criteria for pathogen and
pathogen indicators (including a revised list of testing methods, as appropriate)”. To date,
no official testing methods for the detection of pathogenic viruses in the environment
have been forthcoming. This is despite the fact that it is generally accepted that the
health impact of waterborne diseases due to viruses is greatly underestimated (Leclerc et
al., 2002). Most of the bacterial and protozoal causes of gastroenteritis are well
characterized and detection is generally possible within 24-48 hours. Viruses are harder
and more laborious to detect but yet are believed to survive longer in the environment

and also have a higher probability of causing an infection even at low numbers. Being

obligate intracellular parasites, they have to be cultured in animal cell lines in order to
demonstrate infectivity. Some viruses however, are refractory to cell culture and can only
be detected via molecular methods like reverse transcription PCR or via methods like
electron microscopy or enzyme immuno assays, which are costly and limited to a small

number of central laboratories.

It is currently estimated that there are approximately over 100 individual species
of viruses and an undetermined number of sub-strains of viruses that are capable of
causing infections in humans through drinking and recreational use. These viruses are
excreted in the feces and urine of infected individuals and may survive exposure to the
enviromnent long enough to infect other individuals. In Michigan and the Great Lakes,
while drinking water quality is generally satisfactory, there is growing concern about the
fresh water coastal beaches, especially due to sewage inputs. It is hypothesized that
sewage represents a significant source of viral pathogens and that the detection of viruses
represents a sewage pollution risk to bathers at recreation sites. Hundreds of viruses have
been reported in sewage, yet no approach has been made available to screen widely for

the presence of these viruses in environmental matrices.

1.4 Microarray Detection of Pathogens

Within the past 10 years, microarray technology has been increasingly adapted for
use as a multiple pathogen screen tool. Initially used as a means of performing gene
expression analysis, microarrays now exist for the detection of a diverse list of pathogens

from parasites, to bacteria and viruses. The use of microarrays in an environmental

setting is still in its infancy however and a microarray designed to perform high
throughput analysis of water samples for the presence of waterborne disease-causing

viruses would represent a novel application of this technology.

CHAPTER 2. LITERATURE REVIEW

2.1 Recreational Water Quality Standards

Recreational water quality for safe swimming is regulated at the state level under
the Clean Water Act, as are the quality of ambient waters and sewage discharges. From a
microbiological perspective, public health protection is the goal and fecal indicator
bacteria have been used to develop water quality standards and criteria. For recreational
water use, states have been encouraged to adopt standards that are at least as stringent as
the recommendations set by the EPA under Section 304 of the Clean Water Act (EPA,
2003) and the Beaches Environmental Assessment and Coastal Health (BEACH) act. The
current EPA guideline for safe swimming and full body contact is 126 colony forming
units (cfu) / 100ml for E. coli and for enterococci the level is 33 cfu/ 100ml and 35 cfu/
100ml for freshwater and marine waters respectively. The state of Michigan has adopted
a recreational water quality standard for monthly averages of no more than 130 cfu/ 100
ml for surface waters protected for full body contact with no daily sample average ( 3
samples per site) more than 300 cfu/ 100 ml and a single sample maximum of no more

than 1,000 cfu/ 100 ml for surface waters protected for partial body contact.

In the state of Michigan, the National Pollution Discharge Elimination System
that regulates sewage discharges uses a 30-day geometric mean standard of 200 fecal
coliforrn bacteria per 100 milliliters and a 7 day geometric mean standard of 400 fecal
coliform bacteria per 100 milliliters for treated and untreated sewage discharges. This

standard was originally set up to protect “swimmable” waters where sewage discharges

impacted recreational sites and generally list segments of waterbodies under an
impairment which would require assessment. It has been recognized however that the
presence of pathogenic viruses is not well correlated with the numbers of indicator
bacteria and that the standards for indicator bacteria used to protect recreational waters
are disconnected from the rules that govern sewage discharges and public safety. Thus
characterization of pathogenic viruses in particular may be of particular importance in
these types of waters if in the future better assessment of disease potential from

swimming in sewage polluted waters is to be undertaken.

2.2. The Enteric Viruses

Viruses were first identified by Russian biologist Dmitry Ivanosky in 1892 and
initially referred to as filterable agents. They are very small microorganisms straddling
the border of the biological definition of life. Their sizes range from 20nm to 400nm and
they lack many of the characteristics of a biological cell — they have no cell wall, no
internal membranes and no cellular machinery to reproduce with. They are thus obligate
intracellular parasites, relying upon host machinery in order to multiply. Viruses
historically were classified according to their host specificity, shape and size as seen
under the electron microscope. As more and more viruses were discovered, and more
tools for their characterization were developed, classification began to include their
nature, nucleic acid, and the presence or absence of an envelope. The classification
system developed by David Baltimore lists 7 categories of viruses (Group I to VII) which

includes viruses capable of infecting humans, animals, plants, firngi and bacteria

(Baltimore, 1971). A table is provided below showing the seven categories as well as a

few examples from each (Table 2.1).

The International Committee on Taxonomy of Virus (ICTV) has established a
virus classification scheme based upon a set of 2600 character criteria. The new Linnean-
like classification adopts a virus taxonomy of order, family, subfamily, genus, and
species and viruses are classified based on the relatedness of genome sequence, natural
host range, cell and tissue tropism, pathogenicity and cytopathology, mode of
transmission, physiochemical properties of virions, and antigenic properties of viral
proteins (Buchen-Osmond, 2003). It also allows subspecies, serotype and strain/isolate
classification of viruses. Currently there are close to 2000 virus species in the latest report

from ICTV (Fauquet et al., 2005).

It has been estimated that almost half of all emerging pathogens arising from
infections caused by contact with water are caused by viruses or small infectious protein
particles called prions (Taylor et al., 2001). Viruses are reported to be the main acute
diarrhea outbreaks in infants and young children worldwide (Kapikian, 1996). Mead et
al. have estimated that 80% of the 38.6 million annual cases of gastroenteritis in the
United States are the result of a viral infection (1999). Enteric viruses are viruses spread
by the fecal-oral route as they cause infection through replication in the gastrointestinal
system and are subsequently excreted in feces. They are found thus in sewage and are the
main groups associated with contamination of food and water. The major families of

enteric viruses can roughly be grouped into the following families - the Picornaviridae,

the Adenoviridae, and the Reoviridae. More recently, the noroviruses and other
caliciviruses in the family Caliciviridae have gained prominence for being the causative
agent of several recent outbreaks (Bohnker and Thornton, 2003; Doyle et al., 2004;
Drinka, 2005; Falkenhorst et al., 2005; Fretz et al., 2005; Korsager et al., 2005; Maunula
et al., 2005; Maunula and Von Bonsdorff, 2005; Sakon et al., 2005; Seto et al., 2005;
Hjertqvist et al., 2006; Ike et al., 2006; Koopmans et al., 2006; Takkinen, 2006; Vainio
and Myrrnel, 2006; Vidal et al., 2006; Ljubin-Stemak et al., 2007; Tu et al., 2007; Fukuda

et al., 2008; Makary et al., 2008; Tomer etal., 2008).

Table 2.1. Baltimore Classification of viruses by genome type

 

 

 

 

 

 

 

DNA viruses

Group I — Enterobacteria phage T4

dsDNA viruses (double stranded DNA) Human herpesviruses
Cowpox virus,
Adenoviruses,
Enterobacteria phage 7t,
Polyomavirus

Group II — Parvovirus B19

ssDNA viruses (single stranded DNA)

RNA viruses

Group III — Rotavirus

dsRNA viruses (double stranded RNA)

Group IV — Coronavirus

(+) ssRNA viruses (positive single stranded RNA or
mRNA like)

Norwalk Virus
Hepatitis E virus

 

 

 

 

West Nile Virus
Hepatitis C virus
Poliovirus
Rhinovirus
Hepatitis A virus
Group V — Influenza viruses
(-)ssRN A viruses (negative single-stranded RNA)
DNA and RNA Reverse Transcribing viruses
Group VI -— HIV 1

ssRNA-RT viruses (single stranded RNA)

 

 

Group VII —
dsDNA-RT viruses (double stranded DNA)

 

Hepatitis B virus

 

10

 

2.2.1 The Enteroviruses

Enteroviruses are small, single stranded, positive sense RNA viruses. They are
non-enveloped and have an average diameter of 27-30 nanometers. Enteroviruses are a
large group of viruses from the Picomavirz’dae family that are responsible for many
infections in children. These viruses live in the intestinal tract, but can cause a wide
variety of illnesses. There are sixty-six distinct serotypes of enterovirus known to cause
infection in humans: three polioviruses, twenty-three Coxsackie A viruses, six Coxsackie
B viruses, twenty-eight echoviruses and six other enteroviruses (Bruu, 2003).
Enteroviruses can cause respiratory infections, gastrointestinal infections, skin infections

and neurological infections.

Large numbers of enteroviruses are routinely found in sewage but few waterborne
outbreaks due to this group of viruses have been reported (Metcalf et al., 1995). One of
the suggested reasons is that multiple symptoms are exhibited by enteroviral infections
and thus the lack of a common unifying symptom results in failure to recognize an
outbreak by the medical community. While the entry of polio, coxsackie, echo, or other
enteroviruses through the gut may cause incidental mild diarrheal symptoms, it is the
spread of the virus through the bloodstream to other organs (e. g., central nervous system,
heart, pleura, pancreatic islets) that produces major disease manifestations - examples of
which are hand-foot-and-mouth (HFM) disease, herpangina, myocarditis, neonatal sepsis
and pleurodynia. Research has also indicated that there is a link between recent
enteroviral type 71 infections and type 1 diabetes (Elfaitouri et al., 2007 ; van der Werf et

al., 2007; Frisk et al., 2008; Oikarinen et al., 2008).

ll

2.2.2 The Adenoviruses

Adenoviruses are classified as non enveloped double stranded DNA viruses. Their
average diameter is between 60-90 nanometers. They are medium sized viruses and have
a genome of approximately 30 kilobasepairs. There are 5 groups of human adenoviruses
with more than 40 subtypes. Adenoviruses are widely recognized causes of respiratory,
ocular, and genitourinary infections. However, serotypes 40 and 41 (previously called
fastidious enteric adenoviruses) primarily affect the gut, contributing to 5%-20% of
hospitalizations for childhood diarrhea in developed and developing countries (Giordano
et al., 2001; Rodriguez-Baez et al., 2002; Filho et al., 2007; Sdiri-Loulizi et al., 2008).
Adenovirus is considered to be only second to rotaviruses in causing acute viral
gastroenteritis worldwide. The peak incidence is among children less than 2 years of age,
but older children and adults may be infected, with or without symptoms. Infections
occur throughout the year with no clear peaks (Wong et al., 2008). Other serotypes of
adenovirus, particularly type 2, 7, 12 and 31, have also been associated with diarrhea
(Noel et al., 1994; Harsi et al., 1995; Li et al., 2005). Adenoviruses have frequently been
detected in immuno-compromised patients with diarrhea. A study examining 377 HIV-
positive patients presenting with diarrhea found that adenoviruses were present in 7.2%
of them and accounted for 50% of the patients who were enteric-virus positive (Thomas
et al., 1999). However other studies have made it unclear whether the presence of
adenoviruses in HIV-positive enteric virus-infected patients was associated with the

presentation of diarrhea (Liste et al., 2000).

12

2.2.3 The Rotaviruses

The rotaviruses are double stranded RNA viruses from the family Reoviridae with
a genome consists of 11 segments. Rotaviruses are the most common cause of severe
diarrhea among children. In the United States, approximately 3.5 million cases occur
each year. A review of the national rates, trends, and risk factors for diarrhea- and
rotavirus-associated hospitalizations and deaths among children <5 years of age by
Fischer et al. (2007) found that rotavirus infections remain the most important cause of
pediatric diarrhea throughout the study period of 1993-2003, causing approximately
60,000 hospitalizations and 37 deaths annually. Worldwide, an estimated 140 million

cases occur each year, causing almost 600,000 deaths (Parashar et al., 2003).

In the United States, the peak incidence of rotavirus diarrhea is among children 6
months-2 years of age, although in developing countries younger infants may be affected.
By 4 years of age, most persons have been infected and are immune to this severe
dehydrating syndrome, but a high inoculum or lowered immunity can still produce milder
illness among older children or adults, among travelers to developing nations, the elderly,

and persons with debilitative or immunosuppressive conditions. A person with rotavirus

diarrhea may excrete approximately 1 x 1012 infectious particles/milliliter of stool

(Iturriza-Gomara et al., 1999). Asymptomatic rotavirus excretion has been reported
among half of children the day before diarrhea starts and among one-third during the
week after symptoms end (Pickering et al., 1988). Many children can shed rotavirus and
never become ill (Champsaur et al., 1984; Champsaur et al., 1984). In a small prospective

study in the UK, rotavirus caused 41% of acute diarrhea in adults admitted to hospital

13

(J ewkes et al., 1981). Similarly, 3% of acute diarrhea in Switzerland (Loosli et al., 1985),
3% of infectious diarrhoa pathogens in a Swedish clinic for infectious diseases
(Svenungsson et al., 2000), 5% of adults with gastroenteritis requiring admission in
Thailand (Echeverria et al., 1983), 2—4% of adults older than 15 years with gastroenteritis
presenting to their family physician in the Netherlands (de Wit et al., 2001), and nearly
4% of individuals older than 45 years in Michigan were due to rotavirus (Koopman and
Monto, 1989). Even higher rates of infection have been seen. In Japan, Nakajima et al.
(2001) reported that group A rotavirus had a role in 14% of patients with diarrhea. Pryor
et al. (1987) noted that rotavirus was second only to Campylobacter spp as a cause of
diarrhea among Australian adults, accounting for 17% of all cases. In Indonesia, 42% of
patients presenting with diarrhea had rotavirus-positive stools compared with 11% of
control samples (Oyofo et al., 2002). In a study of Mexican adults, 63% of patients
presenting with acute gastroenteritis during winter months were positive for rotavirus (del

Refugio Gonzalez-Losa et al., 2001).

Nosocomial rotavirus among pediatric populations is common; in one study
among hospitalized children, a nosocomial infection rate of 1.6 per 1000 child-days was
recorded (Snelling et al., 2007). Rotavirus at day-care centers, in both endemic and
outbreak form, is also common (Gabbay et al., 1999; Floret et al., 2006). Outbreaks in
neonatal units are frequently reported, but infection among full-term infants is usually
benign, perhaps because maternal antibody transferred during the third trimester protects
against illness for the first 3-6 months of life (Bishop et al., 1996; Nguyen et al., 2006);

premature infants are at higher risk. Among adults, only one outbreak arising from

14

rotavirus contamination of a municipal water supply has been reported (Gallay et al.,
2006)

It was previously thought that only Group A rotaviruses were capable of causing
an infection in humans while the other antigenic groups (B-E) were zoonotic. In 1982,
however, an epidemic of Group B rotavirus affected millions of persons in China
(including adults, children, and neonates) (Hung et al., 1984), and since then outbreaks
have re-ocurred, although affecting fewer persons. Studies of immunoglobulin pools from
Shanghai suggest that the Chinese population had been exposed to this pathogen in the

past (Penaranda et al., 1989).

2.2.4 The Noroviruses

Noroviruses are single, positive-stranded RNA viruses from the family
Calicivz'ridae. They are only 27 nanometers in diameter which makes them one of the
smallest among the viruses. A British study suggested that approximately 3% of children
hospitalized for diarrhea excrete calicivirus (Ellis et al., 1984), and a US. study found
approximately the same percentage (2.9%) for children with diarrhea in day-care centers
(Matson et al., 1989). On the basis of antibody-prevalence studies of pooled
immunoglobulin and serum samples fiom many parts of the world, most persons appear
to have been infected by age 12 (Sakuma et al., 1981; Parker et al., 1994). Norovirus has
been reported to be highly seasonal and geographic, with peaks in the winter for the
northern hemisphere and peaks in the late spring and summer for the southern
hemisphere (Marshall et al., 2003), though summertime peaks in the northern hemisphere

have been documented (Lopman et al., 2003).

15

Person-to-person transmission is presumed to be essential for endemic disease,
but for enteric viruses, contaminated shellfish, cold foods, and drinking water have been
implicated as vehicles (Wanke and Guerrant, 1987; Stolle and Spemer, 1997). A survey
of foodbome disease outbreaks in Australia between 1995 and 2000 identified six
outbreaks caused by noroviruses or 3% of the 214 documented outbreaks for that period
(Dalton et al., 2004). Most outbreaks of norovirus gastroenteritis were observed to be
community based, affecting nursing homes, schools, cruise ships, camps, restaurants, and
military installations (CDC, 2002; CDC, 2003). A study of 233 US outbreaks occurring
between July 1997 and June 2000 reported that the most common occurrences of
outbreaks were in restaurants and catered meals (39%) and in nursing homes (25%).
Contaminated food was the most common vehicle of transmission (57%) followed by

person-to-person contact (16%) and contaminated water (3%) (Fankhauser et al., 2002).

2.2.5 The Astroviruses

Astroviruses are very similar to noroviruses in that they are small positive
stranded RNA viruses. Astroviruses are from the family Astroviridae and are divided into
eight human species (HAstV-l to HAstV-8). Their genome length is between 6800 and
7900 nucleotides. Initially, astroviruses were regarded as only a minor contributor to the
incidence of childhood diarrhea based on Electro Microscopy (EM) results. More recent
serological testing has shown astroviruses to be more prevalent than previously thought.
Currently, astroviruses are considered second only to rotaviruses as the cause of

gastroenteritis in infants and young children. Studies of hospitalized children suggest that

16

astroviruses may account for 4-10% of admissions for diarrhea (Palombo and Bishop,
1996; Guerrero et al., 1998; Bon et al., 1999; Pang and Vesikari, 1999; Rodriguez-Baez
et al., 2002). Other studies have indicated that the prevalence might be higher among

children evaluated for gastroenteritis in the ambulatory setting (Cruz et al., 1992).

A long term study in Australia over four consecutive years showed that the
pattern of astrovirus infection that was without a statistically significant seasonal peak
(Mustafa et al., 2000). Other studies however have described a predilection of astrovirus
for winter or the rainy season in populations living also in temperate regions (Cruz et al.,
1992; Maldonado et al., 1998). A study by Michell et al. found that the seroprevalence of
393 infants and children to HAstV-l decreased from 67% in infants <3 months of age to
7% by 6 to 8 months of age. Antibodies to HAstV-3 exhibited a lower prevalence, with
26% positive at <3 months, and 0% at 6 to 11 months. This was consistent with loss of
transplacental antibodies. Children acquired HAstV-l antibody with a peak prevalence of
94% at 6 to 9 years of age and HAstV-3 antibodies with a peak prevalence of 42% by 6

to 9 years of age (1999).

2.2.6 The Picobimaviruses

Picobimaviruses are bi-segmented, non-enveloped, double stranded RNA viruses
that are as yet unclassified. Prior to a study done by Pereira and colleagues,
picobimaviruses were only thought to infect animals. Pereira and colleagues identified
picobimaviruses using polyacrylarnide gel electrophoresis among fecal specimens from

children with diarrhea in Brazil (Pereira et al., 1993). Picobimaviruses have also been

17

detected in stool samples from HIV positive individuals both with and without diarrhea
(Grohmann et al., 1993; Gonzalez et al., 1998; Giordano et al., 1999). Though still
primarily associated with young children, the elderly and immunocompromised
individuals, a survey of fecal specimens collected between 1982 and 1993 from the
United Kingdom detected picobimaviruses in patients with and without gastroenteritis

and throughout the age range of 3 to more than 65 years (Gallimore et al., 1995).

2.2.7 The Parvoviruses

Parvoviruses are the smallest of all human viruses, measuring 20-25 nanometers
in diameter. They are single-stranded DNA viruses with a genome size between 3 to 5
kilobasepairs. Only one gastrointestinal parvovirus outbreak has been documented among
humans (Christopher et al., 1978) though several outbreaks have occurred among animals
(Palmer and Thomley, 2004). The relationship of these particles to disease is unclear, but
they have been associated with shellfish-related outbreaks of gastroenteritis (Appleton,

1987).

Human parvovirus 819 is another member of the Paroviridae family that causes
pathogenesis in human but unlike gastrointestinal parvovirus, human parvovirus B19 is
transmitted via the respiratory route. Human parvovirus B19 causes erythema
infectiosum and, particularly in adults, acute symmetric polyarthropathy. In the
immunocompromised host persistent Bl9 infection is manifested as pure red cell aplasia

and chronic anemia. Likewise, the immature immune response of the fetus may render it

18

susceptible to infection, leading to fetal death in utero, hydrops fetalis, or development of

congenital anemia (Heegaard and Brown, 2002).

2.2.8 The Toroviruses

Toroviruses are members of the Coronaviridae family of single-stranded,
positive-sense-strand RNA enveloped viruses. Their envelope is approximately 100
nanometers to 140 nanometers in diameter. The first incidence of detection of toroviruses
in feces from patients with gastroenteritis occurred in 1984 (Beards et al., 1986).
Subsequently, it was realized that toroviruses appear to be associated with persistent and
acute diarrhea in children and may also be readily spread in a hospital (Koopmans et al.,

1997; Jamieson et al., 1998).

Toroviruses are known causes of diarrhea among cattle, and identification in
human specimens has been reported (Koopmans et al., 1993; Koopmans et al., 1997;
Krishnan and Naik, 1997). Studies have also shown that torovirus may be associated with

necrotizing enterocolitis (NEC) in newborn infants (Lodha et al., 2005).

2.2.9 The Coronaviruses

Coronaviruses are enveloped viruses with a positive, single stranded RNA
genome of approximately 27 to 32 kilobasepairs. Prior to 2003, only 2 human
coronaviruses were known — Human coronavirus (HCoV) 229E and HCoV-OC43 from
the group 1 and group 2 coronaviruses respectively. Subsequently, a new strain of

coronavirus, SARS-CoV, was discovered that is either a distinct member of the group 2

19

coronaviruses or the first member of a new group of coronavirus (Rota et al., 2003;
Snijder et al., 2003; Gibbs et al., 2004). In addition, two other human coronaviruses have
also been identified, HCoV-I-IKUI and HCoV-NL63, that cause infections of the upper

and lower respiratory tract (Kahn and McIntosh, 2005).

Coronaviruses are well-established causes of diarrhea in animals especially
among swine, cats, mice, dogs and birds (Squires, 2003; Perlman and Dandekar, 2005 ;
Weiss and Navas-Martin, 2005). The virus also causes respiratory disease in humans
(Kahn and McIntosh, 2005). Coronaviruses are generally not considered to be a cause of
acute gastroenteritis in adult humans as they have been detected in the stools of both
diseased and healthy adults with equal frequency(Maass and Baumeister, 1983). They
have, however, been shown to occur more often in the stools of infants and children with
acute gastroenteritis than in healthy children (Maass and Baumeister, 1983; Kidd et al.,
1989). Kern and colleagues have also shown that coronavirus excretion correlated
significantly with the diagnosis of AIDS or with syndromes belonging to the AIDS-
related complex though they could not establish whether the virus caused any disease in

these individuals (Kern et al., 1985).

2.3. Virus Sampling and concentration methods for water

2.3.1 Concentration of Viruses from the Water Environment
The infectious dose of viruses is not precisely known but infections have been
shown to occur with high probability even at very low doses (Bosch, 1998; Haas et al.,

1999). Thus viruses are capable of causing disease even though they are present in the

20

environment in very low numbers. In order to obtain a representative sample, large
volumes of water (10 to 1000 Liters) need to be concentrated down to smaller sample
volumes for processing and virus detection methods (microliters to milliliters).
Guidelines published by the US EPA recommend that an equivalent volume of 10 Liters
for sewage, 100 Liters for surface and ground-waters and 1000 Liters of finished water be
analyzed in order to achieve a representative sample volume (Fout, 2001). Such large
volumes of water would be unwieldy to store and to transport safely and at the right
temperatures to the laboratory for processing. Several methods have been developed for
the concentration of viruses from water - the virus adsorption/desorption (VIRADEL)
method, cation coated filter method, and ultrafiltration. A ntunber of methods have also
been developed for the concentration of viruses following elution: organic flocculation,

polyethylene glycol precipitation, and Speedvac concentration will be discussed.

2.3.2 Virus Adsorption Elution

The virus adsorption elution method (V IRADEL) was developed to allow the
capture of viruses by passage of large volumes of water through electrostatic filters
(Goyal and Gerba, 1983). Viruses are adsorbed to the filter surfaces, transported to the

laboratory and then desorbed from the filters.

Some of the earliest filters used for the concentration of viruses consisted of
cellulose fibres presented either as a membrane disk or a filter cartridge. The properties
of cellulose acetate meant that when in use, they would adopt a negative electrostatic

charge. Since most virus particles at neutral pH also tended to possess a net negative

21

charge, the use of these filters meant that the pH of the sample would have to be altered
so as to impart a net positive charge to the virus particles and allow their adsorption to the
filters. Commonly this meant lowering the pH of the sample to approximately 3.5 before
passage through the negatively charged filter (Goyal and Gerba, 1983). The addition of
multivalent cations has also been shown to facilitate virus floc formation and adsorption
to the filters (Haramoto et al., 2004). The addition of varying concentrations of
Manganese, Magnesium and Aluminum have all been proposed to enhance adsorption of

viruses to the filters (Lukasik et al., 2000).

In order to avoid having to alter the pH of the water, which could potentially
inactivate viruses (Darnell et al., 2004), positively charged filters were developed by
Sobsey and his colleagues (1973). This also facilitated the collection of large volumes of
sample water. At present, the list of commercially available positively charged filters
include the lMDS filter, Zeta Plus 50-s and 60s (Mehnert and Stewien, 1993; Kittigul et

aL,2001)

Elution of the viruses is accomplished via one of several ways. 1%-2.9% Tryptose
Phosphate Broth (pH 9.0) with or without glycine and arginine, 1.5% - 3% Beef Extract
(pH 9.0 -— 9.5) with or without glycine, 0.05M glycine (pH 10.5 — 11.5) and Urea-
Arginine Phosphate Buffer (1.5M urea, 0.02M arginine, 0.008M phosphate; pH 9.0) have
all been used as eluates with different filters to varying degrees of elution efficiency
(Toranzos and Gerba, 1989; Jothikumar and Cliver, 1997; Kittigul et al., 2001; Dahling,

2002)

22

2.3.3 Cation coated filter method

The cation-coated filter method involves the precoating of conventional cellulose
membranes with cations, commonly aluminium or magnesium followed by filtration,
rinsing with 0.5M H2804 and elution with 1.0mM NaOH. Haramoto and colleagues have
reported recovery efficiencies of greater than 80% using poliovirus-seeded milliQ water
using this method (Preston et al., 1988; Haramoto et al., 2004; Haramoto et al., 2005).
Cellulose membrane filters are inexpensive compared to the electropositive filters,

however their use as reported thus far has only been for volumes of 10 liters or less.

2.3.4 Microfiltration and Ultrafiltration

Microfiltration and ultrafiltration involves the use of specially designed filters that
are capable of denying passage to particles at the sub-micron size. Such filters have been
used either directly, or indirectly to concentrate viruses from water and other
environmental samples (Winona et al., 2001; Rutjes et al., 2005). In addition ultrafilters
can also be used to recover all kinds of microbes, not just viruses which facilitates their
use in assaying unknown microbial contamination events (Morales-Morales et al., 2003;
Hill et al., 2005). The use of ultrafiltration to recover and concentrate viruses has been
demonstrated at both the small (3 2 liters) and at the large scale (100 liters) (Winona et

al., 2001; Olszewski et al., 2005).

23

2.3.5 Organic Flocculation

Organic flocculation is the current EPA-recommended method for the
concentration of viral particles following elution off filters. The addition of organic
material like powdered beef extract has been shown to result in the formation of virus
clumps or flocs around the larger organic molecule. This allows the virus to be pelleted
by centrifugation at relatively low speeds (approximately 2500 x g). Organic flocculation
has been used in a number of surveys to concentrate culturable viruses and
bacteriophages from surface and groundwater, sludge biosolids and wastewater and has
also shown to be compatible with subsequent downstream polymerase chain reaction
detection methods (Pinto et al., 1993; Ma et al., 1994; Ma et al., 1995; Chapron et al.,
2000; Chapron et al., 2000; Sedmak et al., 2003; Laverick et al., 2004; Sedmak et al.,

2005)

2.3.6 Polyethylene Glycol Precipitation

Polyethylene glycol (PEG) precipitation has previously been reported to be more
effective at concentrating some animal viruses compared to organic flocculation (Lewis
and Metcalf, 1988; Boher, 1991). The addition of PEG is believed to sterically block the
protein capsid of the virus from associating with the solvent molecules, resulting in the
precipitation of the viruses, an alternative proposal is that the interaction of the protein
charges on the virus with the PEG polymer causes crystallization to occur. It is believed
that the actual mechanism of precipitation is the result of a combination of both of these
processes (Lewis and Metcalf, 1988). The method has been used to detect rotaviruses,

hepatitis A virus, Norwalk virus and adenoviruses in river, raw and treated drinking water

24

but is primarily used to detect viruses associated with foodstuff, especially shellfish
(Hovi et al., 2001; Kingsley and Richards, 2001; Dubois et al., 2002; Goswami et al.,
2002; Kingsley et al., 2002; van Zyl et al., 2004; van Heerden et al., 2005). A comparison
of PEG precipitation and ultracentrifugation has shown PEG precipitation to be more

effective at removing PCR inhibitors (Sunen et al., 2004).

