United States Office of Water EPA 823-R-10-004
Environmental Protection (4303T) December 2010
Agency
Evaluation of the Suitability of Individual
Combinations of Indicators and Methods
for Different Clean Water Act Programs
-------
Office of Science and Technology
Office of Water
U.S. Environmental Protection Agency
Washington, DC
-------
Evaluation of the Suitability of Individual Combinations of Indicators and Methods
for Different Clean Water Act Programs
Table of Contents
Section
List of Tables 4
List of Acronyms and Abbreviations 4
1 Introduction/Project Description 5
2 Background 6
2.1 Indicator/Methods Evaluated in the 2003-2005 NEEAR Study 7
2.2 Indicator/Methods Evaluated in the 2007/2009 Marine
Epidemiological Studies 8
2.3 Other Indicator Methods 9
3 Quantitative Evaluation of Indicator/Method Combinations Based
on Performance Criteria 11
3.1 Established Health Relationship 12
3.2 Limit of Detect! on 14
3.3 Precision 14
3.4 Sensitivity 15
3.5 Specificity 15
3.6 Ability of Differentiate Fecal Sources 16
3.7 Performance in Different Waterbody Types 16
4 Evaluation of Appropriateness of Indicator/Methods for CWA Purposes 17
5 Study Limitations 18
6 References 18
-------
Evaluation of the Suitability of Individual Combinations of Indicators and Methods
for Different Clean Water Act Programs
List of Tables
Table 1 Indicators/Methods Tested by EPA in NEEAR Epidemiological Studies 9
Table 2 Evaluation of Rapid and Culture Methods 13
Table 3 Evaluation of Rapid and Culture Methods for Clean Water Act Programs
Implementation Purposes 17
List of Acronyms & Abbreviations
BEACH Act
CCE
CE
CPU
CLAT
CSE
CT
CWA
DNA
EPA
GI
Ml
MST
mTEC
NEEAR
NPDES
PBS
PCR
POTW
qPCR
RNA
RSD
SCCWRP
TMDL
WQS
Beaches Environmental Assessment and Coastal Health
Act
calibrator cell equivalents
cell equivalents
colony forming unit
culture, latex agglutination, and typing
calibrator sequence equivalents
Cycle Threshold
Clean Water Act
Deoxyribonucleic acid
Environmental Protection Agency
gastrointestinal
milliliter
Microbial Source Tracking
modified membrane-Thermotolerant Escherichia coli agar
National Epidemiological and Environmental Assessment
of Recreational
National Pollutant Discharge Elimination System
Phosphate Buffered Saline
Polymerase Chain Reaction
Publicly-Owned Treatment Works
Quantitative Polymerase Chain Reaction
Ribonucleic acid
Relative Standard Deviation
Southern California Coastal Water Research Project
Total Maximum Daily Load
Water Quality Standards
-------
Evaluation of the Suitability of Individual Combinations of Indicators and Methods
for Different Clean Water Act Programs
1.0 INTRODUCTION/ PROJECT DESCRIPTION
Project PI8 as described in Environmental Protection Agency's (EPA's) Critical Path Science
Plan is to "identify and evaluate indicator/method combinations for strengths and limitations
with respect to fecal source identification (human, animal); and performance in different
waterbody types. The important features of each indicator/method will be described and the
strengths and weaknesses of those features will be explained and evaluated. The ideal set of
features will be proposed so that the indicator/methods can be compared to a hypothetical ideal
indicator/method. The indicators/methods under consideration will be ranked for each feature
with respect to ability to differentiate fecal sources, performance in different water body types
and appropriateness for different CWA purposes. EPA plans to use the results of this evaluation
to inform the decisions regarding which indicator/methods will be included in the new or revised
criteria and under what conditions those indicator•/methods will be recommended''' The purpose
of this white paper is to report on the result of the evaluation of indicator/methods assessed by
EPA in support of the development of new or revised recreational water quality criteria for
bacteria.
Section 304(a)(l) of the CWA directs EPA to publish criteria for water quality accurately
reflecting the latest scientific knowledge. The criteria published by EPA under section 304(a)
are intended to provide guidance to States in setting water quality standards (WQS) to protect
public health as well as to maintain and restore water quality and ecosystem integrity. CWA
304(a) criteria are typically expressed as numbers (i.e., concentrations of pollutants) or narratives
that EPA recommends that states put into their WQS to protect waters for aquatic life, wildlife,
consumption of aquatic organisms by humans, and primary contact recreation. CWA section 303
requires each state to adopt WQS for all waters of the state and to review, and revise them as
necessary, every 3 years. Once adopted and approved, WQS are binding CWA regulatory
standards and effective for the following CWA purposes:
• Water Quality Assessments. Sections 303(d) and 305(b) provide that states are required
to assess their waters on a regular basis to determine if they are meeting WQS. The
states' water quality criteria are an essential baseline against which states determine
whether particular waters are "impaired."
• Total Maximum Daily Loads (TMDL). TMDL calculations are required for all waters
that have been listed as "impaired" under section 303(d). A TMDL specifies the
maximum amount of a pollutant that a waterbody can receive and still meet WQS, and
"allocates" pollutant loadings among point and non-point pollutant sources. TMDL
calculations for impaired waters must be written to implement the applicable State WQS.
• National Pollutant Discharge Elimination System (NPDES). NPDES permits are
required under section 402 for point source discharges of pollutants to waters of the
United States. NPDES permits must include effluent limitations more stringent than
required by technology regulations, if necessary, to meet water quality standards, which
include state water quality criteria.
-------
Evaluation of the Suitability of Individual Combinations of Indicators and Methods
for Different Clean Water Act Programs
• Non-point Source Program. Water quality standards (including criteria) play a similarly
important role under the CWA section 319 non-point source program as part of the listing
and TMDL processes to determine whether best management practices or other risk
management control strategies are needed to address non-point source pollution.
• Recreational Water Monitoring and Notification. A State's recreational water quality
criteria are used in beach monitoring and notification programs. States and beach
managers typically make decisions about whether to issue advisories or closure notices
by measuring the results of their monitoring against their WQS.
