xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001909537
007 cr mnu|||uuuuu
008 070919s2006 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001666
Brownell, Miriam J.
Use of BOX-PCR subtyping of Escherichia coli and Enterococcus spp. to determine the source of microbial contamination at a Florida beach
h [electronic resource] /
by Miriam J. Brownell.
[Tampa, Fla] :
b University of South Florida,
ABSTRACT: Siesta Key Beach, located on the Gulf Coast of Florida, is frequently mentioned among the top ten beaches in the US. In summer 2004, high levels of indicator bacteria caused health warnings to be posted, and a storm drainage system was implicated as a possible source of microbial contamination. A study was initiated to determine whether indicator bacteria that persisted in the stormwater system could contribute to high microbial loads in receiving waters. Two sampling events, one within 48 hours of a rain event and the other during dry conditions, were conducted. Water and sediment samples were taken at various sites from the storm drainage system to the beach. Fecal coliforms and Enterococcus spp. were enumerated, and genotypic fingerprints of E. coli and Enterococcus spp. were generated by BOX-PCR. Diversity of E. coli and Enterococcus populations was calculated with the Shannon-Weiner diversity index. Similarity of E. coli and Enterococcus populations was calculated^ with the population similarity coefficient. After the rain event, levels of fecal coliforms and Enterococcus spp. were high in sediments and exceeded the regulatory standard for all water samples. In dry conditions, levels were lower in water samples, but still high in sediment samples. Significantly greater population diversity was observed in the rain event compared to the dry event for both E. coli and Enterococcus populations, and greater population similarity was observed in dry conditions. Enterococcus population diversity was significantly higher in untreated sewage and the Siesta Key rain event when compared to dry conditions, and to a site on the Myakka River (no known human input or urban stormwater runoff). Siesta Key populations in dry conditions were most similar to Myakka, and sewage was the least similar to all other populations.Increased population similarity for E. coli and Enterococcus spp. during dry conditions suggests that a portion of the population is composed^ of "survivor" isolates. Persistence of survivor isolates in the storm drainage system, where urban runoff can sit for days, suggests a reservoir for indicator bacteria that can be flushed through the system to the Gulf, causing high levels of indicator bacteria in receiving waters.
Thesis (M.A.)--University of South Florida, 2006.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 51 pages.
Adviser: Valerie Harwood, Ph.D.
Microbial source tracking.
t USF Electronic Theses and Dissertations.
Use of BOX-PCR Subtyping of Escherichia coli and Enterococcus spp. to Determine the Source of Microbial Contamination at a Florida Beach by Miriam J. Brownell A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science Department of Biology College of Arts and Sciences University of South Florida Major Professor: Valerie Harwood, Ph. D. Daniel Lim, Ph.D. John Lisle, Ph.D. Date of Approval: June 13, 2006 Keywords: diversity, population similarit y, indicator bacteria DNA fingerprinting, microbial source tracking Copyright 2006, Miriam J. Brownell
Acknowledgements I would like to thank my family (Mark, Jonathan, and Josh) and friends (Koloseys and Turnages) for seeing me through the hills an d valleys of graduate school. I would like to thank my major professor, Dr. Valeri e (Jody) Harwood, for her excellent guidance, support, and home brew; and my committee me mbers, Dr. Daniel Lim and Dr. John Lisle for their support and encouragement. I also thank the Harwood lab crew for their help, whether it was the processing of samples, or the enjoyment of Dr. Hs excellent home brew. I thank Sarasota County for funding the Si esta Key Beach project and Dr. Ray C. Kurz of PBS & J for managing the project. I would also like to acknowledge Dr. Troy Scott and Dr. George Lukasik of Biological Consulting Services of North Florida, and Shannon McQuaig, for their work with the dete ction of human-associated markers.
