.1;

ralllvliti l.‘ . 0.3-5.9 . .. Alta . . iui‘iiianxvo. r}.c...v\4n.3t Wu§33n¥dsv :11»: n.

3%.? “Sag ‘ V E sssmwggw mi - ifig :3

all- :1! . . . . ..L

 

 

m
This is to certify that the
thesis entitled
TECHNIQUE FOR DETECTlNG SOURCES OF HUMAN
POLLUTION IN WATER USING AN Enterococcus BIOMARKER
0 presented by
H
a
>~ a :2’
a
5:: . . .
m .393 Tracre Michelle Jenkins
H
-= 3 s
2

has been accepted towards fulfillment
of the requirements for the

MS. degree in Crop and Soil Sciences

 

 

:smsg

Major Pfifessor’ 5 Signature
I 0// 0' fl 5‘

Date

MSU is an Affirmative Action/Equal Opportunity Institution

 

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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2/05 c/ClRC/DateDuejndd-p. 1 5

 

TECHNIQUE FOR DETECTING SOURCES OF HUMAN POLLUTION
IN WATER USING AN Enterococcus BIOMARKER

BY

Tracie Michelle Jenkins

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
MASTER OF SCIENCE
Department of Crop and Soil Sciences

2005

ABSTRACT

TECHNIQUE FOR DETECTING SOURCES OF HUMAN POLLUTION
IN WATER USING AN Enterococcus BIOMARKER

By
Tracie Michelle Jenkins

Microbial source tracking is a newly emerging field, which aims to identify sources of
fecal pollution in a body of water. The first portion of this research aimed to develop a
host-specific molecular marker for the identification of human fecal pollution targeting
an Enterococcus spp. The second portion of this research aimed to examine natural
waters for the marker and to develop a preliminary understanding of how the marker
related to other common microbial indicators, including enteric viruses. Two different
methods were examined for developing the molecular marker. The first method targeted
the ace gene in Enterococcusfaecalis and was based on its identification in individually
isolated colonies. While this method did not prove to be host-specific for any of the
human and animal species examined, it did show promise as a marker for the presence of
E. faecalis. The second method targeted the esp gene in Enterococcusfaecium and was
based on evaluating the entire microbial population of a sample. This marker was shown
to be human-specific in 97% (n=69) of the human fecal samples and in 0% (n=102) of
the animal fecal samples analyzed. Nine rivers in the State of Michigan were examined
for the esp marker and it was found in 22% of the samples. Culturable enteric viruses
were found in 33% of the samples. The Grand and Rouge Rivers were positive for both
indicators of human fecal pollution. This preliminary study demonstrates that the esp

marker shows promise as being a reliable indicator of human fecal pollution.

ACKNOWLEDGEMENTS

I would like to thank the Michigan Sea Grant Project Number M/PD-lO for funding the
Michigan river water quality monitoring portion of this research. I would also like to
thank the Homer Nowlin Endowed Chair of Water Research for funding the remainder of

this research.

Many thanks to my advisor Dr. Joan B. Rose, for her support and guidance throughout
my Masters research. Also, thank you to my committee members, Dr. Thomas S.
Whittam, Dr. Syed Anwar Hashsham, and Dr. Frank Dazzo, for their guidance during my

research.

Lastly, I would like to thank the other graduate students and employees in Dr. Rose’s lab

for their help in many aspects of my research. Your time and effort is greatly appreciated.

TABLE OF CONTENTS

LIST OF TABLES ............................................................................... vi

LIST OF FIGURES .............................................................................. vii
CHAPTER 1 ...................................................................................... 1
LITERATURE REVIEW ....................................................................... 1
1.1 Waterbome Diseases and Risks Associated with Exposure to Human Sewage... 1
1.2 Water Quality Indicators .................................................................. 4
1.3 Microbial Source Tracking ............................................................... 7

1.3.1 Phenotypic Library-Based Methods .............................................. 10

1.3.1.1 Antibiotic Resistance Analysis ............................................ 10

1.3.1.2 Multiple Antibiotic Resistance ............................................. 12

1.3.1.3 Carbon Source Utilization .................................................. 13

1.3.2 Genotypic Methods ................................................................. 14

1.3.2.1 Library-Based Methods ..................................................... 14

1.3.2.1.1 Ribotyping .......................................................... 14

1.3.2.1.2 PFGE and AFLP. .................................................. 16

1.3.2.1.3 Rep-PCR ............................................................ 17

1.3.2.2 Library Independent Methods ............................................. 18

1.3.2.2.1 Host-Associated Molecular Markers ........................... 18

1.4 Objectives of this Study .................................................................. 19

CHAPTER 2 ...................................................................................... 21

DEVELOPMENT AND VALIDATION OF A HUMAN-SPECIFIC Enterococcus

MARKER ........................................................................................ 21

2.1 Introduction ................................................................................. 21

2.2 Materials and Methods .................................................................... 22

2.2.1 Sample Collection ................................................................... 23

2.2.2 ace Target ............................................................................ 24

2.2.2.1 Sample Processing .......................................................... 24

2.2.2.2 Whole Cell Extraction ...................................................... 25

2.2.2.3 Polymerase Chain Reaction ................................................ 25

2.2.3 esp Target ............................................................................. 27

2.2.3.1 Bacterial Strains ............................................................. 27

2.2.3.2 Sample Processing .......................................................... 27

2.2.3.3 Membrane Elution ........................................................... 28

2.2.3.4 DNA Extraction ............................................................. 29

2.2.3.5 Polymerase Chain Reaction ................................................ 29

2.2.3.6 Minimum Detection Limit ................................................. 31

2.2.3.7 Persistence in Environmental Samples ................................... 31

2.2.4 Statistical Analysis .................................................................. 31

2.3 Results ....................................................................................... 32

iv

2.3.1 Speciation of Source Isolates ....................................................... 32

2.3.2 ace Target .............................................................................. 33
2.3.3 esp Target .............................................................................. 35
2.4 Discussion .................................................................................. 38
CHAPTER 3 ...................................................................................... 45
COMPARISION OF THE E. faecium esp MARKER WITH MICROBLAL AND
VIRAL INDICATORS IN WATER ........................................................... 45
3.1 Introduction ................................................................................. 45
3.2 Materials and Methods .................................................................... 46
3.2.1 Sample Collection ................................................................... 46
3.2.2 Indicator Analysis ................................................................... 46
3.2.3 Enteric Virus Analysis .............................................................. 48
3.2.4 Microbial Source Tracking ......................................................... 50
3.3 Results ....................................................................................... 50
3.3.1 Presence of Indicators ............................................................... 50
3.3.2 Presence of Enteric Viruses ........................................................ 52
3.3.3 Microbial Source Tracking ......................................................... 53
3.4 Discussion .................................................................................. 55
CHAPTER 4 ................................................. 61
FINAL DISCUSSION ........................................................................... 61
4.1 Introduction ................................................................................. 61
4.2 Future Application of Host-Specific Molecular Markers ............................ 61
4.3 Further Monitoring Surveys .............................................................. 63
APPENDIX I ...................................................................................... 64
APPENDIX 11 .................................................................................... 67
REFERENCES .................................................................................... 68

Table 1-1:

Table 1-2:

Table 1-3:

Table 2-1:

Table 2-2:

Table 2-3:

Table 2—4:

Table 2-5:

Table 2-6:

Table 2-7:

Table 3-1:

Table 3-2:

Table 3-3:

LIST OF TABLES

Key waterborne microorganisms and the symptoms and outbreaks they
caused from 1999-2002 ..........................................................

Surface water quality maximum allowable limits (CF U/ lOOmL) for
recreational use ....................................................................

Characteristics of the recommended and alternative water quality
indicators ...........................................................................

Speciation (api 20 Strep, BioMeriux) of Enterococcus isolates collected
from human and animal sources of fecal contamination ....................

Results of the AceA, AceA-1, and AceA-2 primers, which target the
ace gene, for all the Enterococcus isolates examined .......................

Results of the AceA, AceA-1, and AceA-2 primers, which target the
ace gene, for Enterococcusfaecalis isolates ..................................

Results of AceA primers in E. faecalis, E. faecium, and E. durans from
clinical, sewage, poultry, swine, and dairy cattle samples examined ......

The detection limit for naturally occurring esp in sewage influent,
septic tank effluent and marine waters being impacted with improperly
working septic tanks .............................................................

Detection of the esp marker in composite DNA samples extracted from
enterococci isolated from domestic sewage, livestock, and birds .........

Persistence of the naturally occurring esp gene from raw sewage in
culturable enterococci in fresh water and simulated sea water survival
studies ..............................................................................

Enterovirus and rotavirus primers used for the RT and Nested PCR
reactions ............................................................................

Location of sampling sites with the results from the indicator
microorganisms, enteric viruses, and the microbial source tracking gene

(esp) .................................................................................

Relative pollution ranking of rivers and results of viral analysis and
source tracking marker ...........................................................

vi

32

33

35

35

36

37

38

50

52

55

Figure 2-1:

Figure 2-2:

Figure 3-1:

Figure 3-2:

LIST OF FIGURES

Sample processing outline for the fecal (solid) samples .................. 23
Schematic diagram that outlines how liquid (animal lagoon,

wastewater, and environmental water) samples were analyzed for the
presence of enterococci (A) and the esp marker (B) ...................... 28
Map of the rivers sampled in the State of Michigan ...................... 54
Wastewater treatment locations in the State of Michigan based on

data available from the Michigan Department of Environmental
Quality ........................................................................... 57

vii

CHAPTER 1
LITERATURE REVIEW
1.1 Waterbome Diseases and Risks Associated with Exposure to Human Sewage

Waterbome diseases range from gastroenteritis to respiratory ailments and are caused
by enteric bacteria, viruses, and protozoans. These microorganisms are able to survive
and re-infect upon transmission through saline and fresh waters (Table 1-1) (Lund 1982;
Sliflco et al. 2000; Russell and Jarvis 2001; Leclerc et al. 2002; Adams et al. 2003). The
most severe symptoms occur in the young, elderly, and immunocompromised and
globally, waterborne disease represents a large portion of the total disease burden on
populations. World-wide, approximately four billion cases of diarrhea occur annually
(Murray and Lopez 1996), leading to 2.2 million deaths, mostly in children under the age
of five (World Health Organization 2000).

Fecal material, from animals and humans, and untreated or improperly treated
wastewaters are the main sources of waterborne pathogens (Craun 1986). The pathogens
are transmitted to human hosts via the fecal-oral cycle. The survivability and
reproductive success of the pathogens outside of a host varies from pathogen to pathogen.
Bacterial pathogens survive in both surface and ground waters for several weeks to
months at low (4°C) temperatures, while exposure to elevated (25°C) temperatures result
in reduced logio numbers within a few hours to days (Guan and Holley 2003). Protozoan
pathogens have developed a complex life-cycle with dormant, spore-like structures
termed (oo)cysts, which are able to survive for longer periods (no significant loss of
(oo)cyst integrity was observed after 30 days) as compared with bacteria at both low

(4°C) and elevated (30°C) temperatures (Nasser et al. 2003). Viral pathogens,

Table 1-1. Key waterborne microorganisms and the symptoms and outbreaks they caused

from 1999-2002 (FDA/CF SAN 1992; Lee et al. 2002; Blackburn et al. 2004; Yoder et al.

 

 

 

 

 

 

 

 

 

 

 

 

 

2004)
. . Drinking Recreational
Mrcggzgggrsm Symptoms Water Water
Outbreaks Outbreaks
Watery or sticky diarrhea,
Campylobacterjejuni fever, abdominal pain,
. . 4 l
(campylobacterrosrs) nausea, headache, muscle
pain
Shigella spp. Abdominal paln, cramps,
. . diarrhea, fever, vomiting, 0 5*
(shrgellosrs) .
mucus 1n stools
Rice water diarrhea,
Vibrio cholera abdominal cramps, nausea, O 0
(cholera) vomiting, dehydration,
shock
Pseudomonas F olliculitis (infection of hair 0 30
aerugz'nosa (dermatitis) follicles) or rash
Pathogenrgjfcherzchza Diarrhea, abdominal pain 6 8
Leglonella pneunwphzla Pneumonia 9* 2*
(Legronnarres disease)
Cryptosporidium Severe watery diarrhea, may
. . also be resprratory wrth * *
homlms . 2 27
(cryptosporidiosis) coughing and low-grade
fever
Glaraila Intestinal Diarrhea 8 1
(grardrasrs)
Naegleriafowleri Vanesfiom asymptomatic to 1 12
(meningoencephalitis) m1 gastroenteritis,
dysentery
Norovirus Nausea, vomiting, diarrhea, 8 8
(gastroenteritis) abdominal pain

 

 

 

* Represents outbreaks caused by other species in the same genus.

particularly Norovirus, display a cold weather seasonality phenomenon. This seasonality

trend infers that Norovirus is able to survive for longer periods in the environment in cold

conditions compared with warm, summer conditions (Mounts et al. 2000). Because the

protozoan and viral pathogens require a host in order to reproduce, they will not replicate

 

in the environment. Bacterial pathogens have the ability to replicate outside of a host;
however, there are few field studies to support this hypothesis.

The Centers for Disease Control and Prevention (CDC) defines an outbreak as more
than two people experiencing a similar illness after ingestion of drinking water or
exposure to recreational/occupational waters, with the epidemiological evidence
implicating water as the most probable source (Lee et al. 2002; Blackburn et al. 2004;
Yoder et al. 2004). From January 2001 to December 2002, 31 outbreaks were associated
with drinking water (Blackburn et al. 2004) and 65 with recreational water (Y oder et al.
2004) in the United States for a total of 96 outbreaks. In these two years, over 3,000
people become ill and 15 died after exposure to waterborne pathogens (Table 1-1).

The etiologic agents, including microbialand chemical agents, were identified in 81 of
the outbreaks. The causes of the remaining 15 outbreaks remain unknown. This is the
highest identification rate for the etiological agents for waterborne outbreak surveillance.
However, several key outbreaks were not identified and this continues to be a major
concern for those involved in preventing outbreaks and regulating the sources of potential
contamination. Many factors are involved in the identification of the agent. The outbreak
must be recognized by local health authorities and then reported to the CDC. In order to
identify the agent causing the outbreak, laboratories at the local/state level need to have
methods and equipment available for positive identification (Embrey et al. 2002; Hilbom
et al. 2002). Samples, especially water, need to be collected as soon as the outbreak is
identified to avoid the effects of pathogen dilution, transportation, and mortality, which
hinders their detection. However, a lag time will always exist between the initial period

of contamination and the identification that an outbreak is occurring due to the time

needed to go from the exposure to the illness (incubation period), to physicians reporting
the pathogen/symptom, to health officials identifying that a cluster is occurring in
pathogens/symptoms, and finally to the outbreak investigation. Timely identification of
the pathogens enhances the probability that immediate and appropriate action can be
taken to reduce the severity of the outbreak.

1.2 Water Quality Indicators

Although the microorganisms listed in Table 1-1 represent a large portion of the
causative agents responsible for waterborne disease outbreaks, routine monitoring of
drinking and recreational water quality does not screen for all of them. Instead, water
samples are monitored for the presence of indicator microorganisms, used to infer the risk
of acquiring an acute gastrointestinal illnessupon consuming drinking water or the
accidental ingestion of recreational waters (National Research Council of the National
Academies (NRC) 2004). They indicate the presence of fecal pollution and the possible
presence of enteric pathogens.

Historically, total coliforms were used to monitor water quality; fecal colifonns were
later used to monitor fecal contamination, as they are more specific to feces than total
coliforms (NRC 2004). However, in the 1970’s and early 1980’s, the United States
Environmental Protection Agency (US. EPA) initiated a series of studies designed to
determine if swimming in sewage-contaminated waters carried a health risk for bathers,
and if quantitative relationships between water quality and health could be determined
(Cabelli et al. 1979; Cabelli et al. 1982; Cabelli 1983). An additional goal was to identify
the most appropriate indicator (U .S. EPA 1986). Results led the US. EPA to move from

fecal coliforrns to alternative indicators recommending enterococci as the indicator of

choice in marine environments, and both Escherichia coli and enterococci, with E. coli
being the preferred indicator, in fresh water environments (Table 1-2).
Table 1-2. Surface water quality maximum allowable limits (CPU/lOOmL) for

recreational use.

 

 

Geometric Single Sample

Mean‘ Maximum
US. EPA‘
Marine Water enterococci 35 104
Fresh Water enterococci 33 61
E. coli 126 235
Michigan Department of
Environmental Quality2
Fresh Water E. coli 130 300

 

‘ (us. EPA 1986)
f (MDEQ 2002)
based on 2 5 samples equally spaced over a 30-day period

The US. EPA is currently conducting an extensive epidemiological investigation on
the occurrence of enterococci at marine and fresh water beaches. The results of these
studies will be critical in the re-evaluation of US. EPA’s current recreational water
quality standards. Based on these recommended guidelines, individual states set their
own water quality guidelines or criteria. The State of Michigan has based their criteria on
E. coli concentrations (Table 1-2). Additional microorganisms have also been recognized
as being potential indicators of fecal pollution and the presence of enteric pathogens
(Table 1-3).

Several recommendations have been made in regard to the levels of E. coli and

enterococci in relationship to health risk (U .S. EPA 1986). However, prediction of health

risks in recreational waters continues to be a source of debate among regulators and

 

scientists. Some studies maintain that these criteria are adequate for protecting bathers.

Wade et al. (2003) examined the literature using a meta-analysis. After identifying 27

Table 1-3. Characteristics of the recommended and alternative water quality indicators.

 

 

 

 

 

 

Indicator Definition Tarfit systems Methods
Escherichia Gram-negative, facultative Fresh water (American Public
coli anaerobe that is able to grow at systems used for Health Association

elevated temperatures (45°C) recreation and et al. 1999; US.
by fermenting lactose with the drinking. EPA 2002a; U.S.
production of carbon dioxide. EPA 2002d; U.S.
Member of the fecal coliform EPA 2002c)
group. Colilert®
Enterococci Gram-positive microbes that Marine and fresh (U .8. EPA 2002b;
are able to grow at low and water systems used US. EPA 20020)
elevated temperatures (10°C for recreation. EnterolertTM
and 45°C), at elevated pH
(9.5), and in 6.5% sodium
chloride. Member of the fecal
streptococci. ‘
Clostridium Gram-positive, anaerobic Marine systems in (Bisson and Cabelli
perj‘ringens spore-fonning, rod-shaped Hawaii as an 1979)
bacterium. alternative to E.
coli and
enterococci.
Bacteriophage Viruses that infect only Fresh surface and (U .S. EPA2001a;

 

bacteria.