EPA's current recommended criteria for bacteria are based on bacterial indicator organisms -
Escherichia coll (E. coli) and Enterococcus - which generally do not cause illness, but have
characteristics that make them good indicators of fecal contamination, and thus by inference, of
pathogens capable of causing human illness such as acute gastrointestinal illness. EPA's
recommended bacteria criteria are intended to be adopted by states into state water quality
standards to protect waters designated for recreational use activities such as swimming.
2.0 BACKGROUND
Beginning with the freshwater National Epidemiological and Environmental Assessment of
Recreational (NEEAR) Water Study in 2003, EPA has been evaluating recreational water quality
using traditional culture methods and rapid genetic methods based on qPCR. EPA's research
goal is to determine the indicator/method combination with the best correlation to illness in fresh
and marine recreational waters. EPA was also responding to a provision in the Beaches
Assessment and Coastal Health Act (BEACH Act) of 2000 which required the Agency to
provide additional information for use in developing "appropriate, accurate, and expeditious and
cost-effective methods (including predictive models) for detecting in a timely manner in coastal
recreation waters the presence of pathogens that are harmful to human health."
A culture method for bacteria is one that allows direct propagation of the bacteria, and for the
visible (to the naked eye) increase in the number of cells, which can allow for enumeration via
solid (i.e., membrane filtration technique) or liquid media (i.e., multiple tube fermentation
technique). The currently approved EPA culture methods involve culturing and enumerating
fecal indicator bacteria (E. coli or Enterococcus spp.). Current culture-based methods require 24
to 48 hours to obtain results (Wade et al., 2008).
Rapid methods allow for the direct measurement of genetic material such as DNA without the
need for incubation and can provide results between 2 to 6 hours from time of receipt by the
laboratory to time results are available. Rapid methods are being evaluated for the development
of new or revised criteria because they provide more timely results than traditional culture
methods, allowing beach managers to make beach notification decisions on the same day water
samples are collected.
-------
Evaluation of the Suitability of Individual Combinations of Indicators and Methods
for Different Clean Water Act Programs
2.1 Indicator/Methods Evaluated in the 2003-2005 NEEAR Study
The rapid indicator/methods evaluated in the 2003-2004 freshwater and 2005 (Biloxi, MS)
marine NEEAR studies were commercially-available test methods or validated or experimental
research protocols. The methods included in the NEEAR study met the following criteria:
(1) results could be obtained within a few hours;
(2) Enterococcus or the new potential indicator, Bacteroidales, was detected by the
method;
(3) the detection limit of the method was sufficiently low to allow detection of indicators
in the majority of samples from recreational water environments;
(4) the method was sufficiently resistant to water sample inhibitory effects to allow
detection of indicators in the majority of samples from recreational water
environments.
The four tests included in the 2003-2005 freshwater and marine NEEAR Water Studies are
described below. The following descriptions are based on the methods available at that time and
may not reflect recent changes to these methods.
1. Quantitative Polymerase Chain Reaction (qPCR) Method, two new rapid gene probe
methods developed by Dr. Richard Haugland of the EPA (Haugland et a/., 2005)
(hereafter referred to as the EPA qPCR Assay), were used to detect Enterococcus and
Bacteroidales in water samples based on the collection of these organisms on membrane
filters, extraction of their total DNA, and qPCR amplification (i.e., a process whereby
target DNA strands are doubled in each cycle of amplification) of a genus-specific
(Enterococcus) or order-specific (Bacteroidales) rDNA sequence using the TaqMan™
system.
2. RAPTOR Fiber optic Biosensor, a portable, automated fiber optic biosensor developed
by Research International, Woodinville, Washington that can be used to detect
microbiological and chemical analytes in water samples. The RAPTOR Biosensor, used
in the summer of 2003, was removed from the study in 2004 because the capture and
detection antibodies were not sensitive enough.
3. Luminex 100 System, a compact flow cytometer, developed by the Luminex Corporation,
Austin, Texas and MiraiBio, Alameda, California, that analyzes immunoassays, complex
genetic analyses, and/or enzymatic assays through the use of optics, fluidics, and
advanced signal processing. The Luminex method was removed from further study in
2004 because it lacked adequate sensitivity.
4. EPA Method 1600 is the EPA-approved membrane filter method using mEI Agar for the
detection of Enterococcus in recreational water. This method was included in the study
as a reference method for Enterococcus.
Of the new methods described above, only the qPCR method produced results that were
sufficient to be included for further evaluation in the 2007 marine epidemiological studies in
Fairhope, Alabama and Goddard, Rhode Island. The antibodies used in the RAPTOR and
-------
Evaluation of the Suitability of Individual Combinations of Indicators and Methods
for Different Clean Water Act Programs
Luminex 100 methods lacked sensitivity for Enterococcus and Bacteroidales but these methods
may be useful if new antibodies can be produced to increase the sensitivity and specificity of the
methods. Both the RAPTOR and Luminex methods were dropped from further evaluation.
In the 2003-2005 freshwater and marine studies E. coli methods were not tested. EPA did not
include an E. coli qPCR method because at that time, a method was not available. Since no
qPCR method was available for comparison with the culture method, the culture method was not
included.
2.2 Indicator/Methods Evaluated in the 2007/2009 Marine Epidemiological Studies
In 2007, EPA conducted two marine epidemiological studies in Fairhope, Alabama and Goddard,
Rhode Island at beaches predominantly impacted by treated Publicly-Owned Treatment Works
(POTW) effluent. In 2009, EPA initiated two additional epidemiological studies at a tropical
beach impacted predominantly by human sources of fecal pollution, and a beach impacted by
urban runoff using several rapid and cultural indicator/method combinations. The tropical
epidemiological study was conducted at Boqueron Beach in Puerto Rico; and the urban runoff
study at Surfside Beach in South Carolina. Additional information on the site selection process
for the urban runoff epidemiological study can be found at http://www.epa.gov/nheerl/neear/.