Note to Reader: The original of this document contains color that is necessary for understanding the data. The or iginal thesis is on file with the USF library in Tampa, Florida.
i Table of Contents List of Tables ii List of Figures iii Abstract v Introduction 1 Fecal Contamination in Surface Waters 1 Sources of Indicator Organism other than Fecal Contamination 3 Identifying the Source of Fecal Contam ination: Microbial Source Tracking 6 Diversity/Similarity of Indicator Populations 11 Specific Objectives of the Thesis 13 Materials and Methods 15 Study Site and Sampling Strategy 15 Isolation and Enumeration of Indicator Bacteria 16 BOX-PCR of E. coli and Enterococcus spp. 18 Statistical Analysis 19 Results 25 Enumeration of Indicator Bacteria 25 Diversity Measured by Accumulati on Curves and the Shannon-Weiner Index 29 Similarity Measured by the Population Similarity Coefficient 34 Discussion 38 References 44
ii List of Tables Table 1 Equations for indicator population diversity and similarity 22 Table 2 Sites sampled at Siesta Key Beach after a rain event (08/03/04) and during dry conditions (8/31/04) 24 Table 3 Comparison of the population diversity of E. coli and Enterococcus spp. during the rain even t versus dry conditions 33 Table 4 Comparison of the population diversity of Enterococcus spp. in sewage, Myakka River, rain event, and dry conditions 33
iii List of Figures Figure 1 Sampling locations within th e stormwater system draining to Siesta Key Beach 21 Figure 2 Ditch outfall at Siesta Key Beach 22 Figure 3 Rainfall (inches) during we t conditions sampling at Siesta Key Beach 23 Figure 4 Rainfall (inches) during dr y conditions sampling at Siesta Key Beach 23 Figure 5 Fecal coliform and Enterococcus spp. concentrations from water samples collected during the rain event (log 10 CFU/100ml) 27 Figure 6 Fecal coliform and Enterococcus spp. concentrations from sediment samples collecte d during the rain event (log 10 CFU/100g) 27 Figure 7 Fecal coliform and Enterococcus spp. concentrations from water samples collected duri ng dry conditions (log 10 CFU/100ml) 28 Figure 8 Fecal coliform and Enterococcus spp. concentrations from sediment samples collected during dry conditions (log 10 CFU/100g) 28 Figure 9 Accumulation curves for E. coli populations during the ra in event 31 Figure 10 Accumulation curves for Enterococcus populations during the rain event 31 Figure 11 Accumulation curves for E. coli populations during dry conditions 32 Figure 12 Accumulation curves for Enterococcus populations during dry conditions 32 Figure 13 Averaged accumulation curves for Enterococcus populations 33 Figure 14 Similarity of Enterococcus populations from three water samples collected from a pond, based on BOX-PCR fingerprints 36
iv Figure 15 Similarity of E. coli populations by site during the rain event, based on BOX-PCR fingerprints 36 Figure 16 Similarity of E. coli populations by site during dry conditions, based on BOX-PCR fingerprints 36 Figure 17 Similarity of Enterococcus populations by site during the rain event, based on BOX-PCR fingerprints 37 Figure 18 Similarity of Enterococcus populations by site during dry conditions, based on BOXPCR fingerprints 37 Figure 19 Similarity of Enterococcus populations sampled during the rain event (sampling 1), dry conditions (sampling 2), from sewage, and from Myakka River, ba sed on BOX-PCR fingerprints 37
v Use of BOX-PCR Subtyping of Escherichia coli and Enterococcus spp. to Determine the Source of Microbial Contamination at a Florida Beach Miriam J. Brownell ABSTRACT Siesta Key Beach, located on the Gulf Coas t of Florida, is frequently mentioned among the top ten beaches in the US. In summer 2004, high levels of indicator bacteria caused health warnings to be posted, and a st orm drainage system was implicated as a possible source of microbial contamination. A study was initiated to determine whether indicator bacteria that persisted in the stormwater system could contribute to high microbial loads in receiving waters. Two samp ling events, one within 48 hours of a rain event and the other during dr y conditions, were conducted. Water and sediment samples were taken at various sites from the storm dr ainage system to the beach. Fecal coliforms and Enterococcus spp. were enumerated, and genotypic fingerprints of E. coli and Enterococcus spp. were generated by BOX-PCR. Diversity of E. coli and Enterococcus populations was calculated with the ShannonWeiner diversity index. Similarity of E. coli and Enterococcus populations was calculated with the population similarity coefficient. After the rain event, levels of fecal coliforms and Enterococcus spp. were high in sediments and exceeded the regulatory standard for all water samples. In dry conditions, levels were lower in water samples, but still high in sediment samples. Significantly greater population diversity was observed in the rain event co mpared to the dry event for both E. coli and Enterococcus populations, and greater population similarity was