2.3.7 SpeedVac Concentration

SpeedVac concentration of viruses has not been widely reported. The main
proponents of this method for concentrating viruses have been a group from Mahidol
University in Thailand. Kittigul and colleagues have reported a significantly higher
rotavirus recovery when they compared SpeedVac concentration and PEG precipitation
(Kittigul et al., 2001). The group has successfirlly detected poliovirus, hepatitis A virus

and rotavirus from sewage and other environmental water samples (Kittigul et al., 2000).

2.4 Methods of Viral Detection

2.4.1 Cell Culture for Environmental samples

Cell culture is the most widely used method for the detection of enteric viruses
concentrated from water. Cell culture methods exist for picomaviruses, rotaviruses,
enteric adenoviruses, and astroviruses and are currently being developed for other viruses
(Chapron et al., 2000; Pusch et al., 2005; Shieh et al., 2008). Unfortunately several of the
newly emerging enteric viruses have not been amenable to culture, limiting the use of this

technique. One of the viruses that had previously resisted attempts to culture are the

25

noroviruses. Straub et al. have successfully cultured noroviruses in a three dimensional
fluid shear wall vessel bioreactor and postulate that previous attempts at culturing
noroviruses failed due to the use of monolayers (Straub et al., 2007). Three dimensional
cell culture systems are believed to be superior to two dimensional cell culture systems
because they mimic the morphology and biochemical cell features that are present in the

original cell tissue (Andrei, 2006).

2.4.2 Polymerase chain reaction (PCR)

The use of the polymerase chain reaction for the detection of pathogens was
primarily a research tool that has rapidly gained acceptance in the clinical diagnostic
setting. PCR has been shown to be an extremely sensitive, specific and rapid technique
for the diagnosis of viral pathogens. The sensitivity of PCR has been demonstrated to be
comparable or superior to cell culture (Bae et al., 2003; Raboni et al., 2003) though great
care must be taken to avoid false positive reactions. The chief drawback of the PCR
method is that it is incapable of distinguishing active and inactive targets. Prior to PCR,
the viral nucleic acid must be extracted, several commercial extraction kits are available,
most of them utilizing the Boom method developed by Boom and colleagues (Boom et
al., 1999). One modification of the standard PCR method, Integrated Cell Culture PCR
(ICC PCR), makes use of a cell culture step to enhance sensitivity and demonstrate
infectivity. Another variant of PCR, Real Time PCR (RT PCR) can be used to quantify
the original template concentration of the sample. Primers for the specific detection of
many of the enteric viruses have been published (Beuret, 2004; He and Jiang, 2005;

Jothikumar et al., 2005; Pang et al., 2005; van Heerden et al., 2005; Costafreda et al.,

26

2006; Elia et al., 2006; Oka et al., 2006; Yong et al., 2006; Young et al., 2006; Bird et al.,
2007; Butot et al., 2007; Casas et al., 2007; Escutenaire et al., 2007; Haramoto et al.,

2007; Xagoraraki et al., 2007; Nordgren et al., 2008).

2.4.3 Hybridization probes and microarrays

Dot-hybridization assays have been developed for adenoviruses and enteroviruses
and have been used to detect quantities of target as low as 20 picograms of viral genome
(Kidd et al., 1985; Takiff et al., 1985; Singh-Naz et al., 1988; Rosen et al., 1990; Fong et
al., 2005). It can be useful for screening multiple samples against multiple targets but it is

neither as sensitive nor as specific as PCR or cell culture.

Microarrays were first described in 1995 by Schena et al (1995). They are arrays
of spots on specially prepared glass or silicon surfaces. Each spot on an array serves as a
single test at which either a hybridization of DNA, an immunological attachment or a
chemical reaction can occur. Microarray technology is enabled by the ability to deliver
sub-microliter volumes of material unto an attachment surface or matrix. DNA
microarrays are arrays in which DNA-DNA / DNA-cDNA or DNA-RNA hybridization
reactions are an indication of a positive/negative reaction. Spotted DNA microarrays
consist of either PCR-generated or synthesized oligonucleotides that have been printed or
mechanically spotted on to specially coated glass slides. In situ synthesized arrays have

their probes chemically synthesized directly on to the support matrix.

27

Microarrays have conventionally been used to develop gene expression profiles of
certain targets of interest. Increasingly, research has also focused on adapting microarray
technology to screen clinical specimens against multiple target pathogens in a highly
efficient manner (Zhou, 2003; Bodrossy and Sessitsch, 2004). Microarrays have been
designed for the detection and genotyping of Hepatitis B virus, adenoviruses, Epstein
Barr Virus, Herpes Simplex virus, influenza virus and human papillomavirus (Sengupta
et al., 2003; Boriskin et al., 2004; Korimbocus et al., 2005; Min et al., 2006; Song et al.,

2006)

Proposals have been put forth for using DNA microarrays as an environmental
detection tool and possible biodefense tool (Pannucci et al., 2004; Sergeev et al., 2004).
Only a few examples exist for the application of microarray technology on environmental
samples. For example, Kelly et a1. (2005) have used DNA nricroarrays to analyze the
nitrifying bacterial community in a wastewater treatment plant. Wu et al. (2004)
developed a community genome array that was able to reveal species and strain

differences in microbial community composition in soil, river and marine sediments.

The use of microarrays as an environmental research tool can be divided into two
broad categories. Arrays which serve to detect specific ftmctional genes (e.g toxin
producing genes; pathogenicity islands and catabolic enzymes) sequences irrespective of
which species contains the sequence and phylogenetic arrays which target specific

pathogens or groups of pathogen based on phylogenetic relationships.

28

Straub and Chandler (2003) have proposed that a unified system for the detection
of waterborne pathogens would significantly advance public health and microbiological
water analysis and have indicated that advances in sample collection, on-line sample
processing and purification, and DNA microarray technologies may form the basis of a
universal method to detect known and emerging waterborne pathogens. Table 2.2
summarizes some of the viral microarrays and their applications. Among the viral
microarrays listed in table 2.2 seven array formats were developed for respiratory viruses,
three were for various types of pox and pox-related illnesses (i.e VZV etc), three were for
dengue and other vector-bome transmitted viruses and three were for enterovirus and
other EV diseases (e. g Hand, foot and mouth disease, central nervous system diseases). In
all cases, arrays were targeting viruses extracted from clinical samples. Of special
interest were four arrays described as their developers as being universal or pan-viral
arrays though only one system was actually shown to be able to detect a new unknown

virus (Wang et al., 2003).

A number of commercially available microarray chip platforms are currently
available. Their main differences are the manufacturing method employed and feature
density. Affymetrix GeneChip arrays are able to accommodate up to 1.3 million unique
features on a 5 inch by 5 inch quartz wafer and are manufactured using a
photolithographic masking technique (Pease et al., 1994). Agilent microarrays have a
44000 feature set and are synthesized using an inkjet printing method (Hughes et al.,
2001). Febit’s Geniom microarrays contain only 6 000 features but hybridization can be

carried out with eight chips in parallel allowing multiple sample processing (Guimil et

29

al., 2003). Nimblegen microarrays contain 390 000 probes per array and are
manufactured using a micromirror focusing technique, a technology borrowed from
digital laser projectors in which tiny microscopic mirrors are used to target and focus
beams of light on defined spots on the array, allowing a light dependent chemical
reaction to occur on those selected spots. Less expensive glass slide arrays for smaller
probe sets are also within the in-house fabrication capability of most research institutions
(approximately 16 000 features per slide). Current limitations of the technology include
issues regarding validation and low starting microbial biomass (Wu et al., 2006) and need
to be addressed before the technology may be routinely applied to the analysis of

environmental samples.

30

 

 

 

 

 

 

 

 

 

 

 

 

@833 «582%: mo v.35:
85:63am ammo—28023 Magoo—o8 ESEEmB 293 823 8 8523 _ .5> RESUME
Goon 05 com 352: 333:: one 52:: acumen—Han 33503 < bowoamo cowofiag
9? 8 9326 :96: a Bouncofiov 33mph 2:. 0 one boaflmamoe :oEEoo om 53 oz bowgamom
.mnouoofi: >mO 8 3:86
aoufimooEmE Hog—o 53> maouooofi
was as 3-35 was m 5385 .33? £08er 8.... a308>
890: one .2455 m3? _ x0385 £03250 £8300 £238?on
sea: A>N>o was assessors, eats, @6205 £39 98o
Goon 91.. 88m 36on \EO 3985.86 mo 36on com mo coaaommucoom mono:— £53990
8 Enezmv 3 m: @0323 .5508 2:. 98 cowoowov gougigm owzo o: oo 360% com
Sou—83 3.5805 .mmmocwwao
a 203 3 “co—3 3mg 33333828 mun—ca m .B> Q3083
Goom mo 3:8 039838 5 mam: wcwmou Grammy v "Waggon—mafia owzo ooo.Nm “$5828
:3 Ho :28 E8833 Roam—o Edouoo>oo0 58:: 88803 #02:? <75 183-9“;
@5828
223395 <ZQ
seem .a Bee Em as nausea weed smegma
3 $2on N>Q>m ._>Q>m Sumo mo 385 ov mo aochuaobtmo o5 .838qu mooofi w REE
muo—
aasoam was: o: as an:
68%: Ba 6m .amz .omqmm .m> {moo >N>eoa§o5w manages 9N3 was
8 >oowcomv mega ooaocoooc o Hmcflww Bows—aka wean—=86 8.88 m gamma—95 mo when m Sage—Borg
@on :ouaw: usage 05 8 Boom
GooN BEES mm cogfififlofio £395va \ 853%
is Ho 21: .8893 222.833. REESE wcmoaosoom o5 wcfibnsm ooofi we; < «~5an—
moose—50m
23w
mom ”£65...— uomaz 553554.. no Eon—E. Z momma 2...;

 

 

mnotaom—nnz. #2:. can whatnohom—Z :2; ~.N «Bah.

31

 

.028 25 :85 “0038008

38:52..“

 

 

 

 

 

 

 

 

$883.88.: 08m 0:335 8 <75 8:80
08. E: 523 .2083 5 >00 :5 S .3888 as...
Goon 8825 880200 8m 38m: 2 2:0 0:00 883 E >mm .8 858:8 .0888 0:23
:3 8 So ->mm 05 35 88508820 830 02,—. 0:830: 880298 0:0m 0808A. Mom 2 5 28950825
.8382 05 .8 08—
8m 03::me 0>E898§ 98 .8382
05 8 foe 8m 03:me 03889003.“
£0882 05 .8 $3 8.: b8 25
8280 05 .8382 05 .8 $2 8.:
8888.28 03.80 8280 $33.3:
93 09¢ 82.80 05 “003—02 05
.8 fins .8 0888 :3 8m 839:0 E8 _ 2mm 028:9: 008.8 82808::
Goon :3 8 889 8280 322023 05 “00288 :2>< 23 m .< 028:9: 23— 93 0:00:88
“0:00:30: @808 03: 8.2 3302 8.25800 05. .8 88393. 23 88089 2:800 mm mm 98.2.:
0:23 0:986
Goon :3 8E-nn .8 025080
8 83:5 :28: 088 “053-80: 8:005:82 23 88089 0088 3n 8 .38 80%
83> E 8:002: .8
00.3308 8 83888 05 E 5383
4 082m 880:8: bwcobm 98 22023
Goon :3 .0883. 0882A E 3285088830 8E-mn £28088
8 238023 882 a .8 880E802 880E802 E3 b08006 25> .0088 n02 <75 32> SE

 

 

60:50—80 n.n 03:8

32

 

Ame; Swag Ea .mom§> ofimwaov
£33 0:2 #626 mom—£322.“
26 EB $335530 52:5:

 

 

 

 

 

2953. Mo mombobm =m ma_>ofiwoao§o
A38 93 a 29%. a sea meme: 388-38% Ba .36: .2465 _ 58-8 a6 8358
QOBEEMV W76 8.38 m as 82% 3 2?. 25 was 6386 8&2 €52 8:3 an? .83 m5
.mEo mommommw
<75 92m 05 3 39$ 98 388% 81563 magma 39:3 850 m2: nEo <75
Goom 203 wombeom :38 =m mafiaomoae Bot cozmuaouobmc was muabeom SE 3.3“ ADV/Ev 088%
:3 3 56 g >9: saute .8 SE mo 2:5 as 8%23 £3qu 3 €58 2a sow.
ammo >N> @5888
Goon is 93.23 05 mcofiu waging mo >N> mo .88 commmoaxo m3?
8 385% £26— E mooaogobmv 303m 3 03m 83 050E :ommmoaxo 28w £29me -mh mecca K 568 «:85?
2 .> 33%
$5822
cowoflwm
89:3 bougamom
Goom coca—zeta bowfifimm: .3 coumomtcoE 3853682
in 3 Ed sowofia Dowsing“ 85502.??on _o>o_-fimbm 98 .3609. 83 0 Z x508€<
.moco
woutowoflmnoc: 8 wEwuoEo mo 825m
05 @9586 you 83 :5 meme?
Geom 8505. @ucoE 8 i=0 8: wow: V3350 E mom—d? 8592 9:0
in 6 «5:8 0.8 “m5 monoa 35> uocmmmow 96: 322.30 =a mo coufitowomgo 83 0 Z .83 333:5

 

 

 

 

 

 

3238 3 as:

33

 

mama? 8?on
28 38:8 .magofiwofiorho
“Va 23 2 335.5%
“was Eméaam .5. Ba <0
@3390: 52.5.: K £53092.—

 

 

 

 

 

585: .33 833-3699; mummfiaoa «o

.N 98 _ 25 we? onEE 833 BE,
Agom 9? moans amo.a.=.:>9.~3 50253 2 8m Emma mEo
Ho Exam—emu mouflaflotmv 98 3889 28m mm mZU E? SE
:58 91.. ,- 8:83 9532 9% was a
a 38ng meg m 2:38: 9a a ”$80: Boa 3 2a m 3:30:
mag—260m xmgohomfi
Avoom <29 3 cons—mag 295 358%: >oU-mm<m €320.50 mm<m 8% 888%
:3 Ho weed .8835. 05 53» @93me 8.95% cm mo wcfibocow Ea nosooaom 53083 c .53me

32:3 VA onwaou 05

we; camcon can 0:2 “was? mo 33% Ba Amz waging». 328%.. wag
Goon in B to??? :26 mug—.2802 was $0an— 622 $03 Agv 56% 30:0» 32830:: >380me
EomeoZV 39$ £5322.“ h =a so BEEN, 50253 nflswcummc Ea 830D com £38m w mHEZZNE

 

 

 

 

 

 

355:3 Nd 03:.

34

CHAPTER 3 RESEARCH QUESTIONS, HYPOTHESES, GOALS

AND OBJECTIVES

3.1

Research Questions

Some of the research questions raised during examination of the current state of

environmental virology that form the basis of this thesis are as follows:

3.2

Do viruses represent a potential risk to swimmers in fresh waters such as at Great

Lakes beaches?

How can one determine the array of viruses present in polluted waters?

Can new microarray technology be developed to characterize community viral

infections by monitoring sewage?

Can pathogen detection chip technologies like microarrays be used to determine
which viruses represent the most likely source of risk based on their prevalence in

the environment?

Research Hypotheses and Methodology

The premise of this study is that microarray technology can be adapted and used

to perform high throughput characterizations of various water matrices for the common

waterborne viral pathogens. In the absence of a defined list of target viral pathogen

promulgated by the regulatory authorities or a suitable indicator for viral pathogens,

microarrays are the next best option to screen complex environmental matrices for the

hundreds of potential viruses that may be present. This dissertation will compare the

current methods used to test for viruses in the environment, namely cell culture,

35

polymerase chain reaction (PCR) and integrated cell culture-PCR against the results
obtained using a microarray to detect viruses through labeling and hybridization post cell
culture. We will first characterize the types of viruses present at two typical recreational
beaches in the Great Lakes impacted by human sewage using conventional cell culture
and PCR-based methods. We will also compare the range of viruses present in raw
sewage collected from a typical wastewater treatment plant using both polymerase chain
reaction and the virus microarray and determine whether the virus microarray is able to

detect a wider range of viruses in sewage compared PCR analysis of the same sewage.

3.3 Research Objectives

The objectives of this study were fourfold:

a. To use conventional cell culture and PCR based detection method to demonstrate
the types of viruses that may be detected at recreational beaches that have been

impacted by fecal pollution events.

b. To use freely available sequence information from online gene databases like
GENBANK, to design virus specific probes targeting the known waterborne viral

pathogens

c. To develop a standardized method for the concentration, extraction, labeling and
hybridization of viral nucleic acid, from both DNA and RNA viruses, unto the

viral microarray.

d. To characterize a human wastewater source using high throughput microarray

analysis to screen for the presence of the target viruses of interest and

36

demonstrating that the kinds of viruses detected using a microarray are much

greater than those detectable by conventional methods.

37

CHAPTER 4 EVALUATION OF PUBLIC HEALTH RISKS AT
RECREATIONAL BEACHES IN LAKE MICHIGAN VIA
DETECTION OF ENTERIC VIRUSES AND A HUMAN-SPECIFIC

BACTERIOLOGICAL MARKER

by Mark Wong, Lekha Kumar, Tracie M. Jenkins, Irene Xagoraraki, Phanikumar

Mantha, and Joan B. Rose

Submitted for consideration to Water Research (Elsevier Publications)
4.1 Abstract

In the United States, a total of 3003 days of beach closures and/or advisories were
reported in the 2006 swimming season. Out of the 1445 beaches monitored, 336 were
associated with high risk. In this study, water samples were collected along two Lake
Michigan Beaches over the 2004 swimming season and examined for the presence of
enteric viruses by cell culture and ICC-PCR. In addition, samples were also examined for
the presence of a human sewage marker based on the Enterococcal Surface Protein
(ESP). Results demonstrate that both beaches were impacted by human fecal pollution.
Viruses were detected in 16/30 Silver Beach samples and 9/28 Washington Park samples,
while the human specific ESP marker from cultivatable enterococci was detected in 21
out of 529 samples analyzed. The occurrence of bacterial-derived and viral indicators of
human fecal pollution could not be correlated with each other. Predictive models of virus
pollution were developed for both beaches utilizing physical parameters like wind speed,
wind direction and water temperature. The best models for both beaches were determined

with statistical support. Conclusion: Lake Michigan recreational beaches are being

38

adversely impacted by human fecal pollution. Monitoring for the traditional indicators of
water quality does not address viral risks. Predictive models can be developed and
potentially used as real time water quality forecasting tools. The risk of acquiring a viral
infection at either of the beaches under study was determined to range from 0.2 to

2.4/1000 swimmers.

Key words: water quality; monitoring; predictive models; waterborne pathogens; fecal

pollutants; indicators

4.2 Introduction

The Great Lakes are one of the world’s largest bodies of fieshwater. The volume
of the Great Lakes is estimated at 6 quadrillion gallons and constitutes 95% of the United
State’s freshwater supply (GLIN, 2006). More importantly, the Great Lakes basin is
home to more than 30 million people who use the water for drinking, recreation and
industry. There are more than 1000 Great Lakes beaches along 5500 miles of shoreline
located within the 8 US states and 2 provinces of Canada (Dorfman, 2006). The most
recent National data in 2006, reported that the Great Lakes beaches exceeded standards
14% of the time and had the greatest exceedances compared to all other shorelines in the
US (NRDC, 2007). There were 3003 days of beach closures and/or advisories for the
2006 swimming season and of the 1445 beaches monitored, 336 were identified as high
risk associated with bacterial contamination and rain/runoff/stormwater however the

greatest risks were associated with unknown sources.

39

It has been estimated that the twenty major cities located within the Great Lakes
basin discharge in excess of 92 billion liters of raw sewage into the Great Lakes annually
(MacDonald, 2006). The extent of impact to human health from the discharge of sewage
into the Great Lakes has not been determined. A study in Southern California, a region
that also experiences human health impacts from sewage discharged into recreational
water bodies has estimated that there is between 627,800 and 1,479,200 excess
gastrointestinal illnesses occurring at beaches in Los Angeles and Orange Counties each
year and that this causes an estimated US$21 to $51 million worth of economic loss
(Given et al., 2006). In comparison, Rabinovic et al. (Rabinovici et al., 2004) performed a
study of the economic impact of beach closures on a Lake Michigan freshwater beach
and found that a typical closure causes a net economic loss among would-be swimmers
totaling $1274 — $37030 per day and between $111 088 and $518 415 over a 4 year

period between 1998 to 2001 while avoiding 42% of the predicted illness.

The current US EPA recreational water guidelines are based upon the
enumeration of enterococci and Escherichia coli (EPA, 1986). The guideline for
freshwater is a geometric mean of at least 5 samples per month not exceeding 126 colony
forming units per 100 ml (CFU/100ml) of E. coli or 33 CFU/100ml of enterococci. In
addition, no sample should exceed a one-sided confidence limit (CL) of 75% CL for
designated bathing beaches, 82% CL for moderately used beaches, 90% CL for lightly
used and 95% CL for infrequently used beaches. Three of the biggest drawbacks of the
current indicator methods as used for beach monitoring are the need to incubate the

samples overnight, the inability to attribute the source of pollution and the lack of

40

correlation with pathogens of concern particularly viruses. Previous work has shown that
there is a poor correlation between levels of viruses and bacterial indicators (J iang et al.,
2001; Noble, 2001). This is due to their different levels of excretion from the human
body and their different rates of inactivation upon exposure to the environment (Tree et
al., 2003). Examination of recreational water for the presence of viruses is currently not
mandated by the ambient water quality criteria due primarily to the length of time needed
to obtain results and also the cost of analysis. Also conventional cell culture-based
methods do not provide the identity of the virus. These issues have been partly solved by
the development of numerous PCR primers to allow the identification of the infectious

viruses present in a sample using cell-culture-PCR methods (Chapron et al., 2000).

The polymerase chain reaction (PCR) has also been proposed as a means to
rapidly and specifically detect microorganisms used in source tracking. A set of PCR
primers has been reported by Scott et al. (Scott et al., 2005) to detect the presence of
human enterococci using a specific marker for the enterococcus surface protein (esp)
gene. Other library independent source tracking methods have been reported for
Bacteroides spp. (Bernhard and Field, 2000) and E. coli (Ram et al., 2004). While the
identification of human wastewater / sewage signatures aids in understanding the
potential risk to swimmers, a report by the National Research Council has suggested that
pathogen monitoring is needed to better define the probability of infection. In addition
PCR is limited in regard to risk as it can not distinguish between viable and non viable

microbial particles.

41

Many investigations into recreational outbreaks in freshwater systems fail to
identify the etiological agents responsible, but frequently suggest that noroviruses,
Cryptosporidium and E. coli might be causative agents. Surveillance of recreational
waterborne disease in the United States from 1997 to 2004 showed that of the 94
recreational outbreaks recorded for untreated water, 28.7% were of an unknown etiology
responsible for acute gasteroenteritis infections while E. coli spp. (24.5%), Norovirus
(18.1%), Cryptosporidium spp. (10.6%), Giardia spp. (5.3%), and Shigella spp. (9.6%)
were the most commonly identified microorganisms (Barwick et al., 2000; Lee et al.,
2002; Yoder et al., 2004; Dziuban et al., 2006). Besides outbreaks, unknown
gastroenteritis and generic diarrhea are the most common reported health outcomes in
recreational epidemiological studies (Wade et al., 2006; Colford et al., 2007), and many

believe these are associated with a variety of enteric viruses.

Wastewater has been suspected as the primary source of viral contamination of a
wide range of ambient waters throughout the world. Patti et al. (Patti et al., 2003)
evaluated 196 samples from various sources of water in Italy, detecting Coxsackie B
virus in 35 and enteric- non-entero- viruses in 51 out of 196 samples screened. In Korea,
Japan and the Netherlands untreated or inadequately treated sewage was the source of
noroviruses and enteroviruses. Katayama et al. (Katayama et al., 2004) demonstrated the
presence of enteric viruses (norovirus G1, G2, enteroviruses) in coastal seawater
downstream of a Japanese wastewater treatment plant and up to 4 days following a
combined sewer overflow event. A study in Korea by Lee et al. (Lee et al., 2004) found

that thirty (75.0%) of the 40 samples collected from 4 rivers in a Korean province were

42

positive for enteric viruses based on cell culture. Loder and de Roda Husman (Lodder
and de Roda Husman, 2005) were able to quantify a 2 to 100 fold dilution of norovirus
and rotavirus in river water in the Netherlands using direct PCR impacted by wastewater.
Both in Florida and California, adenoviruses were detected more often than enteroviruses
in waters influenced by non-point sources such as septic tanks and stormwaters with
acceptable levels of microbial water quality indicators (Wetz et al., 2004). An extensive
survey of southern California urban waters by Jiang and Chu (Jiang and Chu, 2004)
detected three types of human viruses (adenoviruses, enteroviruses and hepatitis A
viruses) using nested- and RT-PCR from 11 rivers and creeks. Approximately 50% of the
sites were positive for human adenoviruses and there was no clear relationship between
detection of human pathogenic viruses and the concentration of indicator bacteria and

coliphage.

Several recent epidemiological studies have examined new indicator systems for
recreational water quality and their relationship to the health of swimmers. One of the
larger studies took place in Mission Bay, California (Colford et al., 2007). Diarrhea and
skin rash were related only to coliphage levels and traditional bacterial indicators were
not adequate to address risk. The California study evaluated non-point source and storm
water as the source of the pollution. In another study conducted in the Great Lakes, Lake
Michigan and Lake Erie by Wade et al. determined that enterococci bacterial levels could
be significantly related to gastrointestinal illnesses (Wade et al., 2003). In this case
sewage was one of the likely sources. Time in the water was also associated with

increasing illness rates as well.

43

The study described here addresses the public health risk associated with the
degree of human fecal viral pollution at two Lake Michigan recreational beaches. Water
samples were examined for the presence of viable enteric viruses through cell culture
and integrated cell culture polymerase chain reaction (ICC PCR) and the application of a
new human-specific enterococci bacterial marker was evaluated for its usefulness in
addressing the presence of viral pathogens (Scott et al., 2005). This study was
undertaken during the same time period that the epidemiological investigations were
being implemented at Lake Michigan by the EPA for the National Epidemiological and
Environmental Assessment of Recreational Water Study (NEEAR).
(http://www.epa.gov/NEEAR/) , thus the results of this work will assist in meeting the
goals of the BEACH Act in providing new assessment of pathogens, source tracking
markers and risk assessment approaches for comparison to the traditional epidemiological

investigations.

4.3 Materials and Methods

4.3.1 Area of study

Samples were collected from two public beaches located on Lake Michigan.
Washington Park beach is located at 115 Lakeshore Drive, Michigan City, Indiana. The
public beach measures approximately 3,500 ft. Sampling was carried out at a stretch of
approximately 1,200 ft which experienced the most foot traffic (Figure 4.1). Three
sampling points were chosen, approximately 400 feet apart. Silver Beach is located at St

Joseph, Michigan. The public beach measures approximately 2000 feet. Sampling was

44

canied out at a stretch of approximately 900 ft which had the most foot traffic (Figure
4.2). Physical data for wind speed, wind direction fi'om the Station MCYI3 buoy located
at Michigan City, Indiana (41°43'45" N 86°55'41" W) was downloaded from “Michigan
City Meteorological Data” located at
http://www.gler1.noaa.gov/metdata/mcy/archive/mcy2003.07t.txt.gz (accessed on April
10, 2007) which is owned and maintained by the Great Lakes Environmental Research
Laboratory. These data were used for both Washington Park and Silver Beach Park as it
was the closest buoy. Temperature, pH and the number of people using the beach were
recorded at the time of sample collection using a digital thermometer and pH meter.
Turbidity data was recorded upon return to the laboratory by analyzing an aliquot of the

sample on an Orion AQ4500 Portable Turbidity meter (Thermo Scientific Inc).

4.3.2 Sample Collection

Samples were collected from two public beaches along the shores of Lake
Michigan (Figure 4.1). The sampling was performed in parallel with the EPA’s National
Epidemiological and Environmental Assessment of Recreational Water Study (NEEAR).
In this study, samples for viruses and enterococci were collected every 2 weeks over ten
weekends between July 3 and September 6, 2004. For virus analysis, a total of 58
samples were collected, each representing a composite of 3 spatial locations. At each
spatial location, between 80 and 120 liters of lake water was first pumped into a sterile 20
gallon container, the sample’s pH was lowered using 5N hydrochloric acid to produce an
approximately neutral pH of between 7.0 to 7.5, the sample was then filtered through a

lMDS filer (Cuno, Inc) and the process repeated to give a sample volume total of

45

between 250 and 350 liters. All viral samples were held on ice, transported to the Water
Quality and Health Laboratories at Michigan State University, and eluted within 72
hours. A total of 524 samples were analyzed for enterococci. All of the enterococci
samples were processed through EPA contractors. Plates with membrane filters and

colonies were shipped to MSU and kept at 4°C until processed.