 

ground waters.

 

US. EPA 2001b)

 

studies that met their selection criteria by including data on human exposure to water,

water quality measures, health outcomes, and study design, they concluded that, in

marine waters, enterococci were an adequate indicator of gastrointestinal illness; and that

E. coli was an adequate indicator in fresh waters. They also determined that the

guidelines set by the US. EPA in 1986 are adequate based on epidemiological

information available in both the published and non-published literature.

However, since the original recommendations were made in 1986, concern has been

raised in using E. coli and enterococci as indicators in both marine and flesh water

environments. Griffin et al. (2001) examined the different indicators being used in marine

 

waters, including E. coli, enterococci, Clostridium perfringens, and bacteriophages, and
concluded that the existing recommendations needed to be reevaluated for coastal waters.
The use of E. coli and possible enterococci are being challenged as data show that E. coli
is able to replicate in warm, tropical environments (NRC 2004). There is also a debate on
whether E. coli and enterococci are appropriate indicators of fecal pollution in fresh
waters. Recent studies conducted in the Great Lakes concluded that beach sand (Whitman
and Nevers 2003), gull feces (Fogarty et al. 2003; Haack et al. 2003), and algal mats
(Whitman et al. 2003) are all potential sources of these indicator microorganisms;
thereby providing a source of indicators that does not appropriately reflect the presence
of human pathogens. Various Enterococcus spp. are also known to exist as epiphytes in
the phylloplane of plants (Muller et al. 2001). How and to what degree these non-fecal
sources of indicators affect water quality and therefore the human health risk are
currently unknown.
1.3 Microbial Source Tracking

In 1999, the US. EPA proposed new rules (Total Maximum Daily Load (TMDL)
program) to tackle the problem of non-point sources of pollution that enter waterways.
This program outlines the flamework for regulating the amount and types of pollution
entering the water. First, the maximum amount of pollutant that a waterbody can receive
and still meet water quality standards needs to be calculated. Second, the relative
contribution of the pollutant flom all sources needs to be determined (U .8. EPA 1999).
Since E. coli and enterococci are frequently found in non-sewage sources, new
techniques need to be developed to determine the source of the microorganisms in order

to meet these new regulations.

Historically, the ratio between fecal coliforms and fecal streptococci was used to
determine whether the fecal pollution in a waterway was flom an animal or a human
source. This ratio was used because it was known that animals secreted higher
concentrations of fecal streptococci than fecal coliforms whereas the opposite was seen in
human feces. A ratio > 4.0 would indicate human pollution and a ratio $0.7 would
indicate animal pollution (Geldreich and Kenner 1969). However, this method has not
proven to be accurate at sourcing the pollution. This method flequently fails due to
differences in the survival rates of the two bacterial groups in water, and to differences in
the bacterial concentrations excreted in the feces of various mammals. Pourcher et al.
(1991) found that 39% of the samples flom human origin (feces and wastewater) had
ratios below 4 and the animals samples had ratios over 0.7, 86% of the time.

Recent advances in technology and microbiology have allowed researchers to revisit
the concept of differentiating sources of fecal pollution. Thus, the scientific field of
“Microbial Source Tracking” has emerged. This discipline involves methods, which
utilize the phenotypic and genotypic characteristics of bacteria, viruses, and in some
cases protozoa to assist in the identification of sources of fecal contamination; in addition
the occurrence of some chemicals and the presence of certain microorganisms are also
employed (Sinton et al. 1998; Scott et al. 2002; Simpson et a1. 2002; Meays et al. 2004).
The microbial source tracking methods can be divided into six categories: 1) phenotypic,
2) genotypic, 3) library-based, 4) non-library based or host-specific, 5) chemical, and 6)
the detection of specific microorganisms.

Antibiotic resistance analysis, multiple antibiotic resistance, and carbon source

utilization are the phenotypic methods commonly employed in microbial source tracking.

These methods characterize isolates flom human and animal sources by either the
isolates’ resistance to a series of antibiotics or the carbon sources, which then can
metabolize (Whitlock et al. 2002; Hagedom et al. 2003; Kelsey et al. 2003). Several
genotypic methods (ribotyping, RFLP, AF LP, rep-PCR) have been explored for
microbial source tracking, but all of them are based on cutting and then profiling the
target DNA (Parveen et a1. 1999; Dombek et al. 2000; D'Agata et al. 2001).

The phenotypic and genotypic methods create profiles, which characterizes isolates
flom different fecal sources. The profiles flom known fecal sources are placed together in
a library, which is used to determine the source of isolates flom unknown sources
(Harwood et al. 2003; Myoda et al. 2003). The non-library or host-specific methods
target a section of a microorganism’s genome in a PCR-based assay (Bernhard and Field
2000b; Khatib et al. 2002). The main advantage of the non-library methods over the
library methods is that the non-library methods are not subjected to the pitfalls of having
to create a large dataset, which is useful over an unknown spatial and temporal period.
However, the non-library methods are limited by the number of potential sources, which
can currently be targeted (Field et al. 2003; Stewart et al. 2003).

The chemical methods mainly rely on the presence of caffeine Rogers et al. 1986),
fecal sterols (Leeming et al. 1996), and fluorescent whitening agents (Hayashi et al.
2002) to determine human sewage contamination in water, although the fecal sterols are
also used to determine animal sources as well. It has also been demonstrated that the
occurrence of some microorganisms can be used to determine the presence of fecal
sources. Bacteroidesfragilis bacteriophages (Tartera and Jofle 1987; Tartera et al. 1989)

and human enteric viruses (Noble et al. 2003) have shown the presence of human

pollution, while animal enteric viruses (Maluquer de Motes et al. 2004) have been shown
to determine animal pollution. The F -RNA coliphages are also used to determine human
and animal sources of fecal pollution.
1.3.1 Phenotypic Library-Based Methods
1.3.1.1 Antibiotic Resistance Analysis

Antibiotic resistance analysis (ARA), multiple antibiotic resistance (MAR), and
carbon source utilization are the main phenotypic methods used in microbial source
tracking. These methods rely on the use of microorganisms that can be cultured.
Generally, ARA is developed by isolating bacterial colonies, either E. coli (Parveen et al.
1997) or enterococci (Wiggins 1996), from known sources of fecal pollution and placing
them onto media containing various antibiotics at four concentrations. Antibiotics vary
from study to study; however, they are usually those used in animal feeds and clinical
therapies (DuPont and Steele 1987). The results for each isolate are recorded into a
library based on the greatest antibiotic concentrations that yields colony formation, for
the purpose of creating an antibiotic resistance profile.

Discriminant analysis is often used to determine the classification accuracy of the
library (Wiggins 1996). The average rate of correct classification (ARCC) and
misclassification, which assesses the library’s internal accuracy, must be determined
before the library can be used to classify the source of unknown isolates. ARCC is
obtained by adding the number of known source isolates correctly classified into each
category and dividing by the total number of known source isolates and the rate of
misclassification is obtained by adding the number of known source isolates incorrectly

classified and dividing by the total (Harwood et al. 2000). In some instances, researchers

10

will perform a jackknife analysis to determine if the library contains adequate profiles. A
jackknife is performed by removing a sub-set of known source isolates from the library
and then testing the library with these isolates to determine how well they are classified
(Harwood et al. 2000; Wiggins et al. 2003). The size of a library may partially determine
the correct classification rate as small libraries have been shown to have a higher ARCC
than a larger library (Wiggins et a1. 2003; Johnson et al. 2004). In addition, the removal
of duplicate fingerprints in a library has resulted in a further reduction of the ARCC by
21.7% (Johnson et al. 2004). However, a larger library is more representative of the
microbial diversity present in the watershed (Wiggins et al. 2003).

For ARA analysis, the range of ARCC has been shown to be flom 51% to 95%, which
indicates the presence of variation in the design and evaluation of the libraries being used
to determine sources of fecal pollution (Wiggins 1996; Harwood et al. 2000; Whitlock et
al. 2002; Harwood et al. 2003; Wiggins et al. 2003). Higher ARCC are often obtained
when groups of related animals are pooled together. For example, Wiggins (1996)
obtained an ARCC of 72% when including isolates flom beef and dairy cattle, chicken,
turkey, human (wastewater), and streams. However, when they combined the beef and
dairy cattle and the chicken and turkey, they obtained an ARCC of 82% (cattle, poultry,
human, and streams). The average rate of misclassification has been shown to range flom
7.2% to 30.7% (Harwood et al. 2000; Whitlock et al. 2002; Harwood et al. 2003). During
a blind round-robin study, Harwood et al. (2003) found that the false-positive rate for
ARA ranged flom 39.4% to 54.6%. A false-positive occurs when a method yields a

positive for a fecal source when the fecal source is not present in the sample.

11

After creating a library with 1,435 isolates from known sources of human and animal
(cattle, poultry, and streams), Wiggins (1996) examined two local waterways in Virginia
to determine the largest source of fecal pollution. Of 105 isolates collected flom Cooks
Creek and 88 from Muddy Creek, 96% and 95% (two-way classification human and
animal), respectively, were classified as being of animal origin. Two-way classifications
are performed via discriminant analysis on two sources of fecal contamination. Whitlock
et al. (2002) created a similar library in Florida using isolates flom cattle, canines,
humans, and wild animals to determine that a creek, a puddle, and soil were indeed being
impacted by a failing septic system (52.8% to 91.7% of isolates were two-way classified
as human when comparing human versus animals).
1.3.1.2 Multiple Antibiotic Resistance

Multiple antibiotic resistance distinguishes resistant and sensitive isolates from fecal
sources and is one of the tools in microbial source tracking (Parveen et al. 1997; Kelsey
et al. 2003). The MAR index is determined by calculating the number of antibiotics that
an isolate is resistant to and then dividing by the total number of antibiotics tested
(Kaspar et al. 1990). A high MAR index indicates that the isolate has a high degree of
resistance to the antibiotics tested. Parveen et al. (1997) used MAR to examine point
(wastewater treatment effluent) and non-point (marsh runoff) sources of fecal pollution
and determined the average MAR indices for the sources to be 0.25 and 0.13,
respectively. An isolate with an indice of 1.0 would indicate that it is resistant to all of the
antibiotics that were tested (100%) and an index of 0.1 would indicate resistance to 10%
of the antibiotics. This is consistent with the finding that isolates from point sources have

a higher resistance to single and multiple antibiotics. When the point and non-point

12

isolates were compared with isolates flom humans and animals, it was shown that the
human isolates contained resistance patterns that were more similar to those of point
source isolates.

1.3.1.3 Carbon Source Utilization

Carbon source utilization can be used to determine the source of fecal pollution by
examining an isolate’s biochemical reactions (Pourcher et al. 1991; Hagedom et a1. 2003;
Wallis and Taylor 2003). Hagadom et al. (2003) examined the use of the Biolog GP2
MicroPlateTM for identifying fecal sources of Enterococcus spp. A library created flom
365 known isolates of human and non-human (poultry, cattle, equine, canine, goose, deer,
and muskrat) sources correctly classified the isolates with an ARCC of 92.7% (two-way
classification). Samples flom three bodies of water were then assayed to determine if the
library would correctly identify obvious sources of contamination. At each of the three
sites, the obvious source was correctly identified by a two-way classification (human
versus non-human).

The PhenePlate technique has also been used with Enterococcus spp. to demonstrate
the usefulness of carbon source utilization. Using Simpson’s Diversity Index, which is
used to measure bacterial diversity (ranging flom 0.0 to 1.0 representing low to high
diversity, respectively), it has been shown that Enterococcus spp. flom wastewater
contains a higher diversity of available biochemical pathways (mean diversity of 0.95)
than seen in the different animals (seabirds, cattle, sheep, donkeys, and dogs) examined
(mean diversity of 0.32-0.72) (Wallis and Taylor 2003). Therefore in recreational waters

with a high diversity index, one would expect wastewater to be the dominant factor

13

impacting the water and waters with a low diversity index would indicate an animal
source.

In the round-robin study which examined the efficacy in the different source tracking
methodologies, carbon source utilization was found to have ARCC’s above 80% with a
misclassification rate of approximately 5% (Harwood et al. 2003). However, the methods
false-positive rate was 51.5% to 66.7%. Therefore, the carbon source utilization faired
equally with the ARA methods examined.

1.3.2 Genotypic Methods
1.3.2.1 Library-Based Methods
1.3.2.1.1 Ribotyping

Ribotyping is restriction flagment length polymorphism (RFLP) of rRNA genes and is
used in differentiating between strains of a species as well as in tracking the source of
microorganisms (Gordillo et al. 1993; Parveen et al. 1999; Carson et al. 2001; Svec et al.
2001; Carson et al. 2003; Kuntz et al. 2003; Scott et al. 2003). The methods used in
ribotyping vary; however, the steps include bacterial isolation, digestion with restriction
enzymes, Southern blotting, and then hybridization with probes specific for 168 and 23S
rRNA. As with the phenotypic methods mentioned above, ribotyping requires the
development of a library containing known isolates to which unknown isolates are
compared.

Parveen et al. (1999) were the first to use ribotyping in source tracking. They were
able to demonstrate the utility of using ribotyping by having a two-way classification of
84% for human and animal fecal sources. The library used by Parveen et a1. (1999)

consisted of only 59 E. coli isolates. Carson et al. (2001) examined a total of 287 E. coli

14

isolates from eight different sources of fecal pollution and yielded an ARCC of 97.1% for
a two-way classification of human and non-human sources. When the eight-way
classification was performed, the ARCC was 73.56% demonstrating that ribotyping has a
greater discriminating power compared with the phenotypic methods. In the blind source
study, ribotyping exhibited correct classification of unknown isolates to a small library
between 81% and 100% of the time, with incorrect classification occurring between 19%
and 57% of the time (Myoda et al. 2003).

While ribotyping has been shown to be more accurate than ARA, there are still
problems with it. Hartel et al. (2003) has demonstrated that the ribotype profiles used will
vary depending on the diet of the particular host animal. In deer, they observed more
unique ribotypes in feces flom wild deer compared with penned deer. Temporal
variability has also been examined. Jenkins et al. (2003) examined ribotypes flom
yearling steers and determined that the majority were transient and unique to the specific
sampling event. This implies that all of the isolates used in building a library need to be
collected over a short temporal period and that a library developed in year one could not
be used the following year.

A geographic variability in the ribotype profiles also exists. Hartel et al. (2002)
showed that there is a very small percentage (0%-19%) of shared ribotypes between
geographically distant sources. However, unique ribotypes do exist for each of the host
animals (cattle, swine, chicken, and horses) examined. Based on their results, they
recommended that the library of known isolates needs to be created flom sources within

175km or the watershed being examined.

15

1.3.2.1.2 PFGE and AFLP

Pulse field gel electrophoresis (PFGE) and amplified flagment length polymorphism
(AF LP), like ribotyping, have been used in differentiating bacterial strains and also for
determining sources of fecal pollution (Gordillo et al. 1993; D'Agata et al. 2001; Parveen
et al. 2001; McLellan et al. 2003; Lu et al. 2004). While PFGE has proven to be a useful
method in epidemiological studies (D'Agata et al. 2001; Borchardt et al. 2003), its
usefiilness in source tracking has yet to be determined. Parveen et al. (2001) found that
PFGE was not able to differentiate between sources of fecal pollution and contributed
this to the specificity of the method to detect small sequence differences, which may not
be associated with a specific bacterial characteristic, such as host source. This conclusion
was also reached when PF GE was used to examine genetic variability in irrigation water
and sediments (Lu et al. 2004). In the blind source study, PFGE correctly classified the
unknown isolates 81% of the time with misclassification occurring in 21% of the
unknown isolates (Myoda et al. 2003).

AP LP has been used to show the existence of host specificity of vancomycin-resistant
E. faecium in hospital patients and in a select group of agricultural animals (Willems et
al. 2000). In this same study, AF LP was able to distinguish outbreaks, which occurred at
two separate hospitals. Guan et al. (2002) showed that AF LP can be used to predict
bacterial contamination by using discriminant analysis, instead of cluster analysis (Guan
et al. 2002). Depending on the primers used in the analysis, ARCC (three-way) ranged
flom 52.78% to 80.00%. Recently, AF LP has yielded correct classification rates ranging
flom 90.6% to 97.97% when examining human, bovine, and pig sources of E. coli

(Leung et al. 2004). In addition to high source classification, Leung et al. (2004) was also

16

able to assign isolates to their correct phenotype (non-pathogenic, VTEC, or ETEC) at a
range of 90.9% to 100%.
1.3.2.1.3 Rep-PCR

Repetitive element anchored PCR (rep-PCR) targets repetitive extragenic palindromic
PCR (REP), enterobacterial repetitive intergenic consensus (ERIC), or BOX elements,
and has been used comparing bacterial genomes (McLellan 2004). Dombek et al. (2000)
were the first to take this method and apply it to source tracking. Their studies examined
two different primers, BOX AIR and REP lR-REP 21, for PCR amplification and found
that the BOX AIR primers, while being equally capable of classifying human and sheep
isolates as the REP primers, BOX was superior at classifying other animal sources
(chickens, cows, ducks, geese, and pigs).

The ARCC has been found to be 79.3% when classifying three-ways using gull, cattle,
and sewage (McLellan et al. 2003) and 96.65% when classifying two—ways using human
and non-human (Carson et al. 2003). This difference in classification rates may be due to
differences in the primers being used, McLellan et al. (2003) used REP primers and
Carson et al. (2003) used BOX AlR primers. However, McLellan et al. (2003) also
examined ERIC primers on the same isolates and detemiined that while the REP and
ERIC primers generated comparable, but not identical dendrogram groups that these three
methods yield comparable results. However, ERIC-PCR was found to be poor at
assigning E. coli isolates based on the phenotypic groups of non-pathogenic, VTEC, and
ETEC (Leung et al. 2004). In the blind source study, rep-PCR was performed using the
BOX AIR primers. Unknown isolates were correctly classified between 86% and 90% of

the time with misclassification occurring 38% to 52% (Myoda et al. 2003).