vi observed in dry conditions. Enterococcus population diversity was significantly higher in untreated sewage and the Siesta Key rain ev ent when compared to dry conditions, and to a site on the Myakka River (no known human i nput or urban stormwat er runoff). Siesta Key populations in dry conditions were most similar to Myakka, and sewage was the least similar to all other populations. Increased population similarity for E. coli and Enterococcus spp. during dry conditions suggests that a porti on of the population is composed of survivor isolates. Persistence of survivor isolat es in the storm drainage system, where urban runoff can sit for days, suggests a reservoir for indicator bact eria that can be flushed through the system to the Gulf, causing high levels of i ndicator bacteria in receiving waters.
1 INTRODUCTION Fecal Contamination in Surface Waters Environmental and recreational waters can be impacted by fecal contamination, leading to the risk of pathoge ns infecting the public. This can result in closings of recreational water sites and sh ellfishing beds, and consequently loss of revenue. Sewage from failing infrastructure or onsite septic systems, agricultural runoff, and stormwater discharge can be a potential s ource of pathogens to a wate r body, creating a health risk (29, 35, 88). Diseases affecting the respirator y, ocular, gastrointes tinal and myocardial processes of the human body are caused by human viruses that are excreted in feces (38, 84). Cryptosporidium spp ., which are protozoan parasites, can be excreted in the feces of agricultural livestock, domesticated animals, and wildlife (26). E. coli O157:H7 has been found in cattle feces (44), and Campylobacter jejuni has been found in cattle and poultry feces (2, 78). Thus, the need to protect surface water quality from excessive fecal inputs and remediate impaired watersheds is evident. Fecal coliforms, Escherichia coli and Enterococcus spp. are indicator organisms used as surrogates for waterborne pathogens (3 ). The indicator concep t has been used to gauge water quality since the beginning of the 20 th century (111). These bacteria normally inhabit the gastrointestinal tract of humans and animals and are excreted in feces; therefore, their presence in environm ental and recreational waters indicates the possible presence of pathogens. Characteristi cs of these indicator organisms should
2 include the following: 1) non-pathogenic th emselves, 2) easy and rapid to detect and enumerate, 3) not native to the environment or able to reproduce in the environment, 4) able to survive as long as pathogens and at least as resistant to environmental stressors, and 5) their presence should correlate with the presence of pathogens and the associated health risk (39). The Clean Water Act (1972) addressed re gulation of water quality to protect surface waters in the United States. The US Environmental Protection Agency (USEPA) utilized this legislation and a series of ep idemiological studies to set quality standards based on indicator organism concentrati ons. The USEPA-recommended indicator for fresh water is E. coli or Enterococcus spp., while for marine water the recommended indicator is Enterococcus spp. (99). The Florida Department of Environmental Protection (DEP) recommends fecal coliforms for fres h and marine waters (31). The Florida Department of Health (DOH), which monito rs the beaches of Florida, adopted the USEPA recommendation for Enterococcus spp. and the Florida DEP recommendation for fecal coliforms. A water samp le of 100 milliliters containing 104 Enterococcus spp. and/or 400 fecal coliforms would indicate poor water quality (http://esetappsdoh.doh.state.fl.us/irm00beachwa ter/terms.htm). An increase in concentrations of these indicator bacteria in a water body correlates to an increased probability of exposure to pathogens, and therefor e indicates an increase in health risk. Transmission of disease to swimmers via exposure to contaminated recreational waters has been investigate d. In 1983 the USEPA published a review of epidemiologicalmicrobiological studies it conducted dur ing the 1970s, which found a positive correlation between density of an indicator ( Enterococcus spp.) in marine waters and