Virus elution and concentration was carried out by the organic flocculation method as
described by the US EPA information collection rule (EPA, 1995). Viruses were
desorbed from the filters by two rounds of reverse passage of IL of 1.5% beef extract
solution (1.5% w/v Beef Extract, 0.05M glycine, pH 9.0 — 9.5). Viruses were flocculated
by addition of ferric (III) chloride to a final concentration of 2.5mM and by lowering the
pH of the solution to 3.5 (Payment et al., 1984). Viruses were pelleted by centrifiigation
at 2,500>< g for 15 min and dissolved in 30ml of 0.15 molar sodium phosphate (final pH
9.0). Viruses were purified by centrifiigation at 10,000>< g for 10 min, brought to a neutral

pH, supplemented with 100 units of Penicillin, 100 microgram of Streptomycin and 0.25

microgram of Fungizone and stored in aliquots at —80 °C until analysis.

46

 

 

 

 

rm

 

Figure 4.1. Overhead satellite image of the Washington Park sampling location. Stars
indicate the sampling points.

 

 

 

 

Figure 4.2. Overhead satellite image of the Silver Beach sampling location. Stars indicate
the sampling points.

47

4.3.3 Cell culture

Viruses were cultured on the African Green Monkey Kidney (BGM) animal cell
line using the total culturable virus assay method described in the information collection
rule (EPA, 1995) with minor modifications. Cells were grown in flasks until at least 70%
- 90% confluence was obtained. Virus concentrates were added to the flasks and allowed
2 hours contact with the cells with occasional rocking to ensure complete contact with the
cells and to avoid drying of the cells. Concentrates were decanted and discarded and the
cells were washed with Dulbecco’s Phosphate Buffered Saline. Cells were maintained
with Minimum Essential Media supplemented with L-glutarnine, Earle’s Salts and 2%
Fetal Bovine Serum. Flasks were monitored for up to 14 days for development of
Cytopathic Effects (CPE) indicative of a viral infection. Presence or absence of CPE was

confirmed by performing a secondary passage for each flask.

4.3.4 PCR and RT PCR for viruses

Viral nucleic acid was extracted using the QIAgen viral mRNA mini kit (Qiagen;
Valencia, CA) following the manufacturer’s instructions. Primers developed to screen for
Adenoviruses, Enteroviruses, and Rotaviruses were used to perform PCR and RT PCR on

the cell culture supernatant. The list of primers is given in Table 4.1. For PCR reactions,

thermocycler settings were as follows: 95 °C, 15 minute initial denaturation to activate
the Hotstart Taq polymerase (Qiagen; Valencia, CA), 95°C, 0.5 minute denaturation,

57°C, 0.5 minute annealing and 72°C, 0.5 minute elongation for 35 cycles followed by a

48

final elongation step of 72°C for 5 minutes. The PCR mix consisted of 1 unit of Hotstart
Taq polymerase, 1.5 millimolar MgClz, 1)( PCR buffer, 1X Q solution, 0.5 micromolar of

each primer, 0.5 millimolar of each dNTP. For the RT-PCR reactions, thermocycler

settings were as follows: 50°C, 30 minute first strand synthesis, 95 °C, 15 minute initial
denaturation to activate the Hotstart Taq polymerase (Qiagen; Valencia, CA), 95°C, 0.5
minute denaturation, 57 °C, 0.5 minute annealing and 72°C, 0.5 minute elongation for 35

cycles followed by a final elongation step of 72°C for 5 minutes. The reverse
transcription PCR mixture consisted of 2 microliters of OneStep RT-PCR enzyme mix,

1.5 millimolar MgClz, 1>< PCR buffer, lx Q solution, 0.5 micromolar of each primer, 0.5

millimolar of each dNTP. PCR products were separated on a 1.5% agarose gel stained

with GelStar nucleic acid stain (BioWhittaker) and viewed under UV light.

49

I

 

 

 

 

 

 

 

 

 

 

 

 

A33 :3 Ho UU<<H<E<UU<OGUE< 5%
“22825 _ _N U<H<<O<UE<<UU<UUOE 95m mum 3.563%
UHOOUCUPUUH<EOOUOU<HO<OU<HOE mc3<<xon
m3 OHUUO<UEU<HOU<O<OUU<UUO 33in: Hum @030:
£58 :3 HO<<<UHOH<OGUOUUOU<U©<U Sago:
we coin—UV Sm UH<U<UOH<U<EUHDOHO<UOUUO mam—go: MUAEE mar—30:03.,
<h0<<<UUU<O<OOU<U<<<O mm
mm. <HUOOUOH<<OHUUUUOQUUH 5 Eva @332
Soon in HOHUUOU<HOEUU<PGOU do;
8 coins—UV 3m <UUO<UO<<H<UU<UHOPH< .5— MUAEB mageofim
gov. v88: co
3:20QO 33 fiwfifi cocoa—com 5.5.5 mun—FM 3.: >

 

 

 

 

 

 

 

82:? no 55980.. «5 .5.“ 3:5» 25 5 can: 9.23.:— 10E .5— .3“ 93:.

50

4.3.5 Extraction of Enterococci

The membrane filters and plates (mEI agar) that were used in determining the
levels of Enterococci during the EPA’s National Epidemiological and Environmental
Assessment of Recreational Water Study on Lake Michigan were shipped to Michigan
State University. The filters were initially extracted off the cellulose filter by lifting the
membrane, suspending the membrane in tryptic soy broth (Difco), vortexing vigorously,

and incubating the suspension for 2 hours at 41°C to wash the bacteria from the filters

and partially enrich the culture. Following the incubation, the suspension was used as the
environmental sample from which total DNA was extracted using a QlAamp DNA

extraction kit according to manufacturer’s instructions (Qiagen; Valencia, CA).

4.3.6 PCR Conditions for esp Marker

Primers specific for the esp gene in Enterococcus faecium have been previously
developed and examined for specificity to human fecal pollution (Scott et al., 2005). The
forward primer, which is specific for the E. faecium esp gene is: (5’-TAT GAA AGC
AAC AGC ACA AGT T-3’). A conserved reverse primer (5’-ACG TCG AAA GTT

CGA TTT CC-3’) was used for all reactions.

PCR reactions were performed in a 50 microliter reaction mixture containing 1X

PCR buffer, 1.5 millimolar MgClz, 200 micromolar of each of the four

deoxyribonucleotides, 0.3 micromolar of each primer, 2.5 U of HotStarTaq DNA

polymerase (Qiagen; Valencia, CA), and 5 microliters of template DNA. Amplification

51

was performed with an initial step at 95°C for 15 minutes (to activate Taq polymerase),

followed by 35 cycles of 94°C for l min, 58°C for 1 min, and 72°C for 1 min. PCR

products were separated on a 1.5% agarose gel stained with GelStar nucleic acid stain
(BioWhittaker) and viewed under UV light. The polymerase chain reaction product is

680 base pairs in length.

4.3.7 Statistical Analysis

Multivariate analysis was performed to determine if there was any correlation
between the observed data and the following predictors - wind speed, wind direction,
water temperature, number of swimmers and turbidity recorded. The multiple regression

model used had the following form:

y = 160 + fl1x1+ flzxz + :63x1x2 +° ° ' + :ann + 8 (1)
where the response y (most probable number / 100L) is represented as a combination of a

constant, linear and interactive terms of the predictors x1, 352 a' ‘ ' x" and an error term

5 . Given a set of n observations (35] 9 y, )3 (x2 9 yz ) ' ° ' , equation (1) can be written in a

matrix form whose solution gives the coefficients:
T '1 T

where [I and y are the column vectors of coefficients and the responses respectively. The

design matrix X was based on equation (1) and contained in its columns the model terms

evaluated at the predictors. One of the objectives of our analysis was to identify models

52

with the least number of parameters ( ,6 ’s) and the maximum explanatory power. We

evaluated a large number of models and to aid model selection based on the principle of

parsimony, we used the Akaike Information Criterion (AIC) defined as:

where L denotes the maximum likelihood estimation and k is the number of parameters
(Akaike, 1974). Since the AIC penalizes likelihood with respect to the number of
parameters, the idea of model selection is to select the model with the lowest AIC value.
All computations were performed using MATLAB (Mathworks Inc., Natick, MA) and

SYSTAT (Systat Software Inc., San Jose, CA).

4.4 Results

4.4.1 Viral results

A total of 30 and 28 lake samples were collected from Silver Beach and
Washington Park respectively. Cultivatible viruses were detected by cell culture as
evidence by cell death in 16 of the 30 (53%) water samples collected from Silver beach in
St Joseph, Michigan and 9 of the 28 (45%) water samples collected from Washington
Park beach, Michigan City, Indiana. Data for percentage positive, geometric mean,
minimum-maximum MPN/100L, percentage positive morning and percentage positive

aftemoon data are summarized in Table 4.2.

Most probable number estimation of viruses ranged from <0.6 MPN / 100L to

4.33 MPN/ 100L at Silver Beach with an average of <1 .21 MPN/100L and between <05

53

MPN/100L and 5.70 MPN/100L at Washington Park Beach with an average of <1.30
MPN/100L. The limits of detection were used for non-detects in calculating the mean.
Cell culture PCR or RT-PCR showed the presence of adenoviruses (53.3%),
enteroviruses (3.3%) and rotaviruses (10%) at Silver Beach. For Washington Park beach,
39.3% and 7.1% of the samples were adenoviruses and enteroviruses positive through

RT-PCR analysis of the cell culture supernatant.

4.4.2 Sequencing Confirmation

The presence of adenoviruses in the sample was confirmed through nucleotide
sequencing. In total 5 samples (3 Washington Park samples and 2 Silver Beach samples)
positive for adenovirus were selected at random and sequenced using the primers
hexAA1893 and hexAA1905 (Table 4.1). Sequencing was carried out at the Michigan
State University Research Technology Support Facility. Two different sequences were
obtained from the 5 samples with a single nucleotide mismatch between them and

showing approximately 99% sequence homology with Adenovirus type 40 viruses.

54

Table 4.2. Viral Water

uality at Lake Michigan Beaches

 

Washington Park Silver Beach

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Number of samples 28 30
CELL CULTURE

percentage positive for 32 58
any virus by cell culture

Geometric mean * 0.85 1.0
MPN/100L

Min-Max <0.5-5.7 <0.6-4.33
MPN/ 100L

percentage positive 36 60
morning samples

percentage positive 29 47
afternoon samples

PCR

Percentage positive by 43 60
PCR

Adenovirus percentage 39 49
positive

Enterovirus percentage 7 3
positive

Rotavirus percentage 0 9
positive

ESP percentage positive 32 15

 

 

 

 

 

* non-detects used at limit of detection for mean

55

F._

4.4.3 Enterococcus esp gene PCR results

A total of 294 out of 414 and 235 out of 414 enterococcus membrane filters were
analyzed for the presence of the ESP gene sequence from Silver Beach and Washington
Park respectively. While virus samples were collected twice a day at three near shore
sampling points, Enterococcus samples were collected thrice a day at 6 sampling points,
3 corresponding to the virus sampling points and 3 in further out away from shore. Eight
Silver Beach (32%) and thirteen Washington Park samples tested positive for the esp
gene in at least two separate PCR reactions. Analysis of the fi'equency of detection of the
esp gene was not found to correlate significantly with the presence of either PCR-
detectable or culturable viruses. A Chi-square comparison of ESP PCR results and viral
PCR results generated a Chi value of 1.89 with 1 degree of freedom resulting in a failure
to reject the null hypothesis that the ESP and viral PCR results are independent at the

95% confidence level.

4.4.4 Virus PCR and Cell Culture results correlation

Most probable number estimations of virus using cell culture results were
compared with virus PCR results to determine if a positive cell culture result as evidence
by cell death was statistically more likely to give rise to a positive virus PCR result. Chi
square analysis of the combined virus data for both beaches generated a Chi square value
of 4.66 with 1 degree of freedom. Therefore it was concluded that at the 95% confidence
level that a positive cell culture result was statistically more likely to correlate with a

positive virus PCR result. A similar analysis of Silver Beach samples at the 95%

56

confidence level likewise showed that a positive cell culture result was statistically more
likely to correlate with a positive virus PCR result. Results were inconclusive for
Washington Park data however as the expected value obtained was less than five in some
cases, which violated the conditions for performing a reliable Chi square test. From the
statistical tests, we can conclude that a positive PCR result correlates with a positive cell

culture result and vice versa.

4.4.5 Physical Indicators

Average pH, water temperature, wind speed, wind direction, turbidity and number
of swimmers recorded for both beaches are provided in table 4.3. At Silver Beach, which
is located approximately 40 miles northwest of Washington Park beach, the average
turbidity and number of swimmers was found to be greater than at Washington Park
Beach. The pH, water temperature and windspeed were found to be similar at the two
beaches. The prevailing wind direction for Washington Park beach was found to be
directed mostly towards the shore while the prevailing wind direction for Silver Beach

was generally directed along the shoreline.

57

Table 4.3. Average pH, water temperature, wind speed, wind direction, turbidity
and number of swimmers recorded for Washington Park and Silver Beach samples.
Range is given below in parentheses

 

 

 

 

Average Wind Number

Sample Average . . Water Temp/ Windspeed direction
. turbidity o o _ of
Location pH (NTU) C (m/s) (0 — true swimmers
north)

Washington €2,420 2.71 22.1 (264;; 93.9 17.75
Park 8.72) (05,139) (20, 25) 3.82) (195, 250) (o, 75)
Silver 2,4267 3.45 21.6 (263670 49.6 33%)
Beach 8.67) (0.3, 16) (18.1, 24.5) 4.78) (194, 165)

 

 

 

 

 

 

 

4.4.6 Modeling Virus Contamination at the Beaches
Wind direction is an important variable for beaches situated near river outfalls or
storm drains (Nevers and Whitman, 2005; Liu et al., 2006). However, the influence of

wind on beach water quality can be complex and can vary depending on the orientation

of the beach ( BBeach ) relative to the geographical north and the prevailing wind vector.

For example, if the wind direction is from the north to the south (onshore) and is
perpendicular to the orientation of the beach (East to West), then the river plume can be
forced onshore and travel to the nearby beaches along the shoreline. The opposite is true
for southern winds which may push the river plume out into the lake causing dilution.
Wind directions are generally reported relative to the north (e.g., 0° = wind coming from

the north, 90° = wind coming from the east etc.).

An accurate assessment of the influence of wind direction for different beaches

requires that the effect of the wind direction be evaluated relative to the orientation of the

58

 

beach (Nevers and Whitman, 2005). Therefore, instead of using the wind direction
(angle) as an independent variable in our models, we used a new variable (wind direction
code) that took one of three possible values depending on the action of the wind in the
near-shore region as shown in Figure 4.3 (i.e., onshore winds , offshore winds , and

alongshore winds aiding or opposing the prevailing long-shore current component).

59

w
A

315° 45°
onshore onshore

   
    

  

alongshore

 

O alongshore

alongshore alongshore

offshore offshore
225° 135°

180°

 

Figure 4.3. The effect of wind direction in the near-shore region depends on the
orientation of the beach. Here the beach is oriented in the East-West direction (03 =
90 °C) and so a Northerly wind would blow in the onshore direction. In a beach
oriented in a North-South direction, the same Northerly wind would be blowing in

an alongshore direction. When the beach angle OB changes, then the angles
corresponding to the onshore, offshore and alongshore winds will also change. For

Silver Beach and Washington Park, 9;; = 205°, 72° respectively.

60

Multiple regression results (R2 and adjusted R2 values) indicated that the new

variable (Beach orientation, 639M), ) significantly improved the ability of the model to

explain the observations at the two beaches. 63,300), values for Silver Beach and

Washington Park are 2050 and 720 respectively (measured from the geographical north,

positive clockwise). For Silver Beach, wind vectors between 250 - 340o correspond to

onshore winds, offshore winds occur between 70 and 160 ° and alongshore winds for the
other directions (341 - 69 ° and 161-249°). Values of the wind direction code variable
used in multiple regressions were 1.0 for onshore winds, 0.0 for offshore winds and 0.5
for alongshore winds. An important consideration was to ensure that the variables

included in the model are not redundant or collinear. Collinearity problems were
identified (and resolved) by examining the eigenvalues of the (XTX) matrix in equation

(2). Models for both beaches were found to have condition indices significantly below 15

indicating that collinearity was not an issue (Belsley et al., 1980).

Water temperature was an important predictor; however missing values in the
dataset make it difficult to compute the correlations with confidence. The Bartlett chi-
square test which tests a global hypothesis about the correlations was significant (p <
0.03) for Silver Beach which indicated that there may be some real correlations among
the variables. For analysis with missing data, we used the EM (Expectation
Maximization) algorithm to compute the correlations. For Silver Beach, water

temperature significantly (and negatively) correlated with the virus data (Most Probable

61

Number, MPN) followed by turbidity, wind speed and wind direction. The number of
swimmers correlated least with the viral data. A scatter plot matrix of the associations
between the variables is shown in Figure 4.4 for both beaches with normalized density
plotted along the diagonal of the matrix. The standard deviations of the variables

determine the major axes of the confidence ellipses in Figure 4.4.

For Washington Park, turbidity showed the strongest association with MPN and wind
direction and turbidity were related, however the Bartlett chi-square test failed due to the
small sample size. Pearson correlations based on Bonferroni-adjusted probabilities
(which provide protection for multiple tests) showed that wind direction and turbidity
were related (p < 0.001) for Washington Park. Durbin-Watson statistic for all models

showed that there was no evidence of autocorrelation. The best models for Silver Beach

(R2 =0.92, AIC = -525) and Washington Park (R2 ==0.99, AIC = -17.86) are given below.

Silver Beach:

loglo y = 2.760 - 0.117*(Water Temperature) - 0.019*(Turbidity)*(WindDirCode) (4)
Washington Park:

logm y = 6.208 - 0.303*(Water Temperature) + 0.369*(Turbidity)*(WindDirCode) (5)

62

 

 

 

 

 

 

 

 

 

 

 

 

z
a.
2
BeachCode:
%‘ xx X 1(SilverBeach)
g , OZtWashingtonPark)
9.’
L. a
3 E
’2 .
’X—‘\
2i M
a, ‘* ’
A»
g e e
203% «xexx ‘9 x XXCO x
§ 28 [I
a X,” m." 00
X X x X

 

x x
K xx x x
A .
J. x

x
x x XX... m X e k

0
K
0
ed

 

 

 

 

 

 

 

Number of
Swimmers

Wind ‘
MPN Turbidity “'3‘“ WM Direction WW“
Temperature Speed Code Swammers

 

Figure 4.4. Scatter plot matrix showing the association between the variables for the
two beaches. A narrow ellipse indicates stronger correlation. The orientation and
major axes of the confidence ellipses depend on the standard deviation of the
variables.

63

Computed t-values for all model parameters were significantly greater than 2.0
which was an indication that the independent variables were selected without a high
degree of correlation among them (important for estimating the regression coefficients

with confidence).

Models that best describe the observed variability at the two beaches (equations 4
and 5) were selected based on the principle of parsimony (AIC) and were found to
contain an interaction term between turbidity and wind direction. Turbidity in the near-
shore region is often an indicator of high suspended solids which may transport
(depending on the wind direction) a variety of biological agents including viruses and
bacteria. A comparison of the observed data and the model predictions (Figure 4.5)
shows that a good agreement was obtained. The high R2 values (0.89 and 0.98) indicate
the goodness of the fit; however, these numbers can be expected to decrease as the

sample size increases — typical R2 values for beach models in the Great Lakes region

range from 0.4 to 0.7 (Francy et al., 2003; Olyphant and Whitman, 2004). Although the
present analysis was somewhat limited due to the small sample size of the data and the
missing values, we were able to extract the important predictors and obtain a relation
between them. The models were able to explain a majority of the observed variability at

the two beaches.

64

f-_

 

l l l J T l l l l
0 Silver Beach

4_5__ 0 Washington Park _

3.5 - _

2.5 '— . ”—1

Predicted (MPN/100 L)

1.5— _

0.5 W -«

 

 

O l l l 1 l l l
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Observed (MPN/100 L)

)-

 

Figure 4.5. Observed and simulated viral data plotted on a 1:1 line for Silver Beach
and Washington Park.

65

4.4.7 Risk Estimation

Using the virus MPN data and an exponential probability of infection model [Pi =
l-e’rN] with a r value of 0.4172 (Crabtree et al., 1997) the daily probability of infection

for an assumed dose of 100ml was determined. The daily risk of infection for
Washington Park ranged from 0.24 per 1000 swimmers to 2.4 per 1000 swimmers.
Similarly for Silver Beach, the probability of infection was determined to range from
0.26 per 1000 swimmers and 1.8 per 1000 swimmers when culturable viruses were

detected.

4.5 Discussion

4.5.1 Overall Discussion of Results

Two beaches on Lake Michigan were assessed over a summer season for human
fecal pollution and potential public health risks using two cultivation methods, one for
viruses and one for a genetic marker (esp) detected within culturable Enterococci which
indicates human sewage. The detection of cultivatable viruses and the detection of the
esp gene marker indicated that both of these beaches have been impacted by human fecal

pollution.

The lack of association between the esp gene marker and virus PCR results
highlight the continued problems of using indicator systems to address pathogens.
However, while esp and the virus data seem to give contradictory indications as to which

beach was most heavily impacted by human fecal pollution, it is interesting to note that

66

probability of infection estimates based on the virus MPN values do seem to agree with
the esp gene data in indicating that Washington Park poses a slightly higher risk for

swimming.

The increased presence of the esp at Washington park (32% positive) was also
reflected by the presence of enteroviruses (7% positive), compared to 15% and 3%
positive, respectively at Silver Beach. However, adenoviruses were much more prevalent

and showed an opposite trend. Previous research has reported an inactivation rate
constant, k, for Enterococci of approximately 1.5 d'1 for concentrations of less than 50

colony forming units (CFU)/ 100 milliliters while an inactivation rate constant, k, of less
than 0.5 d'1 was observed for concentrations above 50 CFU / 100 milliliters (Liu et al.,
2006). No comparative data on sunlight inactivation for adenoviruses in water are
available. However, in UV-inactivation laboratory studies, adenoviruses were
demonstrated to be more resistant compared to coliphages, feline caliciviruses and
enteroviruses (Thurston-Enriquez et al., 2003). Inactivation rate constants for adenovirus
were reported to range between 0.018 to 0.040, feline and bovine caliciviruses were
reported to range between 0.106 and 0.293, enteroviruses were reported to range between
0.119 to 0.181, coliphages were reported to range between 0.055 to 0.396 and
Enterococci were reported to have an inactivation rate of 0.312 (Hijnen et al., 2006).
Thus a valid hypothesis is that the Enterococci and enteroviruses are dying off much

more quickly than the adenoviruses.

67

g

4.5.2 Modeling of Pathogens and Indicators

Previous attempts to model Lake Michigan beaches to relate fecal indicator levels
with physical parameters have all enjoyed a measure of success. Nevers and Whitman
showed that their regression model, utilizing parameters like wind direction, wave height,
turbidity, lake chlorophyll, was able to predict E. coli levels and forecast closures more
accurately than the current monitoring scheme (N evers and Whitman, 2005). Olyphant
and Whitman evaluated a number of different physical parameters and finally determined
that wind direction, wind speed, rainfall, insolation, lake stage, water temperature and
turbidity was able to accurately predict the geometric mean E. coli concentration in the
swimming zone of a Lake Michigan beach (Olyphant and Whitman, 2004). Their model
accounted for 71% of the observed variability in the log E. coli concentration. More
importantly, their model was able to accurately predict openings versus closings 88% of
the time. In our study, we showed that predictive modeling could be used to estimate the
levels of actual pathogens. The Most Probable Number (MPN) values for viruses at
Silver Beach was found to be dependent on water temperature followed by turbidity,
wind speed and wind direction respectively. For Washington Park, turbidity showed the

strongest association with MPN then wind speed and then wind direction.

4.5.3 Significance of Detection Methods

The cru'rent study was based on cultivation followed by PCR for both viruses and
Enterococci esp. Direct PCR can potentially improve the speed by which analyses is
undertaken but may affect accuracy as well as fail to address issues of infectivity.

Quantitative methods can begin to address the level of contamination. Realtime PCR

68

analysis of the samples collected in this study for the presence of adenoviruses by
Xagoraraki et al (Xagoraraki et al., 2007) showed that Silver Beach was also more
contaminated with 60% (35 samples) of the adenovirus cultivation-PCR analyses in
agreement with the real-time adenovirus PCR results. In order to choose the best method
one must address a number of issues. Firstly, cell culture typically examines a much
larger volume of analyte compared to PCR, in some cases leading to enhanced
sensitivity, however PCR is able to detect targets at much lower concentrations compared
to cell culture. Secondly, cell culture is able to address the issue of viability which PCR
lacks and is an important factor in assessing human health impacts. Thirdly, PCR and
related methods are able to better assess a wider range of targets compared to cell culture
which are generally restricted in terms of targets. Lastly, in situations where the time-to-

result is crucial, PCR and PCR-related methods are much faster than cultivation methods.

4.5.4 Adenovirus Presence

The detection of adenovirus gene sequences in this study indicates that there
exists a risk of adenovirus infection while swimming and wading. Waterbome outbreaks
caused by or associated with human adenoviruses have been documented, but mostly in
recreational swimming pools (Turner et al., 1987; Papapetropoulou and Vantarakis, 1998;
van Heerden et al., 2005). Even though adenoviruses are included in the US EPA’s
candidate contaminant list, few studies have looked into the occurrence of human
adenovirus in freshwater recreational beaches. This study indicates that adenoviruses
were the most frequently detected enteric viruses among the different virus types

screened and that adenoviruses might make a good indicator of viral fecal pollution. A

69

more comprehensive survey of multiple freshwater locations needs to be canied out to

determine if adenoviruses are similarly prevalent at other recreational locations.

4.5.5 Future Research Needs

There might also be other suitable viral indicators of human fecal pollution that
await discovery. Unfortunately we currently lack adequate tools to perform high
throughput screening of environmental samples in order to characterize the virus
signatures that might be present. To date, several sets of primers have been published for
the detection of human enteric viruses however multiple pathogen detection of viruses
remains an elusive goal. Current research focusing on developing multiple pathogen
screening tools like microarray technology might be able to eventually allow
environmental samples to be screen for many multiple viruses in a cost effective and

rapid manner.

4.6 Conclusions
0 Both these recreational beaches along Lake Michigan were shown to have been
exposed to human fecal pollution at the time of the study.
0 The risk to swimmers at the beach was determined to be low but present.
0 The presence of the viral and bacterial indicators used in this study did not
correlate with each other, demonstrating a possible deficiency in the current

bacteria-based monitoring system to address viral sources of contamination.

70

Adenoviruses were determined to be the most prevalent enteric virus detected at
these beaches as detected by both cell culture-based PCR and real-time PCR with
a high degree of agreement between the two systems.

Statistically supported predictive models could be constructed for the two beaches
using the viral data and physical parameters. Similar models have been developed
and implemented at other recreational beaches for the determination of risk to

swimmers and bathers.

71

CHAPTER 5 DETECTION OF PATHOGENIC VIRUSES
CIRCULATING IN COMMUNITY WASTEWATER USING

OLIGONUCLEOTIDE MICROARRAYS

By Mark Wong, Syed A. Hashsharn, Erdogan Gulari, J can-Marie Rouillard, Joan B. Rose

Submitted for consideration to Environmental Science and Technology

5.1 Abstract

This study describes the novel use of a viral microarray to screen municipal
wastewater for the presence of circulating human pathogenic viruses. A total of 780
unique probes targeting 27 different groups of both DNA and RNA viruses were
designed and tested against laboratory strain viruses and environmental water samples.
Approximately thirty probes were used to target each viral group. Laboratory strains of
poliovirus and adenovirus type 40 and 41 were evaluated initially and indicated that the
probes were highly specific for their targets and that cross hybridization of target nucleic
acid was minimal even when closely related virus species were mixed and co-hybridized
on the array. During 13 months of sampling, RNA viruses were more frequently detected
in a community wastewater compared to DNA viruses. Overall, more viruses were
detected during the winter season compared to the summer months. Conclusion:
Microarrays are capable of screening for a broad number of pathogenic viruses which
may be circulating in the population and excreted in the community wastewater. This is
the first demonstration of an environmental microarray for detection of viruses in water

and could be used to improve public health surveillance.