17

Recently, a new method has been added to the microbial source tracking toolbox, a
horizontal, fluorophore-enhanced, repetitive extragenic palindromic-PCR DNA
fingerprinting technique (HF ERP). This method uses an internal ROX-labeled molecular
weight marker in the gel lanes to align, correct, and normalize the fluorescently labeled,
rep-PCR DNA fingerprint bands. HFERP reduces within-gel clustering of fingerprint
groups, which is normally seen in rep-PCR as well as reducing between- gel variations
due to band migration (Johnson et al. 2004). Johnson et a1. (2004) examining only the
unique bands found in rep-PCR and HFERP analysis were able to determine that the
correct classification rates for human and non-human were 60.9% and 45.8%,
respectively.
1.3.2.2 Library Independent Methods
1.3.2.2.1 Host-Associated Molecular Markers

The use of host-associated molecular markers (HAMM) circumvents all of the issues
associated with using a library-based method. There are two approaches to developing
HAMM in the literature for use in microbial source tracking. The first approach
developed by Bernhard et al. (2000a; 2000b) uses a combination of length heterogeneity
PCR (LH-PCR) and terminal restriction flagrnent length polymorphism (T-RFLP)
analysis to identify unique gene fragments within the desired host. Once the flagrnents
were identified and shown to be host-specific, primers were developed to target them
using standard PCR. This procedure has been used to develop two sets of primers, one for
humans and one for ruminants (cattle). The detection limit for this human marker was 4.2
x 105 bacterial cells per liter and was between 8.4 x 104 and 8.4 x 106 bacterial cells per

liter for cattle.

18

The other approach being used in developing HAMM is to use sequences in public
databases (NCBI or RDP-II) and create primers, which demonstrate host-specificity. This
method has been used by Khatib et al. (2002; 2003) to develop two sets of primers, one
that targets the LTIIa toxin gene in cattle (Khatib et al. 2002) and one that targets the
STH toxin gene in swine (Khatib et al. 2003). For the cattle marker, 100% PCR positives
were obtained when membranes containing an E. coli concentration of 2105 colonies per
filter were analyzed (Khatib et al. 2002). For the swine marker, 100% PCR positives
were obtained when membranes containing an E. coli concentration of 2100 colonies per
filter were analyzed (Khatib et al. 2003).

The library-independent genotypic methods described above were evaluated as part of
the blind source study mentioned several times previously. There was a wide-spread
inability to extract adequate amounts of DNA flom some of the samples received, which
especially hindered the use of T-RLFP for a Bacteroidetes marker (Field et. al.2003). An
additional hindrance in the study was the lack of markers, which could correctly identify
all of the sources in the samples analyzed. However, the ability of these methods used to
correctly classify the appropriate source of fecal material was between 75% and 100%
(Field et al. 2003).

1.4 Objectives of this Study

There is an ongoing need to further the microbial source tracking field, given the
inaccuracies associated with the current methods. Thus, research into new markers is
warranted. The specific goals of this research included:

I. The development of a host-specific molecular marker within Enterococcus

spp. for use in microbial source tracking.

19

II. The development of a method using archived bacterial strains and animal
fecal/lagoon samples to assay environmental samples for the host-specific
molecular marker.

III. The application of the newly developed methods along with conventional

water quality indicators in environmental water samples.

20

CHAPTER 2
DEVELOPMENT AND VALIDATION OF A HUMAN-SPECIFIC Enterococcus
MARKER
2.1 Introduction

Microbial source tracking continues to be a developing field of study where several
approaches and techniques are being explored to answer the same question, “Can the
microorganisms of fecal origin found in a body of water be traced back to their original
host sources?” The approaches examined have been based on both phenotypic and
genotypic characteristics of fecal coliforms, E. coli, and enterococci with each having its
own advantages and disadvantages (Scott et al. 2002). However, it is hypothesized that
the genotypic characteristics are better suited for use in determining the source origin of
microbial contaminants (Simpson et al. 2002). Of the genotypic techniques explored thus
far, it is hypothesized that the PCR-based host-associated molecular markers are more
accurate and precise in determining source origins (Griffith et al. 2003).

With an overall goal of developing and validating a human-specific host molecular
marker for use in microbial source tracking, this section examines the potential of two
surface proteins within Enterococcus. The first protein targeted was Ace (adhesin of
collagen flom enterococci), which is a member of the collagen-binding MSCRAMMs
(microbial surface components recognizing adhesive matrix molecules) and is encoded
by the ace gene (Rich et al. 1999). Ace was first discovered in Enterococcusfaecalis and
primers developed by Nallapareddy et al. (2000) showed specificity to clinical isolates of

E. faecalis. A preliminary assessment in clinical Enterococcusfaecium yielded no

21

evidence of an ace homolog. Duh et al. (2001) explored a range of Enterococcus spp. for
the presence of ace and found it in E. faecalis only. In exploring ace as a source tracking
marker, enterococci isolates were collected flom a variety of host sources and then
examined individually for the presence of the ace markers developed.

The second protein targeted was Esp, the enterococcal surface protein (Shankar et al.
1999). This target was initially discovered in clinical isolates of E. faecalis only, but a
homolog had been identified in isolates of E. faecium (Willems et a1. 2001). The function
of Esp (encoded by esp) is thought to increase virulence, which is supported by evidence
that shows it is located within unique pathogenicity islands within E. faecalis (Shankar et
al. 2001) and E. faecium (Leavis et al. 2004). While Hammerum and Jensen (2002) have
identified esp in E. faecalis in swine and poultry in addition to humans, this gene has not
been explored for a possible use in microbial source tracking. An initial screening of E.
faecium isolates flom clinical and sewage sources following the same methods used for
the ace marker yielded no presence of the esp marker. Therefore, a new method needed
to be explored, which scrutiniZed the whole enterococci population within a single
sarnple.

2.2 Materials and Methods

Two different approaches were undertaken in the development and evaluation of a
host-associated molecular marker for human sources of fecal pollution. Both individual
isolates and total populations grown on a membrane on MEI media were used to screen

for the markers (Figure 2-1).

22

 

Fecal Sample

1

3 grams into 30mL of PBS I

(A) (8)
Serial dilute and filter onto
mEI agar and Incubate at

41°C overnight

 

 

 

 

 

 

 

 

 

Spread plate onto mEl agar and

 

 

 

 

 

 

 

 

 

Incubate at 41°C overnight 1
Extract colonies from filter
by submerging In 10mL
7 TSB, vortexlng, and
Whole cell extraction on incubating 3 hrs at 41°C
Individual colonies 1
using 1M NaCl

 

 

 

Extract total DNA from
1mL of resulting culture

1

PCR using ace markers PCR using esp marker

 

 

 

 

 

 

 

 

 

 

 

 

Figure 2-1. Sample processing outline for the fecal (solid) samples. (A) summarizes the
method used to isolate and extract individual colonies for ace marker analysis. (B)
outlines the method used to examine the microbial population of a sample via membrane
filtration for the presence of the esp marker.
2.2.1 Sample Collection

Composite fecal samples were collected flom animal wastewater lagoons at poultry
(Bushnell, FL), dairy cattle (Hague, FL), and swine (Gainesville, FL) farms. Individual
animal fecal samples were collected for poultry, Canada geese, dairy and beef cattle, and
swine flom Michigan State University’s agricultural farms and wetlands. In addition,
animal scat samples were collected for sea gulls (Grand Haven, MI and Florida Keys),
pelicans (Florida Keys), and canines (canine park in Largo, FL). Human fecal samples

were collected flom domestic septic tanks from the Florida Keys, Gainesville, FL and

flom sites in Colorado. Domestic raw sewage, secondary sewage effluent, and filtered

23

wastewater were collected from wastewater treatment plants in Arizona, Florida, and
Michigan. Dr. Valerie J . Harwood flom the University of South Florida kindly provided
enterococci isolates flom clinical specimens. All samples were collected using aseptic
techniques and were transported to the lab on ice before being processed.

2.2.2 ace Target

The initial gene target for use as a human-specific marker was the ace gene, which is
found only in E. faecalis (Duh et al 2001).
2.2.2.1 Sample Processing

The fecal samples were analyzed according to Figure 2-1A. Three grams of sample
were weighed and placed into a sterile SOmL centrifuge tube. Thirty milliliters of
phosphate buffer saline (0.13M NaCl, 0.0051M NazHPO4, 0.0015M KHZPO4; pH 7.4)
(PBS) was added to the centrifuge tube, which was then vortex to suspend the sample.
Either lOOuL or 1mL of the resulting solution was then spread plated onto mEI agar;
mEnterococcus (mE) agar (Difco) supplemented with indoxyl B-D-glucoside (Sigma-
Aldrich, Co.) and incubated for 24 hours at 41°C. Animal lagoon samples were also
spread plated (l OOuL or 1mL) onto mEI agar. Colonies that were surrounded by a blue
halo were picked for Speciation using the api 20 Strep (bioMerieux, Inc.) biochemical test
strips. All isolates were placed into long-term storage at -80°C in TSB.

Human fecal samples flom septic tanks and wastewater treatment plants were
processed using membrane filtration. Serial dilutions were performed using PBS. The
appropriate dilutions were filtered through a 0.45 pm pore 47mm nitrocellulose-mixed
ester membrane filter (GE Osmonics). Filters were then placed onto mEI agar and

incubated at 41 i 05°C for 24 hours (U .3. EPA 2002). Colonies that were surrounded by

24

a blue halo were picked for Speciation using the api 20 Strep (bioMerieux, Inc.)
biochemical test strips. All isolates were placed into long-term storage at -80 °C.
2.2.2.2 Whole Cell Extraction

Isolates were removed flom long-tenn storage and placed into tryptic soy broth (TSB)
(Difco) and grown overnight at 41 °C. One milliliter of the resulting culture was placed
into a microcentrifuge tube and centrifuged at maximum speed for four minutes. The
supernatant was removed and 1 mL of a 1 M NaCl solution was placed into the
microcentrifuge tube and then vortexed to suspend the pellet. The sample was
centrifirged at maximum speed for four minutes and then was washed a total of three
times with the 1 M NaCl solution. Finally, the pellet was suspended in 50 uL of
DNase/RNase flee water and placed into a -20 °C freezer until analyzed by polymerase
chain reaction (PCR).
2.2.2.3 Polymerase Chain Reaction

The PCR primers used for detecting the ace gene, which is known to be in
Enterococcusfaecalis, were previously developed by Nallapareddy et al. (2000). From
the primers examined by Nallapareddy et al. (2000), a forward primer, AceF2 (aceA
forward 5’-GAG CAA AAG TTC AAT CGT TGA C-3’) and a reverse primer, AceR3
(aceA reverse 5’-GTC TGT CTT TTC ACT TGT TTC T-3’) were chosen.

The PCR reactions were performed in a 20 uL reaction mixture containing 1X PCR
Buffer (Qiagen, Inc.), 200 uM of each of the four deoxyribonucleotides (USB, Co.), 0.3
uM of each primer, 0.5 U of HotStarTaq DNA polymerase (Qiagen, Inc.), and 1 uL of
template DNA (whole cell extractions). The PCR amplification was performed with an

initial step at 95 °C for 15 minutes (to activate Taq), followed by 35 cycles of 94°C for 1

25

minute, 55 °C for 1 minute, 72°C for 1 minute with a final extension at 72°C for 5
minutes. The PCR products were then separated on a 1.5% agarose gel stained with
GeIStar nucleic acid stain (Carnbrex Bio Science, Inc.) and viewed under ultraviolet (UV)
light.

Two additional forward primers were developed in house to assist in distinguishing
human and non-human sources of feces. The DNA flagments from two human and three
poultry E. faecalis isolates, which all contained ace, were extracted and purified flom the
agarose gel using the QIAquick Gel Extraction Kit and Protocol (Qiagen, Inc.). One shot
chemical transformation was performed on the flagments using the TOPO TA Cloning
Kit (Invitrogen, Co.). The cloned DNA was then purified using the Wizard Plus SV
Minipreps DNA Purification System (Promega, Co.) and electrophoresed to ensure that
the plasmid was inserted correctly. All clones were then sequenced.

The sequences obtained were then aligned using Jellyfish Biowire Software. Unique
differences were present between the human and poultry sequences. Two new forward
primers were developed using the same software package. AceA-1, 5’-TGG AAT GAC
CGA GAA CGA TAG T-3’, was designed to identify animal sources. AceA-2, 5’-CGG
AAT GAC CGA GAA CGA TGG C-3’, was designed to identify human sources. The
aceA-reverse primer was used in all reactions.

The PCR reactions were performed in a 20 uL reaction mixture containing 1X PCR
Buffer (Qiagen, Inc.), 200 uM of each of the four deoxyribonucleotides, 0.3 uM of each
primer, 0.5 U of HotStarTaq DNA polymerase (Qiagen, Inc.), and l uL of template DNA
(whole cell extractions). The PCR amplification was performed with an initial step at

95°C for 15 minutes (to activate Taq), followed by 35 cycles of 94°C for 1 minute, 62°C

26

for 1 minute, 72°C for 1 minute with a final extension at 72°C for 5 minutes. The PCR
products were then separated on a 1.5% agarose gel stained with GelStar nucleic acid
stain (Cambrex Bio Science, Inc.) and viewed under UV light.
2.2.3 esp Target

The second gene target examined for use as a human-specific marker was the esp
gene, whose primers were developed to only target the E. faecium sequence.
2.2.3.1 Bacterial Strains

In addition to the isolates provided by Dr. Harwood, other members of
Enterobacteriaceae were evaluated to assess the potential for cross-reactivity of the PCR
primers developed in this portion of this study. The species tested were E. faecium
(ATCC# 19434), E. faecalis (ATCC# 19443), Enterococcus casseliflavus (ATCC#
700327), Enterococcus avium (ATCC# 14025), Enterococcus gallinarum (ATCC#
49573), Enterococcus durans (ATCC# 6056), Streptococcus bovis (ATCC# 15351),
Escherichia coli (ATCC# 15597 and 13706), and Klebsiella pneumoniae (ATCC#
43816). E. faecium strain C68, which contains the esp gene and was used as a positive
control in all PCR reactions, was kindly provided by Dr. Louis B. Rice of the Louis
Stokes Cleveland Veterans Affairs Medical Center in Cleveland, Ohio.
2.2.3.2 Sample Processing

The fecal samples were analyzed according to Figure 2-1B. Three grams of each
sample was weighed and placed into a sterile 50mL centrifuge tube. Thirty milliliters of
PBS was added to the centrifuge tube, which was then vortex to suspend the sample. The
suspended fecal samples (Figure 2-1B) along with the animal lagoon and the human

septic tank and wastewater samples (Figure 2-2A) were filtered through a 0.45um pore

27

47mm nitrocellulose-mixed ester membrane filter (GE Osmonics). Filters were then
placed onto mEI agar and incubated at 41 i 0.5 °C for 24 hours (U .S. EPA 2002).
Membrane filtration was used to concentrate the microorganisms and not as a means of
isolating individual colonies. Serial dilutions were used to estimate bacterial densities for

purposes of enumeration (Figure 2-2B).

 

 

Liquid Sample I

(A) (B)

 

l 6
Filter sample and
Dilute sample (10'1 place onto mEI agar

to 1045) and filter * 1
Incubate plate 5

l 24 hrs at 41°C
Incubate filter on 1 ,

mEl agar at 41 °C

10' 24 hrs 3 5 Transfer filter to 10mL 7
TSB, vortex, Incubate
l 1

3 hrs at 41°C
Determine CFU In Determ'“ 1

minimum detectlon
original sample IImlt Extract DNA from

1mL of resulting
culture

I

Perform PCRJ

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 2-2. Schematic diagram that outlines how liquid (animal lagoon, wastewater, and
environmental water) samples were analyzed for the presence of enterococci (A) and the
esp marker (B).
2.2.3.3 Membrane Elution

After overnight incubation, membranes were lifted off of the agar and placed into a 15

mL centrifuge tube (Figure 2-2A). Ten milliliters of TSB was added to the centrifuge

28

tube, which was then vortexed to ensure that the membrane was submerged with broth.
The tubes were then placed into a 41°C incubator for 3 hours. The centrifirge tubes were
then removed flom the incubator and vortexed again. The resulting bacterial culture
containing both target and non-target microorganisms allowed for the analysis of the
microbial population of the sample, which included both culturable and unculturable
Enterococcus spp.. One milliliter of the resulting culture was then placed into a
microcentrifuge tube to be used as the sample that underwent DNA extraction.
2.2.3.4 DNA Extraction

The microcentrifuge tube from the membrane elution was used as the sample for total
DNA extraction (Figure 2-2A). The tube was centrifuged at maximum speed for five
minutes. The supernatant was removed and 180 uL of lysis buffer (20 mg/ml lysozyme;
20 mM TriSOHCl, pH 8.0; 2 mM EDTA; 1.2% Triton®) was added. The pellet was
resuspended and then placed into a 35°C water bath for 30 minutes. Then 20 [AL of
Proteinase K (Qiagen, Inc.) and 200 uL of Lysis Solution (Qiagen, Inc.) was added to the
tube and vortexed. The tube was then placed into a 45°C water bath for 30 minutes. The
temperature of the water bath was increased to 95°C for 15 minutes. The microcentrifuge
tube was then removed from the bath and kept at room temperature. DNA extraction was
performed using QIAamp DNA Mini Kit according to the manufacture’s instructions
(Qiagen, Inc.).
2.2.3.5 Polymerase Chain Reaction

The PCR primers used for detecting the esp gene, which is in E. faecium and E.
faecalis, flom the DNA extractions were developed in order to detect the E. faecium esp

gene and not the E. faecalis esp gene. Primers were developed by aligning two E. faecium

29

sequences (GenBank accession numbers AF 444000 and AF 443999) with one E. faecalis
sequence (GenBank accession number AF 03477 9). Unique differences were identified
between these genes, and PCR primers were developed to specifically amplify only the E.
faecium variant (Jellyfish Biowire Software). The forward primer designed in this study,
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’)
developed previously by Hammerum and Jensen (2002) was used for all reactions.