3 gastrointestinal symptoms among swimmers ( 98). A series of epid emiological studies published in the 1980s and 1990s, and revi ewed in 1998 (85), examined the link between health risk and indicator organism concentrations in recreational waters by following the health outcomes of groups of people who were exposed to contaminated recreational waters. A majority of the 22 studies were pros pective cohort studies, which have the unfortunate drawback that follow up is not always reliable, and subjects being observed can drop from the study (63). Two st udies (28, 54) were randomized controlled trials, which are considered to be more reli able than prospective cohort studies because they eliminate many biases and sources of e rror (8). Subjects were randomly assigned to either an exposed group or a control ( unexposed) group, and conditions (exposure time, etc.) were pre-determined. Kay et al. (1994) found a significant difference between reported illnesses of the e xposed group versus the control group, and a linear trend between the incidence of gastro enteritis and c oncentration of Enterococcus spp.. In the second study, Fleisher et al. (1996), exposure to water w ith a concentration of > 50 Enterococcus spp.100 ml -1 was predictive of respiratory illness and exposure to water with a concentration of > 100 fecal coliforms100 ml -1 was predictive of ear ailments. Both studies took place in marine waters that were known to be influenced by domestic sewage, thus establishing a link between indicator concentr ations and increased health risk. Sources of Indicator Organisms other than Fecal Contamination Estimating the extent of fecal contamination in a water body and its relationship to human health risk by indicator organism levels relies on many assumptions, including:
4 1) there is no source other than feces for thes e bacteria, 2) all fecal sources pose an equal risk to human health, and 3) pe rsistent survival or regrowth of indicator organisms in the environment does not exist (or mirrors th at of pathogens). Fecal coliforms, E. coli, and Enterococcus spp. have been associated with epiphytic flora (76, 87), insects (37), plankton (69), and green algae (108), as well as effluent from pulp and paper mills. A study by Gauthier and Archibald (2001) measured fecal coliforms and Enterococcus spp. densities at Canadian pulp and paper mills, where water used to process the pulp is clarified and aerated before being released as effluent. Water samples taken at various points during this process harbored concentrations as high as 10 5 CFU100 ml -1 of both fecal coliforms and enterococci. Fecal coli forms were detected on wood chips and bark dust, suggesting a possible source of inoculum material whose growth could be supported by either biofilms in machinery and pipes, or conditions conducive to growth in the primary clarifier (36). Sediment can influence the survival of indicator organisms once they are introduced into the environment by providing nutrients and protection. Previous studies have shown that indicator or ganisms can survive in water and sediment (6, 16, 17, 33, 56) and can possibly propagate in sediment (17, 20, 94). A study by Byappanahalli and Fujioka (17) demonstrated growth of E. coli on 10% soil extract agar. Fecal coliforms and E. coli from sewage also increased in numbers after being inoculated into irradiated soil. Anderson et al. (6) examined indicator su rvival using non-sterile sediment and water in simulated environmental conditions. Se parate experiments were conducted using inoculum from contaminated soil, sewage, or dog feces. Decay ra tes were slower in sediments than in the water columns for fecal coliforms and Enterococcus spp.,
5 indicating greater persistence in sediment. Fu rthermore, the type of inoculum tended to influence persistence, as the bacteria in cubated in mesocosms inoculated with soil inoculum had the lowest decay rate. Bacteria previously ex posed to natural conditions like that of the mesocosms would be better adapted than bacteria from fecal matter accustomed to the gastrointestinal tract where conditions (e.g. temperature, nutrient availability) are far different. Thes e studies suggest the ability for E. coli and Enterococcus spp. to have survivor strains. Stormwater runoff or tidal movement can cause an influx of indicator organisms into surface waters. A study in 2001 (33) co mpared a group of four Hawaiian beaches receiving discharge from streams or storm drai ns to a control group of four beaches that did not receive any discharge. E. coli and Enterococcus spp. concentrations were low (02 CFU100 ml -1 ) for the beaches in the control gr oup. However, the beaches receiving discharge did exceed the State of Hawaii re creational water quality standard of 7 enterococci100 ml -1 Solo-Gabriele et al. (94) sample d a tidally influenced river located in an urban south Florida community. E. coli concentrations were comparatively elevated during rain events and high tide. Concentrations were also higher in water samples taken close to the river bank when compared to water samples taken in the middle of the river. The authors concluded that the elevated E. coli concentrations were not representative of fecal imp act alone, but that the growth of E. coli in riverbanks soils was a contributing factor. The persistence of these bacteria in th e environment and the association with sources other than the gastroin testinal tract of humans and animals strongly suggests that high numbers are not always correlated to th e potential for pathogen presence. Therefore