72

5.2 Introduction

Viruses remain one of the most significant groups of human pathogens associated
with global disease in the let century. While smallpox has been eradicated and vaccines
are available for several viruses, resurgence of poliovirus in the developing world and the
rapid spread globally of newly evolving viruses such as SARS and new strains of
influenza have demonstrated that characterization and detection remain important for
defining public health strategies in the future (Tambyah, 2004; 2006; Wong et al., 2007).
Enteric viruses, including many groups in the picomaviradae, reoviradae and
adenoviradae family also remain a significant cause of global diarrhea, neurological,
respiratory as well as other chronic conditions associated with hepatitis and myocarditis
(Glass et al., 2001; Leclerc et al., 2002; Oberste and Pallansch, 2003). The enteric
noroviruses are associated with major outbreaks on cruise ships, at restaurants and in
nursing homes with a global distribution of new more pathogenic types. Recently, it has
been suggested that high viral excretion loads of Norovirus (NOV) Genogroup II in fact is
associated with its global predominance (Bull et al., 2006). A high infant mortality
(600,000 deaths annually) is still associated with rotavirus infections in the developing
regions of the world (Parashar et al., 2003).Adenoviruses have also been found to be of
significant concern causing gastroenteritis, upper and lower respiratory infections,
conjunctivitis. acute and chronic appendicitis, cystitis, exanthematous disease, and
nervous system diseases (Xagoraraki et al., 2007). Currently, it is estimated that over 150
types of viruses may be excreted in human and animal waste (Tartera and Jofre, 1987).

But a true catalog of the numbers and types of viruses remains elusive. In large part, this

73

is due to the lack of tools with which to screen environmental samples for the presence of

human-pathogenic viruses in a high throughput, high efficiency manner.

At present, cell culture in combination with molecular methods or polymerase
chain reaction (PCR) alone have been used for detection of pathogenic viruses in water
(Metcalf et al., 1995). Nevertheless, it is believed that many viruses exist that are
refractory to cell culture (Seymour and Appleton, 2001). Even if successfully cultured,
cell culture alone still provides no indication as to the virus’ identity. PCR has improved
the sensitivity and specificity of viral analysis of clinical or environmental samples.
However it has been difficult to design multiplex primers for more than 5-8 targets in a
single reaction, invalidating its use as a primary screening tool without some knowledge
as to the potential viral etiology. Secondly, PCR is incapable of targeting new and

emerging infectious viruses for which no known primers exist.

In recent years, virus microarrays have been developed and used to detect and
characterize pathogens from clinical samples allowing for broad screening and pathogen
discovery. Microarrays have been designed to detect and genotype Hepatitis B virus,
adenoviruses, Epstein Barr Virus, Herpes Simplex virus, influenza virus and human
papillomavirus (Wang et al., 2002; Wilson et al., 2002; Sengupta et al., 2003; Wang et
al., 2003; Boriskin et al., 2004; Korimbocus et al., 2005; Baxi et al., 2006; Chen et al.,
2006; Chiu et al., 2006; Liu et al., 2006; Min et al., 2006; Putonti et al., 2006; Song et al.,

2006; Lin et al., 2007; Wong et al., 2007). For this and other reasons, microarrays are

74

being developed and used as a clinical screening tool to detect and subtype a wide array

of viruses (Wang et al., 2002).

One potential application of microarrays as pathogen screening tools is their use
as biosensors for pathogens in the environment. To date however only a handful of
examples of environmentally-applied pathogen detection microarrays have been reported
(Call et al., 2003; Rhee et al., 2004; Sergeev et al., 2004; Wu etal., 2004; Maynard et al.,
2005; Lee et al., 2006; Kostic et al., 2007; Quinones et al., 2007; Miller et al., 2008). One
reason for the dearth of examples of environmentally applied microarrays is the presence
of inhibitory substances in the environment that compromise sample labeling and
hybridization efficiency. Another reason is the lack of sufficient sensitivity to detect the
low concentration of the target pathogens in relation to non-targets in the environment

(Straub and Chandler, 2003; Wagner et al., 2007).

The aim of this project was to design, evaluate and demonstrate the use of
microarrays for the detection of pathogenic viruses in sewage. Through this
characterization of circulating pathogens within a community one can begin to address

broadly an approach for biomonitoring and the screening of community health.

5.3 Materials and Methods
5.3.1 Viruses and Wastewater Sampling
Poliovirus LS—C-l, Adenovirus type 40 and type 41 were obtained from the

American Type Culture Collection (ATCC # VR-59, VR-931 and VR-930 respectively).

75

All viral infections were performed using the African Green Monkey Kidney cell line -
BGM, which were cultured in DMEM supplemented with 10% Fetal Calf Serum and
antibiotics. Viral infections were allowed to proceed until the onset of cytopathic effects
(typically within 24 hours). To recover the viruses, infected cells were freeze-thawed
three times to disrupt the cell integrity and release the viral particles into the cell culture
media. Virus particles were recovered and concentrated fi'om the media by centrifugation
through an Amicon Ultra 100K centrifugation column (Millipore, Billerica, MA). Virus
nucleic acid was extracted using the QIAamp viral RNA mini kit for both DNA and RNA
viruses following the manufacturer’s instructions. The QIAamp viral RNA mini kit has
previously been demonstrated to be capable of isolating viral DNA as well as viral
RNA(Kleines et al., 2003). Briefly, 140 microliters of virus concentrate was added to
tubes containing 560 microliters of lysis buffer and incubated at room temperature for 10
min with intermittent mixing. 560 microliters of ethanol was added and the solution
mixed before passage through a DNA binding column. Columns were washed with
washing solution and eluted with 2 rounds of elution solution using 50 microliters of

DNase and RNase free water each time.

Thirteen 6 liter samples of untreated sewage were collected from August 2006
and September 2007. Samples were brought back from the laboratory and 15% buffered

beef extract (Difco Inc) was added to give a final concentration of 1.5% beef extract and

0.05 molar glycine. The solution was stirred for 30 minutes before 2.5 molar FeCl3 was

added to a final concentration of 2.5 millimolar. The pH of the solution was lowered to

3.5 and the solution was stirred a further 30 minutes. Viruses were pelleted by

76

centrifugation at 2,500 x g for 15 minutes. The viral pellet was resuspended in 90
milliliters of 0.15 molar sodium phosphate dibasic (pH 7.0) solution by agitation on an
orbital shaker set to 160 revolutions per minute. Once the pellet was dissolved, the pH of
the sodium phosphate was raised to between 9.0 and 9.5 and placed on the orbital shaker
for a further 10 minutes. Solid particles were pelleted by centrifugation at 10000 x g
(Beckman model J2-HC). The supernatant was collected and filtered through a 0.22
micrometer syringe filter and supplemented with 100 U/ milliliter of Penicillin, 100
microgram/ milliliter of Streptomycin and 250 nanogram/milliliter of Fungizone. The pH

of the virus concentrate was neutralized to 7 .0 and flown at -80 °C until placed on cell

culture.

Twelve milliliters of virus concentrate was used to infect BGM cells grown to

approximately 80 - 90% confluence. Virus concentrate was allowed 120 minutes of
contact with the BGM cells before being discarded. Cells were incubated at 37 °C until

development of cytopathic effects were observed. Infected cells were harvested by
mechanical lifting using a sterile cell scraper. All cells and free viruses in the media were
concentrated by centrifugation through an Amicon Ultra 100K centrifugation column
(Millipore, Billerica, MA). Virus nucleic acid was extracted using the Qiagen viral RNA

mini kit in the same manner as described above for the ATCC strains of viruses.

5.3.2 Microarray Design and Construction
Viral sequence data were obtained fi'om the curated database of fully sequenced

viral genomes in GenBank. Seventeen RNA virus groups and 10 DNA virus groups were

77

targeted. A list of sequences used to design probes is given in Table 5.1. Genome
sequences were parsed through the Oligoarray 2.1 (Rouillard et al., 2003). The probes
were designed to conform to the following specifications: maximum melting temperature
(Tm) = 75 °C (except for torovirus = 80 °C), minimum Tm = 70 °C (except for torovirus
= 65 °C); maximum guanine + cytosine (GC) content = 60%, minimum GC = 40%,
maximum temperature for secondary structure = 45 °C, maximum temperature for cross
hybridization = 45 °C. Probes were designed from the positive strand of the genetic
sequence. A total of 780 specific probes were designed (approximately 30 probes per
viral family target). Generating multiple probes for each target family was thought to
enhance the reliability of detection by providing multiple binding sites on the chip for

target hybridization.

Microarray chips were synthesized by using the in situ synthesis technology
developed by the University of Michigan as described previously (LeProust et al., 2000;
Gao et al., 2001; Komolpis et al., 2002). The microarray chip format used for the virus
chip has a 7 column 10 by 100 array with a potential for containing a maximum of 7000
spots. Spots were randomly populated with 5 copies of each of the 810 designed probes
representing 48% of the chip capacity. Multiple copies of probes were used to provide
technical replication of the signals. The complete microarray probe layout and annotated
probe information has been deposited at the Gene Expression Omnibus (GEO Accession
# GPL6501). The microarray substrates were made by etching a continuous serpentine
channel ~ one millimeter wide, seventy microns deep and 10 mm long on a 10 mm x 14

mm silicon substrate. The glass cover was bonded to the top of the silicon substrate to

78

provide a closed channel for the in situ synthesis of the oligos. After derivatizing the
silicon and glass surfaces with an aminosilane linker a 15 thymine spacer was
synthesized on top of the linker to reduce the steric hindrance. The final 25-28 mer
probes were in situ synthesized using the phosphoramidite chemistry modified to work
with photogenerated acid. Recirculation of the sample in the closed continuous
serpentine channel of the microarray ensured good exposure of all the probes to the DNA
and RNA targets in the sample during hybridization and reduced the time to reach
equilibrium hybridization. The flow of the sample solution also helps remove weakly
held mismatched targets. The closed nature of the microarray also totally eliminates any

bleaching related to atmospheric ozone.

79

Table 5.1. List of viral targets, type of genomes and Genbank accession numbers

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Virus Type of genome Accession Sequence
no. length (bp)

Hepatitis A vim; ssRNA positive no DNA NC_001489 7478
stage

Hepatitis E vim; ssRNA positive no DNA NC_001434 7176
stage

Human adenovirus A!“ dSDNA NC_001460 34125

Human adenovirus B* dSDNA NC_004001 34794

Human adenovirus C* dSDNA NC_001405 35937

Human adenovirus D,“ dSDNA NC_002067 35100

Human adenovirus E,“ dSDNA NC_003266 35994

Hu‘rnan adenovirus type dsDNA NC_001454 34214

40

Hrunan adenovirus type dsDNA DQ315364 34139

41

Norwalk vim; ssRNA positive no DNA NC_001959 7654
state

Sapovims" ssRNA positive no DNA NC_010624 7458
stage

Human enterovirus A,“ ssRNA positive no DNA NC_001612 7413
stage

Human enterovirus 3* ssRNA positive no DNA NC_001472 7389
stage

Human enterovirus C’“ ssRNA positive no DNA NC_001428 7401
stage

 

80

 

Table 5.1 continued

 

*

 

 

 

 

 

 

 

 

 

 

 

 

 

Human enterovirus D ssRNA positive no DNA NC_00143O 7390
stage
Human enterovirus E" ssRNA positive no DNA NC_003988 7374
stage
Poliovirus!“ ssRNA positive no DNA NC_002058 7440
stage
rotavirus AT dsRNA virus AB077766 2359
(VP4 gene)
AB071404 1062
(VP7 gene)
rotavirus Bl dsRNA virus AY539857 2306
(VP4 gene)
AY539856 814
(VP7 gene)
rotavirus CT dsRNA virus AB008670 2283
(VP4 gene)
ABOO8671 1063
(VP7gene)
Coronavim; ssRNA positive no DNA NC_002645 27317
stage
cytomegalovims (HHS) * dsDNA virus NC_006273 235645
Torovirusi ssRNA positive no DNA AF 1 595 85 1251
stage (Hemagglutinin
esterase gene)
AF 024539 219
(nucleocapsid
gene)

 

 

 

 

81

 

Table 5.] continued

 

 

 

 

 

 

picobimavirus§ dsRNA virus AF246940 1674
(RNA dependent
RNA polymerase)
AF246941 1572
(segment 1 gene)
Astroviruses" ssRNA NC_001943 6813
JC poiyomavims“ ds DNA NC_001699 5130
BK polyomavims'“ ds DNA NC_001538 5153

 

 

 

 

‘ Complete genome
1 VP4 and VP7
i Hemagglutinin esterase and nucleocapsid protein mRNA

§ RNA dependent RNA polymerase and segment 1 gene

82

 

5.3.3 Preparation of Samples for Hybridization

Viral RNA was labeled with fluorescent dyes by a semi-random primed labeling
with Sensiscript III reverse transcriptase as described by Wang et al. (Wang et al., 2003).
Half a microgram (0.5 ug) of viral nucleic acid was used as a template for the generation
of cDNA with a discrete 5’ terminal consisting of the sequence
(5’-GTTTCCCAGTCACGATC-3’) using the semi-random primer A:
(5 ’-GTTTCCCAGTCACGATCNNNNNNNNN-3 ’). Next primer B:
(5’-GTTTCCCAGTCACGATC-3’) and Qiagen Hotstart Taq polymerase was used to

amplify the generated cDNA and label it with amino-ally] dUTP for 40 cycles using the
following profile: 30 seconds at 94°C, 30 seconds at 40°C, 30 seconds at 50°C, 60
seconds at 72°C. To label viral DNA, 0.5 micrograms of virus DNA was first digested

with IU of DPNII restriction endonuclease at 37 °C for 1 hour. The digested DNA was

incubated at 37 °C for 2 hours with Klenow enzyme and Primer A to generate

complementary DNA with a discrete 5’ terminal. Next primer B and Qiagen Hotstart Taq

polymerase was used to amplify the generated cDNA and label it with amino-allyl dUTP

for 40 cycles using the following profile: 30 seconds at 94°C, 30 seconds at 40°C, 30

seconds at 50°C, 60 seconds at 72°C for 40 cycles.

Labeled DNA and RNA were coupled separately with Cyanine dye 3 and Cyanine
dye 5. This reaction was canied out in the absence of light and using 1 molar sodium
bicarbonate (pH 9.5) as a coupling buffer with a 1-hour incubation. Following coupling,
uncoupled dye was removed using the QIAgen PCR pruification kit and the labeled virus

DNA and RNA were dessicated in a DNA 120 SpeedVac (Thermo Fisher Scientific, MA,

83

USA) for 1.5 hours. The sample preparation, labeling and hybridization procedure is

outlined in Figure 5.1.

5.3.3 Hybridization of Samples

Microarray hybridization was performed as described previously (Wick et al.,
2006). The microarrays were hybridized and washed in a M-2 microfluidic station
(Invitrogen, Carlsbad, CA, formerly Xeotron Corporation, Houston, TX) using a flow
rate of 400 microliters / minute. The hybridization buffer was 6x- SSPE, 25% formarnide,
0.4% Triton X-100. Chips were pre-hybridized with 6x- SSPE, 0.2% Triton X-100 and

then with hybridization buffer for 2 min each.

Labeled targets were resuspended in 50 nricroliters hybridization buffer,

denatured at 95°C for 3 minutes, cooled on ice for 1 minute, filtered through a 0.22

micrometer Costar spin filter and then hybridized to the chip for 14-15 hours at 20°C.

The chip was scanned with a GenePix 4000B laser scanner (Axon Instruments,
Union City, CA). All solutions were filtered through a 0.22 micrometer filter to prevent
clogging of the microfluidic channels. The high stringency wash buffer was degassed
under vacuum. All arrays were imaged with a GenePix 4000B scanner (Molecular

Devices, Sunnyvale, CA) and GenePix Pro software.

84

5.3.5 Data Analysis Using DetectiV

The DetectiV software package was used to visualize, normalize and test the
significance of the microarray hybridization data (Watson et al., 2007). DetectiV was
used to generate bar plots of the viral microarray signals following normalization against
the median signal values for each sample. DetectiV was also used to carry out statistical
t-test comparisons of the hybridization signal values between different virus groups in

order to determine which groups of virus target had statistically significant probe signals.

5.3.6 Polymerase chain reaction (PCR) detection of viral targets

Cell culture extracts were also evaluated with PCR for key viruses. PCR was
performed using the QIAgen OneStep Reverse transcription kit for RNA viruses and the
QIAgen Hotstar Taq kit for DNA viruses. The OneStep Reverse Transcription reaction

was carried out as follows: 2 microliters of OneStep RT-PCR enzyme mix, 1.5 millimolar
MgClz, l>< PCR buffer, l>< Q solution, 0.5 micromolar of each primer, 0.5 millimolar of
each dNTP and 0.5 microgram template. Thermocycler settings for the reverse
transcription PCR were as follows: 50°C, 30 minutes first strand synthesis, 95 °C, 15

minutes initial denaturation to activate the Hotstart Taq polymerase (Qiagen; Valencia,

CA), 95°C, 0.5 minutes denaturation, 57 °C, 0.5 minutes armealing and 72°C, 0.5

minutes elongation for 35 cycles followed by a final elongation step of 72°C for 5
minutes. The PCR reaction was carried out as follows: 1 unit of Hotstart Taq polymerase,

1.5 millimolar MgClz, 1>< PCR buffer, 1X Q solution, 0.5 micromolar of each primer, 0.5

86

millimolar of each dNTP and 0.5 microgram of template. Thermocycler settings for the

PCR reaction were as follows: 95 °C, 15 rrrinutes initial denaturation to activate the
Hotstart Taq polymerase (Qiagen; Valencia, CA), 95°C, 0.5 minutes denaturation, 57°C,
0.5 minutes annealing and 72°C, 0.5 minutes elongation for 35 cycles followed by a final

elongation step of 72°C for 5 minutes. PCR primers used are reported in table 5.2.

87

 

Aeoom

8H830020<<0Hm8030 ”NM

 

 

 

 

 

 

 

 

 

 

 

an o 5&9 an om... o~§<oEo~o<E<32m ”NE 9:585
85500035053568 9:5
8 m2 005235520332 ”450
A83 :6 DDDPD<DDP<<§<tDD<D<<Eo ”m5
so aefioaoe 8 m2 85953335055... ED me: 6.3662
3 09¢ 3:30:03 BEBE
ow 25 9:30:03 5853
m 9:30:25 58::
.858850358905948505-.m
”mom ~30: Q 3:30:23 5353
.was865505855850?m
an m: ”mam _<<xos U maggot—s 58::
.v.833358088595-.m
” m _m_<<xon m 3:353 58::
seem an .m-85535555055800-.m
so 8:58 an Sm ” new 2522 < 35828 58$
Goa .m-o< 05 000 5e 50 05 50% “6mm:
.._e so 88: an E .33. <8 con 05 o; 200 65% 8%.: was m 2568:
3% 638m
ooh—om 888:5 $25.5 was

 

 

 

 

 

6qu =2? :93 .8.— wfiouoaaom van .3582. mm: .8323.— Eafi cannon—be: 2: .Su can: Eon—tn he an: .N.m «3:.

88

 

A32 :3

0<OHUOO<<COG<EUHOGHOOEOG Um:

 

 

 

 

 

 

a eofifiuv .5 an <8<oo<<ue<<o§<occooeuo US $5 330338030
A88 .m-<ooo.E<EooE<omofim H 3750
91. Ho 886 .5 EN .mégfigzwiiigokufim ” 5.80 33828
U £5238
m £5238
A88 :3 oo<§<E<uo<oon<
.0 fiancee as :N o§<<o<ofi<<ou<ouwt < 3338
95> < guano:
9:30:9—
m @3885 583;
Q mEFEoEo 385$
U wagohowco 593$
.m-<5<<<ouu<o<coo<o<<<o-.m 382
an we .m-§ooooo_.<<oeouuomouofim 6332 m 33835 :25:
808 :3 .mégouooEwEoofioooJm ”m2
3 Ease 3 «mm .m-<uuu<oo<<e<oo<got<-.m ”E < 53235 525$

 

 

 

 

 

 

 

 

 

 

 

cos—5:3 N.m 03.;

89

 

SOON
f—d H0 mm:

.m-0<:8<U<UUE<O<HO<-.m vv—m

 

 

 

 

-Eomv 3 on 3-50565335833 GE 33833 Mm
:23 3
:3 3 3: .m-c<8<<<t<<tohotugo<<owum H ‘52
4:36 3 o: .m-o<<ooe2bE<oo<o<E<o§m ” am“: 33335on 2
.m-<<uog<tu<<<o<ut<fim HN<
823 gm: .m-<<c<ouo<oooouwtoo-.m ”E
:3 3 .m-o<<uu<<o<<o<oooougoo-.m ”mom
gage .3 SN .m-eoot<o<oout<o<<5-.m ”E 832382
33335553635 “mic:
>an mu: 3 OS 05530358530 ””35 2:385

 

 

 

 

 

32:5qu Nd 033,—:

90

5.3.7 Sequencing confirmation of viral targets

PCR products were purified using the QIAgen QIAquick PCR purification kit
and eluted with molecular grade water. 10 nanograms of purified PCR product and 30
picomoles of each primer were provided to the Michigan State University Research
Technology Support Facility for custom sequencing. Sequencing results were
reassembled from the forward and reverse sequencing runs by hand and the sequences
were used to perform a Basic Local Alignment and Search Tool (BLAST) query
(http://wwwncbi.nlm.nih.gov/blast/Blast.cgi) to determine the closest matches based on

sequence homology.

5.4 Results and Discussion
5.4.1 Hybridization with poliovirus LS-C-l, Adenovirus type 40 and 41
Laboratory Strains

The viral microarray was tested with a laboratory strain of poliovirus LS-C-l
grown in African Green Monkey Kidney cells (BGM). Viral RNA was extracted using
the QIAgen viral RNA kit and labeled with Cy3 dye. All 30 probes generated for
poliovirus hybridized to the cell culture poliovirus samples with zero cross hybridization
above a signal to noise ratio cutoff of 2 with non-poliovirus probes. Hybridizations were
repeated twice with similar results. Hybridization signal values were deposited in GEO
(accession #GSE10566). Hybridization signals were normalized against the median
signal intensity as described by Watson et a1 (2007), log; transformed and averaged
across the three runs. Logz-transformed mean intensity signals were used to generate a

bar plot shown in figure 5.2a. A “t-test” was carried out to test the null hypothesis that

91

hybridization signals between viral target groups were not significantly different. Based

on the t test, at the 95% confidence level, the poliovirus group was found to have a

significantly greater signal than the other viral groups with a p value of 1.74><10-28 and

an average normalized signal value of 3.00, leading to a rejection of the null hypothesis

for poliovirus (Table 5.3).

Adenovirus type 40 strain Dugan (AT CC #VR-931) and Adenovirus type 41
strain Tak (ATCC #VR-930) were extracted directly from the stock vials, labeled with

Cy3 and CyS respectively and co-hybridized on to the viral microarray. Hybridization

signal values were deposited in GEO accession # GSE10569. Logz-transformed and

normalized signals were plotted on a bar graph (Figure 5.2b). Separate “t tests” were
carried out for each channel (Cy3 and CyS) to test the null hypothesis that hybridization
signals between viral target groups were not significantly different. Based on the t tests,
at the 95% confidence level, the Cy3-labeled adenovirus type 40 was found to generate

significant signals with probes designed specifically for adenovirus type 40 (p value =

1.47X10-9, average normalized intensity = 2.776) and also adenovirus type 41 probes (p

value=2.10><10'5) but with a lower average normalized signal intensity (average

normalized intensity=l .034) . Similarly, the CyS-labeled adenovirus type 41 was found to

generate significant signals with probes designed for adenovirus type 41 (p value =

7.09X10-9, average normalized intensity = 2.690) and also adenovirus type 40 probes (p

92

value=7.64><10-6) but with a lower average normalized signal intensity (average

normalized intensity=0.938).

5.4.2 Using a positive fraction criteria for assigning a probable viral target
presence

Based on the fluorescent intensities for each spot, we characterized a probe as
being positive when at least 4 out of the 5 copies of that probe showed a signal to noise
value greater than 2. When adenovirus type 40 and 41 signals were expressed as a
fraction of the maximum possible signal and charted on a histogram together with the
next forty highest non-adenovirus type 40 or 41 signals it was observed that only 12 out
of the possible 30 probes for each virus were positive by our criteria (Figure 5.3). From
this, we concluded that we would expect approximately 12 probes from the target viruses
to show positive signals greater than background levels when any given mixed consortia
of viruses is hybridized to the array. This theoretical number of 12 probes was used to
form the upper limit for determining the positive fraction which is the number of positive

probes divided by 12.

93

 

 

 

Concentration

Environmental Virus Concentrate

sample

r—‘v

r

l
A;
':

 

 

 

 

 

 

 

 

 

 

 

Nucleic Acid
Extraction
. .. .. .. _-...-_......_.... ___ .. n»- r '
i Labeled Target L NUCIB'C AC'd Virus Nucleic Acid
‘ DNA ,
_____ W Labeling
_ Nucleic Acid
: coco-cocoa. Labeling

0000000....
' 00000000000
l 00000000000
1 coco-cocoa. -

% "NH“,"o LabeledVIrusRNA
W‘OOOOOOOOOO.
., . ooooooooooo
Hybridization 00000000000
00000000000 .. _
ooooooooooo Hybridization
00000000000
00000000000

 

 

 

 

 

Virus Microarray

 

 

 

Figure 5.1. Schematic illustrating the steps involved with virus concentration,
nucleic acid extraction, labeling and hybridization on the virus microarray resulting
in a hybridization pattern

85

 

 

Poliovirus control

 

 

3

< 5:39.35 555m
magi—aw

gem—em

55> x1352
55:555.:— Un
53:58 :55:

0 E89. :55:

m 5.555.. 5:5:

< 5525..— :55:
5:35:33:— 555:
m 253932— 555:
a 5:835 :55:
0 5:39:85 :55:
a 55:33:» :55:
4 55:23:”. 55::
E528 :55:
55.5.53 :55:

u: 093 5:30:33 :55:
3. 095 gases :5:
W 55355: 555:
D 355:3: 55::
U 5:553: :55:
m 5:90:96: :55:
< 5555:: :55:
an... a 3:39:

2.5 < 5.3%:
5:95:22— v:—

 

 

Figure 5.2. Bar plots showing the logZ-normalized hybridization signals for various
control experiments. (a) Hybridization of poliovirus LS-C-l RNA extracts labeled

with Cy3.

94

MOHOVINIS control

 

 

or

4 55528:: 5m
5.55:5

5.5559.—

55> 5:352
5.555955 on.
535,—. :55:

U 5.555.. 55:

a 5.555.— :55:

< 5.5.»52 :5:
5.555585: 5:
m 5.55555 :55:
G 55558:: :55:
D 55558:: :55:
a 5.5385 :55:
< 55553:... :55:
5.55.55 :55:
5.55.55 :55:

S. 2:. 5.55:5: 5:
8 5.3 5.55:5: :55:
H 5.55:5: :55:
: 5.55:5: :55:
U 5.55:5: 5::
.— 5.55:5: :55:
< 5.55:5: :55:
55> H 555::

:5 < 55%:
5.55555 :5

(b)

Figure 5.2 cont’d

tion of Cy3-labeled adenovirus type 40 unto the

ridiza
viral array. Each vertical bar represents a different target virus group.

(b) Hyb

95

Adenovirus control

 

 

or

9:: 593.52
9523:52— UH
953...? gm—

0 9:75.: 53...!

a E39. SEE—m

< 9:35.... SEEM
Eggs... SEEM
m 9:392:2— 5.5.:
A 3.328..» .55.;—
U 5.3385 53...:
a 9:39.35 535—.—
< 9:39.35 53...:
9:35.83 53.5w
3.3% 5.5m.

5. 25. 9:32.93 5.5—fl
9. 2:. 9555—... 5.5.!
m 9:32.23 53...!
9 9:32.99. 53...:
0 3.32.33 53...:
a mates“... SEEM
< 9:32.23 53...:
9:... H 95.3%—

uEt < Qua»:
Each—e.— In

(C)

rustype4l untotheviral

(c) Hybridization of CyS-Iabeled adenovi

array. Each vertical bar represents a different target virus group.