Table 1 presents a comprehensive list of the methods tested in the EPA epidemiological studies
conducted to support the development of new or revised recreational water quality criteria.
Several methods were tested at multiple epidemiological sites dating back to the freshwater
Great Lakes studies that were initiated in 2003 as part of the NEEAR Water Study. E. coli
culture was not included in the methods tested in marine waters because E. coli is not stable and
cells lyse easily in saltwater environments.
The Fecal Bacteroides assay was evaluated because of its potential as a human associated
marker. However, the evaluation of this indicator/method combination in the 2007
epidemiological studies resulted in its elimination due to its lack of human source specificity and
the relatively high proportion of sample where target organisms were not detectable. A similar
problem was encountered with theE1. coli qPCR assay. However, evaluation of this assay was
continued in 2009 in conjunction with an added DNA concentration and purification procedure.
The male-specific F+ Coliphage methods were eliminated due to the low density of the indicator
in surface water.
-------
Evaluation of the Suitability of Individual Combinations of Indicators and Methods
for Different Clean Water Act Programs
Table 1 - Indicators/ Methods Tested by EPA in NEEAR Epidemiological
Methods
Enterococcus spp. qPCR
Enter ococcus culture (EPA Method 1600)
Total Bacteroidales spp. qPCR
Fecal Bacteroides spp. qPCR
Human-specific Bacteroidales markers
E. coli qPCR
Clostridium spp. qPCR
F+ RNA Coliphage CLAT assay
F+ DNA Coliphage CLAT assay
F+ Coliphage 24-hr SPOT assay
Studies
EPA Epidemiological Studies Where Methods
Were Used
Freshwater: 2003, 2004 l
Marine: 2005, 2007, 20091
Freshwater: 2003, 20042
Marine: 2005, 2007, 20092
Freshwater: 2003, 2004 j
Marine: 2005, 2007, 20094
Marine: 20075
Marine: 2009M
Marine: 2007s, 2009*'y
Marine: 2007, 2009s
Marine: 20071U
Marine: 20071U
Marine: 200711
^Haughlande/a/.,2005
2USEPA 2002
3Dick and Field, 2004
4Siefringe/a/.,2008
5Conversee/a/., 2009
6 Shanks et al, 2009
7Haugland et al, 2010.
8Cherne/a/.,2009
9Chern et al., nd
1 ° Love and Sobsey, 2007
nUSEPA2001
2.3 Other Indicator Methods
2.3.1 Microbial Source Tracking Methods
There is uncertainty about the risk to human health associated with non-human sources of fecal
pollution. The ability to differentiate sources of fecal contamination site-specifically may be
important for an accurate assessment of the risks to human health from domestic, agricultural
animals and wildlife. Source differentiation is also important with regard to CWA monitoring
and assessment (§303[d] and §305[b]) and TMDL programs. Some states have expressed a
desire to be able to adjust the applicable criteria based on data that the indicator levels present in
the waterbody are not due to human sources of fecal pollution. The stated concern is that once
waters are listed as impaired, states must then expend resources to develop a TMDL to restore
waters that pose less of a risk to human health. In this context, microbial source tracking (MST)
methods are useful to supplement sanitary survey investigations (i.e., to identify sources of
contaminants or TMDL sources) and for risk analysis (human vs. non-human vs. domestic
animals). Most MST methods attempt to identify specific fecal sources to assist with prioritizing
polluted areas for restoration.
-------
Evaluation of the Suitability of Individual Combinations of Indicators and Methods
for Different Clean Water Act Programs
EPA has been conducting research on rapid MST methods and has evaluated qPCR and PCR
genetic markers of human and bovine pollution. This research includes a performance evaluation
of (1) seven PCR and qPCR assays targeting Bacteroidales genes reported to be associated with
either ruminant (goat, sheep, deer and others) or bovine feces (Shanks et a/., 2010a); and (2) five
PCR and ten qPCR assays targeting Bacteroidales genes reported to be associated with human
feces (Shanks etal., 2010b). The bovine assay study found large discrepancies in the
performance of qPCR assays across different bovine populations; and recommended that the use
of bovine-associated MST applications require an a priori characterization of each watershed due
to variability in genetic marker abundance and prevalence between populations. Study results
also suggest that some assays are more suitable for the characterization of fecal contamination
than others, and that the assay of choice can vary from one bovine population to another.
The human assay performance study included the evaluation of two qPCR assays for
quantification of recently developed human-associated genetic markers targeting putative
Bacteroidales-like cell surface-associated genes (Shanks 2009), seven qPCR assays for
quantification of 16s rRNA gene markers from human-associated Bacteroidales species
(Haugland et a/., 2010), and several other published assays. Some assays showed human source
specificity levels exceeding 97% when tested against a panel of reference fecal samples
originating from cattle, poultry, swine, and various wildlife animal sources. Based on assay
performance and the prevalence of DNA targets in a collection of reference untreated sewage
samples collected from 54 different waste water treatment facilities in the United States, this
research suggests a potential application for human-associated quantitative methods for
monitoring fecal pollution in ambient waters.
EPA is also conducting research to identify genetic sequences that could form the basis of
chicken and seagull-associated MST methods as specific fecal source assays. EPA has selected
PCR-based assays that could uniquely identify avian sources (primarily chicken and seagull) of
fecal pollution. Four assays are currently being evaluated and additional molecular data are being
collected to further validate existing assays and determine if additional assays can be developed.
If EPA determines that chicken and seagull specific assays can be developed, EPA will evaluate
chicken and seagull-associated MST PCR-based assays for sensitivity and specificity using
reference fecal samples and environmental water samples with known sources of fecal
contamination.
2.3.2 Chemical Methods
In addition to the microbial indicators discussed in sections 2.2 and 2.3, EPA also tested water
samples for chemical constituents. The chemical constituents tested included coprostanol, a
product of cholesterol metabolism in feces; and urobilin, a bile pigment found in human feces
and urine. Additionally, 48 different chemicals distinctly associated with humans that could be
markers of human sewage or human fecal contamination (e.g., caffeine, cotininine,
Acetaminophen, and codeine) were tested. Preliminary data suggest that chemical detections
were not very frequent at freshwaters beaches, but were more frequent and occurred at higher
concentrations at marine beaches tested in 2005 and 2007. Preliminary analyses did not show a
consistent relationship with health outcomes, and the chemical measurements were discontinued
in the 2009 epidemiological studies due to a lack of funding.