6 the ability to determine the source of indi cator organisms would be beneficial in establishing the risk to human health of environmental and recreational waters that are classified as poor quality. Identifying the Source of Fecal Contamination: Microbial Source Tracking Microbial source tracking (MST) is a recently developed concept that includes a group of methodologies that provide information used to identify the dominant source(s) of fecal contamination. Its many methods us e phenotypic or genotypic characteristics of an indicator or target organi sm to differentiate fecal s ources. Phenotypic schemes are typically based on characteristics such as anti biotic resistance or carbon source utilization (40, 45). A genotypic characteristic is a specific component of the genome that is identified by a probe or amplified by th e polymerase chain reaction (PCR) (77, 97). Methods of MST can be grouped into two broad categories, library-dependent and library-independent. Library-dependent methods rely upon a database of fingerprints or patterns created from the phenotypic or ge notypic traits of indicator organisms (e.g., E. coli or Enterococcus spp.) isolated from feces of specific host sources, i.e. human, cow, dog and seagull (101). This creates a library of patterns from known sources. Fingerprints of the indicator orga nism found in a contaminated water body are then compared to the library to determ ine the probable source. An example of a phenotypic library method is antibiotic resistance analysis (ARA), in which a pure culture of a bacterium is grown in the presence of different antibiot ics at several concentrations and scored for resistance. The underlying hypothe sis behind this method is that different host types are exposed to differe nt antibiotics at different levels, ranging from clinical
7 treatment to no exposure, which results in va riation in ARA patterns. Many studies using this method to identify sour ces of contamination have been published, showing discrimination between human and non-human sources (41, 45, 75, 107, 109). An example of a genotypic librarybased method is the PCR-mediated amplification of several diffe rent genetic repeating elemen ts, collectively known as repPCR. Some of the repeating elements target ed are repetitive extragenic palindromes (REP), enterobacterial repetitive intergenic concensus (ERIC), and the Box sequences (BOX) believed to be part of a gene re gulatory element (68). Rep-PCR has been primarily used to type pathogen strains ( 23, 55, 102) and has recently been applied to MST using E. coli strains (18, 21, 53, 73). Primers are designed to read outward from the genetic element so that segments of DNA be tween the repeating elements are amplified, creating amplicons of varying lengths. The am plicons are then elec trophoresed, creating a visual fingerprint or patter n. A study in 2000 (21) constr ucted MST libraries containing human and nonhuman sources generated by RE P and BOX primer(s). Using Jackknife analysis, the library generated by the BOX prim er was shown to have a higher percentage of isolates correctly assigned to the source groups. A possible reason for the discriminative ability of the BOX primer wa s the increased number of bands it generated in the pattern when compared to the REP-patterns. A more recent study published in 2005 (46) also compared BOX and REP-generated libraries for E. coli and observed both libraries to have the same overall correct classification rate. This MST study was the first peer-reviewed publication to include rep-PCR libraries of Enterococcus spp. and demonstrated that BOX-generated patterns for Enterococcus spp. had the highest overall