Figure 5.2 cont’d

96

Table 5.3. t test results for the hybridization of control viruses and sewage extracted
samples to the viral microarray showing the most significant results with p values

less than 0.01 and their correspondingygz mean intensity

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sample Sample Top targets p values logz mean
type intensity
Control Poliovirus Poliovirus 1741 x10'28 3 .0056
LS-C-l
Control Adenovirus Human adenovirus type 40 1.472x10'9 2.7757
type 40 Human adenov1rus type 41 2.099x10-5 1.0340
Control Adenovirus Human adenovirus type 41 7.037x10‘9 2.6902
type 41 Human adenovirus type 40 7.641 x10-6 0.9376
Sample August Human Herpesvirus 5 0.00291 0.05398
2006 Hepatitis B Virus 0.00466 0.07704
sewage Human Enterovirus A 0.01002 0.12030
Sample September Human Adenovirus C 0.00257 0.14788
2006 Human Adenovirus A 0.00310 0.12599
sewage Hepatitis B Virus 0.01361 0.08063
Sample October Human enterovirus D 0.00411 0.21707
2006 Human enterovirus E 0.00563 0.24263
sewage Human Adenovirus B 0.01652 0.18649
Sample November Human enterovirus E 0.00558 0.18725
2006 Human enterovirus A 0.01134 0.22106
sewage Human Astrovirus 0.01587 0.27328
Sample December Human enterovirus A 0.00114 1.15508
2006 Human Astrovirus 0.00121 1.09681
sewage Human enterovirus E 0.00161 0.97785
Human Adenovirus type 41 0.00253 0.61050
Sample January Human Adenovirus type 41 0.00029 1.24453
2007 Human Astrovirus 0.00032 1.48500
sewage Human Enterovirus A 0.00044 1.33010
Human Enterovirus E 0.00049 1.19024
Norwalk virus 0.00457 0.89360
Human Enterovirus D 0.00516 0.84998
Human Enterovirus B 0.00585 0.75816
Sapovirus 0.00615 0.70991
Human Adenovirus B 0.00761 0.98407

 

97

 

Table 5.3 Continued

 

 

 

 

 

 

 

 

Sample February Human adenovirus type 41 0.00033 1.10309
2007 Human astrovirus 0.00095 1.29553

sewage Human enterovirus B 0.00417 0.75738

Human enterovirus A 0.00753 0.78814

Sample March Human Herpesvirus 5 2.290x10‘05 0.41377
2007 Human Enterovirus B 0.00081 0.22016

sewage Hepatitis B Virus 0.00434 0.11590

Sample April 2007 Sapovirus 7.36><10m 0.25954
sewage Human Herpesvirus 5 2.15x10*’5 0.41552

Hepatitis B Virus 3.86x10"3 0.11773

Sample May 2007 Sapovirus 7.46><10'07 0.25990
sewage Human Herpesvirus 5 2.13><10‘°5 0.41574

Human Enterovirus B 1.11><10‘03 0.21538

Sample June 2007 Sapovirus 7.46><10*’7 0.25912
sewage Human Herpesvirus 5 2.25><10*’5 0.41541

Human Enterovirus A 6.75X10'05 0.26078

Sample July 2007 Sapovirus 7.52x10‘” 0.25938
sewage Human Herpesvirus 5 2.85><10*’5 0.41257

Human Enterovirus A 6.75><10*’5 0.26011

Sample August Sapovirus 7.65X10'07 0.25731
2007 Human Herpesvirus 5 2.67x10"5 0.32781

sewage Human Enterovirus B 1.05X10'03 0.21546

 

 

 

 

 

98

 

5.4 3 Hybridization of Sewage Extracts

Cy3 (DNA) and Cy5-labeled (RNA) nucleic acids derived from sewage samples
collected between August 2006 and August 2007 were hybridized on to the viral
microarray. Hybridization signals were deposited at the Gene Expression Omnibus
(GEO) as accession #GSE11195. Based on a positive fraction cut off of at least 0.33
representing 4 out of the arbitrary upper limit of 12 positive probes, the potential virus
detected by the microarray are listed in table 5.4. Using the software package DetectiV,
bar plots, illustrating the hybridization intensities for the various sewage samples were
created and are Shown in figure 5.4. It was observed that much higher normalized signal
intensities were obtained during the winter months compared to the summer months —
average signal intensities for the top five most significant virus groups were
approximately 1 in December, January and February while in August, September, April,
May, June and July they were approximately 0.1, except for a few strong positives that

were observed (Table 5.3). In the August 2006 sample, the most significant t value was

found to belong to the Human Herpesvirus group 5 (Table 5.3). However, the logz mean
signal intensity for that group of viruses was only 0.05398, thus it should not be
considered as the most likely virus in that sample. As discussed by Watson et a1 (Watson

et al., 2007), it is more likely a Human enterovirus group A virus (p value = 0.01002, log

mean Signal intensity = 0.1203).

99

Adenovirus type 40/41 hybridization profile

 

 

 

 

.0
N

 

 

 

Fraction of max signal
0

 

 

 

 

 

 

 

 

L‘Adenovirus type 41 I Adenovirus type 40

Figure 5.3 Adenovirus type 40 and 41 hybridization results. Adenovirus
type 40 signals are shown above the x-axis line and adenovirus type 41
signals are shown below the x-axls line.

100

Table 5.4. Comparison of positive signal fractions, software analysis, group specific
PCR results and sequencing data for the identification of viruses within a sample

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sample Potential Potential PCR Sequenced identities
targets via targets via results
positive software
fraction analysis
analysis
August + Human adenovirus type 41
2006 + Human astrovirus type 7
September + Human astrovirus type 7
2006
October + Human adenovirus type 32
2006 + Human enterovirus A-2 plaque
November Human -
2006 coronavirus
+ Human astrovirus type 7
December Human Human + Human astrovirus type 7
2006 astrovirus astrovirus
Human -
coronavirus
Human Human + Human coxsaclde A10
enterovirus A enterovirus A
Human Human
enterovirus E enterovirus E
Norovirus + Norovirus isolate
Human + Human rotavirus C isolate
rotavirus C
Human + Human adenovirus type 7
Adenovirus B
Human
adenovirus
type 41
+ BK polyomavirus strain AS

 

 

 

 

 

101

 

Table 5.4 Continued

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

December + Hepatitis E virus
2006
+ (human No homology
torovirus)

January Human Human + Human astrovirus type 7
2007 astrovirus astrovirus

Human -

coronavirus

Human Human + Human enterovirus A-2 plaque

enterovirus A enterovirus A

Human

enterovirus B

Human

enterovirus C

Human Human

enterovirus D enterovirus D

Human Human

enterovirus E enterovirus B

Hepatitis E + Hepatitis E virus

virus

Norovirus Norovirus + Norovirus isolate

Sapovirus Sapovirus

Human + Human rotavirus B

Rotavirus B

Human

Rotavirus C

 

 

 

 

 

 

102

 

Table 5.4 Continued

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

January Human + Human adenovirus type 41
2007 Adenovirus A

Human Human

Adenovirus B adenovirus B

Human

Adenovirus D

Human Human

Adenovirus adenovirus

type 41 type 41

BK + BK polyomavirus strain AS

polyomavirus

+ (human No homology
torovirus)

February Human Human + Human astrovirus type 7
2007 astrovirus astrovirus

Human -

coronavirus

Human Human + Human coxsackie A10

enterovirus A enterovirus A

Human
enterovirus B

Human + Human adenovirus type 41

adenovirus B

Human Human

adenovirus adenovirus

type 41 type 41

BK + BK polyomavirus strain AS

polyomavirus
March + Human coxsackie virus A20
2007

 

 

 

 

 

 

103

 

Table 5.4 Continued

 

 

 

 

 

 

April + Human coxsackie virus A 20
2007
May 2007
June 2007 + Human astrovirus type 7

+ Human adenovirus type 41
July 2007
August + Human coxsackie virus A16
2007

 

 

 

 

 

104

 

August 2006 Sewage

 

 

o_.

< mE_>o.cEm c556
mason—mm

953%...

95> 52:02
«522020.. 0..
955.8. 55.:

0 952.8”. 55::
m 9:398. cmEsI
< 9:330”. cues:
955E300... 55::
m $3380: 56::
0 955.25 :55::
0 953.25 cuss:
m 9:3ng 5.5::
< 95>2ofiw 5:5:
wE_>mco.oo 5.5::
95352 5.5:...

S. 8.: 955:8... :25:
av 83 95.653. ana...
m 9.._>occu< 55::
o ma_>ocou< 55::
0 9:30.53. 55::
m 95>ocou< 55::
< 95>ocou< 58::
3.5 m. 2589.
95> < 2580...
m5>mEozoa v5

(3)

lized hybridization signals for the different

labeled DNA and Cy5-label RNA extracted from

logz-norma

ing

Figure 5.4. Bar plot show

hen Cy3
sewage is hybridized unto the array. (a) August 2006 sample. Each vertical bar

represents a different target virus group.

target virus groups w

105

September 2006 Sewage

 

 

< o5>EoEw c555
9535

953:0...

95> 533.02
«2.383.. 0..
953.8. 55::

o 953.5 55::
m 953.5 :oEnz
< 9538a :95:
n53EBooE 55::
n 9530901 55::
a «53.25 55.:
o ua_>EoEw 55::
m 953.25 55::
< 953.25 5.5::
«535.00 55::
9.53:2 55:...

S. on... 95.682 :25:
9. on». 95353. 55::
m «.5353. 55::
o «53:25.. 5.5::
0 95353.. cuss:
m 953cou< 55::
.4 95353. 5.5::
95> w 9582..
95> < £58:
3.55028 xm

(b)

(b) September 2006 sample. Each vertical bar represents a

different target virus group

Figure 5.4 (Continued)

106

October 2006 Sewage

 

 

o_.

4. 953.25 0556
95305

953:5

9:3 3.3.3.02
9535020.... 0..
953.0 ._. 55::

o 9:325 5:5:
m 9:325 55::
< 9:325 55...:
95352050 :03...
m 9:302:01 :35:
a 953.25 5:5...
0 9:30.25 55::
m 9:30.25 :25:
< 9:30.25 55::
9533.00 005::
953...... 55::

3 00>. 953:2}. 5.5:...
9. 00... 953003. :95...
m 0:30:03. 55::
0 9:30:03. 55.:
0 9:30:03. 55:...
m 9:30:03. :25...
< 9:30:03. :95:
9:5 m 05.000...
9:3 < 05.000...
9:383. gm

(C)

(c) October 2006 sample. Each vertical bar represents a

Figure 5.4 (Continued)

different target virus group.

107

November 2006 Sewage

 

 

< 3.3055 5.5m

953E020... 0..
9:320... :95:

o 9:320: 55::
m 9:320: 5:5:
< 9:320: 5:5:
9:39.505: :mEa:
m 953080: :mEa:
o 953.2:w 5:5:
0 9:30.25 :25:
m 953.2:w :mEa:
< 9:30:25 :mEa:
9:330:00 :aEs:
mE3ES< 58::

S. on». 953:2.< :95:
9. 95 953cou< 58::
w 9:30:23 :35:
o m53cou< :nEs:
0 95352 55::
m 9:30:22 5&3:
< 9:30:02 :25:
95> w 9580:
95> < 3582..
“.3553: gm

((1)

(d) November 2006 sample. Each vertical bar represents a

Figure 5.4 (Continued)

different target virus group

108

December 2006 Sewage

 

 

< 953.2:w 555
95350

023 0.53.02
35:53.. 0..
953.0 5 5:5:

0 9:320: 5:5:
m 9:320: 55:
< 9:320: 5E0:
9535000.: 5.5:
m 03330.0: 5:5:
0 9:30.25. 5:5:
0 953.2:m 5:5:
m 953.2:m 5:5:
< 9:30:25 55.:
9535.00 5:5:
9:30.22 5E0:

S. 005 95353. 5:5:
9 00.: 95353. 5:5:
w 95350.... 5:5:
0 9:30:02 55;:
0 9:30:03. 5E0:
m 9:30:02 5:5:
< 9:30:23 5:5:
95> m 9.500:
95> < 258%
053E020... xm

(e)

(e) December 2006 sample. Each vertical bar represents a

different target virus group

Figure 5.4 (Continued)

109

January 2007 Sewage

 

 

 

or

< 9:39.25 5.50
95350

9:30:01

9:3 0:52:02
9:35.020: 0..
953.8. 5:5:

0 9:320: 55.:
m 9:320: 5E0:
< 9:320: 5:5:
9:32:58... 55.:
m 95300.0: 5:5:
0 9:30.25 :25:
0 9:30.25 5.5:
m 9:30.25 5:5:
.0. 95.6.2.5 5:5:
9:305:00 5E0:
9:30.204. 5:5:

S. 00.: 03305.0( 5.5:
ov 00.: 95353.. 5.5:
5 9:30:03. 5:5:
0 9:30:00... 5:5:
0 9:305..< 5.5:
m 9:30:03. 55;:
< 95350:. 5:5:
n95> m 9.500:
9:3 < 05500:
055E020: X:

m

(0 January 2007 sample. Each vertical bar represents a

Figure 5.4 (Continued)

different target virus group

110

February 2007 Sewage

 

 

< 3.5.25 55.0
953000

3.5.0:

8..., 5:82
5.553.. 0..

0 3.56: 5.5:

< 3.5.3. 5.5...
3.538... :25...

0 9538.0: :05:

0 053.25 :25:

0 95325 5:5:

0 953.25 :05:

< 953.25 :05:
9:305:00 :05:
95352 :05:

3. 8.. 2:553. 5.5...
9 a... 3.583. 5.5:
m. 95353. 5:5:

0 3.553 5:5...

0 9535.2 :25:

0 9:30:03. 5E0:

< 3.352 5:5...
3.3 m. 0532

3.... < 9.58:
9:30.520: :0

(g)

(g) February 2007 sample. Each vertical bar represents a

Figure 5.4 (Continued)

different target virus group.

111

March 2007 Sewage

 

 

 

< 3322...". .525
25285

3.32.9.

3.3 .382
3.8.8.3“. 0..
3.320.. 5.5.:

o 3.323. 5.5...
m .2333. 5.5...
< 3.359.. 5.5...
02.3538... :35:
n 0.53090: :25:
o 3.3.2.... 5.5..
o 5.35.... :35:
m 5.5.2.... 5......
< 5.52% 5......
02.3.3.8 c053:
0.53:2 5.5:

s. a... 3.553. 5.5..
3 2.... 5.853. 5.5...
m. 3.853. 5.5...
o 3.3.8.. 5.5...
o 5.5; 5.5:
m 3.3203 5:5...
< 3.3.8.. 5:5...
3..> m 3508..
3.3 < 3.3%..
3.5.5.3 gm

(h)

(1)) March 2007 sample. Each vertical bar represents a

Virus group.

Figure 5.4 (Continued)

different target

112

April 2007 Sewage

 

 

 

< 5.5.2.... 5.....0
3.33m

3.6.3..

3.3 5332
5.5.8.3.. 0..
02.3.0... 006::

0 3.5.3. 5.5..
m 3.3.3. 5.5:
< 3.333. 5.5:
02.3638... 005:...
m 25300.0: 58::
o 3.8.2.... 5.5:
U 0.53.2.5 008::
0 5.5.3.0 5.5..
< 3.3955 5.5...
02.33.00 008::
3352.~ 5.5..

S 8... 3.388.. 5.5:
3 a... 3.353 5.3...
w 3.....52 5.5..
o 5.353 5.3:
0 3.3803 5.5:
m 3.353 5.5...
< 0:53:03. 00.5.:
3.5 m 35%:
3.3 < 35%:
3.5.53.0... v.0

(i)

) April 2007 sample. Each vertical bar represents a

different target virus group.

(

Figure 5.4 (Continued)

113

May 2007 Sewage

 

 

 

< 25.6.35m 5.86
92.3w

3.3.6...

3.... 5.2202
3325...... 0..
353.3 58:...

o 3.383. 5.5...
m 3.3.8. 5.5...
< 255.3". 5.5....
«2.3633... 5.5:
n Q5385... 5.5...
a 253.25 5.5....
o 355.05.". 55:...
m 3.535m 5.5....
< 3.39.2.5 5E3...
3.5.5.00 5.5....
3.52.3. 5.6:...

S. 0%. «3.5.52 5:5...
9. 8.. 3.353.. 5.5....
m. 5.353. 5E3...
o 3.383% 5.5....
0 «5.6.53. 5E3...
m 3.32.03. 55:...
< 3.3.53. 5:5...
«ES m 55.32..
95> < 55.31..
3.3.8.6... V...

0)

Each vertical bar represents a

(j) May 2007 sample.

different target virus group

Figure 5.4 (Continued)

114

June 2007 Sewage

 

 

 

< 3.....55 5.56
0.5285

3.323..

35> 52:02
355.85.... 0..
3.3.05. can—2...

0 3.5.3. 5.5:
m 3.5.5 5:5...
< 3.3-.3. 5.5...
3.2.550... :35...
m 3.3890... 5.5:
o 3.....55 5.5.:
0 3.355 5.5...
m 5.835 5:5...
< 3.55m 56......
255.580 55.:
253:3 5:3...

3. 8... 3.353. 5:5...
8 2.... 3.853. 5.5....
m. 3.353. 5.5...
a 0.5353. 5.5....
0 3.859.. 5.5...
m 3...,28< 5.5...
< 3.39.03. 5:5...
3.5 5 5.5.2.9...
3.3 < 2.58:
3.3.8.3.. 5

(k)

Each vertical bar represents a

. (k) June 2007 sample.

Figure 5.4 (Continued)

different target virus group.

115

July 2007 Sewage

 

 

< 3.355 5.5....
25an

356......

3.3 5332
3.6.5.3.. 0..
«5.20... .325...

0 3.358. 5.5..
m 3.323. :25...
< «2.6.3. 5.5..
3.6538... 5......
m 3.3890... =95...
0 3.55m :aEax
0 3.55.5 5......
m 3.6.25 5.5..
< 3.30.8.5 :25:
52.55.00 58:...
3.652 55...

S 8.. 3.653. 5.5:
9. 8.. 3.553. 5.5..
m 3.65.2 5.5..
o 5.653. 5.5..
0 3.653. 5.5..
m 5.653. 5.5..
< 3.65.... 5.5..
3.3 m 258$
3.3 < 5.582.
3.65.3. v...

(1)

(1) July 2007 sample. Each vertical bar represents a

different target virus group

Figure 5.4 (Continued)

116

August 2007 Sewage

 

 

< 3.32mEN cflfiw
3.385

3.32.3.

3.3 £53.02
322.82.... 0..
3.39.0... caEsx

o 3.330”. 5.5:
m 3.353. 5E3...
< 3.3on 58::
«Eggnog... :25:
m 3.3380: 55::
o 3.3225 5:5...
o 3.30.2..m :95:
m 3.322..m SE3:
< 3.30.2.5 :25:
3.39.200 cuss...
3.32.3 5:5:

3 on». 3:59.03~ :uEax
ow 0%. 3.39.03. 58::
w 3.39.31 52:...
o m230cou< caEnz
0 3.32.03. 5&3...
m ma_>o..ou< 5E3:
< 3.39.03. cups...
3.5 w 3:30:
3.3 < 3.53..
3.32.820... xm

(In)

Each vertical bar represents a

(In) August 2007 sample.

Figure 5.4 (Continued).

different target virus group

117

For the other samples, identification of the putative pathogen(s) present was more

evident. In the September 2006 sample, the most likely viruses to have been present were

a member of the Human Adenovirus C group (p value = 0.00257, logz mean signal

intensity = 0.14788) followed by a member of the Human Adenovirus A group (p value =

0.0031, logz mean intensity = 0.12599). For the October 2006 sample, the most likely

viruses present were a member of the Human enterovirus group D group (p value

0.00411, logz mean intensity = 0.217) followed by viruses bearing a homology with

Simian enterovirus group A (which includes members of the Human enterovirus group
E). In the November 2006 sample, the most likely viruses present were a member of the
Human enterovirus group B viruses (represented on the array by Simian enterovirus

group A) with Human enterovirus group A and astroviruses being potential agents as

well (p values = 0.00113, 0.0159 and logz mean intensity = 0.221 and 0.273

respectively). Multiple viruses were found to be present in the sample from December
2006 — Human enterovirus group A, Human astroviruses, Human enterovirus group B
(represented on the array by Simian enterovirus group A sequences) Human adenovirus

type 41 and Human enterovirus group B.

Similarly, in the January 2007 sample, the putative viruses present as indicated by

DetectiV were Human Adenovirus type 41, Human Astrovirus, Human Enterovirus group

A, Human enterovirus group B (represented on the array by Simian Enterovirus group A),

118

Human Herpesvirus type 5, Norwalk virus, Human Enterovirus D, Human Enterovirus B,

Sapovirus, and Human Adenovirus B.

5.4.4 Comparison of probable viral target detection using positive fraction
analyses and software analyses with PCR and sequencing confirmation

Based on the t test results, the most probable virus present in each sample, as
determined by the DetectiV software package and utilizing a p value cut off of less than
0.01 and a log mean intensity value of at least 0.5, are listed in table 5.4 and compared
against the probable identification obtained through positive fraction analyses. In general,
DetectiV was more conservative in proposing the presence of a viral target compared to
positive fraction analyses. Nevertheless there is agreement between the two methods as
shown in table 5.4, particularly for adenovirus and astrovirus targets. Polymerase chain
reaction primers and amplified products appear to be more sensitive, detecting viruses
that the microarray could not pick up in several months and identified 13 different types
of viruses with Adenovirus type 41 and Astroviruses type 7, identified most frequently
(in 4 and 7 months, respectively, table 5.4). The enteroviruses (including enterovirus A-
2, Coxsackieviruses A10, A20 and A16) were identified during 7 of the 13 months

surveyed.

Proposals have been put forth for using DNA microarrays as an environmental
detection tool and possible biodefense tool (Pannucci et al., 2004; Sergeev et al., 2006).
To date, only a few examples exist for the application of microarray technology on

environmental samples and even fewer as a pathogen detection tool (Call et al., 2003;

119

Zhou, 2003; Sergeev et al., 2004; Maynard et al., 2005; Lee et al., 2006; Quinones et al.,
2007; Miller et al., 2008). One drawback of the use of microarrays for pathogen detection
is the lengthy hybridization time, ofien more than 12 hours. In order to reduce the time to
result, several modifications have been suggested. Examples of modified microarray
technology include Barlaan et al. who combined electric field-driven migration of nucleic
acid targets to specific test sites with a detection microarray. Using this electronic
microarray they are able to achieve hybridization results in minutes as opposed to the
usual hours. Barlaan et al. applied this technology to the detection of harmful algal
blooms in coastal and microcosm environments (Barlaan et al., 2007). Ahn et a1. likewise
examine harmful algal blooms with a modified microarray that is visualized through fibre

optic bundles, allowing the hybridization assay to be carried out in-situ (Ahn et al., 2006).

5.5 Conclusion

Based on the results obtained, we conclude that there is a great diversity in the
types of human enteric viruses circulating among a given community. Generally, the
most frequently detected family of viruses year round fiom sewage were the group A
Human enteroviruses from the picomaviridae family. The second most frequently
detected virus family were the Human adenoviruses from the adenoviridae family.
Human caliciviruses as represented on the array by the Norovirus, human astrovirus and
sapoviruses were detected during the winter months, indicating a possible seasonality to
these groups of viruses as has been previously reported (Mounts et al., 2000). From our
results it appears that human pathogenic viruses were present among community during

the periods of sampling at levels high enough to be detected by hybridization unto an

120

array even though no outbreaks of these diseases were reported among the community.
This corroborates research indicating that viral gastroenteritis diseases are frequently
under-reported (Majowicz et al., 2005; Day and Sutton, 2007). The presence of viruses in
the sewage without reports of outbreaks could also represent asymptomatic shedding of
viruses (Gallimore et al., 2004; Mendez-Toss et al., 2004; Nwachuku and Gerba, 2006;

Monica et al., 2007).

To our knowledge, this study demonstrates the first use of a microarray to
characterize human pathogenic viruses in the environment. Target detection microarrays
are increasingly being accepted as a viable tool for environmental monitoring. Several
hurdles still remain that prevent the widespread use of arrays in environmental
application. The first has to do with the presence of inhibitors in the sample which can
lower labeling efficiency. Second, is the need for extensive validation of the signals
observed via alternative means. Third is the lack of software to analyze, visualize,
normalize and carry out significance testing on pathogen array data. Also, the cost
associated with designing and implementing a microarray detection system remains too
high for routine use. Despite these drawbacks, the potential applications for these
microarrays are tremendous and include multiple pathogen detection, community health
screening and monitoring, bioterrorism surveillance, monitoring the expression of key
metabolic genes, and pathogen discovery. The ability to screen community health via
excretion of viral pathogens in urine and feces, means that we could improve our

understanding of exposure, disease and ultimately prevention strategies. In addition, we

121

are working on reducing the cost and hybridization time by an order of magnitude as well

as eliminating the need for a scanner for yes or no type of diagnosis uses.

CHAPTER 6 TESTING THE SENSITIVITY OF THE

MICROARRAY WITH POLIOVIRUS SPIKED INTO SEWAGE

6.1 Introduction

In order to determine the level of sensitivity of detection for the virus microarray,
a series of hybridizations were carried out to determine the signal response when
increasing levels of poliovirus LS-C-l was spiked into the cell culture supernatant of
flasks that had developed cytopathic effects after being exposed to raw sewage
concentrates. A range of concentrations of poliovirus LS-C-l between 12 and 12,000
plaque forming units was chosen to express levels of virus reflective of those thought to
be present in sewage and the environment. Poliovirus was chosen for its ease of culture as
well as enumeration. In addition, vaccination using the live attenuated strain of poliovirus
had been discontinued since 2000 and a previous year long survey of the same plant’s

sewage never indicated the presence of poliovirus above the detection criteria.

6.2 Literature Survey

6.2.1 Levels of Human Viruses in Water

Several studies have been conducted to determine the concentration of human
viruses in various environmental matrices. Concentrations of human viruses are highest
in raw sewage, approximately 10 to 100 times lower in wastewater effluent and hundreds

to tens of thousands of times lower in receiving waters. A study by Albinana-Gimenez et

122

a1 (2006) measured levels of human polyomavirus, adenovirus and hepatitis E virus and
found that the average concentration of JC polyomavirus in a Spanish river was 26
genome copies per liter and that human adenoviruses have an average concentration of
400 genome copies per liter using quantitative polymerase chain reaction detection.
Hepatitis E virus was frequently detected at low levels in urban sewage, biosolids and
sewage containing swine feces but was not observed in the river water samples

(Albinana—Gimenez et al., 2006).

Another study of treated wastewater by Gantzer et al. (1998) found that the levels
of infectious enterovirus ranged from <1 most probable number of cytopathic units
(MPNCU) per liter to 4 MPNCU / liter. The investigators also observed that the
percentages of samples testing positive for the enterovirus genome were significantly
higher than those for infectious enteroviruses and attributed this finding to either the
presence of noninfectious enteroviruses or to the presence of infectious enteroviruses that

do not multiply in BGM cell cultures (Gantzer et al., 1998).

Two studies by Haramoto et al. (2007, 2008) which looked at the concentrations
of human adenoviruses and sapoviruses in the aquatic environment found that the enteric
serotypes of HuAst were detected at the concentration of 7.3 -1500 PCR-detection
units (PDU) per milliliter in raw sewage, 0.00060 - 4.1 PDU per milliliter in secondary-
treated sewage before chlorination, 0.0018-7.0 PDU per milliliter in river water, and

0.032-6.l PDU per milliliter in seawater (Haramoto et al., 2007). On the other hand, the

. . . . 3 .
concentration of sapovuuses 1n influent ranged from 2.8 x 10 to 1.3 x 105 copies per

123

liter, showing a higher value in winter. In all, seven (58%) of 12 effluent samples tested
were positive for sapoviruses, as were 23 (64%) of 36 river water samples collected from

three sites along the Tamagawa River (Haramoto et al., 2008).

One study by Jiang et al. (2001) which looked at the levels of human
adenoviruses in coastal waters detected similar levels of virus as those determined by
Haramoto et al. (2007). J iang et al.(2001) detected 880 to 7,500 most probable numbers
of adenovirus genomes per liter of water. The investigators concluded that the prevalence
of adenoviruses made it a useful indicator of human viral fecal pollution in surface and

environmental waters.

6.2.2 Sensitivity of Various Virus Detection Methods

Several papers have been published looking at the method sensitivity of various
PCR-based virus detection methods. A detection sensitivity of 0.04 PFU has been
reported for hepatitis A virus using immunomagnetic capture reverse transcription
polymerase chain reaction in both spiked finished water and environmental samples
(J othikurnar et al., 1998). In comparison, the detection sensitivity using conventional
reverse transcription polymerase chain reaction was found to be ten times less sensitive at
0.4 PFU (Jothikumar et al., 1998). Another studying using reverse transcription
polymerase chain reaction reported that poliovirus could be detected even in the presence
of 0.5 milligrams of humic acid or 5.0 milligrams of fulvic at the detection limit of 0.06
plaque forming units (Ijzerman et al., 1997). This study however was carried out using

spiked poliovirus in sodium phosphate buffer and might not be indicative of the detection

124

sensitivity in environmental matrices. A study by Green and Lewis (Green and Lewis,
1995) reported that the sensitivity of detection of enteroviruses in wastewater was
between 0.02 and 0.2 plaque forming units per sample. Wyn-Jones et al. found that
detection using the polymerase chain reaction gave comparable results to cell culture and
in a much shorter time period. Their detection limit was reported to be 5 plaque forming
units (pfu)/sample for enteroviruses in river and marine waters (Wyn-Jones et al., 1995).
Schwab et al. reported a detection limit of 2 pfu / sample using immunoaffinity
concentration and reverse transcription for enteroviruses in fecal-contaminated surface
waters (Schwab et al., 1996). Using dot blot hybridization, Pinto et al reported a lower
sensitivity of detection compared to polymerase chain reaction-based detection methods.
Their limit of detection for astroviruses in environmental water was determined to be 3

pfu per sample (Pinto et al., 1996).

6.3 Materials and Methods

6.3.] Poliovirus Plaque Assay

Agar overlay plaque assays were carried out to determine the titres of poliovirus
for use in the spiking assay to test the sensitivity of the microarray. African Green
Monkey (BGM) cells were grown to 80-90% confluence, washed and then exposed to
serial dilutions of stock poliovirus LS-C-l. Diluted viruses were allowed contact with the
cells for 1 hour with occasional rocking to prevent cells from drying out. Excess diluent
was decanted. 10 milliliters of virus agar overlay solution was added to each 25 square
centimeter flask and allowed to set. Virus agar overlay solution consists of 1x MEM, 100

U/ milliliter of Penicillin, 100 microgram/ milliliter of Streptomycin and 250

125

nanogram/milliliter of F ungizone, 0.003% neutral red, 1% sodium bicarbonate, 1.1 % w/v

agarose. Cells were incubated at 37 °C and examined daily for the presence of plaques.