The PCR reactions were performed in a 20 uL reaction mixture containing 1X PCR
Buffer (Qiagen, Inc.), 200 uM of each of the four deoxyribonucleotides (U SB, Co.), 0.3
uM of each primer, 0.5 U of HotStarTaq DNA polymerase (Qiagen, Inc.), and 1 uL of
template DNA. The PCR amplification was performed with an initial step at 95°C for 15
minutes (to activate Taq), followed by 35 cycles of 94°C for 1 minute, 58°C for 1 minute,
72°C for 1 minute with a final extension at 72°C for 5 minutes. The PCR products were
then separated on a 1.5% agarose gel stained with GelStar nucleic acid stain (Cambrex
Bio Science, Inc.) and viewed under UV light. The PCR product is 680 base pairs in
length.

The PCR product flom the E. faecium C68 was purified using a QIAquick PCR
Purification Kit (Qiagen, Inc.). The purified PCR product was cloned using the TOPO
TA Cloning ® Kit (Invitrogen, Co.). Sequencing of the product was performed at the
Genomics Technology Support Facility at Michigan State University using an ABI Prism
® 3100 Genetic Analyzer to verify the sequence of the positive control and specificity of

the primers.

30

2.2.3.6 Minimum Detection Limit

Five sewage, septic tank, and marine environmental samples flom the Florida Keys
suspected of containing human sewage contamination were collected. The samples were
serially diluted in PBS, and 1 mL of each dilution was membrane filtered thru a 0.45pm
pore 47mm nitrocellulose-mixed ester membrane filter (GE Osmonics) and placed onto
mEI media as described above (Figure 2-2B). Each filter was enumerated for total
enterococci and was prepared for analysis by PCR as described previously. Total viable
enterococci were then compared to PCR results in order to estimate the colony densities
that must be present to ensure detection of the human associated marker.
2.2.3.7 Persistence in Environmental Samples

In order to associate the presence of the esp gene and its detection with direct colony
counts, autoclaved flesh water and simulated marine (Instant Ocean) water samples were
spiked with primary sewage influent containing approximately 104 enterococci. Each
type of water sample (n=3) was then divided equally (25 ml) into 10 separate 50 ml
polypropylene tubes and were incubated in a water bath at 30°C. At intervals of 0, 3, 5,
7, 10, and 15 days, individual sample tubes were processed and analyzed for total
enterococci and the esp gene by plate counts and PCR, respectively, as described
previously.
2.2.4 Statistical analysis.

All statistical analyses (Chi-squared, t-tests) were performed using Pro-Stat statistical

software (Polysoftware International, Inc.).

31

2.3 Results

2.3.] Speciation of Source Isolates

A total of 337 enterococci isolates that were surrounded by a blue halo when

cultivated on mEI agar were speciated using api 20 Strep. Four species, E. faecalis, E.

faecium, E. durans, and E. gallinarum were identified flom seven different fecal sources

as outlined in Table 2—1. Overall, the human sources contained similar amounts of E.

faecalis (50%) and E. faecium (43%). However, the clinical sources were dominated by

E. faecalis (76%) and the sewage sources contained a proportionately higher

concentration of E. faecium (56%) compared with E. faecalis (35%).

Table 2-1. Speciation (api 20 Strep, BioMeriux) of Enterococcus isolates collected flom

human and animal sources of fecal contamination.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

S Number (%)) of Isolates Speciated Per Source
ource . . -
E. faecalis E. Eecmm E. durans E. M
Clinical 35 (76%) 10 (22%) 1 (2%) 0 (0%)
Sewage 28 (35%) 44 (56%) 7 (9%) 0 (0%)
Poultry 57 (97%) 2 (3%) 0 (0%) 0 (0%)
Swine 20 (5 7%) 14 (40%) 1 (3%) 0 (0%)
Dairy Cattle 14 (15%) 34 (36%) 46 (48%) 1 (1%)
Beef Cattle 2 (29%) 0 (0%) 5 (71%) 0 (0%)
Canine 0 0% 0 (0%) 16 (100% 0 0%
Total Human 63 (50%) 54 (43%) 8 (7%) 0 (0%)
| Total Animal 93 (44%) 50 (23.5%) 68 (32%) 1 (0.5%) |

 

 

 

 

Overall, the animal sources examined contained a greater number of isolates flom E.

faecalis (44%) > E. durans (32%) > E. faecium (23.5%) (Table 2-1). E. gallinarum was

only isolated once flom dairy cattle feces. The poultry samples examined yield the

greatest concentration (97%) of E. faecalis from the seven sources analyzed. While swine

32

contained a more equal ratio between E. faecalis and E. faecium isolates. The dairy cattle
had higher proportions of E. faecium (35.8%) and E. durans (48.4%) than any of the
other animal sources. While beef cattle contained a high proportion of E. durans (71.4%),
the canines examined contained exclusively E. durans.
2.3.2 ace Target

Table 2-2 shows the results of a total of 145 Enterococcus isolates flom six sources
screened using the AceA primers previously developed by Nallapareddy et al. (2000).
For the animals, only poultry, swine, and dairy cattle were further characterized for the
presence of the ace gene. The percentage of positive results varied between 21% (n=29)
for the dairy cattle and 67% (n=6) for the swine. Overall, 17 out of 47 or 36% of the
animal source isolates examined were positive for the ace gene. The results flom
Nallapareddy’s primers showed no specificity to source type. Therefore, the PCR band
flom two human and three poultry isolates was cloned and sequenced (Appendix I). From
these sequences two new forward primers were designed (AceA-l and AceA-2) and all
samples positive with the AceA primers were then characterized with the new primers.
Table 2-2. Results of the AceA, AceA-1, and AceA—2 primers, which target the ace gene,

for all the Enterococcus isolates examined.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Source Number of Isolates Positive
AceA AceA-1 AceA-2
Clinical 27 (n=41) 12 (n=27) 10 (n=27)
Sewage 9 (n=29) 2 (n=9) 6 (n=9)
Septic 28 (n=28) 0 (n=28) 21 (n=28)
Poultry 7 (n=12) 5 (n=7) 2 (n=7)
Swine 4 (n=6) 0 (n=4) 2 (n=4)
Dairy Cattle 6 (n=29) 0 (n=6) 1 (n=6)
Total Human 64 (n=98) 14 (n=64) 37 (n=64)
Total Animal 17 (n=47) 5 (n=l7) 5 (n=l7) |

 

33

Of the seven poultry samples that were positive with the AceA primers, five (71%)
were positive with AceA-1 and two (29%) with AceA-2 (Table 2-2). The swine isolates
were all negative using the AceA—1 primer, but two (50%) were positive with the AceA-2
primer. Again, all of the dairy cattle isolates were negative with the AceA-1 primer and
one (17%) isolate was positive with AceA-2. The above results demonstrate that not all
of the isolates, which were positive with AceA were also positive with AceA-1 and
AceA-2 describing a reduction in the sensitivity of the two newly developed primers.

Clinical, sewage, and septic tanks were examined as sources of human fecal
contamination. Table 2-2 shows that a total of 98 samples were characterized for the ace
gene and 65% were positive with AceA. The 28 isolates flom septic tanks were all
positive. However, only nine (31%) sewage isolates were positive. The AceA primers
were originally developed for use in detecting ace in clinical samples (N allapareddy et al.
2000). In this study, 66% (27 out of 41) of the clinical isolates were positive.

When the primers developed in this study were examined in the human sources, 22%
were positive with AceA-1 and 58% were positive with AceA-2. This difference is due to
zero AceA-1 positives flom the septic tank isolates while 21 were positive with AceA-2.
In the clinical isolates, 12 of 27 (44%) were positive with AceA-1 and 10 out of 27 (37%)
were positive with AceA-2. For the sewage isolates, two out of nine (22%) were positive
with AceA-l and six out of nine (67%) were positive with AceA-2.

Table 2-3 shows the results of the AceA, AceA-1, and AceA-2 primers for source
isolates that were identified by api 20 Strep as E. faecalis. Overall, 78% of the human
source isolates were positive with AceA, 14 of 35 for AceA-1, and 15 of 35 for AceA-2.

Twenty-six of the clinical isolates (n=30) were positive for AceA. Comparing this result

34

with the clinical results showed in Table2-2 reveals that one isolate that was identified as

E. faecium (Table 2-4) was positive using the AceA primers. However, this E. faecium

isolate was negative with AceA-1 and AceA-2.

Table 2-3. Results of the AceA, AceA-l, and AceA-2 primers, which target the ace gene,

for Enterococcusfaecalis isolates.

 

 

 

 

 

 

 

 

 

 

 

Number of Isolates Positive

Source AceA AceA-1 AceA-2
Clinical 26 (n=30) 12 (n=26) 9 (n=26)

Sewage 9 (n=15) 2 (n=9) 6 (n=9)

Poultry 7 (n=12) 5 (n=7) 2 (n=7)

Swine 4 (n=5) 0 (n=4) 2 (n=4)

Dairy Cattle 6 (n=8) 0 (n=6) 1 (n=6)
Total Human 35 (n=45) 14 (n=35) 15 (n=35)
Total Animal 17 (n=25) 5 (n=17) 5 (n=17)

 

 

 

 

Table 2-4. Results of AceA primers in E. faecalis, E. faecium, and E. durans flom

clinical, sewage, poultry, swine, and dairy cattle samples examined.

 

PCR Result
- (Isolates)

 

Positive

Number (%) of Species

1 (3%)

 
 

 

I Negative

 

18 (26%)

 

29 (97%)

 

17 (100%) |

 

2.3.3 esp Target

The detection limit of the naturally occurring esp gene in five separate sewage, septic,

and ambient water samples was determined, based on the coupled membrane filtration

and PCR method developed. As shown in Table 2-5, an average of 58 i 24 enterococci

colony forming units (CFU) per membrane filter was necessary to ensure the detection of

the esp gene and to identify the presence of the marker. The marine water samples being

35

influenced by failing septic tanks contained lower (< 104 CF U) overall concentrations of
culturable enterococci and therefore was only evaluated at those lower concentrations.
Table 2-5. The detection limit for naturally occurring esp in sewage influent, septic tank

effluent and marine waters being impacted with improperly working septic tanks.

 

 

 

 

 

 

Sample Number’I of enterococci screenedb (Number of Positive Samples)
Type _
(Numberof >104-<105 >103-<104 100-1000 13"1000‘ 1—10(u-5
— 58 :t 24) :I: 3)
Samples)
Primary
Influent (5) + (5) + (5) + (5) + (5) -
SePt1(c5 $I‘ank + (5) + (5) + (5) + (5) -
Water
Samplesc NAc + (5) + (5) + (5) -
(5)

 

 

 

 

 

 

 

" Higher values extrapolated from plates in the countable range (30-300 CFU).
b CPU/100 mL. ~

° Water samples contained ~ 1.5 x 103 CFU/lOO mL and were collected flom a marine
environment (Florida Keys) with a defined sewage input. NA indicates that the counts
were not in this range for any of these samples.

Primers specific for the E. faecium esp gene developed in this study as a marker of
human source fecal pollution were used to screen composite fecal samples flom humans,
livestock, and birds to test for potential cross-reactivity in various sources of fecal
contamination. A total of 171 samples containing 101 — 107 enterococci CFUs per
membrane filter were screened. The gene was detected only in samples from primary
sewage influent (n=40), secondary sewage effluent (n=10), filtered wastewater (n=5), and
septic tanks (n=14). The marker was not detected in any samples flom livestock or birds
(n=102) (Table 2-6). Overall, this gene was detected in 97% of the sewage and septic

tank samples. The two septic tank samples, which were negative for the marker,

contained low levels of culturable enterococci.

36

Six reference strains of enterococci along with three reference strains from other
Enterobacteriaceae were screened to test for further cross-reactivity. The primers did not
amplify any PCR products in any of these reference strains (data not shown). In addition,
16 E. faecium isolates flom clinical samples were evaluated for the presence of the esp
gene using the developed primers. These isolates were also negative (data not shown). In
addition, the E. faecium C68 strain positive by these primers was sequenced for
verification (Appendix II).

Table 2-6. Detection of the esp marker in composite DNA samples extracted flom

enterococci isolated flom domestic sewage, livestock, and birds.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

S # of CFUs # of esp Results
ource Sample Type , ,
Screened Samples Posrtlve Negative
Sewage Egg 106 - 107 40 40 0
Sewage 5:83:31? 103 - 104 10 10 o
Sewage Filter Effluent 10l — 102 5 5 0
Human Septic Tank 102 - 105 14 12 2a
Poultry Lagoon/Feces 105 - 106 6 0 6
Swine Lagoon/Feces 105 — 106 9 0 9
Dairy Cattle Lagoon/Feces 105 - 106 28 0 28
Beef Cattle Feces 105 -106 4 0 4
Canada Geese Feces 102 - 103 12 0 12
Seagull Feces 104 - 10S 28 0 28
Pelican Feces 105 - 106 7 0 7
Wild Birds Feces 104 - 105 8 0 8
Total Human 69 o7F 2
Total Animal 102 015 102

 

 

 

 

 

 

“ Total Enterococcus levels were < 1 x 102 CFU and showed atypical colony morphology.
b Significantly different statistically as shown by a chi square analysis p < 0.001.

Survival studies, which used naturally-occurring enterococci flom raw sewage
inoculated into flesh water and simulated sea water, were conducted and the results are

shown in Table 2-7. These studies were performed under elevated temperatures (30°C) to

37

enhance the die-off rates. The number of culturable enterococci dropped rapidly within
three days; falling to 19 CFU per mL by Day 7 in flesh water and were no longer
detectable by Day 15. In simulated sea water, the level of culturable enterococci declined
to 132 CFU by Day 7, to 71 CFU by Day 10 and were no longer detectable by Day 15.
The esp marker was only detectable up to Day 5, when levels of enterococci were
between 19 and 122 CFU/mL in fresh water and up to Day 10 at levels above 70
CFU/mL in simulated sea water.

Table 2-7. Persistence of the naturally occurring esp gene flom raw sewage in culturable

enterococci in flesh water and simulated sea water survival studies conducted at 37°C.

 

 

 

 

 

 

 

Water Total Culturable Enterococci (CFU/mL, u :1: sd)
Type Day 0 Day 3 Day 5 Day 7 Day 10 Day 15—
gfig (11.53;: 2.51 (a: 2 ‘ 1.22 (a 2 1.9 (a l 9.0 (a (M) < 6.37.}
Q=3) 104,. 0.64) x 10 0.30) x 10 0.30) x 10 x 10 10
PCR
Results + + + - - -
(+/-)
Simulated 1.34 (at 5.01 (:l: 2.65 (:t 1.32 (i
Sea Water 0.42) x 0.81) x 0.46) x 0.41) x 7':(1i09.;3) :31."
(“:31 104a 1021) 102b 102b
PCR
Results + + + + + -
(+/-)

 

 

 

 

 

 

 

 

j Values in the same column followed by the same letter are not statistically different at
f<0iii<51icates the detection limit of the samples
2.4 Discussion
In order for microbial source tracking (MST) methods to be most useful, they should
be specific and applicable over a broad geographic region. Recent research has indicated

that methods requiring a reference database may have limited utility for identification of

specific sources of fecal pollution when isolates are collected from multiple watersheds

38

(Griffith et al. 2003; Scott et a1. 2003; Johnson et al. 2004). Molecular methods currently
exist that identify specific sources of fecal pollution without relying on a database. These
methods can circumvent the inherent drawbacks of existing, library-dependent MST
methodologies (Bernhard and Field 2000a, 2000b; Khatib et al. 2002; Khatib et al. 2003;
Noble et al. 2003). These targeted methods also generally cost less and can produce
results within 1-2 days, which is significantly better than methods such as ribotyping,
which can take 7-14 days.

Most MST methods attempt to characterize indicator microorganisms (either
genotypically or phenotypically) on the basis of the host flom which they were isolated.
However, the ubiquitous nature of most of these indicators makes this task particularly
challenging. The Enterococcus genus contains multiple species and several researchers
have reported on the variable distribution of species of enterococci in different animal
hosts (Pourcher et al. 1991; Wheeler et al. 2002).

In this study, Speciation of Enterococcus was carried out using cultivation and
biochemical methods. These methods resulted in the identification of only four species;
E. faecalis, E. faecium, E. durans, and E. gallinarum flom a total of seven human and
non-human sources. The use of biochemical assays to speciate Enterococcus has been
addressed previously (Harwood et al. 2004; Velasco et al. 2004), and the use of
molecular based assays for identification have been suggested as being superior (Ozawa
et al. 2000; Manero et al. 2002). More specific biochemical tests for Speciation of
Enterococcus and further development of species or source-specific molecular markers
would be extremely useful for identifying sources of fecal pollution impacting a water

body.

39

The ace gene of E. faecalis is composed of a single A and repeated B domains. The
primers developed and used in this study target the A domain, which is 46% similar to
the Staphylococcus aureus Cna (collagen adhesion) A domain (Rich et a1. 1999). Since
the Ace A domain has been shown to be responsible for an isolate’s ability to bind to
collagen type I (Rich et al. 1999), it was the target of the primers developed initially by
Nallaparredy et a1. (2000) to characterize the gene and applied in this study for its
potential use as a source tracking marker. However, as this study examined the
occurrence of this gene in other sources beyond not only the clinical arena, but also
examined its occurrence in multiple Enterococcus spp. it has been concluded that it is
found in a variety of sources both human and non-human. Its occurrence in other
Enterococcus spp. suggests that it may be on a transposable element as a positive was
detected in a clinical E. faecium isolate. However, this conclusion has not been confirmed
in the literature. Duh et al. (2001) examined 103 E. faecium and 62 E. faecalis isolates all
flom clinical specimens and detected ace only in the E. faecalis isolates. However, the
primers used in this study and by Duh et al. (2001) targeted different base pair locations
based on the sequence from E. faecalis strain V583. In addition, it is possible that this
clinical isolate was incorrectly speciated. E. faecalis and E. faecium have very similar
biochemical characteristics differing most predominately in their ability to ferment
pyruvate (E. faecalis is positive and E. faecium is negative) (Wood and Holzapfel 1995).