10
-------
Evaluation of the Suitability of Individual Combinations of Indicators and Methods
for Different Clean Water Act Programs
2.3.3 Southern California Coastal Water Research Project Research
Southern California Coastal Water Research Project (SCCWRP) has been conducting research
on indicator/methods with the goal of developing rapid methods that can augment or replace
existing traditional culture methods for one or more types of indicator bacteria. SCCWRP is
assessing water quality by measuring both traditional and non-traditional indicators. Traditional
indicator methods include total coliform, fecal coliform, Enterococcus using membrane filtration
and Enterolert chromogenic substrate method, E. coli using Colilert chromogenic substrate
method and membrane filtration, and Coliphage. Nontraditional measurements include rapid
methods for quantifying Enterococcus and E. coli., Bacteroides, Bacteroides thetaiotamicron,
adenovirus, norovirus, enterovirus and Coliphage (somatic and F+), among others. These
indicator methods are being tested by SCCWRP as part of a 4-year project on Rapid Bacterial
Indicator Development and then validated in three large-scale epidemiological studies conducted
at Avalon Bay Beach, Malibu Surfrider Beach and Doheny State Beach in southern California.
As part of the epidemiological studies, SCCWRP has analyzed more than 4,000 water samples
using 36 different analytical methods (SCCWRP, 2010).
As part of its 2010/2011 Research Plan, SCCWRP is conducting a 3-year study to assess which
source identification methods are optimal for differentiating fecal sources with the goal of
bringing together a team of water quality experts experienced in source identification methods to
create a source identification manual, implement selected protocols at several beaches of high
interest to California, and then transition source identification capabilities to local laboratories.
3.0 QUANTITATIVE EVALUATION OF INDICATOR/METHOD COMBINATIONS
BASED ON PERFORMANCE CRITERIA
EPA's approach to evaluate the suitability of individual combinations of indicators and methods
for different CWA programs (PI8) was to conduct two separate analyses - a quantitative analysis
of indicator/method combinations based on a set of performance criteria; and a qualitative
evaluation of the appropriateness of the qPCR and culture methods for each CWA program.
Enterococcus qPCR, Bacteroidales qPCR, Method 1600 (membrane filtration method for
Enterococci,) and Method 1603 (membrane filtration method for E. coli using modified mTEC
agar) were evaluated because they have an association with illness in swimmers (Fleisher et a/.,
2010; Wade et al., 2008; Wade et al., 2006; and USEPA 1986) (Table 2). The performance
criteria used to evaluate these methods are as follows: (1) established health relationship, (2)
limit of detection, (3) sensitivity, (4) specificity, (5) precision, (6) percent false positive and (7)
percent false negative. Table 2 shows the results of the quantitative evaluation of the qPCR and
culture methods. The methods were also evaluated on their ability to differentiate fecal sources
and performance in different waterbody types.
11
-------
Evaluation of the Suitability of Individual Combinations of Indicators and Methods
for Different Clean Water Act Programs
3.1 Established Health Relationship
3.1.1 Enterococcus Culture and qPCR
A demonstrated relationship between indicator concentration and gastrointestinal (GI) illness is
the most important criterion for selection of the method for new or revised ambient water quality
criteria. The method must demonstrate a positive relationship between illness in humans and
with indicator levels (i.e., as indicator levels increase, risk of illness increases) in recreational
waters.
Enterococci are commonly found in the feces of humans and other warm-blooded animals.
Although some strains are ubiquitous and not related to fecal pollution, the presence of
enterococci in water is an indication of fecal pollution and the possible presence of enteric
pathogens. Results from EPA's recent epidemiological studies showed a correlation between
qPCR measured Enterococcus levels to GI illness at freshwater beaches impacted by POTW
sources (Wade et a/., 2008). Additionally, the recent studies point to significantly increased
illness rates among swimmers 10 years and younger exposed above 35 CPU of Enterococcus
using the culture method compared to non-swimmers in the Great Lakes study, but the trend is
not as strong (or evident) at low exposures. QPCR cell equivalent (CE) levels were a stronger
predictor of GI illness than the CPU measure in the NEEAR study (Wade et al, 2008). Other
studies have shown enterococci levels measured by culture methods correlated with GI illness
levels in marine and fresh waters (Au-Yeung et a/., unpub; Zmirou et a/., 2003). The meta-
analysis of the Epibathe studies found an increased risk of gastroenteritis in both marine and
freshwater water sites when bathers were exposed to enterococci concentrations higher than 100
enterococci/100 ml in marine water and 200 enterococci/100 ml in fresh waters (Au-Yeung et
a/., unpub).
3.1.2 E. coli Culture and qPCR
Epidemiological studies conducted by EPA and SCCWRP separately found that there was no
relationship between E. coli qPCR and GI illness in marine waters. There are no recent EPA
data on the health relationship between E. coli culture and swimming-related illness. E. coli
culture was not included in the list of methods tested by EPA in the NEEAR epidemiological
studies. However, in epidemiological studies conducted by EPA in the 1970s, there was an
association between E. coli culture and GI illness in fresh water but not marine waters.
More recent data from randomized controlled trial epidemiological studies conducted in Europe
found a weak relationship between E. coli culture and GI illness in freshwater only. A meta-
analysis of the (Epibathe) studies found an increased risk of gastroenteritis in both marine and
fresh water sites in bathers exposed to E. coli concentrations higher than the level that represents
the 2006/7/EC1 "excellent quality" criteria (< 500 E.coli/100 ml) (Au-Yeung etal., unpub).