8 correct classification rate when comp ared to REP-generated patterns for Enterococcus spp. and to both BOX and REP-generated patterns for E. coli. A library-independent method does not require a database of patterns for comparison, but instead has a sp ecific target which, when pr esent, would indicate fecal contamination from a particular source. The ta rget could be a gene, virus, or a bacterium associated with a specific host, and is usua lly detected by a molecular method such as PCR. An example of a specific gene w ould be the enterococcal surface protein ( esp ) gene, a putative virulence factor found in human-associated E. faecium and E. faecalis subtypes (43, 91) Scott et al (89) develope d a PCR assay to target the E. faecium variant, which was detected in 97% of se wage samples (n=65), and not in bird or livestock fecal samples (n=102). Detection of the esp gene is based on absence/presence and has not been modified for quantification. Host-associated viruses have also been investigated as possible MST markers. Hsu et al (49) developed oligonucleotide probe s to differentiate between the four classes (serotypes) of F+ coliphages. A distinct ion was made between coliphages associated with human feces (class II and III) and coliphages associated with animal feces (class I and IV), but there is a questi on about the distribution of F+ coliphages in all individuals (47, 79) and serotype crossspecificity between human and animal hosts has been reported (83). Adenoviruses (32, 52, 66, 84) and enteroviruses (32, 62), have been targeted by PCR to detect human and non-human fecal contamination, and more recently polyomaviruses (74) have been used fo r the detection of human contribution. Polyomaviruses are secreted th rough the urine of an infected individual in concentrations as high as 10 5 ml -1 (14). Serological studies estimat e that 27 to 80% of the human
9 population is infected in early childhood (10, 57) with what is normally an asymptomatic infection unless the individual is immunocom promised. Behzad-Behbahani et al (10) demonstrated that shedding through urine was significantly higher in immunocompromised cohorts than in immunoc ompetent ones, which would suggest that distribution/contribution would be limited to a portion of the population. However, two studies (13, 14) have detected polyomaviruses in sewage from the US, Europe, and Africa. Bofill-Mas et al (2000) used nest ed-PCR to target JCV and BKV, two human strains from the genus Polyomavirus in sewage samples collected from Spain, France, Sweden, and South Africa. Ninety-six percent of the samples (n=28) were positive for JCV and 77.8% were positive for BKV. In 2001, the authors detected both strains in all sewage samples (n=15) collected from Egypt, Greece, and Washington, D.C. at concentrations of 10 2 to 10 3 JCV particlesml -1 and 10 1 to 10 2 BKV particlesml -1 (13). The concentrations found would indicate th at even though a portion of the human population is secreting the viruses, sewage as a composite sample generally contains the polyomavirus-markers. An example of a host-specific bacterial group utilized to determine sources of fecal contamination is the Bacteroides-Prevotella group ( Bacteroidales ). They are noncoliform, anaerobic bacteria th at are highly concentrated in feces. Bernhard et al (11) designed primers to distinguish between human-associated and ruminant-associated species. The PCR assay does not require cultu ring, but uses DNA extracted from fecal or water samples as template. A study in 2003 (12) te sted coastal sites in southern California for human impact using the human-associat ed primers. No correlation was found between positive reactions (presence of marker) and levels of indicator bacteria (total
10 coliforms, E. coli and Enterococcus spp.). There was no exceedance of regulatory standards at the sites testing positive for the human-associated marker, but one site tested negative for the marker and exceeded the en terococci standard. Quantification of the marker would help in determining correlati on to enumerated indicator bacteria in contaminated waters. A study in 2005 (90) developed a SYBR Green PCR assay for quantification using a previous ly published human-specific forward primer (11) and a novel reverse primer. The limit of detection was one nanogram of human feces seeded into one liter of freshwater and the limit of quantification was 10 5 markers per liter of seeded freshwater. A phylogenic approach used by Di ck et al (2005) analyzed Bacteroidales 16S rRNA gene sequences from the feces of ma ny animal hosts. Human, cat, dog, and gull sequences clustered together with known culturable specie s, while ruminant, pig, and horse formed unique clusters of uncultivated bacteria from Bacteroidales. Primers were developed for pig and horse that amplified ta rget DNA from the feces of those hosts and not from other species. Such an approach could be useful for identification of other hostspecific markers. Microbial source tracking includes a wide array of phenotypic, genotypic, librarydependent, or library-independent methods that together represent a toolbox approach. Currently, no single tool or method can pred ict the source of fecal contamination with great confidence. There are stil l questions about the distribu tion of host-specific patterns, fingerprints, and markers; e.g., are they distribu ted in all individuals of that host, and only for that particular host? As methods continue to develop, and are combined for validation