Based on the cloning results flom seven isolates, two new forward primers were
developed to enhance the host identification of E. faecalis isolates. These two new
primers were shown to be too restrictive in their ability to identify the ace gene, which is

shown by a reduction in positives achieved with these primers after being positive with

40

the less restrictive primers developed by Nallaparreddy et al. (2000). In addition, these
primers were not shown to be more specific for a human vs. a non-human source.

Much like the approach of Khatib et al. in their identifications of a bovine and a
porcine-specific biomarkers in E. coli (2002; 2003), the intent of this research was to
target potential enterococcal virulence factors that have been associated only with human
clinical disease (Nallapareddy et al. 2000; Nallapareddy et al. 2003; Rice et al. 2003).
Multiple targets were evaluated and fecal samples flom domestic sewage, septic tanks,
livestock, and birds were screened. The research has demonstrated the esp method
presented here is more useful for tracking sources of human fecal pollution than the ace
methods.

The esp gene target is believed to be locatedon the bacterial chromosome (personal
communication, Dr. Louis B. Rice); therefore, the target is assumed to be stable and
would be less likely to be transferred (via horizontal transfer) to non-target
microorganisms in the environment. However, recent laboratory-based research indicated
that conjugative transfer of this gene between species is a possibility (Oancea et al. 2004)
as it has been found within a pathogenicity island (Leavis et al. 2004).

This test method circumvented the need to isolate and characterize individual
colonies of Enterococcus by examining the entire population of enterococci in a water or
fecal sample and specifically targeting a single HAMM, and allowed for the screening of
millions of bacteria flom multiple hosts. In agreement with previous literature (Shankar
et al. 1999; Willems et al. 2001; Eaton and Gasson 2002; Hammerum and Jensen 2002),
the primers specific for the E. faecium esp gene produced a product only in populations

of microorganisms cultured flom fecal samples of human origin.

41

The marker was consistently detected when levels of human derived enterococci in a
water or diluted fecal sample totaled approximately 50 to 100 colony-forming units
(CFU). This is useful, as the current recommendation for permissible levels of
enterococci as a fecal indicator in marine waters is less than 104 CFU/ 100ml for a single
sampling event (US. EPA 1986). Current federal flesh water recommendations of less
than 61 CFU/lOOmL for a single sample event (US. EPA 1986) also falls within the
detection limit of 58 i 24 CFU/membrane filter determined flom naturally occurring
sources of esp. The current water quality guidelines are based on total numbers of
enterococci, regardless of composition. While these guidelines have been shown to be
reliable as general indicators of water quality and health risks (Cabelli et al. 1979; Cabelli
et al. 1982; Wade et al. 2003), microbial source tracking methods should seek to better
characterize the constituent species within this group in order to develop specific tests for
specific sources of fecal pollution.

Survival studies showed that the marker was detectable in both fresh water and
simulated seawater for as long as total numbers of culturable enterococci were above 70
CFU, thus indicating that differential survival of enterococci populations in primary
sewage influent did not significantly affect the utility of this test. The study was
conducted at a relatively high environmental temperature (30°C). Therefore, the
experiment was a worst-case scenario of temperature and matrix (i.e. predation by
sewage microorganisms) and was designed only to indicate the stability of the marker
relative to culturable enterococci. In addition, the marker was no longer detected in filter

concentrated cells once enterococci were no longer able to be cultured flom the sample.

42

A literature review by Wade et al. supports culturable enterococci as indicators of
human health risk in recreational waters (2003) and culturable enterococci were also
shown to be superior markers for risk of disease due to contamination of groundwater
used for drinking when septic tanks were the source of contamination (Borchardt et a1.
2003). Finally, culturable enterococci have been shown to be extremely useful indicators
of the efficacy of wastewater treatment for purposes of reclamation (Rose et al. 2001).

Following along flom the above mentioned studies, I propose that the presence of
culturable enterococci and the detection of the E. faecium esp gene in a water sample
indicates the presence of human fecal contamination and potential human health risk,
while the absence of the marker in association with high counts of enterococci could then
also indicate an alternative source of contamination. Both results therefore will aid in the
fiirther investigation of the contamination and use of the water body. The method
developed herein allows for the identification of viable microorganisms, as well as
quantification and is compatible with current methods utilizing US. EPA approved
membrane filtration methods.

The ideal source-tracking tool is one that employs multiple targets for each individual
source group/animal. Overall, the results of the current study suggest that better genetic
characterization of the communities of enterococci present in different animal hosts
warrants further investigation. Future studies may reveal valuable information, which
could subsequently be used to design more specific tests for the identification of multiple
sources of fecal pollution. While the primers designed in this study for the ace gene have
thus far not demonstrated to be specific to a single source, it does show specificity to E.

faecalis. This specificity to E. faecalis is important flom a public health aspect in that E

43

faecalis is a major cause of nosacomial infections (Koch et al. 2004). The use of PCR
primers specific for the esp gene in Enterococcusfaecium as an index of human fecal
pollution may be a usefirl addition to the ever expanding microbial source tracking
toolbox. As with all source tracking methods, as they are utilized in new regions of the
country and with new potential sources of fecal pollution it is recommended that a set of
field collected Quality Assurance/Quality Control samples (known sources of fecal
contamination) be evaluated. The enterococci are becoming more useful indicators of
water pollution and of public health risk and the source of this risk should continue to be

investigated by research directed at further characterization of these microorganisms.

44

CHAPTER 3
COMPARISION OF THE E. faecium esp MARKER WITH MICROBIAL AND
VIRAL INDICATORS IN WATER
3.1 Introduction

The State of Michigan is a peninsula surrounded by three of the five Great Lakes. This
coupled with 49,141 miles of rivers and streams and over 11,000 ponds and in-land lakes
makes the state an attractive area for enjoying recreational water activities (MDEQ
2002). Therefore, these waters need to be monitored and protected flom a public health
perspective. Studies are being done within the Great Lakes region to monitor waters for
fecal indicators (Byappanahalli et al. 2003; Kinzelman et al. 2003; Whitman and Nevers
2003). However, studies within Michigan have been limited in scope to the beach areas
along the coast (Cabelli et al. 1982; Haack et al. 2003; Wheeler Alrn et al. 2003). In
addition while these studies enumerate the fecal bacterial concentrations, they do not
address the issues of bacterial sources and human health risks.

Microbial source tracking approaches need to be undertaken in order to better
understand where the microbial population is originating as this relates directly to hmnan
health risks. The source tracking tools available are wide and varied. However, an
enterococci host-specific molecular marker in humans has been developed and could
serve as a first approach in understanding sources within Michigan waters. In addition,
monitoring for human pathogens themselves assists in grasping the risk of being exposed
to human pathogens during recreational activities. While it is expensive to monitor for all
pathogens, viruses serve as a good choice because of their host specificity and low

infectious doses (Payment et al. 2000).

45

The purpose of this field study was to survey a range of recreating waters to determine
if the esp marker in enterococci was detectable under field conditions and to examine
how the presence/absence of the esp marker related to other microbial and viral indicators
commonly used in monitoring recreational water quality.

3.2 Materials and Methods
3.2.1 Sample Collection

Nine rivers (AuSable, Clinton, Grand, Kalamazoo, Raisin, Rouge, Saginaw,
Shiawassee, and Thunder Bay) that drain into the Great Lakes were chosen for analysis
of bacterial indicators and enteric viruses (see page , Figure 3-1). One site from each river
was sampled once during July 2003 under normal base-flow conditions. The location of
the sampling sites was based on previous data collected by the Michigan Department of
Environmental Quality (MDEQ), if available; otherwise, sampling occurred near the
river’s outflow into the receiving Great Lake (Table 3-2). One-liter grab samples were
collected for processing the bacterial indicators. The enteric viruses were collected via
filtration through Virusorb l MDS filters (Cuno Inc.) according to the US. EPA Manual
of Methods for Virology (2001a). At each site a minimum of 40 gallons (151.2 liters) of
surface water was filtered for enteric virus analysis. All samples were transported on ice
to the Water Quality and Health Laboratories at Michigan State University and processed
within six hours.

3.2.2 Indicator Analysis

Grab samples were analyzed for the presence of fecal coliforms, Escherichia coli,

enterococci and coliphages. Bacterial analysis was performed by aseptic membrane

filtration of water samples through a 0.45 pm pore 47mm nitrocellulose-mixed ester

46

membrane filter (GE Osmonics). Membranes for fecal coliform analysis were placed onto
mF C agar (Difco) sealed in a water-proof whirl-pak bag and submerged into a 44.5 :I:
0.2°C waterbath for 24 i 2 hours according to Method 9222D (American Public Health
Association et al. 1999). Colonies, which yielded a blue morphology color, were counted
as fecal coliforms. For E. coli analysis, membranes flom the fecal coliform analysis were
aseptically transferred to EC with MUG (Difco) supplemented with 1.5% agar (Difco)
plates and were incubated at 44.5 d: 02°C for 24 :t 2 hours according to Method 9222G
(American Public Health Association et a1. 1999). Colonies that fluorescenced upon
exposure to ultraviolet light at 365 nm were counted as E. coli. Membranes for
enterococci analysis were placed onto mEI (Difco) agar supplemented with indoxyl-fi-D-
glucoside (Sigma Inc.) and were incubated at 41_ :1: 05°C for 24 hours according to
Method 1600 (U .8. EPA 2002). Colonies that were surrounded by a blue halo were
counted as enterococci.

Coliphage analysis was done using a modified double-agar overlay method. Two host
bacteria, E. coli F+amp and E. coli C3000 (ATCC 15597) were used for coliphage
analysis. Samples for coliphage analysis were flom the IMDS filter elutions, except for
the Shiawassee River, which used a one-liter grab sample for its analysis. Either I or 2
mls of sample were mixed with 0.5 mls of a 4-hour log phase bacterial host in 1% tryptic
soy agar (Difco) overlays and then poured using aseptic techniques onto 1.5% agar
tryptic soy plates (Difco) and incubated at 37 i 1°C for 20 :t 4 hours according to Method
1602 (U .S. EPA 2001b). Clear zones (plaques) on the bacterial lawn were counted as

coliphage plaque forming units (PFU).

47

3.2.3 Enteric Virus Analysis

Samples were analyzed for enteric viruses according to the US. EPA’s Manual of
Methods for Virology (2001a). The Virusorb 1MDS filters (Curio, Inc.) were eluted with
1 L of 1.5% beef extract (BBL) with 0.05M glycine (Sigma, Inc.) (pH 9.2, 25°C). The
elution was reconcentrated using organic flocculation by the addition of 1M HCl until the
pH reached 3.5. However at pH 7.0, 50 mLs of the elution was removed and used in the
coliphage analysis (except for Shiawassee River). The remaining elution solution was
further concentrated by centrifugation and the floc containing the viruses was
resuspended in 0.15 M sodium phosphate, dibasic (NazHP04-7H20) and an antibiotic-
antimycotic solution (Cellgro) was added before being filter sterilized. The concentrated
samples were then placed onto MA-104, BS-C-l, and RD cell lines. The cells (samples
and negative controls) were observed for cytopathic effect (CPE) as an indication of the
presence of viable, infectious enteric viruses for 14 days. All positive results were
confirmed with a second passage.

Integrated cell culture reverse transcription polymerase chain reaction (ICC RT-PCR)
was performed on samples that were positive via cell culture to further characterize the
infectious viruses present. Viral RNA was extracted and purified flom the cell culture
supernatant using the QIAamp Viral RNA Mini Kit according to the manufacture’s
instructions (Qiagen, Inc.). QIAGEN OneStep RT-PCR Kit was used according to the
manufacture’s instructions for the RT-PCR reactions (Qiagen, Inc.). For the detection of
enteroviruses, primers (Table 3-1) for the RT and Nested PCR were previously developed

by Puig er al. (1994) based on human clinical specimens. For the detection of rotavirus,

48

the primers (Table 3-1) for the RT and Nested PCR were previously developed by

LeGuyader et al. (1994).

The RT-PCR reactions were run using a 50 11L reaction mixture containing 1X
QIAGEN OneStep RT-PCR Buffer (Qiagen, Inc.), 400 M of each of the four
deoxyribonucleotides (U SB, Co.), 0.6 11M of each primer, 2.0 [L QIAGEN One—Step
RT-PCR Enzyme Mix, and 10 11L of viral RNA extraction. For enterovirus detection, the
RT-PCR amplifications were performed with an initial step at 50°C for 30 minutes (RT)
followed by 95°C for 15 minutes (to activate Taq polymerase), followed by 35 cycles of
94°C for 30 seconds, 55°C for 30 seconds, and 72°C for 30 seconds with a final
extension at 72°C for 5 minutes. For rotavirus detection, the same amplification was
performed with an annealing temperature of 50°C.

The nested-PCR reactions were run using a 50 11L reaction mixture containing 10X
PCR buffer (Qiagen, Inc.), 1.5 mM MgC12 (Qiagen, Inc.), 200 uM of each of the four
deoxyribonucleotides (U SB, Co.), 0.3 uM of each primer, 2.5 U of HotStarTaq DNA
polymerase (Qiagen, Inc.) and 2.5 uL of the RT-PCR product. For enterovirus detection,
the nested-PCR amplifications were performed with an initial step at 95°C for 15 minutes
(to activate T aq polymerase), followed by 35 cycles of 94°C for 30 seconds, 55°C for 30
seconds, and 72°C for 30 seconds with a final extension at 72°C for 5 minutes. For
rotavirus detection, the same amplification was performed with an annealing temperature
of 50°C. PCR products were then separated on a 1.5% agarose gel stained with GelStar

nucleic acid stain (Cambrex Bio Science, Inc.) and viewed under a UV light.

49

Table 3-1. Enterovirus and rotavirus primers used for the RT and Nested PCR reactions.

 

 
 

PCR Prrime 7 , _ Seuence

  
     

 

Enterovirus“ RT Entl 5’-CGG TAC CTT TGT ACG CCT GT-3’
RT Ent2 5’-ATI‘ GTC ACC ATA AGC AGC CA-3’
Nested neEntl 5’-TCC GGC CCC TGA ATG CGG CTA-3’
Nested neEnt2 5’-GAA ACA CGG ACA CCC AAA GTA-3’
RotavirusT RT R1 5’-GGC TTT AAA AGA GAG AAT TTC CGT CTG (3-3:
RT R2 5’-GAT CCT GTT GGC CAT CC-3’
Nested R3 5’-GTA TGG TAT TGA ATA TAC CAC-3’
Nested Rp 5’-TCC ATT GAT CCT GTT ATT GG-3’

 

 

 

° Puig et al. (1994)
b Le Guyader et al. (1994)
3.2.4 Microbial Source Tracking

Sites that were positive for culturable enterococci using membrane filtration were
further evaluated to determine whether the fecal pollution present was flom humans. This
evaluation was based on the presence of the target gene, esp, which codes for the
enterococcal surface protein found in Enterococcusfaecium. Membrane filters flom the
enumeration process on mEI agar, which contained greater than 58 CFUs were
aseptically lifted, suspended in tryptic soy broth (Difco), vortexed vigorously and
incubated at 41 °C for two hours. DNA extraction was performed on 1 ml of the resulting
culture of bacteria using the QIAamp DNA Mini Kit (Qiagen, Inc.) as described in
Chapter 2. PCR was then performed as described in Chapter 2 for the detection of the
human esp marker. All negative samples were analyzed for PCR inhibitors by spiking the
sample DNA with DNA extracted flom the positive control, E. faecium strain C68.
3.3 Results
3.3.1 Presence of Indicators

Nine rivers in Michigan were sampled for the presence of microbial fecal indicators

(Figure 3-1; Table 3-2). The Rouge River had the greatest fecal coliform concentration of

50

155,000 CFU/ 100mL compared with the AuSable River, which had the lowest fecal
coliform concentration of 83 CFU/ 100mL (Table 3-2). If an 800 CFU/ 100mL fecal
coliform level is used, then 56% (5 of 9) of the samples exceeded the water quality
guidelines. Six of the rivers (67%) exceeded U.S. EPA’s single sample maximum of 235
CFU/100mL for swimmable water based on E. coli concentrations (U .S. EPA 1986). If
Michigan’s single sample limit of 300 CFU/lOOmL (MDEQ 2002) was used, five rivers
(56%) exceeded allowable limits. The largest number of exceedances of seven rivers
(78%) occurred when US. EPA’s single sample maximum of 61 CFU/ 100mL based on
enterococci (U .S. EPA 1986) concentrations were used. The AuSable and Thunder Bay
Rivers, which are considered pristine by the State of Michigan, were the only two rivers
analyzed in this study that did not exceed any of the bacterial water quality standards
currently in use.

The presence of two types of coliphage, bacteriophages that infect E. coli, was
examined in each of the rivers (Table 3-2). In each river, both somatic (C3000 bacterial
host) and F+-specific phage, which attached to the F pili of E. coli, (F+amp bacterial host)
were detected. While the numbers of phage present in the rivers varied flom less than 64
plaque forming units (PFU)/100L to greater than 63,000 PFU/100L, there was a general
increase in the presence of the coliphages from the rivers not listed as Areas of Concern
(AOCs) on EPA’S Great Lakes web site (AuSable, Grand, Shiawassee, and Thunder Bay
Rivers) to the rivers listed as AOCs (Clinton, Kalamazoo, Raisin, Rouge, and Saginaw

Rivers).

51

3.3.2 Presence of Enteric Viruses

Three rivers (Grand, Kalamazoo and Rouge Rivers) were positive for CPE indicating
the presence of viable enteric viruses, which was confirmed with a second cell line
passage (Table 3-2). The Grand and Kalamazoo Rivers contained viable viruses, which
generated CPE on MA-104 cells and the Rouge River presented CPE on MA-104 and
BS-C-l cells. The concentrations of viable enteric viruses ranged from 1 to 3 most
probable number (MPN)/100L. Table 3-2 presents the MPN results of the three positive
rivers as well as the detection limits for the samples, which were negative for CPE after

14 days.