Flasks with plaque counts of between 10 and 100 were averaged and the concentration of

poliovirus LS-C-l in stock was determined in plaque forming units (pfu) per milliliter.

6.3.2 Wastewater Sample Collection and Cell Culture

A 6 liter samples of untreated sewage was collected in January 2008 from the East
Lansing Wastewater Treatment plant. The sample was brought back from the laboratory
and 15% buffered beef extract (Difco Inc) was added to give a final concentration of

1.5% beef extract and 0.05 molar glycine. The solution was stirred for 30 minutes before

2.5 molar lFeCl3 was added to a final concentration of 2.5 millimolar. The pH of the

solution was lowered to 3.5 and the solution was stirred a further 30 minutes. Viruses
were pelleted by centrifugation at 2,500 x g for 15 minutes. The viral pellet was
resuspended in 90 milliliters of0. 15 molar sodium phosphate dibasic (pH 7.0) solution by
agitation on an orbital shaker set to 160 revolutions per minute. Once the pellet was
dissolved, the pH of the sodium phosphate was raised to between 9.0 and 9.5 and placed
on the orbital shaker for a further 10 minutes. Solid particles were pelleted by
centrifugation at 10000 x g (Beckman model J2-HC). The supernatant was collected and
filtered through a 0.22 micrometer syringe filter and supplemented with 100 U/ milliliter

of Penicillin, 100 microgram] milliliter of Streptomycin and 250 nanogram/milliliter of
Fungizone. The pH of the virus concentrate was neutralized to 7 .0 and frozen at —80 °C

until placed on cell culture.

126

Twelve milliliters of virus concentrate was used to infect BGM cells grown to

approximately 80 — 90% confluence. Virus concentrate was allowed 120 minutes of
contact with the BGM cells before being discarded. Cells were incubated at 37 °C until

development of cytopathic effects were observed. Infected cells were harvested by
mechanical lifting using a sterile cell scraper. All cells and free viruses in the media were
concentrated by centrifugation through an Amicon Ultra 100K centrifugation column

(Millipore, Billerica, MA).

Virus nucleic acid was extracted using the Qiagen viral RNA mini kit following
the manufacturer’s instructions. in the same manner as described above for the ATCC
strains of viruses. Briefly, 140 microliters of virus concentrate was added to tubes
containing 560 microliters of lysis buffer and incubated at room temperature for 10 min
with intermittent mixing. 560 microliters of ethanol was added and the solution mixed
before passage through a DNA binding column. Columns were washed with washing
solution and eluted with 2 rounds of elution solution using 50 microliters of DNase and

RNase free water each time.

6.3.3 Poliovirus spiking

A range of concentrations of poliovirus as determined from the plaque assay
described above was used to spike into the cell culture supematants from the wastewater
sample cell culture. Virus particles were recovered and concentrated from the media by

centrifugation through an Amicon Ultra 100K centrifugation column (Millipore,

127

Billerica, MA). Virus nucleic acid was extracted using the QIAamp viral RNA mini kit
for both DNA and RNA viruses following the manufacturer’s instructions. The QIAamp
viral RNA mini kit has previously been demonstrated to be capable of isolating viral
DNA as well as viral RNA(Kleines et al., 2003). Briefly, 140 microliters of virus
concentrate was added to tubes containing 560 microliters of lysis buffer and incubated at
room temperature for 10 min with intermittent mixing. 560 microliters of ethanol was
added and the solution mixed before passage through a DNA binding column. Columns
were washed with washing solution and eluted with 2 rounds of elution solution using 50

microliters of DNase and Rnase free water each time.

6.3.4 Preparation of Samples for Hybridization

Viral RNA was labeled with fluorescent dyes by a semi-random primed labeling
with Sensiscript III reverse transcriptase as described by Wang et al. (Wang et al., 2003).
Half a microgram (0.5 pg) of viral nucleic acid was used as a template for the generation
of cDNA with a discrete 5’ terminal consisting of the sequence
(5 ’-GTTTCCCAGTCACGATC-3’) using the semi-random primer A:
(5 ’-GTTTCCCAGTCACGATCNNNNNNNNN-3 ’). Next primer B:
(5’-GTTTCCCAGTCACGATC-3’) and Qiagen Hotstart Taq polymerase was used to

amplify the generated cDNA and label it with amino-ally] dUTP for 40 cycles using the

following profile: 30 seconds at 94°C, 30 seconds at 40°C, 30 seconds at 50°C, 60
seconds at 72°C. To label viral DNA, 0.5 micrograms of virus DNA was first digested

with IU of DPNII restriction endonuclease at 37 °C for 1 hour. The digested DNA was

incubated at 37 0C for 2 hours with Klenow enzyme and Primer A to generate

128

complementary DNA with a discrete 5’ terminal. Next primer B and Qiagen Hotstart Taq

polymerase was used to amplify the generated cDNA and label it with amino-ally] dUTP

for 40 cycles using the following profile: 30 seconds at 94°C, 30 seconds at 40°C, 30

seconds at 50°C, 60 seconds at 72°C for 40 cycles.

Labeled DNA and RNA were coupled separately with Cyanine dye 3 and Cyanine
dye 5. This reaction was carried out in the absence of light and using 1 molar sodium
bicarbonate (pH 9.5) as a coupling buffer with a l-hour incubation. Following coupling,
uncoupled dye was removed using the QIAgen PCR purification kit and the labeled virus
DNA and RNA were dessicated in a DNA 120 SpeedVac (Thenno Fisher Scientific, MA,
USA) for 1.5 hours. The sample preparation, labeling and hybridization procedure is

outlined in Figure 5.1.

6.3.5 Hybridization of Samples

Microarray hybridization was performed as described previously (Wick et al.,
2006). The microarrays were hybridized and washed in a M-2 microfluidic station
(Invitrogen, Carlsbad, CA, formerly Xeotron Corporation, Houston, TX) using a flow
rate of 400 microliters / minute. The hybridization buffer was 6x- SSPE, 25% formamide,
0.4% Triton X-100. Chips were pre-hybridized with 6x- SSPE, 0.2% Triton X-100 and

then with hybridization buffer for 2 min each.

129

Labeled targets were resuspended in 50 microliters hybridization buffer,

denatured at 95°C for 3 minutes, cooled on ice for 1 minute, filtered through a 0.22

micrometer Costar spin filter and then hybridized to the chip for 14—15 hours at 20°C.

The chip was scanned with a GenePix 4000B laser scanner (Axon Instruments,
Union City, CA). All solutions were filtered through a 0.22 micrometer filter to prevent
clogging of the microfluidic channels. The high stringency wash buffer was degassed
under vacuum. All arrays were imaged with a GenePix 40003 scanner (Molecular

Devices, Sunnyvale, CA) and GenePix Pro software.

6.3.6 Data Analysis Using DetectiV

The DetectiV software package was used to visualize, normalize and test the
significance of the microarray hybridization data (Watson et al., 2007). DetectiV was
used to generate bar plots of the viral microarray signals following normalization against
the median signal values for each sample. DetectiV was also used to carry out statistical
t-test comparisons of the hybridization signal values between different virus groups in

order to determine which groups of virus target had statistically significant probe signals.

6.4 Results

6.4.1 Viruses Present in January 2008 Control Sample
Using DetectiV and applying the detection criteria of p value less than or equal to

0.01 and log average signal greater than or equal to 0.5, five virus groups were identified

130

in the January 2008 control sample (sample with no poliovirus). The five groups of
viruses were Human adenovirus type 41, Human astrovirus, Human adenovirus group B,
Human enterovirus group D and Human enterovirus group A. Table 6.1 lists the p values

and log mean signal values for the ten most likely targets for the January 2008 sample.

6.4.2 Sensitivity Testing using Spiked Poliovirus LS-C-l
The sensitivity of the microarray was tested using a range of concentrations of
poliovirus LS-C-l. A ten-fold serial dilution of poliovirus stock to give final

concentrations of poliovirus of between 1.2 and 12000 plaque forming units resulted in

logz average signal readings for poliovirus of between 0.258 and 1.237 when DetectiV

was used to analyze the hybridization results. This translated into a detection sensitivity

for the array of approximately 59 poliovirus plaque forming units in order to achieve a

logz mean signal intensity of 0.5 in this sample. Figure 6.1 shows the plot of logz average

signal against poliovirus plaque forming units.

131

Table 6.1 p-values and log; average signal for the ten most likely viruses present in
the January 2008 sewage sample

 

Virus p value

 

 

 

 

 

 

 

 

 

 

 

Logz average signal
Human adenovirus type 41 0.0003939283 0.7321386
Human Astrovirus 0.0008543337 0.9832780
Human Adenovirus E 0.0038398798 0.3784977
Human Enterovirus D 0.0056843549 0.5509884
Human Enterovirus A 0.0062747645 0.7452414
J C Polyomavirus 0.0076384783 0.2057239
BK Polyomavirus 0.0141987829 0.2330732
Human Enterovirus E 0.0160536991 0.3271534
Sapovirus 0.0162622777 0.2643265
Human Adenovirus D 0.0174830642 0.3644236

 

 

 

132

 

 

Log-transformed average poliovirus signal versus
number of poliovirus plaque forming units added

 

 

 

 

 

 

 

 

Log transformed average
srgnal
O O
O) CD —I
W a
. ‘ 1

l
i I
l
l
\
l
L -

 

 

1 10 100 1000 10000 100000
PFU virus

 

 

 

Figure 6.1. logz-transformed average poliovirus signal versus number of poliovirus
plaque-forming units (PFU). PFU virus (x axis) is provided in logarithmic scale.

133

6.5 Discussion

6.5.1 Viruses present in January 2008 sewage

From the results, the January 2008 sewage sample contained 5 types of viruses —
Human adenovirus type 41, Human astrovirus, Human adenovirus group B, Human
enterovirus group D and Human enterovirus group A. This was similar, though no
identical, to the numbers and types of viruses obtained in the previous year (January
2007) — Human astrovirus, Human enterovirus group A, Human enterovirus group D,
Human enterovirus group B, norovirus, sapovirus, Human adenovirus group B and
Human adenovirus type 41 signals were observed in that sample. Out of the 5 groups of
viruses present in the January 2008 sample, 4 were also present the previous year. Human
adenovirus group B was present in the January 2008 sample but absent in the January
2007 sample. Human enterovirus group B, Human adenovirus group B, noroviruses and
sapoviruses were present in the January 2007 sample but absent in the January 2008

sample.

Three groups of viruses, Human enterovirus group A, Human astrovirus and
Human adenovirus type 41, were present in 4 samples collected during winter months
(December to February). These three groups of viruses might represent viruses that are
either constantly being circulating within the community or else representing an endemic
presence in the community. Table 6.2 illustrates the distribution of virus groups among
the 4 samples collected during winter months. Two groups of viruses were present in two
of the four months, Human enterovirus group B was present in the December 2006 and

January 2007 sample. Human enterovirus group D was present in the January 2007 and

134

January 2008 sample. These might reflect groups of viruses that are less widespread
among the community during winter months. F our groups of viruses were present in only
one of the 4 samples, norovirus, sapovirus, Human adenovirus group B and human
enterovirus group B. These might represent viruses that are only transiently present in the

community.

6.5.2 Spiking sensitivity
Based on the results fi'om spiking poliovirus in raw sewage, the microarray

showed a sensitivity of approximately 59 plaque forming units in order to achieve a log

mean signal intensity of 0.5, which was the detection criteria used in this study.
Compared to reverse transcription polymerase chain reaction and other reported methods,
the detection limits reported in this study were low. Polymerase chain reaction detection
of viruses in environmental water was widely reported to be at least a hundred-fold
higher. The sensitivity of the array could potentially be improved (i.e < 59 plaque
forming units) if the sample contained fewer types of target viruses as there would be less
competition for labeling and hence more signal from the viruses that are present.
Alternatively, if specific primers were used for labeling target individual target viruses.
Sensitivity tests would have to be carried out for different water matrices to determine the
effect that target concentration vis-a-vis background concentration of viruses would

affect signals on the array.

135

6.5.3 Conclusions

Based on the sensitivity test using poliovirus spiked in sewage, the virus
microarray is able to detect the presence of viruses at concentrations commonly present
in sewage.The intensity of the signal however is potentially affected by the presence of
inhibitors for the labeling of viral nucleic acid, the coupling of the florescent dye to the
label nucleic acid, and the proportional concentration of the target virus in relation to the
background nucleic acid levels which serve as competition for labeling. Inhibition issues
were partly avoided by performing a cell culture incubation step prior to virus labeling
and hybridization. This would help to biologically amplify the virus and dilute any

inhibitors that might be present in a sample.

In addition, it was observed that three of the winter month samples analysed
during the the year-long survey of raw sewage and one other sample collected in January
2008 used for the sensitivity test for poliovirus spiked in sewage showed the presence of
three groups of viruses that were present in all four samples. This might indicate that
these viruses were constantly circulating among the community during the winter months
and also potentially endemic to the community since they were present in a sample taken
a year later. This illustrates one potential use of the virus microarray as a means of
monitoring a community’s wastewater virus signature in order to gain insights into the
range of viruses that are continuously circulating within a community and groups of

viruses that are found more transiently within human wastewater.

136

Table 6.2

Distribution of virus groups among winter month samples

 

 

 

 

 

 

 

 

 

Month Virus group present in Virus group present in Virus group present in
all 4 winter months 2 out of 4 winter 1 out of 4 winter
months months
December Human enterovirus
2006 group B
January . Human enterovirus norovirus
2007 Adenovrrus type 41 group B .
Human astrovirus . sapovrrus
Human enterovrrus .
Human enterovirus group D HrérlrlranBadenovrrus
group A gr p
February Human enterovirus
2007 group B
January Human enterovirus
2008 group D

 

 

 

 

137

 

CHAPTER 7 SUMMARY AND CONCLUSION

Medically important viruses were first noted as part of the environmental
“malaises” early in human history (often described with symptoms such as jaundice) but
it was advances in cell culture, electron microscopy and immunology that spurred the
discovery and characterization of human viruses. The first isolations of viruses from
water came in the 19505 and 1960s for surface waters and drinking waters, respectively
(Gerba, 1989). Yet it had long been understood that enteric viruses such as poliovirus
were shed in the feces and thus by association present in sewage and sewage-polluted
waters. Our conventional definition of environmental virology has primarily focused on
enteric viruses and contaminated drinking water, fecal-oral transmission, and associated
person-to-person transmission. Likewise, the management and control of waterborne
viruses has focused on disinfection of drinking water and vaccinations. Despite the
tremendous improvements in water and sanitation management fueled by a better
understanding of the nature of viruses, emerging viruses such as the polyomaviruses and
reemerging epidemics of age-old viruses such as poliovirus as well as concerns
associated with intentional use of eradicated viruses such as the smallpox virus, challenge
our conventional definition of “environmental virology” and traditional approaches to

control.

The global outbreaks of Severe Acute Respiratory Syndrome (SARS) and Avian
Influenza (AI) highlight the degree of vulnerability that high-density urban populations
face when threatened by novel, unanticipated viral pathogens. This is further

underscored by security fears brought on by recent acts of terrorism both in the United

138

States and abroad. Less sensational but equally serious outbreaks of many other viruses
like norovirus, hantavirus, and West Nile virus have been documented worldwide and are
on the rise. Rotavirus-induced diarrhea is still the most prevalent infant killer in many
developing nations causing an estimated 140 million cases worldwide and killing almost

600 000 people annually (Parashar et al., 2003).

Historically, the importance of protecting one’s drinking water supply has been
well documented and recognized. Poisoning or contaminating an enemies’ water supply
has been practiced in warfare since at least the fourth century BC. Less deliberate acts of
contamination occur more frequently due to industrial accidents, inclement weather, poor
infrastructure, weak enforcement of regulations or operator neglect. While animal
wastes have been implicated in bacterial and parasitic outbreaks the viruses remain
associated with some of the most serious health consequences such as the outbreak of
viral hepatitis E in Kanpur India in 1991 which affected an estimated 79 091 people
(Naik et al., 1992) with 30% mortality in pregnant women in the first trimester. The
recent widespread poliovirus outbreaks throughout Africa are likely in part due to
contaminated water and the inadequate sewerage and wastewater treatment (Pavlov et al.,

2005)

More recently, attention has also focused on the need to protect recreational water
sources (Wade et al., 2003; Standish-Lee and Loboschefsky, 2006). Fresh and salt-water
sources represent an important recreational resource, especially to economies that rely

heavily on tourism. In addition, the increasing scarcity of pristine water sources has

139

meant that the water cycle is being short circuited in order to provide adequate water for
drinking, recreation, power generation, agriculture and industrial processing. The
assessment of the impairment of waterways for the various uses based on the “indicator
bacteria” and Escherichia coli have not provided enough specificity in regard to health
risk, sources of the pollution, identification of the responsible party and control. This has

fueled a demand for advanced pathogen detection.

In the twenty-first century, viral diseases have changed the landscape of medicine.
Acquired Irnmuno Deficiency Syndrome (AIDS) now infects millions of people
worldwide and up to 30% of the populations in Afiica. Waterbome diseases will be
particularly devastating to these individuals and the list of potential waterborne viral
agents is growing. Certain microbiological advances like the polymerase chain reaction
(PCR) and microarray technology may provide the tools necessary for monitoring any

new agent of interest.

The detection of viruses in water and other environmental samples constitutes
special challenges. The standard method of detection of viral pathogens in environmental
samples uses assays in mammalian cell culture. The infected cell cultures undergo
observable morphological changes called cytopathogenic effects (CPEs) which are used
for the detection of viruses. Even though many viruses are culturable in several cell lines
and are thus detectable by their development of CPE cell culture, there are several
viruses, like enteric waterborne adenoviruses types 40 and 41, which do not produce clear

and consistent CPE. Other viruses, like waterborne caliciviruses, have not yet been

140

successfully grown in cell cultures. Conventional cell culture assays for the detection of
viruses in environmental samples have limited in some cases the specificity and can be
labor-intensive and time-consruning. However, detection of infectious viruses and ability
to process relatively large volumes of concentrates (up to 30 ml) means greater

sensitivity.

The limitations of the cell culture method were highlighted in Chapter 4. We
found that PCR detected more positive virus samples than conventional cell culture (30
positives and 26 positives respectively out of 58 samples tested). We attributed this
disparity to the presence of non-CPE forming enteric viruses in the sample. In addition, in
our study we observed that while higher number of adenovirus and rotavirus-positive
samples were detected at the Silver Beach location compared to the Washington Park
location, the enteroviruses and enterococcus surface protein (ESP) PCR results would
have led to the opposite conclusion. In our study we proposed that the contradictory
results could be explained by the different survival rates of the different indicators,
underscoring the need to screen any environmental sample against multiple target

indicators before reasonable conclusions can be made. Lastly we observed that while a

positive virus PCR result could be correlated with a positive cell culture result (x2 = 4.66;

1 degree of freedom), not all the cell culture positive samples could be identified through
PCR. We believe that while the inability to assign identities to these samples using the
PCR primer sets we had available could be attributed to some extent by the presence of
inhibitors during the polymerization chain reaction in the sample (as documented through

realtime PCR analysis of the same set of samples by Xagoraraki et al. (2007)) it is also

141

likely that the inability to achieve a PCR identification of these samples could also be due

to the presence of cultivatable viruses that we did not have primers for.

At present, the monitoring of public health occurs at the individual patient level.
The highly disseminated nature of the public health system, however, means that it takes
either a long time or a massive influx of cases before a disease outbreak is recognized.
The trend towards increasingly urbanized and dense city living and the more frequent
travel between communities necessitates that community health monitoring adopt a more
proactive preventative role instead of merely recording and reporting disease data. In
order to do so, there needs to be tools which are able to screen for the large panel of
possible viral pathogens which are representative of the pathogen loads present in the
larger community. To meet this requirement it becomes logical to monitor the
community’s sewage using microarrays designed to detect the presence of waterborne

pathogens.

In the microarray experiments (Chapter 5), it was demonstrated that
oligonucleotide microarrays provide an approach for screening hundreds of pathogens in
environmental samples. While microarrays have previously been used to perform
environmental analyses, namely describing microbial communities and elucidating levels
of gene expression among microbial consortia, use of this platform for the detection of
viruses in polluted waters, which cause human disease represents a novel use of this
technology. This is especially important because unlike many of the other fecal

indicators, no suitable indicators of viral fecal pollution have been forthcoming. In this

142

research, a community’s virus infection profile was characterized through a series of
samples taken over a period of 13 months. RNA viruses were more prevalent than DNA
viruses, and there was some degree of seasonality. The next steps are to begin examining
disease in the community in relationship to this biomonitoring of the infection in the

population.

In addition, we were able to validate two pieces of software, one used to design
the probes for the microarray and the second used to visualize and validate the
hybridization signals generated by the arrays. The high degree of specificity
demonstrated by the virus microarray chip from the very onset in hybridization
experiments with laboratory and commercially purchased strains of viruses demonstrates
that the OligoArray 2.1 software is extremely proficient at selecting probes with a high
degree of specificity under the conditions set. The other piece of software, DetectiV, has
previously only been applied to analyze hybridization signals generated during the
analysis of clinical specimens for the presence of viruses. Nevertheless the broad
applicability of the program has been demonstrated by our use of this software, with only
a few modifications to the acceptance criteria, to analyze signals generated from
environmental samples which are inherently more complex compared to patient clinical

samples.

Chapter 6 describes the study in which poliovirus was spiked into the cell culture
supernatant of Afiican Green Monkey kidney cells that had been exposed to raw sewage

concentrates. This study was used to determine the level of sensitivity needed to produce

143

a positive signal on the microarray. In this study, the sensitivity of the virus microarrayto

poliovirus spiked into sewage was found to be approximately 59 plaque forming units in

order to produce a log; mean signal of 0.5. This level of sensitivity is reflective of the

concentration of viruses commonly found in less than 100 milliliters of raw sewage. Also
Chapter 6 reports the observation that three groups of viruses, Human adenovirus type
41, human astroviruses and Human enterovirus group A, were present in all 4 samples
collected between December 2006 and February 2007 as well as January 2008. This
suggests that these three groups of viruses are either endemic, since they occurred in
samples that were taken as much as a year apart or that they are widely and continually

being spread among the community.

The objective of this dissertation is to compare the current methods used to test
for viruses in the environment, namely cell culture, polymerase chain reaction (PCR) and
integrated cell culture-PCR against the results obtained using a microarray to detect
viruses through labeling and hybridization post cell culture. Our analysis of the
cultivatable and PCR-identifiable viruses collected from two recreational beaches along
the Great Lakes indicates that this is likely to be so. We have shown that conventional
cell culture is limited in its ability to detect a wide range of viruses due to the host
specificity of most animal viruses. We have also documented that conventional molecular
detection methods like the polymerase chain reaction suffer from inhibition problems and
are extremely specific in their target range, making high throughput screening of samples
for any one or a combination of the more than a hundred different varieties of human and

animal viruses impossible. Modifications to the conventional cell culture method have

144

helped to overcome some of these issues. One such modification is to pair cell culture
with the polymerase chain reaction (PCR) in order to over come inhibition issues as well
as to increase the speed and sensitivity of the detection of viruses in environmental
samples. This thesis describes the next evolution for virus detection in which a
microarray hybridization step is incorporated post cell culture in order to increase the
throughput for virus testing, overcoming the current limitation of the cell culture-PCR
method. Using the virus microarray, we have demonstrated that oligonucleotide arrays
are able to hybridize with complex, mixed environmental samples and may be used as a

multiple pathogen detection/screening tool.

Our analysis of human wastewater samples with the microarray indicate that a
wide range of DNA and RNA viruses are present in human feces and could potentially
survive to reach the environment if improperly treated. Viruses remain a public health
concern and should remain a priorityfor the water and health community. These bio-
nano particles are excreted in high concentration by infected individuals, have high
potency (probability of infection is high with low numbers (Haas et al., 1999)) and are
environmentally robust. The ability of both DNA viruses and RNA viruses to rapidly
evolve means new and emerging viral pathogens will need to be addressed. Pathogen
discovery and characterization, occurrence in the environment, exposure pathways and
health outcomes via environmental exposure are all issues that deserve future attention
and elucidation. This will likely follow a new microbial risk fi'amework which will
require focused research on some important properties of viral disease transmission. The

future will require models that examine community risks and provide explicit links

145

between the models currently under development for environmental exposure and

infectious disease.

146

REFERENCES

147

Ahn, S., D. M. Kulis, et al. (2006). "Fiber-optic microarray for simultaneous detection of
multiple harmful algal bloom species." Appl Environ Microbiol 72(9): 5742-9.

Akaike, H. (1974). "A new look at the statistical model identification." Transactions on
Automatic Control AC 19: 716-723.

Albinana-Gimenez, N., P. Clemente-Casares, et al. (2006). "Distribution of human
polyomaviruses, adenoviruses, and hepatitis E virus in the environment and in a drinking-
water treatment plant." Environ Sci Technol 40(23): 7416-22.

Andrei, G. (2006). "Three-dimensional culture models for human viral diseases and
antiviral drug development." Antiviral Res 71(2-3): 96-107.

Appleton, H. (1987). "Small round viruses: classification and role in food-bome
infections." Ciba Found Svmp 128: 108-25.

Bae, H. G., A. Nitsche, et al. (2003). "Detection of yellow fever virus: a comparison of
quantitative real-time PCR and plaque assay." J Virol Methods 110(2): 185-91.

Baltimore, D. (1971 ). "Expression of animal virus genomes." Bacteriol Rev 35(3): 235-
41.

Barlaan, E. A., S. Furukawa, et al. (2007). "Detection of bacteria associated with harmfirl
algal blooms from coastal and microcosm environments using electronic microarrays."
Environ Microbiol 9(3): 690-702.

Barnes, G. L., E. Uren, et al. (1998). "Etiology of acute gastroenteritis in hospitalized
children in Melbourne, Australia, from April 1980 to March 1993." J Clin Microbiol
36(1): 133-8.

Barwick, R. S., D. A. Levy, et a1. (2000). "Surveillance for waterbome-disease outbreaks-
-United States, 1997-1998." MMWR CDC Surveill Summ 49(4): 1-21.

Baxi, M. K., S. Baxi, et al. (2006). "Microarray-based detection and typing of foot—and-
mouth disease virus." Vet J 172(3): 473-81.

Beards, G. M., D. W. Brown, et al. (1986). "Preliminary characterisation of torovirus-like
particles of humans: comparison with Beme virus of horses and Breda virus of calves." J
Med Virol 20(1): 67-78.

Belsley, D. A., E. Kuh, et al. (1980). Regression diagnostics: Identifying influential data
and sources of collinearity. New York, John Wiley & Sons.

Bernhard, A. E. and K. G. Field (2000). "A PCR assay To discriminate human and
ruminant feces on the basis of host differences in Bacteroides-Prevotella genes encoding
16S rRNA." Appl Environ Microbiol 66(10): 4571-4.

148

Beuret, C. (2004). "Simultaneous detection of enteric viruses by multiplex real-time RT-
PCR." J Virol Methods 115(1): 1-8.

Bird, B. H., D. A. Bawiec, et a1. (2007). "Highly sensitive and broadly reactive
quantitative reverse transcription-PCR assay for high-throughput detection of Rift Valley
fever virus." J Clin Microbiol 45(11): 3506-13.

Bishop, R. F ., H. C. Bugg, et a1. (1996). "Serum, fecal, and breast milk rotavirus

antibodies as indices of infection in mother-infant pairs." J Infect Dis 174 Suppl 1: S22-
9.

Blackburn, B. G., G. F. Craun, et al. (2004). "Surveillance for waterbome-disease
outbreaks associated with drinking water--United States, 2001-2002." MMWR Surveill
Summ 53(8): 23-45.

Bodrossy, L. and A. Sessitsch (2004). "Oligonucleotide microarrays in microbial
diagnostics." Curr Opin Microbiol 7(3): 245-54.

Bofill-Mas, S., M. Formiga-Cruz, et a1. (2001). "Potential transmission of human
polyomaviruses through the gastrointestinal tract after exposure to virions or viral DNA."
J Virol 75(21): 10290-9.

Boher, S. B., C; Terver, D; Schwartzbrod, L (1991). " Comparison of two methods for
the recovery of rotavirus from mussels and oysters. ." Water Sciencefi and Technology
24(2): 423-426.

Bohnker, B. K. and S. Thornton (2003). "Explosive outbreaks of gastroenteritis in the
shipboard environment attributed to Norovirus." Mil Med 168(5): iv.

Bon, F., P. Fascia, et a1. (1999). "Prevalence of group A rotavirus, human calicivirus,
astrovirus, and adenovirus type 40 and 41 infections among children with acute
gastroenteritis in Dijon, France." J Clin Microbiol 37(9): 3055-8.