1 Directive 2006/7/EC of the European Parliament and of the Council of 15 February 2006 concerning the
management of bathing water quality and repealing Directive 76/160/EEC. The purpose of this Directive is to
preserve, protect and improve the quality of the environment and to protect human health by complementing
Directive 2000/60/EC. This Directive lays down provisions for: (a) the monitoring and classification of
bathing water quality; (b) the management of bathing water quality; and (c) the provision of information to
the public on bathing water quality.
"12
-------
Evaluation of the Suitability of Individual Combinations of Indicators and Methods
for Different Clean Water Act Programs
Table 2 -Evaluation of Rapid and Culture Methods
Performance
Evaluation Criteria
50% Limit of detection
(CCE)
50% Limit of detection
(CSE)
Sensitivity
Specificity
Precision (log 10
standard deviation)
False positives
False negatives
Health relationship
established
Rapid Methods
Enter ococcus 23 S
qPCR
Diluted crude extract
65 L
9041
100%2
100%'
0.244
0%\ 19%'
Ll-4.5%1, 2.0%v
Yes8'y
Bacteroidales 16S qPCR
Diluted crude extract
991
1,380'
NR
100%4
0.204
0%'
Ll-4.5%1, 2.0%v
Yes (Marine)9
Culture Methods
Method 1600
(Enterococcus by MF)
NR
NR
NR
2.2%- 18.9%5
6%5
6.5%5
Yes
Method 1603
(E. coli by MF)
NR
NR
NR
25.9%b
6%b
5%b
Yes (Freshwater)
Notes:
Results presented in this table may be modified as additional information is collected and peer reviewed.
For Method 1600 (Enterococcus byMF) and Method 1603(E. coli byMF), performance criteria are based on the results of the inter-laboratory validation in
disinfected wastewater matrices.
NR = Not reported.
iChern eta/., 2009
2 Ludwig and Schleifer, 2000
3 Frahm and Obst, 2003
4 Siefring et a/., 2008
5USEPA,2006a
6USEPA2006b
7 Haughland, eta/., 2005
8 Wade eta/., 2008
9 Wade eta/., 2010
13
-------
Evaluation of the Suitability of Individual Combinations of Indicators and Methods
for Different Clean Water Act Programs
3.1.3 Bacteroidales qPCR
Bacteroidales is a group of anaerobic bacteria commonly found in the gut of humans.
Bacteroidales densities, as measured by qPCR, are orders of magnitude higher in raw waste
streams than enterococci or E. coli densities due in part to its greater abundance in feces. There
are qPCR primers available that target the general Bacteriodales population and other primers
that are thought to target human-specific strains. A positive relationship has been observed
between general Bacteroidales and GI illness in marine waters in the NEEAR epidemiological
studies (Wade et a/., 2010).
3.2 Limit of detection
3.2.1 qPCR
The limits of detection of the Enterococcus and Bacteroidales qPCR methods have been reported
as the estimated number of target organism calibrator cell equivalents (CCE) or target organism
calibrator sequence equivalents (CSE) that need to be present per total water sample filter extract
in order to allow detection in 50% of analyses of that extract by the specified method (Chern et
al., 2009). Chern etal. (2009) reported mean limits of detection of 65 CCE (904 CSE) for
Enterococcus qPCR and 99 CCE (1,380 CSE) for Bacteroidales qPCR.
3.2.2 Culture
The limits of detection of the Enterococcus andE. coli culture methods (Method 1600 and
Method 1603, respectively) have not been reported.
3.3 Precision
3.3.1 qPCR
Precision of a qPCR method is the estimate of total variability of Cycle Threshold (CT)
measurements obtained in analyses of common water sample filter extracts by single or multiple
laboratories. It can be expressed in terms of standard deviation, incorporating between lab,
between run and random error estimates.
At this time, EPA has completed a Single Laboratory Validation Study for Enterococcus qPCR
and Bacteroidales qPCR. A Multi-Laboratory Validation Study is on-going; however, the results
will not be available until 2012 and are therefore, not presented or discussed in this report.
Estimates of method precision (expressed as logic standard deviations of CCE) from analyses of
10,000 Enterococcus and Bacteroides cells spiked into 51 different fresh and marine water
samples have been reported as 0.24 and 0.20, respectively (Siefring et al., 2008).
3.3.2 Culture
Precision for Method 1603 has been characterized by laboratory-specific relative standard
deviations (RSDs) from disinfected wastewater samples spiked with laboratory-prepared. For
Method 1603, the within-laboratory pooled RSD was 25.9% (USEPA 2006b). Precision
estimates in the range of 2.2% - 18.9% has been reported for Method 1600 (USEPA 2006a).
"14
-------
Evaluation of the Suitability of Individual Combinations of Indicators and Methods
for Different Clean Water Act Programs
3.4 Sensitivity
Sensitivity can be expressed as a percentage of targeted species that are detected by a method.
3.4.1 qPCR
It has been reported that the Enterococcus qPCR method detects all validly described species of
Enterococcus (Ludwig et al., 2000). For the Bacteroidales qPCR method, there are no publish
data available on the sensitivity of the method.
3.4.2 Culture
For the Enterococcus and E. coll culture methods there are no published reports available on
method sensitivity. However, as part of the formal QA process for each laboratory that uses
Methods 1600 and 1603, the methods recommend that laboratories develop a statement of
accuracy for method by calculating the average percent recovery and the standard deviation of
the percent recovery (USEPA 2006a, b).
3.5 Specificity
Specificity is the ability of a method to select and or distinguish the target bacteria under test
from other bacteria in the same water sample. The specificity characteristic of a method is
usually reported as the percent of false positive and false negative results. In the following
sections, data are provided on the specificity, false positive and false negative rates for
Enterococcus qPCR, Bacteroidales qPCR, Method 1600, and Method 1603.
3.5.1 qPCR
The specificity of the Enterococcus qPCR method was estimated from experimental analyses of
five closely related non-Enterococcus species showing at least 10,000 times higher limit of
detection when compared to Enterococcus species (Frahm and Obst, 2003). For the
Bacteroidales qPCR method specificity was experimentally assessed using representative species
of the related bacterial classes, Flavobacteria and Sphingobacteria (Siefring et a/., 2008).