Table 3-2. Location of sampling sites with the results flom the indicator microorganisms,

enteric viruses, and the microbial source tracking gene (esp).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

River‘l Fecal Escherichia En terococci Coli ha e Enteric es
(city) coliforms coli p g viruses” p
PFU/ 100 L MPN/ 100
CFU/10° "‘L C3000 F‘amp L°

AuSable'

(OscodaL 83 27 18 580 < 64 < 0.64 -

Clintonr

(Mt. 1200 700 250 265 397 < 0.53 -

Clemens)

Grand 1950 900 780 3075 1872 3.12 +

(Jackson)

Kalamazoo
(Galcsburg) 4050 375 216 1443 1924 3 .5 -
Raisin
@lin ton) 310 235 290 245 489 < 0.98 -
R°“g° 155000 8500 250 21,179 63,536 1.55 +
earbom)

Saginaw. _
(Bay City) 895 400 780 1605 988 < 0.62
Shlawass“ 515 165 370 20,000d < 10,000d < 0.62 -

(Owosso)

Thunder

Bay. 325 1 10 l l 1175 365 < 0.41 -

(Alpena)

 

 

 

 

a Rivers with an asterisk indicates that the sampling sites were located near the river’s
outfall into the surrounding Great Lake.

52

 

° Values with an < indicates that the microorganism was not detected based upon the

limits of detection.

° Most probable number (MPN)/ 100 L.

d Results are based on a one-liter grab sample and not the Virusorb lMDS filter elution
Cell culture supernatant containing viable enteric viruses was further evaluated for the

presence of enterovirus or rotavirus via RT and Nested PCR. The Rouge River was

positive for enteroviruses and negative for rotavirus. The Grand and Kalamazoo Rivers

were negative for both enterovirus and rotavirus (data not shown). Thus, it is likely that

other enteric viruses were associated with the cell culture CPE.

3.3.3 Microbial Source Tracking

The Grand and Rouge Rivers, which contained elevated concentrations of E. coli and
viable enteric viruses, tested positive for human fecal pollution based upon the esp
enterococci marker (Table 3-2). The remaining river samples tested negative for the
human fecal pollution marker and did not contain PCR inhibitors. The AuSable and
Thunder Bay Rivers were negative for the enterococci marker due to low concentrations
of enterococci in the samples, which were below the detection limit of 58 CFU for the

esp marker. Figure 3-1 shows the sites and results of the sampling in Michigan.

53

 

Figure 3-1. Map of the rivers sampled in the State of Michigan.
C sampling locations
+ rivers were positive for enteric viruses via cell culture
* rivers had E. coli levels which exceeded US. EPA guidelines (235 CFU/lOOmL)
I rivers had enterococci levels which exceeded US. EPA guidelines (61 CFU/lOOmL)
Grand and Rouge rivers positive for esp marker

54

3.4 Discussion

Recent publications examining the occurrence of indicator bacteria in the water and
sediments at Great Lake beaches have implied an increase in potential public health
impacts (Murray et al. 2001; Byappanahalli et al. 2003; Kinzehnan et al. 2003; Wheeler
Alrn et al. 2003; Whitman and Nevers 2003). Relative pollution rankings for each river
were assigned based upon the bacterial and somatic coliphage data collected in this
preliminary study (Table 3-3) (Griffin et al. 1999). These rankings show the Rouge River
(rank 9) as being the most polluted and the AuSable River (rank 1) as being the least
polluted. The Shiawassee (rank 5) and the Raisin (rank 3) Rivers, while listed as Areas
of Concern (AOC), had relatively low concentrations of fecal contamination present in
this single sample assessment.

Table 3-3. Relative pollution ranking of rivers and results of viral analysis and source

tracking marker.

 

 

 

 

 

 

 

 

 

 

 

Rivers Rankinga AOCb Cell Enteroviruses esp Wastewater
Culture (RT-PCR) Marker Facilitiesc

Rogge 9 Yes + + + 13
Grand 8 N0 + - + 64
Saginaw 7 Yes - - - 4
Kalamazoo 6 Yes + - - 22
Shiawassee 5 No - - - 14
Clinton 4 Yes - - - 4
Raisin 3 Yes - - - 24
Thunder 2 No - - - 2
Bay

AuSable 1 No - - - 4

 

 

 

 

 

 

 

 

a Ranking is based upon bacterial and somatic coliphage indicator data (1 is cleanest to 9
is the most contaminated).

° Areas of Concern (AOC).

° Data courtesy of the Michigan Department of Environmental Quality and includes all
wastewater treatment facilities along the rivers (Figure 2).

55

A variety of pollution sources, including wastewater treatment plants, septic tanks,
combined sewer overflows (CSOs), and agricultural and wild animals, may be impacting
a river. The microbial indicators used in determining water quality do not define the
source of the contamination. The relationship of the pollution ranking and the presence of
a human source (esp and/or enteric virus positives) were compared to the number of
sewage treatment facilities located along the rivers. Figure 3-2 shows the location of
wastewater treatment facilities (based on data provided by MDEQ). The Grand River
(rank 8) had the highest number of sewage treatment facilities and was positive for
viruses and the esp human marker (Table 3-3). The Saginaw River (rank 7), which
ranked high for fecal contamination, had fewer treatment plants located on it and was
negative for the enterococci marker and humanenteric viruses. In contrast, the
Kalamazoo River (rank 6) did have a higher number of treatment plants and was positive
for enteric viruses. The Shiawassee and Raisin Rivers showed relatively less pollution
compared with the other rivers, even with more sewage treatment facilities discharging to
them. Further assessment of discharge volumes, the types of wastewater disinfection,
CSOs, rainfall and distances between discharge and sampling sites is needed in future

comprehensive monitoring studies.

56

   

' Wastewater
Treatment Plants
far". Major Rivers
- Major Lakes
-- Michigan
Data Courtesy of MDEQ
Figure 3-2. Wastewater treatment locations in the State of Michigan based on data

available from the Michigan Department of Environmental Quality.

The presence of enteric viruses in three of these rivers suggests that wastewaters with
inadequate sewage treatment or faulty septic systems are impacting these waterways.
Payment et al. (2000) examined the occurrence of bacterial indictors and pathogens in the
Saint Lawrence River in Canada. Human enteric viruses were found in 39% of the 381

samples analyzed. These results are mimicked in this preliminary study in that 33% of the

57

Michigan samples were also positive. In addition to examining the source water used by
drinking water plants along the Saint Lawrence River, Payment et al. (2001) examined
the current treatment process used by the Montreal Urban Community wastewater
treatment plant before discharging into the river. The current treatment being used is a
physico-chemical treatment based on flocculation and sedimentation. Based on 104
untreated and 104 treated wastewater samples, Payment et al. (2001) showed a 0%
removal of human enteric viruses. The impact of the discharge on recreational users
needs to be further evaluated.

The widespread occurrence of bacteriophages, particularly the F +-specific
bacteriophages, in this study indicates that improperly treated sewage is impacting these
waters (Noble et al. 2003). However, the coliphage results for this study are not likely to
be an accurate measure of the mean concentrations that would be found normally in these
waters. The sampling method used, collection of bacteriophages using Virusorb 1MDS
filters, while allowing a large volume to be sampled, has a recovery efficiency of less
than two percent (Alonso et al. 1994). This low recovery efficiency greatly
underestimates the levels of coliphage present in these waterways. The use of the eluent
flom the lMDS filter allows only for a quick assessment of bacteriophage presence. This
preliminary study indicates that it would be advantageous to examine the waters directly
for bacteriophages in the future.

Enterococci is currently being used by the US. EPA for its’ National Epidemiological
and Environmental Assessment of Recreational Water Study being conducted at Great
Lake beaches as well as in marine waters and may be preferred as a human health risk

marker based on the outcome of this large-scale study. In this preliminary study, the

58

enterococci levels did not correlate well with the E. coli levels with a correlation value of
—0.038.Yet, two of the top three most contaminated sites, as measured by E. coli levels,
were positive for the human enterococci marker. In addition, these same two sites were
also positive for enteric viruses. Therefore, the MST methods bring a more
comprehensive understanding of the risk involved when recreating in areas impacted by
these waters.

Kinzelman et al. (2003) examined indicator levels in the water column and found that
by using enterococci as the indicator for beach closures instead of E. coli, there would
have been 56 more unsafe-recreational-water-quality advisories for the cities of
Milwaukee and Racine along Lake Michigan. This study found that while E. coli was
generally present in higher concentrations than enterococci, the stricter enterococci
guideline would have precipitated more closures for recreational activities. These results
are reflected in this study with enterococci levels exceeding the guidelines in 78% of the
samples and E. coli exceeding State standards 56% of the time.

This preliminary survey suggests that human sewage is impacting waters of the State
of Michigan and the surrounding Great Lakes. The findings in these single samples
indicate a need to further examine the water quality and public health impacts associated
with the designated uses of Michigan waters. This study demonstrates that the US. EPA
enterococci guidelines result in more warnings of a human health risk than do the
Michigan E. coli criteria. Therefore, it is recommended that the states in the Great Lakes
region begin to monitor recreational waters for enterococci. High levels of coliphage, the
presence of the enterococci human marker, and the detection of viral pathogens suggest

that a more comprehensive microbial/patho gen monitoring program is needed to evaluate

59

the impact of sewage discharges to Michigan waterways on public health. This should
include source tracking, alternative indicators and pathogen monitoring. These types of

studies will assist in prioritizing impaired waters and watershed management strategies to

maximize water quality and public health protection.

60

CHAPTER 4
FINAL DISCUSSION
4.] Introduction
The field of microbial source tracking is new and evolving and will continue to do so
in the near future as the methods are adapted from research laboratories to those
monitoring recreational waters. Those applying these methods want to know the variety
of sources that are impacting their waters, thus causing them to exceed the US. EPA’s
criteria. However, the methods are currently limited in their ability to answer these
questions accurately and because of their expense and/or time requirements (library-
dependent) (Scott et al. 2002) or their sensitivity (amount of marker present) or ability to
detect only one-specific source of contamination (library-independent) (Field et al. 2003).
In addition, the range of methods available makes it cumbersome for managers to
implement the necessary method(s) that will answer their specific questions. The US.
EPA is addressing this very issue by drafting a guidance document, which is designed to
assist managers in developing sample schemes tailored to their specific situation
(personal communication Jorge Santo-Domingo, US. EPA).
4.2 Future Application of Host-Specific Molecular Markers
The major goal of this research was to develop a host-specific molecular marker that
could be added to the microbial source tracking toolbox. Because the presence of human
fecal pollution in water indicates a higher potential for the presence of human pathogens
than the presence of animal feces, such as seagulls, the research focused around

developing a human-specific marker. In addition, the ability to develop a marker in a

61

bacterial group already being used in routine monitoring allows for the method to be
easily implemented in routine testing laboratories.

While two different markers were examined, only the esp marker demonstrated host-
specificity. Further research and development needs to be undertaken with the ace gene
in order for it to be used as a source tracking marker. This research targeted only a small
region of the ace protein. Further primer development and research may yet lead to a
host-specific molecular marker using the ace gene. In addition, the ace gene could still be
a valuable marker as it is specific to Enterococcusfaecalis.

While this research has shown that the esp marker is specific to human fecal pollution,
several areas need to be further addressed in order for this marker to reach its full
potential in source tracking. The validation of the esp marker examined fecal sources
flom several geographic locations including Florida, Michigan, Colorado, and Arizona:,
however, a more comprehensive study of potential geographic variability needs to be
undertaken. Until this is done, additional tests on known sources should be performed by
monitoring programs that are using this marker for identifying human fecal impacts.

Our initial research in the esp marker showed that it was not present in our library of
clinical isolates. In addition, two of the septic samples examined were negative for the
marker. Therefore, the prevalence of the esp gene in the human population needs to be
determined. Thus, a negative result cannot be interpreted as meaning there is no human
impact. Research also needs to be undertaken to examine how the esp gene behaves in the
environment. For example, do strains with the marker regrow in the water column or

sand?

62

4.3 Further Monitoring Surveys

The monitoring survey undertaken in Michigan was very preliminary. A more
comprehensive study including multiple sites along each river and sample collections
would allow for a more complete understanding of the water quality in these rivers. It
would be especially valuable in the three rivers that were positive for either the esp
marker or the enteric viruses. A comprehensive study would increase the understanding
of how the marker is transported in a natural system and further relate it to the occurrence

of human pathogens back as viruses.

63

APPENDIX I
Human clinical isolate H13 ace gene sequence

> GACAAAATTCAATCGTTGACAAAGTAGAATTAGATCACACTACTTTATAT
CAAGGGGAGATGACCTCCATTAAAGTATCTTTTAGTGACAAAGAAAATCAGA
AAATAAAACCCGGAGATACAATTACTTTAACCTTACCAGACGCACTAGTCGG
AATGACCGAGAACGATGGCTCACCACGAAAAATCAATTTAAATGGTTTAGGA
GAAGTTTTCATTTATAAAGATCATGTTGTAGCAACATTTAACGAAAAAGTTGA
ATCTTTACATAATGTGAATGGGCATTTTTCTTTCGGGATTAAAACGCTTATCA
CCAATAGTTCGCAACCGAATGTGATAGAAACGGATTTCGGAACAGCAACGG
CGACTCAACGTTTGACGATTGAAGGAGTGACTAACACAGAGACTGGTCAAA
TTGAGCGAGACTATCCTTTTTTTATAAAGTAGGCGATTTGGCTGGAGAGTCA
AATCAAGTACGTTGGTTTTTAAATGTGAACCTCAATAAATCCGATGTCACAGA
AGATATTTCAATTGCGGATCGACAAGGAAGTGGTCAACAATTAAATAAAGAG
AGTTTTACATTTGATATTGTGAATGACAAAGAAACTAAATATATTTCACTTGC
CGAGTTTGAACAACAAGGTTATGGCAAAATTGACTTTGTGACAGATAACGAC
TTTAATTTACGTTTTTATCGGGATAAAGCACGCTTTACTTCCTTTATCGTCCGT
TACACTTCGACAATCACGGAAGCGGGGCAACATCAAGCGACATTTGATAATA
GTTATGACATCAATTATCAACTAAACAATCAAGACGCAACGAATGAAAAAAA
TACATCACAGGTTAAAAATGTTTTTGTAGACGGTGAGGCAAGCGGCAATCA
AAATGTGGAAATGCCAACAGAAGAAAGTCTAGACATTCCTTTAGAGACAATA
GAAGAATGGGACCAAAGACACCTACTTCGGAACAGGCAACAGGAACAAGTG
AAA

Human clinical isolate H15 ace gene sequence

> GAGCAAAATTCAATCGTTGCCAAAGTAGAATTAGATCACACTACTTTATAT
CAAGGGGAGATGACCTCCATTAAAGTATCTTTTAGTGACAAAGAAAATCAGA
AAATAAAACCCGGAGATACAATTACTTTAACCTTACCAGACGCACTAGTCGG
AATGACCGAGAACGATGGCTCACCACGAAAAATCAATTTAAATGGTTTAGGA
GAAGCTTTCATTTATAAAGATCATGTTGTAGCAACATTTAACGAAAAAGTTG
AATCTTTACATAATGTGAATGGGCATTTTTCTTTCGGGATCAAAACGCTTATC
ACCAATAGTTCGCAACCGAATGTGATAGAAACGGATTTCGGAACAGCAACGG
CGACTCAACGTTTGACGATTGAAGGAGTGACTAACACAGAGACTGGTCAAA
TTGAGCGAGACTATCCTTTTTTTATAAAGTAGGCGATTTGGCTGGAGAGTCA
AATCAAGTACGTTGGTTTTTAAATGTGAACCTCAATAAATCCGATGTCACAGA
AGATATTTCAATTGCGGATCGACAAGGAAGTGGTCAACAATTAAATAAAGAG
AGTTTTACATTTGATATTGTGAATGACAAAGAAACTAAATATATTTCACTTGC
CGAGTTTGAACAACAAGGTTATGGCAAAATTGACTTTGTGACAGATAACGAC
TTTAATTTACGTTTTTATCGGGATAAAGCACGCTTTACTTCCTTTATCGTCCGT
TACACTTCGACAATCACGGAAGCGGGGCAACATCAAGCGACATTTGATAATA
GTTATGACATCAATTATCAACTAAACAATCAAGACGCAACGAATGAAAAAA
TACATCACAGGTTAAAAATGTTTTTTGTAGACGGTGAGGCAAGCGGCAATCA

64

AAATGTGGAAATGCCAA
Poultry isolate C15 ace gene sequence

> GAGCAAAATTCAATCGTTACCAAAGTAGAATTAGATCACACTACTTTATAT
CAAGGAGAGATGACCTCAATTAAAGTATCTTTTAGTGACAAAGAAAATCAGA
AAATAAAACCTGGAGATACTATTACTTTAACTTTACCAGACGCACTAGTTGG
AATGACCGAGAACGATAGTTCACCACGAAAAATCAATTTAAATGGTTTAGGG
GAAGTTTTTATCTATAAAGATCATGTTGTAGCAACATTTAACGAAAAAGTTGA
ATCTTTACATAATGTGAATGGGCATT'I‘TTC'I‘TTCGGGATTAAAACGCTTATCA
CCAATAGTTCTCAACCGAATGTGATAGAAACGGATTTCGGAACAGCAACGG
CGACTCAACGTTTGACGATTGAAGGAGTGACTAACACAGAGACTGGCCAAA
TTGAGCGAGACTATCCTTTTTTTATAAAGTAGGCGATTTGGCTGGAGAGTCA
AATCAAGTACGTTGGTTTTTAAATGTGAACCTCAATAAATCCGATGTCACAGA
AGATATTTCAATTGCGGATCGACAAGGAAGTGGTCAACAATTAAATAAAGAG
AGTTTTACATTTGATATTGTGAATGACAAAGAAACTAAATATATTTCACTTGC
CGAGTTTGAGCAACAAGGTTATGGCAAAATTGACTTCGTAACAGATAATGAC
TTTAACTTACGTTTTTATCGGGATAAAGCACGCTTTACTTCCTTTATCGTCCGT
TACACTTCGACAATCACAGAAGCAGGCCAACATCAAGCGACATTTGAAAATA
GTTATGACATCAATTATCAACTAAACAATCAAGACGCAACGAATGAAAAAAA
TACATCACAGGTTAAATATGTTTTTGTAGAAGGCGAGGCAAGCGGCAATCA
AAATGTGGAAATGCCAACAGAAGAAAGTCTAGACATTCCTTTAGAGACAATA
GATGAATGGGA