Boom, R., C. Sol, et al. (1999). "Improved silica-guarridiniumthiocyanate DNA isolation
procedure based on selective binding of bovine alpha-casein to silica particles." J Clin
Microbiol 37(3): 615-9.

Borchardt, M. A., P. H. Chyou, et al. (2003). "Septic system density and infectious
diarrhea in a defined population of children." Environ Health Perspect 1 11(5): 742-8.

Borchardt, M. A., N. L. Haas, et a1. (2004). "Vulnerability of drinking-water wells in La
Crosse, Wisconsin, to enteric-virus contamination from surface water contributions."
Appl Environ Microbiol 70(10): 5937-46.

Boriskin, Y. S., P. S. Rice, et al. (2004). "DNA microarrays for virus detection in cases of
central nervous system infection." J Clin Microbiol 42(12): 5811-8.

149

Bosch, A. (1998). "Human enteric viruses in the water environment: a minireview." _Im
Microbiol 1(3): 191-6.

Bruu, A. L. (2003). Enteroviruses: Polioviruses, Coxsackieviruses, Echoviruses and
Newer Enteroviruses. A Practical Guide to Clinical Virology (Second Editiom. P. S. J. R.
P. P. R. J. W. L. R. Haaheim: 44-45.

Buchen-Osmond, C. (2003). Taxonorrrvé and Classification of Viruses. Washington DC,
ASM Press.

Bull, R. A., E. T. Tu, et al. (2006). "Emergence of a new norovirus genotype II.4 variant
associated with global outbreaks of gastroenteritis." J Clin Microbiol 44(2): 327-33.

Butot, S., T. Putallaz, et al. (2007). "Procedure for rapid concentration and detection of
enteric viruses from berries and vegetables." Apnl Environ Microbiol 73(1): 186-92.

Caballero, O. L., C. L. Menezes, et al. (1997). "Highly sensitive single-step PCR protocol
for diagnosis and monitoring of human cytomegalovirus infection in renal transplant
recipients." J C lin Microbiol 35(12): 3192-7.

Call, D. R., M. K. Borucki, et al. (2003). "Detection of bacterial pathogens in
environmental samples using DNA microarrays." J Microbiol Methoda 53(2): 235-43.

Casas, N., F. Amarita, et al. (2007). "Evaluation of an extracting method for the detection
of Hepatitis A virus in shellfish by SYBR-Green real-time RT-PCR." Int J Food
Microbiol 120(1-2): 179-85.

CDC (2002). "Outbreak of acute gastroenteritis associated with Norwalk-like viruses
among British military personnel--Afghanistan, May 2002." MMWR Morb Mortal Wk_ly
R_qp 51(22): 477-9.

CDC (2003). "Norovirus activity--United States, 2002." MMWR Morb Mortal WJSU Rep
52(3): 41-5.

CDC (2006). "Resurgence of wild poliovirus type 1 transmission and consequences of
importation--21 countries, 2002-2005." Morb Mortal Wkly Rap 55(6): 145-50.

Champsaur, H., M. Henry-Amar, et al. (1984). "Rotavirus carriage, asymptomatic
infection, and disease in the first two years of life. II. Serological response." J Infect Dis
149(5): 675-82.

Champsaur, H., E. Questiaux, et al. (1984). "Rotavirus carriage, asymptomatic infection,
and disease in the first two years of life. 1. Virus shedding." J Infect Dis 149(5): 667-74.

Chapron, C. D., N. A. Ballester, 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(6): 2520-5.

150

Chapron, C. D., N. A. Ballester, et a1. (2000). "The detection of astrovirus in sludge
biosolids using an integrated cell culture nested PCR technique." J Ann] Microbiol 89(1):
11-5.

Chen, T. C., G. W. Chen, et a]. (2006). "Combining multiplex reverse transcription-PCR

and a diagnostic microarray to detect and differentiate enterovirus 71 and coxsackievirus
A16." J Clin Microbiol 44(6): 2212-9.

Chiu, C. Y., S. Rouskin, et al. (2006). "Microarray detection of human parainfluenzavirus
4 infection associated with respiratory failure in an immunocompetent adult." Clin Infect
D_is 43(8): e71-6.

Chou, C. C., T. T. Lee, et al. (2006). "Design of microarray probes for virus identification
and detection of emerging viruses at the genus level." BMC Bioinforrnatics 7: 232.

Christopher, P. J ., G. S. Grohmann, et a]. (1978). "Parvovirus gastroenteritis--a new
entity for Australia." Med J Aust 1(3): 121-4.

Colford, J. M., J r., T. J. Wade, et a1. (2007). "Water quality indicators and the risk of
illness at beaches with nonpoint sources of fecal contamination." Epidemiology 18(1):
27-35.

Costafreda, M. I., A. Bosch, eta]. (2006). "Development, evaluation, and standardization
of a real-time TaqMan reverse transcription-PCR assay for quantification of hepatitis A
virus in clinical and shellfish samples." Appl Environ Microbiol 72(6): 3846-55.

Crabtree, K. D., C. P. Gerba, et a1. (1997). "Waterbome adenovirus: a risk assessment."
Water Sci Technol 35(11-12): 1-6.

Cruz, J. R., A. V. Bartlett, et al. (1992). "Astrovirus-associated diarrhea among
Guatemalan ambulatory rural children." J Clin Microbiol 30(5): 1140-4.

Dahling, D. R. (2002). "An improved filter elution and cell culture assay procedure for
evaluating public groundwater systems for culturable enteroviruses." Miter Environ Res
74(6): 564-8.

Dalton, C. B., J. Gregory, et al. (2004). "Foodborne disease outbreaks in Australia, 1995
to 2000." Commun Dis Intell 28(2): 211-24.

Darnell, M. E., K. Subbarao, et al. (2004). "Inactivation of the coronavirus that induces
severe acute respiratory syndrome, SARS-CoV." J Virol Methods 121(1): 85-91.

Day, F. and G. Sutton (2007). "General practitioner notifications of gastroenteritis and
food poisoning: cause for concern." J Public Health (Oxfl 29(3): 288-91.

de Wit, M. A., M. P. Koopmans, et al. (2001). "Etiology of gastroenteritis in sentinel
general practices in the netherlands." Clin Infect Dis 33(3): 280-8.

151

del Refugio Gonzalez-Losa, M., G. G. Polanco-Marin, eta]. (2001). "Acute
gastroenteritis associated with rotavirus in adults." Arch Med Res 32(2): 164-7.

Deregt, D., S. A. Gilbert, et a1. (2006). "A multiplex DNA suspension microarray for
simultaneous detection and differentiation of classical swine fever virus and other
pestiviruses." J Virol Methods 136(1-2): 17-23.

Dorfinan, M. (2006). Testing the Waters 2006. L. Goffredi, National Resources Defense
Council.

Dorfman, M. and N. Stoner (2007). Testing the Waters. A Guide to Water Quality at
Vacation Beachs, National Resources Defence Council.

Doyle, A., D. Barataud, et a1. (2004). "Norovirus foodbome outbreaks associated with the
consumption of oysters from the Etang de Thau, France, December 2002." Euro Surveill
9(3): 24-6.

 

Drinka, P. J. (2005). "Norovirus outbreaks in nursing homes." J Am Geriatr Soc 53(10):
1839-40.

Dubois, E., C. Agier, et a]. (2002). "Modified concentration method for the detection of
enteric viruses on fruits and vegetables by reverse transcriptase-polymerase chain
reaction or cell culture." J Food Prot 65(12): 1962-9.

Dziuban, E. J ., J. L. Liang, et a]. (2006). "Surveillance for waterborne disease and
outbreaks associated with recreational water--United States, 2003-2004." MMWR
Surveill Summ 55(12): 1-30.

Echeverria, P., N. R. Blacklow, et al. (1983). "Rotavirus as a cause of severe
gastroenteritis in adults." J Clin Microbiol 18(3): 663-7.

Elfaitouri, A., A. K. Berg, et al. (2007). "Recent enterovirus infection in type 1 diabetes:
evidence with a novel IgM method." J Med Virol 79(12): 1861-7.

Elia, G., N. Decaro, et a]. (2006). "Detection of canine distemper virus in dogs by real-
time RT-PCR." J Virol Methods 136(1-2): 171-6.

Ellis, M. E., B. Watson, et al. (1984). "Micro-organisms in gastroenteritis." Arch Dis
Child 59(9): 848-55.

EPA (1986). Ambient Water Quality Criteria for Bacteria.

EPA (1995). Virus monitoring protocol for the Information Collection Requirements
Rule. US Environment. U. E. P. Agency, Government Printing Office, Cincinnati, Ohio.

EPA (2003). Bacteria] Water Quality Standards for Recreational Waters (Freshwater and
Marine Waters) US-EPA.

152

Escutenaire, S., N. Mohamed, et al. (2007). "SYBR Green real-time reverse transcription-

polymerase chain reaction assay for the generic detection of coronaviruses." Arch Virol
152(1): 41-58.

F alkenhorst, G., L. Krusell, et al. (2005). "Imported fiozen raspberries cause a series of
norovirus outbreaks in Denmark, 2005." Euro Surveill 10(9): E050922 2.

Fankhauser, R. L., S. S. Monroe, eta]. (2002). "Epidemiologic and molecular trends of
"Norwalk-like viruses" associated with outbreaks of gastroenteritis in the United States."
J Infect Dis 186(1): 1-7.

Fauquet, C. M., M. A. Mayo, et a]. (2005). Virus Taxonomy: VIIIth Report of the
International Committee on Taxonomy of Viruses Academic Press.

Filho, E. P., N. R. da Costa Faria, et al. (2007). "Adenoviruses associated with acute
gastroenteritis in hospitalized and community children up to 5 years old in Rio de J aneiro
and Salvador, Brazil." J Med Microbiol 56(Pt 3): 313-9.

Fischer, T. K., C. Viboud, et al. (2007). "Hospitalizations and deaths from diarrhea and
rotavirus among children <5 years of age in the United States, 1993-2003." J Infect Dis
195(8): 1117-25.

Floret, D., B. Lina, et al. (2006). "Epidemiology and burden of rotavirus diarrhea in day
care centers in Lyon, France." Eur J Pediatr 165(12): 905-6.

Fong, T. T., D. W. Griffin, et al. (2005). "Molecular assays for targeting human and
bovine enteric viruses in coastal waters and their application for library-independent
source tracking." Appl Environ Microbiol 71(4): 2070-8.

Fout, G., Dahling, D.R., Safferman, RS. (2001). Total Culturable Virus Quanta] Assay.
Methods for Virology. Washington, DC, US EPA Office of Research and Development.

Francy, D. S., A. M. Gifford, et a1. (2003). Escherichia coli at Ohio bathing beaches——
distribution, sources, wastewater indicators, and predictive modeling. Columbus, Ohio.,
US Geological Survey.

Fretz, R., P. Svoboda, et a]. (2005). "Outbreaks of gastroenteritis due to infections with
Norovirus in Switzerland, 2001-2003." Epidemiol Infect 133(3): 429-37.

Frisk, G., T. Hansson, et al. (2008). "A unifying hypothesis on the development of type 1
diabetes and celiac disease: Gluten consumption may be a shared causative factor." Med
Hypotheses 70(6): 1207-1209.

Fukuda, S., Y. Sasaki, eta]. (2008). "Recombinant norovirus implicated in gastroenteritis
outbreaks in Hiroshima prefecture, Japan." J Med Virol 80(5): 921-8.

153

Gabbay, Y. B., B. J iang, et al. (1999). "An outbreak of group C rotavirus gastroenteritis
among children attending a day-care centre in Belem, Brazil." J DLmhoea] Dis Res
17(2): 69-74.

Gallay, A., H. De Valk, et al. (2006). "A large multi-pathogen waterborne community
outbreak linked to faecal contamination of a groundwater system, France, 2000." m
Microbiol Infect 12(6): 561-70.

Gallimore, C. 1., H. Appleton, et a]. (1995). "Detection and characterisation of
bisegrnented double-stranded RNA viruses (picobimaviruses) in human faecal
specimens." J Med Virol 45(2): 135-40.

Gallimore, C. I., D. Cubitt, et al. (2004). "Asymptomatic and symptomatic excretion of
noroviruses during a hospital outbreak of gastroenteritis." J Clin Microbiol 42(5): 2271-4.

Gantzer, C., A. Maul, et a]. (1998). "Detection of infectious enteroviruses, enterovirus
genomes, somatic coliphages, and Bacteroides fragilis phages in treated wastewater."
Appl Environ Microbiol 64(11): 4307-12.

Gao, X., E. LeProust, et a]. (2001). "A flexible light-directed DNA chip synthesis gated
by deprotection using solution photogenerated acids." Nucleic Acids Res 29(22): 4744-
50.

Gerba, C. P. a. R., J. B. (1989). Viruses in Source and Drinking Water. Advances in
Drinking Water Microbiology Research. G. A. McFeters. Madison, WI, Science Tech.

Gibbs, A. J ., M. J. Gibbs, et al. (2004). "The phylogeny of SARS coronavirus." Arch
Virol 149(3): 621-4.

Giordano, M. O., L. J. Ferreyra, et a1. (2001). "The epidemiology of acute viral
gastroenteritis in hospitalized children in Cordoba City, Argentina: an insight of disease
burden." Rev Inst Med Trop Sao PM 43(4): 193-7.

Giordano, M. 0., L. C. Martinez, et a]. (1999). "Diarrhea and enteric emerging viruses in
HIV-infected patients." AIDS Res Hum Retroviruses 15(16): 1427-32.

Given, S., L. H. Pendleton, et al. (2006). "Regional public health cost estimates of
contaminated coastal waters: a case study of gastroenteritis at southern California
beaches." Environ Sci Technol 40(16): 485 1-8.

Glass, R. I., J. Bresee, et al. (2001). "Gastroenteritis viruses: an overview." Novartis
Found Symp 238: 5-19; discussion 19-25.

GLIN. (2006, 01/04/2006). "The Great Lakes." Retrieved 05/06/2006, 2006, fiom
http://wwwgreat-lakes.net/lakes/.

Gonzalez, G. G., F. H. Pujol, et al. (1998). "Prevalence of enteric viruses in human
immunodeficiency virus seropositive patients in Venezuela. " J Med Virol 55(4): 288-92.

154

Goswami, B. B., M. Kulka, et al. (2002). "A polymerase chain reaction-based method for
the detection of hepatitis A virus in produce and shellfish. " J Food Prot 65(2): 393-402.

Goya], S. M. and C. P. Gerba (1983). "Viradel method for detection of rotavirus from
seawater." J Virol Methods 7(5-6): 279-85.

Green, D. H. and G. D. Lewis (1995). "Enzymatic amplification of enteric viruses from
wastewaters." _Vlater Science and Technology 31: 329-336.

Grohmann, G. S., R. 1. Glass, et a]. (1993). "Enteric viruses and diarrhea in HIV -infected
patients. Enteric Opportunistic Infections Working Group." N Engl J Med 329(1): 14-20.

Guerrero, M. L., J. S. Noe], et a]. (1998). "A prospective study of astrovirus diarrhea of
infancy in Mexico City." Pediatr Infect Dis J 17(8): 723-7.

Guimil, R., M. Beier, et al. (2003). "Geniom technology--the benchtop array facility."
Nucleosides Nucleotides Nucleic Acidg 22(5-8): 1721-3.

Haas, C. H., J. B. Rose, et al. (1999). Quantitative Microbial Riik Assessment, John
Wiley & Sons.

Haramoto, E., H. Katayama, et al. (2005). "Application of Cation-Coated Filter Method
to Detection of Noroviruses, Enteroviruses, Adenoviruses, and Torque Teno Viruses in
the Tamagawa River in Japan." Appl. Environ. Microbiol. 71(5): 2403-2411.

Haramoto, E., H. Katayama, et a]. (2007). "Quantitative analysis of human enteric
adenoviruses in aquatic environments." J Appl Microbiol 103(6): 2153-9.

Haramoto, E., H. Katayama, et a]. (2004). "Detection of noroviruses in tap water in Japan
by means of a new method for concentrating enteric viruses in large volumes of
freshwater." Appl Environ Microbiol 70(4): 2154-60.

Haramoto, E., H. Katayama, et al. (2008). "Quantitative detection of sapoviruses in
wastewater and river water in Japan." Lett Appl Microbiol 46(3): 408-13.

Harsi, C. M., D. P. Rolirn, et al. (1995). "Adenovirus genome types isolated fiom stools
of children with gastroenteritis in Sao Paulo, Brazil." J Med Virol 45(2): 127-34.

He, J. W. and S. J iang (2005). "Quantification of enterococci and human adenoviruses in
environmental samples by real-time PCR." Appl Environ Microbiol 71(5): 2250-5.

Heegaard, E. D. and K. E. Brown (2002). "Human parvovirus B19." Clin Microbiol Rev
15(3): 485-505.

Hijnen, W. A., E. F. Beerendonk, et al. (2006). "Inactivation credit of UV radiation for
viruses, bacteria and protozoan (oo)cysts in water: a review." Water Res 40(1): 3-22.

155

Hill, V. R., A. L. Polaczyk, et a]. (2005). "Development of a rapid method for
simultaneous recovery of diverse microbes in drinking water by ultrafiltration with
sodium polyphosphate and surfactants." Appl Environ Microbiol 71(11): 6878-84.

Hj ertqvist, M., A. J ohansson, et a]. (2006). "Four outbreaks of norovirus gastroenteritis
after consuming raspberries, Sweden, June-August 2006." Euro Surveill 11(9): E060907
l. '

Hovi, T., M. Stenvik, et al. (2001). "Poliovirus surveillance by examining sewage
specimens. Quantitative recovery of virus after introduction into sewerage at remote
upstream location." Epidemiol Infect 127(1): 101-6.

Hughes, T. R., M. Mao, et al. (2001). "Expression profiling using microarrays fabricated
by an ink-jet oligonucleotide synthesizer." Nat Biotechnol 19(4): 342-7.

Hung, T., G. M. Chen, et al. (1984). "Waterbome outbreak of rotavirus diarrhoea in
adults in China caused by a novel rotavirus." Lancet 1(83 87): 1139-42.

Ijzennan, M. M., D. R. Dahling, et a1. (1997). "A method to remove environmental
inhibitors prior to the detection of waterborne enteric viruses by reverse transcription-
polymerase chain reaction." J Virol Methoda 63(1-2): 145-53.

Ike, A. C., S. O. Brockmann, et al. (2006). "Molecular epidemiology of norovirus in
outbreaks of gastroenteritis in southwest Germany from 2001 to 2004." J Clin Microbiol
44(4): 1262-7.

Inoue, J ., M. Takahashi, et al. (2006). "Deve10pment and validation of an improved RT-
PCR assay with nested universal primers for detection of hepatitis E virus strains with
significant sequence divergence." J Virol Method_§ 137(2): 325-33.

Iturriza-Gomara, M., J. Green, et al. (1999). "Comparison of specific and random priming
in the reverse transcriptase polymerase chain reaction for genotyping group A
rotaviruses." J Virol Methods 78(1-2): 93-103.

Jamieson, F. B., E. B. Wang, et al. (1998). "Human torovirus: a new nosocomial
gastrointestinal pathogen." J Infect Dis 178(5): 1263-9.

Jewkes, J ., H. E. Larson, et al. (1981). "Aetiology of acute diarrhoea in adults." Ga;
22(5): 388-92.

Jiang, S., R. Noble, et a]. (2001). "Human adenoviruses and coliphages in urban runoff-
impacted coastal waters of Southern California." Appl Environ Microbiol 67(1): 179-84.

J iang, S. C. and W. Chu (2004). "PCR detection of pathogenic viruses in southern
California urban rivers." J Appl Microbiol 97(1): 17-28.

156

 

J othikumar, N. and D. O. Cliver (1997). "Elution and reconcentration of coliphages in

water from positively charged membrane filters with urea-arginine phosphate buffer. " J
Virol Methods 65(2): 281-6.

J othikumar, N., D. O. Cliver, et al. (1998). "Irnmunomagnetic capture PCR for rapid
concentration and detection of hepatitis A virus from environmental samples." Appl
Environ Microbiol 64(2): 504-8.

J othikumar, N., J. A. Lowther, et a]. (2005). "Rapid and sensitive detection of noroviruses
by using TaqMan-based one-step reverse transcription-PCR assays and application to
naturally contaminated shellfish samples." Appl Environ Microbiol 71(4): 1870-5.

Kalm, J. S. and K. McIntosh (2005). "History and recent advances in coronavirus
discovery." Pediatr Infect Dis J 24(11 Suppl): S223-7, discussion S226.

Kapikian, A. Z. (1996). "Overview of viral gastroenteritis." Arch Virol Suppl 12: 7-19.

Katayama, H., K. Okuma, et a]. (2004). "Series of surveys for enteric viruses and
indicator organisms in Tokyo Bay after an event of combined sewer overflow." Water Sci
Technol 50(1): 259-62.

Kelly, J. J ., S. Siripong, et al. (2005). "DNA microarray detection of nitrifying bacterial
16S rRNA in wastewater treatment plant samples." Water Res 39(14): 3229-38.

Kennedy, P. G., E. Grinfeld, et a1. (2005). "Transcriptomal analysis of varicella-zoster
virus infection using long oligonucleotide-based microarrays." J Gen Virol 86(Pt 10):
2673-84.

Kern, P., G. Muller, et al. (1985). "Detection of coronavirus-like particles in homosexual
men with acquired immunodeficiency and related lymphadenopathy syndrome." Klin
Wochenschr 63(2): 68-72.

Kidd, A. H., S. A. Esrey, et a]. (1989). "Shedding of coronavirus-like particles by
children in Lesotho." J Med Vir0127(2): 164-9.

Kidd, A. H., E. H. Harley, et al. (1985). "Specific detection and typing of adenovirus
types 40 and 41 in stool specimens by dot-blot hybridization." J Clin Microbiol 22(6):
934-9.

Kingsley, D. H., G. K. Meade, et a]. (2002). "Detection of both hepatitis A virus and
Norwalk-like virus in imported clams associated with food-bome illness." Appl Environ
Microbiol 68(8): 3914-8.

Kingsley, D. H. and G. P. Richards (2001). "Rapid and efficient extraction method for

reverse transcription-PCR detection of hepatitis A and Norwalk-like viruses in shellfish."
Appl Environ Microbiol 67(9): 4152-7.

157

Kittigul, L., P. Khamoun, et al. (2001). "An improved method for concentrating rotavirus
from water samples." Mem In_st Oswaldo Cruz 96(6): 815-21.

Kittigul, L., B. Raengsakulrach, et al. (2000). "Detection of poliovirus, hepatitis A virus
and rotavirus from sewage and water samples." Southeag Asian J Trop Med Public
Health 31(1): 41-6.

Kleines, M., K. Schellenberg, et al. (2003). "Efficient extraction of viral DNA and viral
RNA by the Chemagic viral DNA/RNA kit allows sensitive detection of
cytomegalovirus, hepatitis B virus, and hepatitis G virus by PCR." J Clin Microbiol
41(11): 5273-6.

Komolpis, K., O. Srivannavit, et a]. (2002). "Light-directed simultaneous synthesis of
oligopeptides on microarray substrate using a photogenerated acid." Biotechnol Prog
18(3): 641-6.

Koopman, J. S. and A. S. Monto (1989). "The Tecumseh Study. XV: Rotavirus infection
and pathogenicity." Am J Epidemiol 130(4): 750-9.

Koopmans, M., J. Harris, et al. (2006). "European investigation into recent norovirus
outbreaks on cruise ships: update." Euro Surveill 11(7): E060706 5.

Koopmans, M., M. Petric, et a]. (1993). "Enzyme-linked irnmunosorbent assay reactivity
of torovirus-like particles in fecal specimens fi'om humans with diarrhea." J Clin
Microbiol 31(10): 2738-44.

Koopmans, M. P., E. S. Goosen, et a1. (1997). "Association of torovirus with acute and
persistent diarrhea in children." Pediatr Infect Dis J 16(5): 504-7.

Korimbocus, J ., N. Scaramozzino, et a1. (2005). "DNA probe array for the simultaneous
identification of herpesviruses, enteroviruses, and flaviviruses." J Clin Microbiol 43(8):
3779-87.

Korsager, B., S. Hede, et a]. (2005). "Two outbreaks of norovirus infections associated
with the consumption of imported frozen raspben’ies, Denmark, May-June 2005." Euro
Surveill 10(6): E050623 1.

Kostic, T., A. Weilharter, et al. (2007). "A microbial diagnostic microarray technique for
the sensitive detection and identification of pathogenic bacteria in a background of
nonpathogens." Anal Biochem 360(2): 244-54.

Krishnan, T. and T. N. Naik (1997). "Electronmicroscopic evidence of torovirus like
particles in children with diarrhoea." Indian J Med Res 105: 108-10.

Laverick, M. A., A. P. Wyn-J ones, et al. (2004). "Quantitative RT-PCR for the
enumeration of noroviruses (Norwalk-like viruses) in water and sewage." Lett Appl
Microbiol 39(2): 127-36.

158

Leclerc, H., L. Schwartzbrod, et a1. (2002). "Microbial agents associated with waterborne
diseases." Crit Rev Microbiol 28(4): 37 l -409.

Lee, C., S. H. Lee, et al. (2004). "Use of cell culture-PCR assay based on combination of
A549 and BGMK cell lines and molecular identification as a tool to monitor infectious

adenoviruses and enteroviruses in river water." Appl Environ Microbiol 70(11): 6695-
705.

Lee, D. Y., K. Shannon, et al. (2006). "Detection of bacterial pathogens in municipal
wastewater using an oligonucleotide microarray and real-time quantitative PCR." J
Microbiol Methods 65(3): 453-67.

Lee, S. H., D. A. Levy, et al. (2002). "Surveillance for waterbome-disease outbreaks--
United States, 1999-2000." MMWR Surveill Summ 51(8): 1-47.

LeProust, E., J. P. Pellois, et a1. (2000). "Digital light-directed synthesis. A microarray

platform that permits rapid reaction optimization on a combinatorial basis." J Comb
Chem 2(4): 349-54.

Lewis, G. D. and T. G. Metcalf (1988). "Polyethylene glycol precipitation for recovery of
pathogenic viruses, including hepatitis A virus and human rotavirus, from oyster, water,
and sediment samples." Appl Environ Microbi0154(8): 1983-8.

Li, C., R. S. Chen, et al. (2006). "Detection of Epstein-Barr virus infection and gene
expression in human tumors by microarray analysis." J Virol Methods 133(2): 158-66.

Li, L., T. G. Phan, et al. (2005). "Molecular epidemiology of adenovirus infection among
pediatric population with diarrhea in Asia." Microbiol Irnmunol 49(2): 121-8.

Liang, J. L., E. J. Dziuban, et a1. (2006). "Surveillance for waterborne disease and
outbreaks associated with drinking water and water not intended for drinking-United
States, 2003-2004." MMWR Surveill Summ 55(12): 31-65.

Lin, B., A. P. Malanoski, et al. (2007). "Application of broad-spectrum, sequence-based
pathogen identification in an urban population." PLoS ONE 2(5): e419.

Lin, B., Z. Wang, et al. (2006). "Broad-spectrum respiratory tract pathogen identification
using resequencing DNA microarrays." Genome Res 16(4): 527-35.

Liste, M. B., I. Natera, et a1. (2000). "Enteric virus infections and diarrhea in healthy and
human immunodeficiency virus-infected children." J Clin Microbiol 38(8): 2873-7.

Liu, L., M. S. Phanikumar, et al. (2006). "Modeling the transport and inactivation of E.
coli and enterococci in the near-shore region of Lake Michigan." Environ Sci Technol
40(16): 5022-8.

159

Liu, R. H., M. J. Lodes, et al. (2006). "Validation of a fully integrated microfluidic array
device for influenza a subtype identification and sequencing." Anal Chem 78(12): 4184-
93.

Liu, Y. C., G. S. Huang, et a]. (2006). "Detection of foot and mouth disease and porcine
reproductive and respiratory syndrome viral genes using microarray chip." Vet Res
Commun 30(2): 191-204.

Ljubin-Stemak, S., A. Lukic-Grlic, et a]. (2007). "Norovirus genotypes involved in the
outbreaks of gastroenteritis in Croatia during the winter season 2004--2005." Acta Virol
51(3): 189-94.

Lodder, W. J. and A. M. de Roda Husman (2005). "Presence of noroviruses and other
enteric viruses in sewage and surface waters in The Netherlands." Appl Environ
Microbiol 71(3): 1453-61.

Lodha, A., N. de Silva, et al. (2005). "Human torovirus: a new virus associated with
neonatal necrotizing enterocolitis." Acta Paediatr 94(8): 1085-8.

Long, W. H., H. S. Xiao, et a]. (2004). "A universal microarray for detection of SARS
coronavirus." J Virol Methods 121(1): 57-63.

Loosli, J ., K. Gyr, et al. (1985). "Etiology of acute infectious diarrhea in a highly
industrialized area of Switzerland." Gastroenterology 88(1 Pt 1): 75-9.

Lopman, B. A., M. Reacher, et al. (2003). "A summertime peak of "winter vomiting
disease": surveillance of noroviruses in England and Wales, 1995 to 2002." BMC Public
Health 3: l3.