The percent false negatives for the Enterococcus and Bacteroidales qPCR methods have been
reported as the percentage of water sample filter extracts that fail salmon DNA sample
processing control assay quality control criterion of > 3 CT units higher than the mean of
associated calibrator samples in the analysis of NEEAR study samples (Haugland etal., 2005;
Chern et al., 2009). Additionally, Griffith et al. (2007) reported false negative results of 58% for
enteroccocci qPCR extracted using a multi-step purification and concentration process after bead
beating and 29% for enterococci qPCR bead beaten method. The analysis for false negative rates
was conducted relative to the State of California's (AB411) standard of 104 cells/100 ml for
enterococci (Griffith et al., 2007).
Percent false positives for the qPCR methods have been reported as the percentages of negative
control samples, consisting of clean filters that were subjected to the entire method including
DNA extraction and qPCR analysis, that gave positive detection of target sequences (i.e., a true
logarithmic amplification trace) during analyses of NEEAR study samples (Haugland etal.,
15"
-------
Evaluation of the Suitability of Individual Combinations of Indicators and Methods
for Different Clean Water Act Programs
2005; Chern et a/., 2009). It should be noted that the mean CT value from the false positive
analyses shown for the study cited in footnote 7 in Table 2 was 43.65 which would not be
detected as positive in a more recent version of the method where only 40 thermal cycles are run.
Additionally, Griffith et al. (2007) reported false positive results of 3% for enteroccocci qPCR
extracted using a multi-step purification and concentration process after bead beating and 9% for
enterococci qPCR bead beaten method. For E. coli qPCR, the false positive results were 0%
before and 27% after adjustment for amplification efficiency. The analysis for false positive rates
was conducted relative to the State of California's (AB411) standard of 104 cells/100 ml for
enterococci and 400 cells/100 ml for E. coli (Griffith etal, 2007).
3.5.2 Culture
For the culture methods, the percent false negative is the percentage of the samples that had a
negative result but were actually positive. For Method 1600 and 1603, the percent false negatives
are reported from analyses of various environmental water samples (USEPA 2000). Five percent
of the E. coli colonies observed gave a false negative reaction (USEPA, 2006). In unspiked CSO
samples, false negative rates of 37.5% and 1.7% have been reported for methods 1600 and 1603,
respectively (USEPA, 2008). Additionally, Francy et al. (2000) reported false negative results of
11% for the modified membrane-Thermotolerant Escherichia coli agar (mTEC) (Method 1603)
when compared to the mTEC (EPA Method 1103.1). Griffith et al. (2007) reported false
negative results of 10% for Method 1600 with respect to the AB411 standard of 104
enterococci/100 ml.
For Method 1603, the percent false positives reported from analyses of various environmental
water samples were <1% (USEPA, 2000) and averaged 6% for marine and fresh water samples
(USEPA, 2006). False positive confirmation rates were 0% and 6.7% for Method 1600 and 1603,
respectively (USEPA, 2008) were reported in unspiked CSO samples. Francy et al. (2000)
reported false positive results of 0% for the modified mTEC (Method 1603) when compared to
the mTEC (Method 1103.1). Griffith et al. (2007) reported false positive results of 4% for
Method 1600 with respect to the AB411 standard of 104 enterococci/100 ml.
3.6 Ability to Differentiate Fecal Sources
This criterion evaluates the ability of the indicator/method combination to identify the sources
contributing to fecal pollution in a waterbody. None of the indicator/methods evaluated are able
to differentiate between fecal contamination sources. However, current research by EPA on
microbial source tracking assays for use in detecting bovine and human fecal pollution suggests
that these assays maybe useful in monitoring fecal contamination in ambient waters (see Section
2.4.1).
3.7 Performance in Different Waterbody Types
This criterion evaluates how well the method performs (i.e., does it work or provide usable
results) in different types of recreational waters (e.g., freshwater, marine water, temperate,
tropical, and subtropical waters). Recent epidemiological studies conducted by EPA have shown
that Enterococcus qPCR and Bacteroidales qPCR performed well in both marine and
freshwaters.
"16
-------
Evaluation of the Suitability of Individual Combinations of Indicators and Methods
for Different Clean Water Act Programs
Results from earlier EPA epidemiological studies found that Enterococcus culture works well in
both marine and freshwater, whereas E. coli only had good correlation (R2 > 0.5) with GI illness
in freshwater (Cabelli, 1983; Dufour, 1984).
4.0 EVALUATION OF APPROPRIATENESS OF INDICATOR/METHODS FOR
CWA PURPOSES
The appropriateness of the method for the various CWA programs is important for state adoption
and implementation of EPA's new or revised criteria recommendations. To assess the potential
use and applicability of new fecal indicators and methods to meet CWA program needs, EPA's
Monitoring and Assessment, TMDLs, and NPDES programs evaluated the impact of new
indicators/methods on their programs. Each program identified attributes necessary for indicator-
methods to meet their programs' CWA needs. These attributes are shown in Table 4 and ranked
on whether they are important (Yes) or not important (No) for each CWA program.
Table 3 -Evaluation of Rapid and Culture Methods for
Clean Water Act Programs Implementation Purposes
Method Attributes
Low Cost
Ease of Use
Time to results (rapid/results within 4 hours of
sample process/analysis)
Allows for use of historical data for model
development, etc.
Demonstrates effectiveness of treatment from
source to beach
Count/signal associated with human health
risk pathogens
Precise/accurate
Beach
Program
Yes
Yes
Yes
Yes
No
Yes
Yes
NPDES
Yes
Yes
No
Yes
Yes
Yes
Yes
TMDL
Yes
Yes
No
Yes
No
Yes
Yes
Assessment
Yes
Yes
No
Yes
No
Yes
Yes
While beach advisory decisions and closures require same day results to protect public health,
for the Assessment, TMDLs and the NPDES programs same day results are not necessary.