Poultry isolate C17 ace gene sequence

> GAGCAAAATTCAATCTTGACAAAGTAGAATTAGATCACACTACTTTATAT
CAAGGAGAGATGACCTCAATTAAAGTATCTTTTAGTGACAAAGAAAATCAGA
AAATAAAACCTGGAGATACTATTACTTTAACTTTACCAGACGCACTAGTTGG
AATGACCGAGAACGATAGTTCACCACGAAAAATCAATTTAAATGGTTTAGGG
GAAGTTTTTATCTATAAAGATCATGTTGTAGCAACATTTAACGAAAAAGTTGA
ATCTTTACATAATGTGAATGGGCATTTTTCTTTCGGGATTAAAACGCTTATCA
CCAATAGTTCTCAACCGAATGTGATAGAAACGGATTTCGGAACAGCAACGG
CGACTCAACGTTTGACGATTGAAGGAGTGACTAACACAGAGACTGGCCAAA
TTGAGCGAGACTATCCTTTTTTTATAAAGTAGGCGATTTGGCTGGAGAGTCA
AATCAAGTACGTTGGTTTTTAAATGTGAACCTCAATAAATCCGATGTCACAGA
AGATATTTCAATTGCGGATCGACAAGGAAGTGGTCAACAACTAAATAAAGAG
AGTTTTACATTTGATATTGTGAATGACAAAGAAACTAAATATATTTCACTTGC
CGAGTTTGAGCAACAAGGTTATGGCAAAATTGACTTCGTAACAGATAATGAC
TTTAACTTACGTTTTTATCGGGATAAAGCACGCTTTACTTCCTTTATCGTCCGT
TACACTTCGACAATCACAGAAGCAGGCCAACATCAAGCAACATTTGAAAATA
GTTATGCCATCAATTATCAACTAAACAATCAAGACGCAACGAATGAAAAAAA
TACATCACAGGTTAAAAATG'ITTTTGTAGGAGGCGAGGCAAGCGGCAATCA
AAATGTGGGAATGCCAACAGAAGAAAGTCTAGACATTCCTTTAGAGACAATA
GATGAGTGGG

65

Poultry isolate C18 ace gene sequence

> GAGCAAAAGTTCAATCTTACCAAAGTAGAATTAGATCACACTACTTTATAT
CAAGGAGAGATGACCTCAATTAAAGTATCTTTTAGTGACAAAGAAAATCAGA
AAATAAAACCTGGAGATACTATTACTTTAACTTTACCAGACGCACTAGTTGG
AATGACCGAGAACGATAGTTCACCACGAAAAATCAATTTAAATGGTTTAGGG
GAAGTTTTTATCTATAAAGATCATGTTGTAGCAACATTTAACGAAAAAGTTGA
ATCTTTACATAATGTGAATGGGCATTTTTCT’ITCGGGATTAAAACGCTTATCA
CCAATAGTTCTCAACCGAATGTGATAGAAACGGATTTCGGAACAGCAACGG
CGACTCAACGTTTGACGATTGAAGGAGTGACTAACACAGAGACTGGCCAAA
TTGAGCGAGACTATCCTTTTTTTATAAAGTAGGCGATTTGGCTGGAGAGTCA
AATCAAGTACGTTGGTTTTTAAATGTGAACCTCAATAAATCCGATGTCACAGA
AGATATTTCAATTGCGGATCGACAAGGAAGTGGTCAACAATTAAATAAAGAG
AGTTTTACATTTGATATTGTGAATGACAAAGAAACTAAATATATTTCACTTGC
CGAGTTTGAGCAACAAGGTTATGGCAAAATTGACTTCGTAACAGATAATGAC
TTTAACTTACGTTTTTATCGGGATAAAGCACGCTTTACTTCCTTTATCGTCCGT
TACACTTCGACAATCACAGAAGCAGGCCAACATCAAGCAACATTTGAAAATA
GTTATGTCATCAATTATCAACTAAACAATCAAGACGCAACGAATGAAAAAAA
TACATCACAGGTTAAAAATGTTTTTGTAGAAGGCGAGGCAAGCGGCAATCA
AAATGTGGAAATGCCACCAGAAGTAAGTCTAGACATTCCTTTAGAGACAATA
GATGAATGGGACCAAAGACACCTACTTCGG

66

APPENDIX II
Enterococcusfaecium strain C68 esp gene sequence

> CGTCGAAAGTTCGATTTCCGATCTTAACAGTTACAATACCTTTCGTACTAT
CAGTAACGTTAGAAGTATCAGGCTCTGTTTTCCATGAATACTCTGCATCTTCA
GGTAAGTTTTTATTATTCCGAATTGATTCTTTAGCATCTGGTTTAAATCCAACT
ACCACGGTTTGTTTATCAACCGCTTTTGGTGATTCCTTAATAACGGTTGAACC
TTCTTCTGGTTTATCAAAACCTGGAGAAACGATTTGGATTTTTAATTCATCTTT
CGCGATTAATTTGCTTGAATCTACACCCGTAAATTCAAATTCTACAGTATCTG
AAATTGGAGCCCCATCTTTTTCATTTGGAGCGATAGTTTTTTCTGCTAATACA
GTACCTTCTTTATCAACAAGTTGGGCTTTGTGACCTGTTCCCGCTAACTCGTG
GATGAATACTTTCCCCTTAACTGTTGTGTCAACATCATATACTTCTTCTAATTC
AGGTTTCCCGATATATGGATTCGTGTCTCCGCTCTCTTCTTTATCGTCAGGGTT
TCCATCACCATCAATATCTTGCMATAATAATAACCTGTTCCATAAGTGTTCT
GGATTAATTTTTTATTCGTATCC

67

REFERENCES

Adams, B. L., T. C. Bates and J. D. Oliver (2003). Survival of Helicobacter pylori in a
natural freshwater environment. Appl. Environ. Microbiol. 69(12): 7462-7466.

Alonso, M. C., J. M. Sanchez, M. A. Morinigo and J. J. Borrego (1994). A direct
membrane filter method for enumerating somatic coliphages in drinking water.
Microbiologia 10(3): 285-296.

American Public Health Association, American Water Works Association and Water
Environment Federation, Eds. (1999). Standard methods for the examination of
water and wastewater.

Bernhard, A. E. and K. G. Field (2000a). Identification of nonpoint sources of fecal
pollution in coastal waters by using host-specific 16S ribosomal DNA genetic
markers flom fecal anaerobes. Appl. Environ. Microbiol. 66(4): 1587-1594.

Bernhard, A. E. and K. G. Field (2000b). 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): 4571-4574.

Bisson, J. W. and V. J. Cabelli (1979). Membrane filter enumeration method for
Clostridium perfringens. Appl. Environ. Microbiol. 37(1): 55-66.

Blackburn, B. G., G. F. Craun, J. S. Yoder, V. Hill, R. Calderon, N. Chen, S. H. Lee, D.
A. Levy and M. J. Beach (2004). Surveillance for waterbome-disease outbreaks
associated with drinking water-United States, 2001-2002. Morb. Mortal. Wkly.
Rep. 53(SS-8): 23-44.

Borchardt, M. A., M. E. Stemper and J. H. Standridge (2003a). Aeromonas isolates flom
human diarrheic stool and groundwater compared by pulsed-field gel
electrophoresis. Emerging Infect. Dis. 9(2): 224-228.

68

Borchardt, M. A., P. D. Bertz, S. K. Spencer and D. A. Battigelli (2003b). Incidence of
Enteric Viruses in Groundwater flom Household Wells in Wisconsin. Appl.
Environ. Microbiol. 69(2): 1172-1 180.

Byappanahalli, M., M. Fowler, D. Shively and R. Whitman (2003). Ubiquity and
persistence of Escherichia coli in a midwestem coastal stream. Appl. Environ.
Microbiol. 69(8): 4549-4555.

Cabelli, V. J. (1983). Health effects for marine recreation waters. United States
Environmental Protection Agency 600: 1-80-031.

Cabelli, V. J ., A. P. Dufour, M. A. Levin, L. J. McCabe and P. W. Haberrnan ( 1979).
Relationship of microbial indicators to health effects at marine bathing beaches.
Am. J. Public Health 69(7): 690-696.

Cabelli, V. J., A. P. Dufour, M. A. Levin and L. J. McCabe (1982). Swimming-associated
gastroenteritis and water quality. Am. J. Epidemiol. 115: 606-616.

Carson, C. A., B. L. Shear, M. R. Ellersieck and J. D. Schnell (2003). Comparison of
ribotyping and repetitive extragenic palindromic-PCR for identification of fecal
Escherichia coli flom humans and animals. Appl. Environ. Microbiol. 69(3):

1 836-1 839.

Carson, C. A., B. L. Shear, M. R. Ellersieck and A. Asfaw (2001). Identification of fecal
Escherichia coli flom humans and animals by ribotyping. Appl. Environ.
Microbiol. 67: 1503-1507.

Craun, G. F. (1986). Waterbome Diseases in the United States. Boca Raton, FL, CRC
Press, Inc.

D'Agata, E. M. C., M. M. Gerrits, Y.-W. Tang, M. Samore and J. G. Kusters (2001).
Comparison of pulsed-field gel electrophoresis and amplified flagrnent-length
polymorphism for epidemiological investigations of common nosacomial
pathogens. Infection Control & Hospital Epidemiology 22: 550-554.

69

Dombek, P. E., L. K. Johnson, S. T. Zimmerley and M. J. Sadowsky (2000). Use of
repetitive DNA sequences and the PCR to differentiate Escherichia coli isolates
from human and animal sources. Appl. Environ. Microbiol. 66: 2572-2577.

Duh, R.-W., K. V. Singh, K. Malathum and B. E. Murray (2001). In vitro activity of 19
antimicrobial agents against enterococci from healthy subjects and hospitalized
patients and use of an ace gene probe flom Enterococcusfaecalis for species
identification. Microb. Drug Resist. 7(1): 39-46.

DuPont, H. L. and J. H. Steele (1987). Use of antimicrobial agents in animal feeds:
implications for human health. Rev. Infect. Dis. 9: 447-460.

Eaton, T. J. and M. J. Gasson (2002). A variant enterococcal surface protein espfm in
Enterococcus faecium; distribution among food, commensal, medical, and
environmental isolates. FEMS Microbiol. Lett. 216(2): 269-275.

Embrey, M. A., R. T. Parkin and J. M. Balbus, Eds. (2002). Handbook of CCL Microbes
in Drinking Water.

F DA/CFSAN . 1992. Foodbome Pathogenic Microorganisms and Natural Toxins
Handbook. US. Food and Drug Administration/Center for Food Safety and
Applied Nutrition.

Field, K. G., E. C. Chem, L. K. Dick, J. Fuhrrnan, J. F. Griffith, P. A. Holden, M. G.
LaMontagne, B. H. Olson and M. T. Simonich (2003). A comparative study of
culture-independent, library-independent genotypic methods of fecal source
tracking. J. Water Health 1(4): 181-194.

F ogarty, L. R., S. K. Haack, M. J. Wolcott and R. L. Whitman (2003). Abundance and
characteristics of the recreational water quality indicator bacteria Escherichia coli
and enterococci in gull faeces. J. Appl. Microbiol. 94(5): 865-878.

Geldreich, E. E. and B. A. Kenner (1969). Concepts of fecal streptococci in stream
pollution. J. Water Pollut. Control Fed. 41: R335-R352.

70

Gordillo, M. E., K. V. Singh and B. E. Murray (1993). Comparison of ribotyping and
pulsed-field gel electrophoresis for subspieces differentiation of strains of
Enterococcusfaecalis. J. Clin. Microbiol. 31(6): 1570-1574.

Griffin, D. W., C. J. 1. Gibson, E. K. Lipp, K. Riley, J. H. Paul and J. B. Rose (1999).
Detection of viral pathogens by reverse-transcriptase PCR and of microbial

indicators by standard methods in the canals of the Florida Keys. Appl. Environ.
Microbiol. 65(9): 4118-4125.

Griffin, D. W., E. K. Lipp, M. R. McLaughlin and J. B. Rose (2001). Marine recreation
and public health microbiology: quest for the ideal indicator. BioScience 51(10):
81 7-825.

Griffith, J. F., S. B. Weisberg and C. D. McGee (2003). Evaluation of microbial source
tracking methods using mixed fecal sources in aqueous test samples. J. Water
Health 1(4): 141-151.

Guan, S., R. Xu, S. Chen, J. Odumeru and C. Gyles (2002). Development of a Procedure
for Discriminating among Escherichia coli Isolates flom Animal and Human
Sources. Appl. Environ. Microbiol. 68(6): 2690-2698.

Guan, T. Y. and R. A. Holley (2003). Pathogen Survival in Swine Manure Environments
and Transmission of Human Enteric Illness--A Review. J Environ Qual 32(2):
383-392.

Haack, S. K., L. R. Fogarty and C. Wright (2003). Escherichia coli and enterococci at
beaches in the Grand Traverse Bay, Lake Michigan: Sources, characteristics, and
environmental pathways. Environ. Sci. Technol. 37(15): 3275-3282.

Hagedom, C., J. B. Crozier, K. A. Mentz, A. M. Booth, A. K. Graves, N. J. Nelson and
R. B. Reneau (2003). Carbon source utilization profiles as a method to identify
sources of faecal pollution in water. J. Appl. Microbiol. 94(5): 792-799.

Hammerum, A. M. and L. B. Jensen (2002). Prevalence of esp, encoding the enterococcal
surface protein, in Enterococcusfaecalis and Enterococcusfaecium isolates from
hospital patients, poultry, and pigs in Denmark. J. Clin. Microbiol. 40(11): 4396.

71

Hartel, P. G., J. D. Summer and W. I. Segars (2003). Deer diet affects ribotype diversity
of Escherichia coli for bacterial source tracking. Water Res. 37: 3263-3268.

Hartel, P. G., J. D. Summer, J. L. Hill, J. V. Collins, J. A. Entry and W. I. Segars (2002).
Geographic variability of Escherichia coli ribotypes flom animals in Idaho and
Georgia. J. Environ. Qual. 31(4): 1273- 1 278.

Harwood, V. J ., J. E. Whitlock and V. Withington (2000). Classification of antibiotic
resistance patterns of indicator bacteria by discriminant analysis: use in prediction

the source of fecal contamination in subtropical Florida waters. Appl. Environ.
Microbiol. 66: 3698-3704.

Harwood, V. J ., N. C. Delahoya, R. M. Ulrich, M. F. Kramer, J. E. Whitlock, J. R. Garey
and D. V. Lim (2004). Molecular confirmation of Enterococcusfaecalis and E.

faecium flom clinical, faecal and environmental sources. Lett. Appl. Microbiol.
38(6): 476-482.

Harwood, V. J., B. Wiggins, C. Hagedom, R. D. Ellender, J. Gooch, J. Kern, M.
Samadpour, A. C. H. Chapman, B. J. Robinson and B. C. Thompson (2003).
Phenotypic library-based microbial source tracking methods: Efficacy in the
California collaborative study. J. Water Health 1(4): 153-166.

Hayashi, Y., S. Managaki and H. Takada (2002). Fluorescent whitening agents in Tokyo
Bay and adjacent rivers: Their application as anthropogenic molecular markers in
coastal environments. Environ. Sci. Technol. 36: 3556-3563.

Hilbom, E. D., M. O. Royster and D. J. Drabkowski (2002). Survey of US public health
laboratories: microbial pathogens on the CCL. Journal AWWA 94(6): 88-95.

Jenkins, M. B., P. G. Hartel, T. J. Olexa and J. A. Stuedemann (2003). Putative temporal
variability of Escherichia coli ribotypes flom yearling steers. J. Environ. Qual.
32(1): 305-309.

Johnson, L. K., M. B. Brown, E. A. Carruthers, J. A. Ferguson, P. E. Dombek and M. J.
Sadowsky (2004). Sample size, library composition, and genotypic diversity
among natural populations of Escherichia coli flom different animals influence

accurarcy of determining sources of fecal pollution. Appl. Environ. Microbiol.
70(8): 4478-4485.

72

Kaspar, C. W., J. L. Burgess, I. T. Knight and R. R. Colwell (1990). Antibiotic resistance
indexing of Escherichia coli to identify sources of fecal contamination in water.
Can. J. Microbiol. 36: 891-894.

Kelsey, R. H., G. 1. Scott, D. E. Porter, B. Thompson and L. Webster (2003). Using
multiple antibiotic resistance and land use characteristics to determine sources of
fecal coliform bacterial pollution. Environ. Monit. Assess. 81: 337-348.

Khatib, L. A., Y. L. Tsai and B. H. Olson (2002). A biomarker for the identification of
cattle fecal pollution in water using the LTIIa toxin gene flom enterotoxigenic
Escherichia coli. Appl. Microbiol. Biotechnol. 59: 97-104.

Khatib, L. A., Y. L. Tsai and B. H. Olson (2003). A biomarker for the identification of
swine fecal pollution in water, using the STII toxin gene flom enterotoxigenic
Escherichia coli. Appl. Microbiol. Biotechnol. 63: 231-23 8.

Kinzehnan, J., C. Ng, E. Jackson, S. Gradus and R. Bagley (2003). Enterococci as
indicators of Lake Michigan recreational water quality: Comparison of two

methodologies and their impacts on public health regulatory events. Appl.
Environ. Microbiol. 69(1): 92-96.