Lukasik, J., T. M. Scott, et al. (2000). "Influence of salts on virus adsorption to
microporous filters." Appl Environ Microbiol 66(7): 2914-20.

Ma, J. F ., C. P. Gerba, et a1. (1995). "Increased sensitivity of poliovirus detection in tap

water concentrates by reverse transcriptase-polymerase chain reaction." J Virol Methods
55(3): 295-302.

Ma, J. F., J. Naranjo, et al. (1994). "Evaluation of MK filters for recovery of
enteroviruses fi'om tap water. " Appl Environ Microbiol 60(6): 1974-7.

Maass, G. and H. G. Baumeister (1983). "Coronavirus-like particles as aetiologica]
agents of acute non-bacterial gastroenteritis in humans." Dev Biol Sta_ngl_ 53: 319-24.

MacDonald, E. (2006). The Great Lakes Sewage Report Card. S. Legal, Sierra Legal.

Majowicz, S. E., V. L. Edge, et a]. (2005). "Estimating the under-reporting rate for
infectious gastrointestinal illness in Ontario." Can J Public Health 96(3): 178-81.

160

Makary, P., L. Maunula, et a]. (2008). "Multiple norovirus outbreaks among workplace
canteen users in Finland, July 2006." Epidemiol Infect: 1-6.

Maldonado, Y., M. Cantwell, et al. (1998). "Population-based prevalence of symptomatic
and asymptomatic astrovirus infection in rural Mayan infants." J Infect Dis 178(2): 334-
9.

Marshall, J. A., M. E. Hellard, et al. (2003). "Incidence and characteristics of endemic
Norwalk-like virus-associated gastroenteritis." J Med Virol 69(4): 568-78.

Matson, D. O., M. K. Estes, et al. (1989). "Human calicivirus-associated diarrhea in
children attending day care centers." J Infect Dis 159(1): 71-8.

Maunula, L., I. T. Miettinen, et al. (2005). "Norovirus outbreaks from drinking water. "
Emgg Infect Dis 11(11): 1716-21.

Maunula, L. and C. H. Von Bonsdorff (2005). "Norovirus genotypes causing
gastroenteritis outbreaks in Finland 1998-2002." J Clin Virol 34(3): 186-94.

Maynard, C., F. Berthiaume, et al. (2005). "Waterbome pathogen detection by use of
oligonucleotide-based microarrays." Appl Environ Microbiol 71(12): 8548-57.

Mead, P. S., L. Slutsker, et al. (1999). "Food-related illness and death in the United
States." Emerglnfect Dis 5(5): 607-25.

Mehnert, D. U. and K. E. Stewien (1993). "Detection and distribution of rotavirus in raw
sewage and creeks in Sao Paulo, Brazil." Appl Environ Microbiol 59(1): 140-3.

Mendez-Toss, M., D. D. Griffin, et a]. (2004). "Prevalence and genetic diversity of
human astroviruses in Mexican children with symptomatic and asymptomatic infections."
J Clin Microbiol 42(1): 151-7.

Metcalf, T. G., J. L. Melnick, et a]. (1995). "Environmental virology: from detection of

virus in sewage and water by isolation to identification by molecular biology--a trip of
over 50 years." Annu Rev Microbiol 49: 461-87.

Miller, S. M., D. M. Tourlousse, et al. (2008). "An In-situ Synthesized Virulence and
Marker Gene (VMG) Biochip for the Detection of Bacterial Pathogens in Water." Appl
Environ Microbiol.

Min, W., M. Wen-Li, et a]. (2006). "Oligonucleotide Microarray with RD-PCR Labeling
Technique for Detection and Typing of Human Papillomavirus." Curr Microbiol 52(3):
204-9.

Mitchell, D. K., D. O. Matson, et a]. (1999). "Prevalence of antibodies to astrovirus types

1 and 3 in children and adolescents in Norfolk, Virginia." Pediatr Infect Dis J 18(3): 249-
54.

161

Moes, E., L. Vij gen, et a]. (2005). "A novel pancoronavirus RT-PCR assay: frequent
detection of human coronavirus NL63 in children hospitalized with respiratory tract
infections in Belgium." BMC Infect Dis 5(1): 6.

Monica, B., S. Ramani, et al. (2007). "Human caliciviruses in symptomatic and
asymptomatic infections in children in Vellore, South India." J Med Virol 79(5): 544-51.

Morales-Morales, H. A., G. Vidal, et a]. (2003). "Optimization of a reusable hollow-fiber
ultrafilter for simultaneous concentration of enteric bacteria, protozoa, and viruses from
water." Appl Environ Microbiol 69(7): 4098-102.

Mounts, A. W., T. Ando, et al. (2000). "Cold weather seasonality of gastroenteritis
associated with Norwalk-like viruses." J Infect Dis 181 Suppl 2: 8284-7.

Mustafa, H., E. A. Palombo, et al. (2000). "Epidemiology of astrovirus infection in young
children hospitalized with acute gastroenteritis in Melbourne, Australia, over a period of
four consecutive years, 1995 to 1998." J Clin Microbiol 38(3): 1058-62.

Naik, S. R., R. Aggarwal, et a]. (1992). "A large waterborne viral hepatitis E epidemic in
Kanpur, India." Bull World Health Organ 70(5): 597-604.

Nakajima, H., T. Nakagomi, et al. (2001). "Winter seasonality and rotavirus diarrhoea in
adults." Lancet 357(9272): 1950.

Nevers, M. B. and R. L. Whitman (2005). "Nowcast modeling of Escherichia coli
concentrations at multiple urban beaches of southern Lake Michigan." Water Res 39(20):
5250-60.

Nguyen, T. V., L. Yuan, et a1. (2006). "High titers of circulating maternal antibodies
suppress effector and memory B-cell responses induced by an attenuated rotavirus
priming and rotavirus-like particle-immunostimulating complex boosting vaccine
regimen." Clin Vaccine Immunol 13(4): 475-85.

Noble, R. T. a. J. A. F. (2001). "Enteroviruses detected by reverse transcriptase
polymerase chain reaction from the coastal waters of Santa Monica Bay, California: Low
correlation to bacterial indicator levels. ." Hydrobiologia 460: 175-184.

Noe], J ., A. Mansoor, et al. (1994). "Identification of adenoviruses in faeces from patients
with diarrhoea at the Hospitals for Sick Children, London, 1989-1992." J Med Virol
43(1): 84-90.

Nordgren, J ., F. Bucardo, et al. (2008). "Novel light-upon-extension real-time PCR
assays for detection and quantification of genogroup I and H noroviruses in clinical
specimens." J Clin Microbiol 46(1): 164-70.

Nordstrom, H., K. I. Falk, et al. (2005). "DNA microarray technique for detection and
identification of seven flaviviruses pathogenic for man." J Med Virol 77(4): 528-40.

162

NRDC (2007). Testing the Waters. A Guide to Water Quality at Vacation Beaches.
Seventeenth Edition, National Resources Defense Council.

Nwachuku, N. and C. P. Gerba (2006). "Health risks of enteric viral infections in
children." Rev Environ Contam Toxicol 186: 1-56.

Oberste, M. S. and M. A. Pallansch (2003). "Establishing evidence for enterovirus
infection in chronic disease." Ann N Y Acad Sci 1005: 23-31.

Oikarinen, M., S. T auriainen, et al. (2008). "Detection of enteroviruses in the intestine of
type 1 diabetic patients." Clin Exp Irnmuno] 151(1): 71-5.

Oka, T., K. Katayama, et a1. (2006). "Detection of human sapovirus by real-time reverse
transcription-polymerase chain reaction." J Med Virol 78(10): 1347-53.

Okada, M., Y. Yamashita, et al. (2006). "The detection of human sapoviruses with
universal and genogroup-specific primers." Arch Virol 151(12): 2503-9.

Olszewski, J ., L. Winona, et a]. (2005). "Comparison of 2 ultrafiltration systems for the
concentration of seeded viruses fi'om environmental waters." Can J Microbiol 51(4): 295-
303.

Olyphant, G. A. and R. L. Whitman (2004). "Elements of a predictive model for
determining beach closures on a real-time basis: the case of 63rd Street Beach Chicago."
Environ Monit Assess 98: 175-190.

Oyofo, B. A., D. Subekti, et al. (2002). "Enteropathogens associated with acute diarrhea
in community and hospital patients in Jakarta, Indonesia." FEMS Immunol Med
Microbiol 34(2): 139-46.

Palmer, J. and M. Thomley (2004). "Canine parvovirus outbreaks." Aust Vet J 82(12):
720.

Palombo, E. A. and R. F. Bishop (1996). "Annual incidence, serotype distribution, and

genetic diversity of human astrovirus isolates from hospitalized children in Melbourne,
Australia." J Clin Microbiol 34(7): 1750-3.

Pang, X. L., J. K. Preiksaitis, et a1. (2005). "Multiplex real time RT-PCR for the detection
and quantitation of norovirus geno groups I and II in patients with acute gastroenteritis." J
Clin Virol 33(2): 168-71.

Pang, X. L. and T. Vesikari (1999). "Human astrovirus-associated gastroenteritis in
children under 2 years of age followed prospectively during a rotavirus vaccine trial."
Acta Paediatr 88(5): 532-6.

Pannucci, J ., H. Cai, et al. (2004). "Virulence signatures: microarray-based approaches to
discovery and analysis." Biosefins Bioelectron 20(4): 706-18.

163

Papapetropoulou, M. and A. C. Vantarakis (1998). "Detection of adenovirus outbreak at a
municipal swimming pool by nested PCR amplification." J Infect 36(1): 101-3.

Parashar, U. D., E. G. Hummelman, et a]. (2003). "Global illness and deaths caused by
rotavirus disease in children." Emerg Infect Dis 9(5): 565-72.

Parker, S. P., W. D. Cubitt, et al. (1994). "Seroprevalence studies using a recombinant
Norwalk virus protein enzyme immunoassay." J Med Virol 42(2): 146-50.

Patti, A. M., A. L. Santi, et al. (2003). "Enviromnental surveillance of poliovirus in Italy:
pilot study." Ann Ig 15(2): 97-105.

Pavlov, D. N., W. B. Van Zyl, et al. (2005). "Prevalence of vaccine-derived polioviruses
in sewage and river water in South Africa." Water Res 39(14): 3309-19.

Payment, P., S. Fortin, et al. (1984). "Ferric chloride flocculation for nonflocculating beef
extract preparations." Appl Environ Microbiol 47(3): 591-2.

Pease, A. C., D. Solas, et a1. (1994). "Light-generated oligonucleotide arrays for rapid
DNA sequence analysis." Proc Natl Acad Sci U S A 91(11): 5022-6.

Penaranda, M. E., M. S. Ho, et al. (1989). "Seroepidemiology of adult diarrhea rotavirus
in China, 1977 to 1987." J Clin Microbiol 27(10): 2180-3.

Pereira, H. G., A. C. Linhares, et a]. (1993). "National laboratory surveillance of viral
agents of gastroenteritis in Brazil." Bull Pan Am Health Orga_n 27(3): 224-33.

Perlman, S. and A. A. Dandekar (2005). "Irnmunopathogenesis of coronavirus infections:
implications for SARS." Nat Rev Immunol 5(12): 917-27.

Pickering, L. K., A. V. Bartlett, 3rd, et al. (1988). "Asymptomatic excretion of rotavirus

before and after rotavirus diarrhea in children in day care centers." J Pediatr 112(3): 361-
5.

 

Pinto, R. M., F. X. Abad, et al. (1996). "Detection of Infectious Astroviruses in Water."
Appl Environ Microbiol 62(5): 1811-1813.

Pinto, R. M., J. Jofre, et al. (1993). "Concentration of fish enveloped viruses from large
volumes of water." J Virol Methods 43(1): 31-40.

Preston, D. R., T. V. Vasudevan, et a]. (1988). "Novel approach for modifying
microporous filters for virus concentration from water." Appl Environ Microbiol 54(6):
1325-9.

Pruss, A. and A. Havelaar (2001). The Global Burden of Disease Study and Applications
in Water, Sanitation and Hygiene. _Water Quality: Guidelines. Standards and Health.
World Health Organization Water Series. L. Fewtrell and J. Bartram. London, UK, IWA
publishing.

164

Pryor, W. M., W. A. Bye, et a]. (1987). "Acute diarrhoea in adults: a prospective study."
Med J Aust 147(10): 490-3.

Pusch, D., D. Y. Oh, et a]. (2005). "Detection of enteric viruses and bacterial indicators in
German environmental waters." Arch Virol 150(5): 929-47.

Putonti, C., S. Chumakov, et al. (2006). "Human-blind probes and primers for dengue
virus identification." Febs J 273(2): 398-408.

Quinones, B., C. T. Parker, et a1. (2007). "Detection and genotyping of Arcobacter and
Carnpylobacter isolates from retail chicken samples by use of DNA oligonucleotide
arrays." Appl Environ Microbiol 73(11): 3645-55.

Rabinovici, S. J ., R. L. Bemknopf, et a]. (2004). "Economic and health risk trade-offs of
swim closures at a Lake Michigan beach." Environ Sci T echno] 38(10): 2737-45.

Raboni, S. M., M. M. Siqueira, et a1. (2003). "Comparison of PCR, enzyme immunoassay
and conventional culture for adenovirus detection in bone marrow transplant patients
with hemorrhagic cystitis." J Clin Virol 27(3): 270-5.

Ram, J. L., R. P. Ritchie, et al. (2004). "Sequence-based source tracking of Escherichia
coli based on genetic diversity of beta-glucuronidase." J Environ Oual 33(3): 1024-32.

Reynolds, K. A., K. D. Mena, et al. (2008). "Risk of waterborne illness via drinking
water in the United States." Rev Environ Contam Toxicol 192: 117-58.

Rhee, S. K., X. Liu, et al. (2004). "Detection of genes involved in biodegradation and
biotransformation in microbial communities by using 50-mer oligonucleotide
microarrays." App] Environ Microbiol 70(7): 4303-17.

Rodriguez-Baez, N., R. O'Brien, et al. (2002). "Astrovirus, adenovirus, and rotavirus in
hospitalized children: prevalence and association with gastroenteritis." J Pediatr
Gastroenterol Nutr 35(1): 64-8.

Rosen, B. I., L. J. Saif, et a1. (1990). "Serotypic differentiation of group A rotaviruses
with porcine rotavirus gene 9 probes." J Clin Microbiol 28(11): 2526-33.

Rota, P. A., M. S. Oberste, et a]. (2003). "Characterization of a novel coronavirus
associated with severe acute respiratory syndrome." Science 300(5624): 1394-9.

 

Rouillard, J. M., M. Zuker, et al. (2003). OligoArray 2.0: design of oligonucleotide
probes for DNA microarrays using a thermodynamic approach. Nucleic Acids Res. 31:
305 7-62.

Rutj es, S. A., R. Italiaander, et a1. (2005). "Isolation and detection of enterovirus RNA

from large-volume water samples by using the NucliSens miniMAG system and real-time
nucleic acid sequence-based amplification." Appl Environ Microbiol 71(7): 3734-40.

165

Ryabinin, V. A., L. A. Shundrin, et al. (2006). "Microarray assay for detection and
discrimination of Orthopoxvirus species." J Med Virol 78(10): 1325-40.

Sakon, N., K. Yarnazaki, et al. (2005). "Norovirus outbreaks at nursing homes in osaka,
Japan." Jpn J Infect Dis 58(4): 254-5.

Sakuma, Y., S. Chiba, et al. (1981). "Prevalence of antibody to human calicivirus in
general population of northern Japan." J Med Virol 7(3): 221-5.

Schena, M., D. Shalon, et al. (1995). "Quantitative monitoring of gene expression
patterns with a complementary DNA microarray." Science 270(5235): 467-70.

Schwab, K. J ., R. De Leon, et al. (1996). "Irnmunoaffinity concentration and purification

of waterborne enteric viruses for detection by reverse transcriptase PCR." Appl Environ
Microbiol 62(6): 2086-94.

Scott, T. M., T. M. Jenkins, et a1. (2005). "Potential use of a host associated molecular
marker in Enterococcus faecium as an index of human fecal pollution." Environ Sci
Technol 39(1): 283-7.

Sdiri-Loulizi, K., H. Gharbi-Khelifi, et al. (2008). "Acute infantile gastroenteritis
associated with human enteric viruses in Ttmisia." J Clin Microbiol 46(4): 1349-55.

Sedmak, G., D. Bina, et al. (2003). "Assessment of an enterovirus sewage surveillance
system by comparison of clinical isolates with sewage isolates from milwaukee,
Wisconsin, collected august 1994 to december 2002." Appl E_nviron Microbiol 69(12):
71 81-7.

Sedmak, G., D. Bina, et a]. (2005). "Nine-year study of the occurrence of culturable
viruses in source water for two drinking water treatment plants and the influent and
effluent of a Wastewater Treatment Plant in Milwaukee, Wisconsin (August 1994
through July 2003)." Appl Environ Microbiol 71(2): 1042-50.

Sengupta, S., K. Onodera, et al. (2003). "Molecular detection and identification of

influenza viruses by oligonucleotide microarray hybridization." J Clin Microbiol 41(10):
4542-50.

Sergeev, N., M. Distler, et al. (2004). "Multipathogen oligonucleotide microarray for
environmental and biodefense applications." Biosens Bioelect;o_n 20(4): 684-98.

Sergeev, N., E. Rubtcova, et al. (2006). "New mosaic subgenotype of varicella-zoster
virus in the USA: VZV detection and genotyping by oligonucleotide-microarray." J Virol
Methods 136(1-2): 8-16.

Seto, Y., N. Iritani, et al. (2005). "Genotyping of Norovirus strains detected in outbreaks

between April 2002 and March 2003 in Osaka City, Japan." Microbiol Immunol 49(3):
275-83.

166

Seymour, 1. J. and H. Appleton (2001). "Foodbome viruses and fresh produce." J Appl
Microbiol 91(5): 759-73.

Shieh, Y. C., C. I. Wong, et al. (2008). "Detection of naturally occurring enteroviruses in
waters using direct RT-PCR and integrated cell culture-RT-PCR." J Virol Methods
149(1): 184-9.

Singh-Naz, N., W. J. Rodriguez, et a1. (1988). "Monoclonal antibody enzyme-linked
immunosorbent assay for specific identification and typing of subgroup F adenoviruses."
J Clin Microbiol 26(2): 297-300.

Snelling, T., T. Cripps, et al. (2007). "Nosocomial rotavirus infection in an Australian
children's hospital." J Paediatr Child Health 43(4): 318.

Snijder, E. J ., P. J. Bredenbeek, et al. (2003). "Unique and conserved features of genome
and proteome of SARS-coronavirus, an early split-off from the coronavirus group 2
lineage." J Mol Biol 331(5): 991-1004.

Sobsey, M. D., C. Wallis, et al. (1973). "Concentration of Enteroviruses from Large
Volumes of Water." Appl Microbiol 26(4): 529-534.

Song, Y., E. Dai, et al. (2006). "Genotyping of hepatitis B virus (HBV) by
oligonucleotides microarray." Mol Cell Probes.

Squires, R. A. (2003). "An update on aspects of viral gastrointestinal diseases of dogs and
cats." N Z Vet J 51(6): 252-61.

Standish-Lee, P. and E. Loboschefsky (2006). "Protecting public health from the impact
of body-contact recreation." Water Sci Technol 53(10): 201-7.

Stolle, A. and B. Spemer (1997). "Viral infections transmitted by food of animal origin:
the present situation in the European Union." Arch Virol Suppl 13: 219-28.

Straub, T. M. and D. P. Chandler (2003). "Towards a unified system for detecting
waterborne pathogens." J Microbiol Methods 53(2): 185-97.

Straub, T. M., K. Honer zu Bentrup, et a]. (2007). "In vitro cell culture infectivity assay
for human noroviruses." Emerg Infect Dis 13(3): 396-403.

Sunen, E., N. Casas, et al. (2004). "Comparison of two methods for the detection of

hepatitis A virus in clam samples (Tapes spp.) by reverse transcription-nested PCR." I_n_t_J
Food Microbiol 91(2): 147-54.

Svenungsson, B., A. Lagergren, et al. (2000). "Enteropathogens in adult patients with

diarrhea and healthy control subjects: a 1-year prospective study in a Swedish clinic for
infectious diseases." Clin Infect Dis 30(5): 770-8.

167

Takiff, H. E., M. Seidlin, et al. (1985). "Detection of enteric adenoviruses by dot-blot
hybridization using a molecularly cloned viral DNA probe." J Med Virol 16(2): 107-18.

Takkinen, J. (2006). "Recent norovirus outbreaks on river and seagoing cruise ships in
Europe." Euro Surveill 11(6): E060615 2.

Tarnbyah, P. A. (2004). "SARS: responding to an unknown virus." Eur J Clin Microbiol
Infect Dis 23(8): 589-95.

Tartera, C. and J. J ofre (1987). "Bacteriophages active against Bacteroides fragilis in
sewage-polluted waters." Appl Environ Microbiol 53(7): 1632-7.

Taylor, L. H., S. M. Latharn, et a]. (2001). "Risk factors for human disease emergence."
Philos Trans R Soc Lond B Biol Sci 356(1411): 983—9.

Thomas, P. D., R. C. Pollok, et al. (1999). "Enteric viral infections as a cause of diarrhoea
in the acquired immunodeficiency syndrome." HIV Med 1(1): 19-24.

Thurston-Enriquez, J. A., C. N. Haas, et al. (2003). "Inactivation of feline calicivirus and
adenovirus type 40 by UV radiation." Appl Environ Microbiol 69(1): 577-82.

Toranzos, G. A. and C. P. Gerba (1989). "An improved method for the concentration of
rotaviruses from large volumes of water." J Virol Methoda 24(1-2): 131-40.

Tomer, N., A. Dominguez, et al. (2008). "Acute gastroenteritis outbreaks in Catalonia,
Spain: norovirus versus Salmonella." _Sflrnd J Gastroenterol 43(5): 567-73.

Townsend, M. B., E. D. Dawson, et al. (2006). "Experimental evaluation of the FluChip
diagnostic microarray for influenza virus surveillance." J Clin Microbiol 44(8): 2863-71.

Tree, J. A., M. R. Adams, et a1. (2003). "Chlorination of indicator bacteria and viruses in
primary sewage effluent." Appl Environ Microbiol 69(4): 2038-43.

Tu, E. T., T. Nguyen, et a]. (2007). "Norovirus GII.4 strains and outbreaks, Australia. "
Emerg Infect Dis 13(7): 1128-30.

Turner, M., G. R. Istre, et a1. (1987). "Community outbreak of adenovirus type 7a
infections associated with a swimming pool." South Med J 80(6): 712-5.

Urisman, A., R. J. Molinaro, et a]. (2006). "Identification of a Novel Garnmaretrovirus in
Prostate Tumors of Patients Homozygous for R462Q RNASEL Variant." PLoS Pathog
2(3): e25.

Vainio, K. and M. Myrmel (2006). "Molecular epidemiology of norovirus outbreaks in

Norway during 2000 to 2005 and comparison of four norovirus real-time reverse
transcriptase PCR assays." J Clin Microbiol 44(10): 3695-702.

168

van der Werf, N., F. G. Kroese, et a]. (2007). "Viral infections as potential triggers of
type 1 diabetes." Diabetes Metab Res Rev 23(3): 169-83.

van Heerden, J ., M. M. Ehlers, et a1. (2005). "Detection and risk assessment of
adenoviruses in swimming pool water." J Appl Microbiol 99(5): 1256-64.

van Heerden, J ., M. M. Ehlers, et a]. (2005). "Prevalence, quantification and typing of
adenoviruses detected in river and treated drinking water in South Afi'ica." J Appl
Microbiol 99(2): 234-42.

van Zyl, W. B., P. J. Williams, et al. (2004). "Application of a molecular method for the
detection of group A rotaviruses in raw and treated water." Water Sci Technol 50(1):
223-8.

Vidal, R., P. Roessler, et al. (2006). "Novel recombinant norovirus causing outbreaks of
gastroenteritis in Santiago, Chile." J Clin Microbiol 44(6): 2271-5.

Wade, T. J., R. L. Calderon, et al. (2006). "Rapidly measured indicators of recreational
water quality are predictive of swimming-associated gastrointestinal illness." Environ
Health Perspect 114(1): 24-8.

Wade, T. J ., N. Pai, et al. (2003). "Do U.S. Enviromnental Protection Agency water
quality guidelines for recreational waters prevent gastrointestinal illness? A systematic
review and meta-analysis." Environ Health Perspect 111(8): 1102-9.

Wagner, M., H. Smidt, et al. (2007). "Unravelling microbial communities with DNA-
microarrays: challenges and future directions." Microb Eco] 53(3): 498-506.

Wang, D., L. Coscoy, et al. (2002). "Microarray-based detection and genotyping of viral
pathogens." Proc Natl Acad Sci U S A 99(24): 15687-92.

Wang, D., A. Urisman, et al. (2003). "Viral discovery and sequence recovery using DNA
microarrays." PLoS Biol 1(2): E2.

Wang, Z., L. T. Daum, et al. (2006). "Identifying influenza viruses with resequencing
microarrays." Emerg Infect Dis 12(4): 638-46.

Wanke, C. A. and R. L. Guerrant (1987). "Viral hepatitis and gastroenteritis transmitted
by shellfish and water." Infect Dis Clin North Am 1(3): 649-64.

Watson, M., J. Dukes, et a1. (2007). "DetectiV: visualization, normalization and
significance testing for pathogen-detection microarray data." Genome Biol 8(9): R190.

Weiss, S. R. and S. Navas-Martin (2005). "Coronavirus pathogenesis and the emerging
pathogen severe acute respiratory syndrome coronavirus." Microbiol Mol Biol Rev 69(4):
635-64.

169

Wetz, J. J ., E. K. Lipp, et al. (2004). "Presence, infectivity, and stability of enteric viruses
in seawater: relationship to marine water quality in the Florida Keys." Mar Pollut Bull
48(7-8): 698-704.

 

WHO (2002). The World Health Report 2002: Reducing Risks, Promoting Healthy Life.
Geneva, World Health Organization.

Wick, L. M., J. M. Rouillard, et a]. (2006). "On-chip non-equilibrium dissociation curves
and dissociation rate constants as methods to assess specificity of oligonucleotide
probes." Nucleic Acids Res 34(3): e26.

Wilson, W. J ., C. L. Strout, et a1. (2002). "Sequence-specific identification of 18

pathogenic microorganisms using microarray technology." Mol Cell Probes 16(2): 119-
27.

Winona, L. J., A. W. Ommani, et a]. (2001). "Efficient and predictable recovery of
viruses from water by small scale ultrafiltration systems." Can J Microbiol 47(11): 1033-
41.

Wong, C. W., C. L. Heng, et al. (2007). "Optimization and clinical validation of a
pathogen detection microarray." Genome Bio] 8(5): R93.

Wong, M., I. Xagoraraki, et al. (2007). Recent Advances and Future Needs in
Environmental Virology. Human Viruses in Watar. A. Bosch. Oxford, UK, Elsevier. 17:
259 - 284.

Wong, S., K. Pabbaraju, et al. (2008). "Detection of a broad range of human adenoviruses
in respiratory tract samples using a sensitive multiplex real-time PCR assay." J Med Virol
80(5): 856-65.

Wu, L., X. Liu, et al. (2006). "Microarray-based analysis of subnanograrn quantities of
microbial community DNAs by using whole-community genome amplification." Appl
Environ Microbiol 72(7): 4931-41.

Wu, L., D. K. Thompson, et al. (2004). "Development and evaluation of microarray-
based whole-genome hybridization for detection of microorganisms within the context of
environmental applications." Environ Sci Technol 38(24): 6775-82.

Wyn-J ones, A. P., R. Pallin, et a1. (1995). "Use of the polymerase chain reaction for the
detection of enteroviruses in river and marine recreational waters." Water Sci Technol 31:
337 - 344.

Xagoraraki, I., D. H. Kuo, et a]. (2007). "Occurrence of human adenoviruses at two
recreational beaches of the great lakes." Appl Environ Microbiol 73(24): 7874-81.

Yoder, J. S., B. G. Blackburn, et al. (2004). "Surveillance for waterbome-disease
outbreaks associated with recreational water--United States, 2001-2002." MMWR
Surveill Summ 53(8): 1-22.

170

Yong, S. F., Y. F. Ngeow, et a1. (2006). "Real-time PCR detection of male-specific
coliphages." Malays J Path0128(2): 79-82.

Young, N. J ., C. J. Thomas, et a]. (2006). "Real-time RT-PCR detection of Bovine Viral

Diarrhoea virus in whole blood using an external RNA reference." J Virol Methoda
138(1-2): 218-22.

 

Zhaohui, S., Z. Wenling, et a]. (2004). "Microarrays for the detection of HBV and HDV."
J Biochem Mol Biol 37(5): 546-51.

Zhou, J. (2003). "Microarrays for bacterial detection and microbial community analysis."
Curr Opin Microbiol 6(3): 288-94.

171

    

  

urill/11111111111111]191111111]

7