Therefore, rapid methods are suggested as a key component of the criteria for the Beach Program
only. Other important attributes for indicator/methods for the Beach Program include ease of use
of the method and cost. Ease of use refers to how easily the method can be applied by those
skilled in analysis. The cost associated with new indicator/method combination in criteria could
include capital cost, training cost, per sample cost and additional sampling requirements. Capital
costs include the upfront cost such as equipment purchase and space required to conduct the test.
For example, when performing genetic testing, aside from the equipment needed (e.g., platform
[i.e., specific machine]), laminar flow hoods, dedicated pipettes), space is needed, ideally in
separate rooms, for reagent preparation (material not containing any genetic materials). Space is
also needed for the two types of sample preparation, those containing high target sequence DNA
concentrations such as DNA standards and calibrator samples, and those containing expected
low target sequence DNA concentrations (e.g., filter blanks and water samples) - the latter of
which should also be in separate laminar flow hoods (USEPA, 2007).
17
-------
Evaluation of the Suitability of Individual Combinations of Indicators and Methods
for Different Clean Water Act Programs
Training costs are incurred prior to routine testing so that the user can perform the test within the
performance criteria of the test (USEPA, 2007). Training cost would be dependent on the types
of training available, i.e., whether workshop-type training with hands-on experience or
completing a training module.
Sample cost can vary and becomes an important cost consideration with a new method
depending on the volume of tests needed to be completed on a routine basis. Additional
sampling is generally an effort that results from rapid testing. For example, if an early morning
sample yields a positive result resulting in beach closures, it may then lead to additional
sampling to determine if the beach still needs to be closed in the mid-afternoon (USEPA, 2007).
Therefore, depending on the number of designated bathing beaches that need to be monitored,
the monitoring frequency, number of samples, and the need for additional sample analysis per
beach, the sample costs associated with the use of rapid method could become cost prohibitive or
result in beaches being sampled less frequently or not re-sampled to revise advisory or closure
decision during the day.
Ease of use and cost are also important factors for Assessment, TMDL and NPDES programs.
For the TMDL and Assessment programs the ability to use prior historical data for modeling
loadings is also important. Current historical data for E. coli, enterococci, and fecal coliform are
derived from culture methods data. Criteria based on a rapid method without the necessary
linkage between qPCR and the culture method would not allow for the use of historical data.
For the NPDES Program, the ability of the indicator/method to reflect the treatment efficacy
(e.g., chlorination) is crucial. Traditional culture methods for fecal indicator bacteria detect only
a subset of the total viable population within any given water sample. The qPCR method will
theoretically detect all intact cells in a water sample whether they are viable or not.
Precision and accuracy of methods are also important to each program as well as the ability to
associate indicator counts/concentrations to health risk.
5.0 STUDY LIMITATIONS
There are several weaknesses in this evaluation with regard to how the project was defined in the
Critical Path Science Plan. Due to limited data on various water body types, EPA is unable to
evaluate method performance in different water body types based on the performance criteria in
Table 2. Additionally, method performance with respect to fecal source identification (i.e.,
human vs. animal) could not be evaluated because none of the current indicator/method
combinations can distinguish between sources.
6.0 REFERENCES
Au-Yeung, H.C. , D. Rh. Thomas, and R.L. Salmon. Nd. Appendix 2 Section 2 Risk of
gastroenteritis from recreational bathing water: A meta-analysis of randomized exposure trials in
four European countries. From http://www.ies.aber.ac.uk/en/epibathe. Accessed on July 2009.
18
-------
Evaluation of the Suitability of Individual Combinations of Indicators and Methods
for Different Clean Water Act Programs
Cabelli, VJ. 1983. Health Effects Criteria for Marine Recreational Waters. EPA 600/1-80-031,
U.S. Environmental Protection Agency, Cincinnati, OH.
Chern E.G., K. Brenner, L. Wymer, and R.A. Haugland. 2009. Comparison of Fecal Indicator
Bacteria Densities in Marine Recreational Waters by QPCR. Water Qual. Expo. Health 1(3),
203-214.
Chern E.G., S.C. Siefring, J. Paar, M. Doolittle, and R.A. Haugland. Comparison of quantitative
PCR assays for Escherichia coli targeting ribosomal RNA and single copy genes. Lett. Appl.
Microbiol. Manuscript Submitted.
Converse R.R., A.D. Blackwood, M. Kirs, J.F. Griffith, and R.T. Noble. 2009. Rapid QPCR-
based assay for fecal Bacteroides spp. as a tool for assessing fecal contamination in recreational
waters. Water Res 43:4828-4837. From http://dx.doi.Org/10.1016/j.watres.2009.06.036.
Dick L.K. and K.G. Field. 2004. Rapid estimation of numbers of fecal Bacteroidetes by use of a
quantitative PCR assay for 16SrRNA genes. Appl Environ Microbiol 70(9):5695-5697.
DuFour, A.P. 1984. Health Effects Criteria for Marine Recreational Waters. EPA 600/1-84-004.
U.S. Environmental Protection Agency, Cincinnati, OH.
Fleisher, J.M., L.E. Fleming, H.M. Solo-Gabriele, J.K. Kish, C.D. Sinigalliano et al. 2010. The
Beaches Study: health effects and exposures from non-point source microbial contaminants in
subtropical recreational marine waters. Int. J. Epidemiol. 1-8. From http://ije.oxfordjournals.org.
Frahm, E. and U. Obst. 2003 Application of the fluorogenic probe technique (TaqMan PCR) to
the detection of Enterococcus spp. and Escherichia coli in water samples. Journal
Microbiological Methods 52,123-131.
Francy, D.S. and R.A. Darner. Comparison of methods for determining Esherichia coli
concentrations in recreational water. 2000. Water Res.., 34:2770-2778.
Griffith, J.F., Y. Cao, C. D. McGee, S. B. Weisberg. 2009. Evaluation of rapid methods and
novel indicators for assessing microbiological beach water quality. Water Research., 43: 4900-
4907.
Griffith, J.F., D. Moore, C. McGee, and S.B. Weisberg. 2007. Beta Testing of Rapid Methods for
Measuring Beach Water Quality. Southern California Coast Water Research Project Technical
Report 506. February.