Koch, S., M. Hufnagel, C. Theilacker and J. Huebner (2004). Enterococcal infections:
host response, therapeutic, and prophylactic possiblities. Vaccine 22: 822-830.

Kuntz, R. L., P. G. Hartel, D. G. Godfley, J. L. McDonald, K. W. Gates and W. I. Segars
(2003). Targeted sampling protocol as prelude to bacterial source tracking with
Enterococcusfaecalis. J. Environ. Qual. 32(6): 23 1 1-2318.

Le Guyader, F ., E. Dubois, D. Menard and M. Pommepuy (1994). Detection of hepatitis
A virus, rotavirus, and enterovirus in naturally contaminated shellfish and

sediment by reverse transcription-seminested PCR. Appl. Environ. Microbiol.
60(10): 3665-3671.

Leavis, H., J. Top, N. Shankar, K. Borgen, M. Bonten, J. van Embden and R. J. L.
Willems (2004). A vovel putative enterococcal pathogenicity island linked to the

esp virulence gene of Enterococcusfaecium and associated with epidemicity. J.
Bacteriol. 186(3): 672-682.

73

Leclerc, H., L. Schwartzbrod and E. Dei-Cas (2002). Microbial agents associated with
waterborne diseases. Crit. Rev. Microbiol. 28(4): 371-409.

Lee, S. H., D. A. Levy, G. F. Craun, M. J. Beach and R. L. Calderon (2002). Surveillance
for Waterbome-Disease Outbreaks-«United States, 1999-2000. Morb. Mortal.
Wkly. Rep. 51(SS-8): 1-28.

Leeming, R., A. Ball, N. Ashbolt and P. Nichols (1996). Using faecal sterols flom
humans and animals to distinguish faecal pollution in receiving waters. Water
Res. 30(12): 2893-2900.

Leung, K. T., R. Mackereth, Y.-C. Tien and E. Topp (2004). A comparison of AF LP and
ERIC-PCR analysis for discriminating Escherichia coli flom cattle, pig and
human sources. FEMS Microbiol. Ecol. 47: 111-119.

Lu, L., M. E. Hume, K. L. Stemes and S. D. Pillai (2004). Genetic diversity of
Escherichia coli isolates in irrigation water and associated sediments:
implications for source tracking. Water Res. 38(18): 3899-3908.

Lund, E. (1982). Waterbome virus diseases. Ecol. Dis. 1(1): 27-35.

Maluquer de Motes, C., P. Clemente-Casares, A. Hundesa, M. Martin and R. Girones
(2004). Detection of bovine and porcine adenoviruses for tracing the source of
fecal contamination. Appl. Environ. Microbiol. 70(3): 1448-1454.

Manero, A., X. Vilanova, M. Cerda-Cuellar and A. R. Blanch (2002). Characterization of
sewage waters by biochemical fingerprinting of Enterococci. Water Res. 36(11):
2831-2835.

McLellan, S. L., A. D. Daniels and A. K. Salmore (2003). Genetic characterization of
Escherichia coli populations flom host sources of fecal pollution by using DNA
fingerprinting. Appl. Environ. Microbiol. 69(5): 2587-2594.

McLellan, S. L. (2004). Genetic diversity of Escherichia coli isolated flom urban rivers
and beach water. Appl. Environ. Microbiol. 70(8): 4658-4665.

74

Meays, C. L., K. Broersma, R. Nordin and A. Mazumder (2004). Source tracking fecal
bacteria in water: a critical review of current methods. J. Environ. Manage. 73(1):
71-79.

Michigan Department of Environmental Quality Surface Water Division. 2002. Water
quality and pollution control in Michigan: 2002 Section 305(b) report. 323.1062
Microorganisms (Rule 62).

Mounts, A. W., T. Ando, M. Koopmans, J. S. Bresee, J. Noel and R. 1. Glass (2000).
Cold weather seasonality of gastroenteritis associated with Norwalk-like viruses.
J. Infect. Dis. 181: $284-$287.

Muller, T., A. Ulrich, E.-M. Ott and M. Muller (2001). Identification of plant-associated
enterococci. J. Appl. Microbiol. 91(2): 268-278.

Murray, C. and A. Lopez (1996). Global burden of disease and injury. Global health
statistics. Cambridge, MA, Harvard School of Public Health on behalf of the
World Health Organization and the World Bank. Series Vol. 2.

Murray, K. S., L. E. Fisher, J. Therrien, B. George and J. Gillespie (2001). Assessment
and use of indicator bacteria to determine sources of pollution to an urban river. J.
Great Lakes Res. 27(2): 220-229.

Myoda, S. P., A. Carson, J. J. Fuhrrnann, B.-K. Hahm, P. G. Hartel, H. Yampara-Iquise,
L. Johnson, R. L. Kuntz, C. H. Nakatsu, M. J. Sadowsky and M. Sarnadpour
(2003). Comparison of genotypic-based microbial source tracking methods
requiring a host origin database. J. Water Health 1(4): 167-180.

Nallapareddy, S. R., G. M. Weinstock and B. E. Murray (2003). Clinical isolates of
Enterococcusfaecium exhibit strain-specific collagen binding mediated by Acm,
a new member of the MSCRAMM family. Mol. Microbiol. 47(6): 1733-1747.

Nallapareddy, S. R., K. V. Singh, R.-W. Duh, G. M. Weinstock and B. E. Murray (2000).
Diversity of ace, a Gene Encoding a Microbial Surface Component Recognizing
Adhesive Matrix Molecules, from Different Strains of Enterococcusfaecalis and
Evidence for Production of Ace during Human Infections. Infect. Irnmun. 68(9):
5210-5217.

75

 

Nasser, A. M., N. Zaruk, L. Tenenbaum and Y. Netzan (2003). Comparative survival of
Cryptosporidium, coxsackievrius A9 and Escherichia coli in stream, brackish and
sea waters. Water Sci. Technol. 47(3): 91-96.

National Research Council of the National Academies, Ed. (2004). Indicators for
Waterbome Pathogens. Washington DC, National Academies Press.

Noble, R. T., S. M. Allen, A. D. Blackwood, W. Chu, S. C. Jiang, G. L. Lovelace, M. D.
Sobsey, J. R. Stewart and D. A. Wait (2003). Use of viral pathogens and
indicators to differentiate between human and non-human fecal contamination in
a microbial source tracking comparison study. J. Water Health 1(4): 195 -208.

Organization, W. H. 2000. The world health report: making a difference. Geneva, World
Health Organization.

Ozawa, Y., P. Courvalin and M. Galimand (2000). Identification of enterococci at the
species level by sequencing of the genes for D-alanine:D-alanine ligases. Syst.
Appl. Microbiol. 23: 230-237.

Parveen, S., R. L. Murphree, L. Edmiston, C. W. Kaspar, K. M. Portier and M. L.
Tamplin (1997). Association of multiple-antibiotic-resistance profiles with point

and nonpoint sources of Escherichia coli in Apalachicola Bay. Appl. Environ.
Microbiol. 63: 2607-2612.

Parveen, S., K. M. Portier, K. Robinson, L. Edmiston and M. L. Tamplin (1999).
Discriminant analysis of ribotype profiles of Escherichia coli for differentiating

human and nonhuman sources of fecal pollution. Appl. Environ. Microbiol. 65(7):
3142-3147.

Parveen, S., N. C. Hodge, R. E. Stall, S. R. Farrah and M. L. Tamplin (2001). Phenotypic
and genotypic characterization of human and nonhuman Escherichia coli. Water
Res. 35(2): 379-386.

Payment, P., A. Berte, M. Prevost, B. Menard and B. Barbeau (2000). Occurrence of
pathogenic microorganisms in the Saint Lawrence River (Canada) and

comparison of health risks for populations using it as their source of drinking
water. Can. J. Microbiol. 46(6): 565-576.

76

Payment, P., R. Plante and P. Cejka (2001). Removal of indicator bacteria, human enteric
viruses, Giardia cysts, and Cryptosporidium oocysts at a large wastewater
primary treatment facility. Can. J. Microbiol. 47(3): 188-193.

Pourcher, A.-M., L. A. Devriese, J. F. Hernandez and J. M. Delattre (1991).
Enumberation by a miniaturized method of Escherichia coli, Streptococcus bovis

and enterococci as indicators of the origin of faecal pollution of waters. J. Appl.
Bacteriol. 70(6): 525-530.

Puig, M., J. Jofle, F. Lucena, A. Allard, G. Wadell and R. Girones (1994). Detection of
adenoviruses and enteroviruses in polluted waters by nested PCR amplification.
Appl. Environ. Microbiol. 60(8): 2963-2970.

Rice, L. B., L. Carias, S. Rudin, C. Vael, H. Goossens, C. Konstabel, I. Klare, S. R.
Nallapareddy, W. Huang and B. E. Murray (2003). A Potential Virulence Gene,

hYlEfm, Predomiantes in Enterococcusfaecium of Clinical Origin. J. Infect. Dis.
187(1): 508-512.

Rich, R. L., B. Kreikemeyer, R. T. Owens, S. LaBrenz, S. V. L. Narayana, G. M.
Weinstock, B. E. Murray and M. Hook (1999). Ace is a collagen-binding
MSCRAMM flom Enterococcusfaecalis. J. Biol. Chem. 274(38): 26939-26945.

Rose, J. B., D. E. Huffman, K. Riley, S. A. Farrah, J. Lukasik and C. L. Hamann (2001).
Reduction of enteric microorganisms at the Upper Occoquan Sewage Authority
Water Reclamation Plant. Water Environ. Res. 73: 711-720.

Russell, J. B. and G. N. Jarvis (2001). Practical mechanisms for interrupting the oral-
fecal lifecycle of Escherichia coli. J. Mol. Microbiol. Biotechnol. 3(2): 265-272.

Scott, T. M., S. Parveen, K. M. Portier, J. B. Rose, M. L. Tamplin, S. A. Farrah, A. K00
and J. Lukasik (2003). Geographical variation in ribotype profiles of Escherichia
coli isolates from humans, swine, poultry, beef, and dairy cattle in Florida. Appl.
Environ. Microbiol. 69(2): 1089-1092.

Scott, T. M., J. B. Rose, T. M. Jenkins, S. R. Farrah and J. Lukasik (2002). Microbial
Source Tracking: Current Methodology and Future Directions. Appl. Environ.
Microbiol. 68(12): 5796-5803.

77

Shankar, N., C. V. Lockatell, A. S. Baghdayan, C. Drachenberg, M. S. Gilmore and D. E.
Johnson (2001). Role of Enterococcusfaecalis surface protein esp in the
pathogenesis of ascending urinary tract infection. Infect. Irnmun. 69(7): 4366-
4372.

Shankar, V., A. S. Baghdayan, M. M. Huycke, G. Lindahl and M. S. Gilmore (1999).
Infection-derived Enterococcusfaecalis strains are enriched in esp, a gene
encoding a novel surface protein. Infect. Immun. 67(1): 193-200.

Simpson, J. M., J. W. Santo Domingo and D. J. Reasoner (2002). Microbial source
tracking: state of the science. Environ. Sci. Technol. 36(24): 5279-88.

Sinton, L. W., R. K. Finlay and D. J. Hannah (1998). Distinguishing human flom animal
faecal contamination in water: a review. N. Z. J. Mar. Freshwater Res. 32: 323-
348.

x"
‘-

Slifl<o, T. R., H. V. Smith and J. B. Rose (2000). Emerging parasite zoonoses associated
with water and food. Int. J. Parasitol. 30(12-13): 1379-1393.

Stewart, J. R., R. D. Ellender, J. A. Gooch, S. C. Jiang, S. P. Myoda and S. B. Weisberg
(2003). Recommendations for microbial source tracking: Lessons flom a methods
comparison study. J. Water Health 1(4): 225-231.

Svec, P., I. Sedlacek, R. Pantucek, L. A. Devriese and J. Doskar (2001). Evaluation of
ribotyping for characterization and identification of Enterococcus
haemoperoxidus and Enterococcus moraviensis strains. FEMS Microbiol. Lett.

203(1): 23-27.

Tartera, C. and J. J ofle (1987). Bacteriophages active against Bacteroidesfragilis in
sewage-polluted waters. Appl. Environ. Microbiol. 53(7): 1632-1637.

Tartera, C., F. Lucena and J. Jofre (1989). Human origin of Bacteroides fragilis
bacteriophages present in the environment. Appl. Environ. Microbiol. 55(10):
2696-2701.

United States Environmental Protection Agency. 1986. Ambient Water Quality Criteria
for Bacteria-1986. EPA440/5-84-002.

78

United States Environmental Protection Agency. 2001a. Manual of Methods for
Virology. EPA 600/4-84/013 (N14).

United States Environmental Protection Agency. 2002a. Method 1103.1: Escherichia coli

(E. coli) in water by membrane filtration using membrane-thermotolerant
Escherichia coli agar (mTEC). EPA 821-R-02-020.

United States Environmental Protection Agency. 2002b. Method 1106.1: enterococci in
water by membrane filtration using membrane-Enterococcus-esculin iron agar
(mE-EIA). EPA 821 -R-02-021 .

United States Environmental Protection Agency. 2002c. Method 1600: enterococci in

water by membrane filtration using membrane-Enterococcus indoxyl-fi-D-
glucoside agar (mEI). EPA 821-R-02-022.

United States Environmental Protection Agency. 2001b. Method 1601: male-specific (F)
and somatic coliphage in water by two-step enrichment procedure. EPA 821-R-
01-030. '

United States Environmental Protection Agency. 2001c. Method 1602: male-specific (FL)

and somatic coliphage in water by single agar layer (SAL) procedure. EPA 821-
R-01-029.

United States Environmental Protection Agency. 2002d. Method 1603: Escherichia coli
(E. coli) in water by membrane filtration using modified membrane-
thermotolerant Escherichia coli agar (modified mTEC). EPA 821-R-02-023.

United States Environmental Protection Agency. 2002c. Method 1604: Total coliforms
and Escherichia coli in water by membrane filtration using a simultaneous
detection technique (M1 medium). EPA 821-R-02-024.

United States Environmental Protection Agency. 1999. Total maximum daily load
(TMDL) program. EPA 84l-F-99-003A.

Velasco, D., S. Perez, F. Pena, M. A. Dominguez, M. Cartelle, F. Molina, R. Moure, R.
Villanueva and G. Bou (2004). Lack of correlation between phenotypic

techniques and PCR-based genotypic methods for identification of Enterococcus
spp. Diagn. Microbiol. Infect. Dis. 49: 151-156.

79

 

Wade, T. J., N. Pai, J. N. S. Eisenberg and J. J. Colford (2003). Do US. Environmental
Protection Agency water quality guidelines for recreational waters prevent

gastrointestinal illness? A systematic review and meta-analysis. Environ. Health
Perspect. 111(8): 1102-1109.

Wallis, J. L. and H. D. Taylor (2003). Phenotypic population characteristics of the
enterococci in wastewater and animal faeces: implications for the new European
directive on the quality of bathing waters. Water Sci. Technol. 47(3): 27-32.

Wheeler, A. L., P. G. Hartel, D. G. Godfrey, J. L. Hill and W. I. Segars (2002). Potential
of Enterococcusfaecalis as a human fecal indicator for microbial source tracking.
J. Environ. Qual. 31(4): 1286-1293.

Wheeler Alrn, E., J. Burke and A. Spain (2003). Fecal indicator bacteria are abundant in
wet sand at freshwater beaches. Water Res. 37(16): 3978-3982.

Whitlock, J. E., D. T. Jones and V. J. Harwood (2002). Identification of the sources of
fecal coliforms in an urban watershed using antibiotic resistance anlaysis. Water
Res. 36: 4273-4282.

Whitman, R. L. and M. B. Nevers (2003). Foreshore sand as a source of Escherichia coli
in nearshore water of a Lake Michigan beach. Appl. Environ. Microbiol. 69(9):
5555-5562.

Whitman, R. L., D. A. Shively, H. Pawlik, M. B. Nevers and M. N. Byappanahalli
(2003). Occurrence of Escherichia coli and enterococci in Cladophora
(Chlorophyta) in nearshore water and beach sand of Lake Michigan. Appl.
Environ. Microbiol. 69(8): 4714-4719.

Wiggins, B. A. (1996). Discriminant analysis of antibiotic resistance patterns in fecal
streptococci, a method to differentiate human and animal sources of fecal
pollution in natural waters. Appl. Environ. Microbiol. 62: 3997-4002.

Wiggins, B. A., P. W. Cash, W. S. Creamer, S. E. Dart, P. P. Garcia, T. M. Gerecke, J.
Han, B. L. Henry, K. B. Hoover, E. L. Johnson, K. C. Jones, J. G. McCarthy, J. A.
McDonough, S. A. Mercer, M. J. Noto, H. Park, M. S. Phillips, S. M. Pumer, B.
M. Smith, E. N. Stevens and A. K. Vamer (2003). Use of Antibiotic Resistance
Analysis for Representativeness Testing of Multiwatershed Libraries. Appl.
Environ. Microbiol. 69(6): 3399-3405.

80

 

Willems, R. J. L., J. Top, N. van den Braak, A. van Belkum, H. Endtz, D. Mevius, E.
Stobberingh, A. van den Bogaard and J. D. A. van Embden (2000). Host

specificity of vancomycin-resistant Enterococcusfaecium. J. Infect. Dis. 182(3):
8 16-823.

Willems, R. J. L., W. L. Homan, J. Top, M. van Santen-Verheuvel, D. Tribe, X.
Manzioros, C. Gaillard, C. M. J. E. Vandenbroucke-Grauls, E. M. Mascini, E. van
Kregten, J. D. A. van Embden and M. J. M. Bonten (2001). Variant esp gene as a
marker of a distinct genetic lineage of vancomycin-reistant Enterococcusfaecium
spreading in hospitals. Lancet 357: 853-855.

Yoder, J. S., B. G. Blackburn, G. F. Craun, V. Hill, D. A. Levy, N. Chen, S. H. Lee, R.
Calderon and M. J. Beach (2004). Surveillance for waterbome-disease outbreaks

associated with recreational water-United States, 2001-2002. Morb. Mortal.
Wkly. Rep. 53(SS-8): 1-20.

81