Haugland, R.A., Siefring, S.C., Wymer, L.J., Brenner, K.P. and Dufour, A.P. 2005.
Comparison of Enterococcus density measurements by quantitative polymerase chain reaction
and membrane filter culture analysis at two freshwater recreational beaches. Water Research.,
39:559-568.
Haugland, R.A., M. Varma, M. Sivaganesan, C. Kelty, L. Peed, and O.C. Shanks. 2010.
Evaluation of genetic markers from the 16S rRNA gene V2 region for use in quantitative
detection of selected Bacteroidales species and human fecal waste by real time PCR. Syst. Appl.
Microbiol. 33:348-357.
19
-------
Evaluation of the Suitability of Individual Combinations of Indicators and Methods
for Different Clean Water Act Programs
Love D.C., and M.D. Sobsey M.D. Simple and rapid F+ Coliphage culture, latex agglutination,
and typing assay to detect and source track fecal contamination. Appl. Environ. Microbiol.
73:4110-8.
Ludwig, W. and K.-H. Schleifer. (2000) How quantitative is quantitative PCR with respect to
cell counts? Systematic AppliedMicrobiology 23: 556-562.
Shanks, O.C., C. Kelty, M. Sivaganesan, M. Varma, andR.A. Haugland. 2009. Quantitative PCR
for genetic markers of human fecal pollution. Applied and Environmental Microbiology,
75:5507-5513
Shanks, O.C., K. White, C.A. Kelty, S. Hayes, M. Sivaganesan, M. Jenkins, M. Varma, and
R.A. Haugland. 2010a. Performance assessment PCR-based assays targeting Bacteroidales
genetic markers of bovine fecal pollution. Appl. Environ Microbiol., Mar;76(5): 13 59-66.
Shanks, O.C., White, K., Kelty, C.A., Sivaganesan, M., Blannon, I, Meckes, M., Varma, M.,
Haugland, R.A. 201 Ob. Performance of PCR-based assays targeting Bacteroidales genetic
markers of human fecal pollution in sewage and fecal samples. Environ Sci Technol. 44:6281-8.
Siefring, S., M. Varma, E. Atikovic, L. Wymer and R.A. Haugland. 2008. Improved real-time
PCR assays for the detection of fecal indicator bacteria in surface waters with different
instrument and reagent systems. J. Water and Health 6:225-237.
Southern California Coastal Water Research Project (SCCWRP). FY 2010/2011 Research Plan -
Commissioned Approved. June 2010.
USEPA. 1986. Ambient Water Quality Criteria for Bacteria - 1986. EPA 440/5-84-002. United
States Environmental Protection Agency, Office of Water, Washington, DC.
USEPA. 2000. Improved Enumeration Methods for the Recreational Water Quality Indicators:
Enterococci and Escherichia coli. EPA-821/R-97/004. U.S. Environmental Protection Agency,
Office of Water (4303T), Washington, DC.
USEPA. 2001. Method 1601: Male-specific (F+) and somatic Coliphage in water by two-
step enrichment procedure. Technical Report EPA 821-R-01-030. U.S. Environmental
Protection Agency, Office of Water, Washington, DC.
USEPA. 2002. Method 1600: Enterococci in water by membrane filtration using membrane-
Enterococcus indoxyl-D-glucoside agar (mEI), EPA 821/R-02/022. U.S. Environmental
Protection Agency, Office of Water, Washington, DC.
USEPA. 2006a. Method 1600: Enterococci in water by membrane filtration using membrane-
Enterococcus indoxyl-D-glucoside agar (mEI), EPA 821-R-06-009. U.S. Environmental
Protection Agency, Office of Water, Washington, DC.
20
-------
Evaluation of the Suitability of Individual Combinations of Indicators and Methods
for Different Clean Water Act Programs
USEPA. 2006b. Method 1603: Escherichia coll (E. coli) in Water by Membrane Filtration Using
Modified membrane-Thermotolerant Escherichia coli Agar (Modified mTEC). EPA-821-R-06-
011. U.S. Environmental Protection Agency, Office of Water, Washington, DC.
USEPA. 2007. Guidelines establishing test procedures for the analysis of pollutants; Analytical
methods for biological pollutants in wastewater and sewage sludge: Final rule. Federal Register,
Vol. 72, No. 57, March 26, 2007.
USEPA. 2008. Results of the inter-laboratory testing study for the comparison of methods for
detection and enumeration of Enterococci and E. coli in combined sewer overflows. EPA 821-
R-08-006. U.S. Environmental Protection Agency, Office of Water, Washington, DC.
Wade, T.J., N. Pai, J.N.S. Eisenberg, and J.M. Colford. 2003. Do U.S. Environmental Protection
Agency water quality guidelines for recreational waters prevent gastrointestinal illness? A
systematic review and meta-analysis. Environmental Health Perspectives 111(8): 1102-1109.
Wade, T.I, R. L. Calderon, E. Sams, M. Beach, K.P. Brenner, A. H. Williams, and A P. Dufour.
2006. Rapidly Measured Indicators of Recreational Water Quality Are Predictive
of Swimming-Associated Gastrointestinal Illness. Environmental Health Perspectives,
114(l):24-28.
Wade, T. J., R.L. Calderon, KP Brenner, E. Sams, M. Beach, R. Haugland, L. Wymer, and A. P.
Dufour. 2008. High Sensitivity of Children to Swimming-Associated Gastrointestinal Illness
Results Using a Rapid Assay of Recreational Water Quality. Epidemiology, 19: 375-383.
Wade, T.J., E. Sams , KP Brenner, R. Haugland , E. Chern, et al. 2010. Rapidly measured
indicators of recreational water quality and swimming-associated illness at marine beaches: a
prospective cohort study. Environmental Health, 9:66. Available online at
http ://www. ehj ournal. net/content/9/1766.
Zmirou, D., Pena, L., Ledrans, M., Letertre, A. 2003. Risks associated with the microbiological
quality of bodies of fresh and marine water used for recreational purposes: summary estimates
based on published epidemiological studies. Archives of Environmental Health 58(11): 703-711
21
------- |