11 and robustness, this will aid in identifying s ources of fecal contam ination and therefore aid in the restoration of impacted recreational and environmental waters. Diversity/Similarity of Indicator Populations The eighteenth century biologist, Carolus Li nnaeus, created a system to classify all living organisms based on the differences and similarities of the organisms (95). This system, which still exists today, used morphol ogical characteristics to name and separate large, visible organisms into a hierarchy of gr oups or taxa. To further define species, the smallest unit of the classification system a biological species concept was first formulated in 1942 by Ernst Mayr (70). Accord ing to the concept, species are populations that can reproduce amongst themselves, but not with other groups, th erefore keeping their gene pools separate. Applying the classical species concept to prokaryotic organisms is quite problematic. Not only are prokaryotes asexual, but many can participate in lateral transfer of DNA from other species (22, 96). A molecular approach is used to circumvent the classic species definition for one more accommodating to microorganisms. DNA: DNA hybridization is one method used to determine relatedness between bacterial isolates. There is no set rule, but in general, an outcome of 70% hybridization between the genomic DNA of two isolates would mean they were of the same species (64). Another approach to identifying species is to use a molecular chronometer to measure evolutiona ry genetic changes. Among prokaryotes the 16S rRNA sequence is considered to be hi ghly conserved and can therefore measure long-term evolutionary relationships. Variab le regions within the conserved sequences can be translated into the phylogenic distances that are used to determine genera and
12 species (110). When comparing isolates, le ss than 97%similarity in 16S rRNA sequences would infer different species and is usua lly coupled with less than 70% DNA:DNA hybridization (64). Genetic differences within bacterial species are also common, and are utilized for library-based MST methods that use molecular typing of b acterial groups (e.g. E. coli and Enterococcus spp.) from different hosts (46, 53, 1 01). Various genetic typing methods can be used to generate a DNA fingerprint for a given isolate, which can be matched for identity, or a pre-determined level of similarity, to other fingerprints (e.g. unknowns to host sources). These same fingerprints can be compared in terms of their genetic variability to determine how diverse E. coli or Enterococcus spp. subtypes are in a particular population. Measuring the diversity of an E. coli or Enterococcus population by typing the finite number of individuals in that community is an impossible task. Measuring a sample or subset of that population to estimat e its diversity is more plausible and can be done with diversity indices such as Shannon-Weiner. This diversity index takes into account the number of subtypes as well as the frequency of those subtypes (9), and has been previously used to measure microbial population diversity in habitats such as rhizospheres, artesian spring sediments, and microbial mats (25, 71, 82). Another method used to measure population diversity is the accumulation curve, which plots the number of new subtypes observed versus sampling effo rt. This gives information about how well a population has been sampled; as the curve r eaches an asymptote a larger portion of the total population has been sampled (50). The accumulation curve has been previously used to estimate diversity in animal popul ations (15) and more recently applied to E. coli
13 populations in horse, cattle, and human feces (7 ). Accumulation curves can be useful in comparing relative diversities of populations th at have been affected by an environmental change (50). Similarity between E. coli or Enterococcus populations can be measured with the population similarity coefficient, which measures the proportion of identical subtypes in two populations (60). This has been previous ly used to compare phenotypic subtypes of coliforms in environmental water samples (60), and phenotypic subtypes of fecal coliforms and enterococci in sewage ( 67, 104, 105), and in the feces of livestock, seabirds, and dogs (61, 106). Population similari ty can be used to explore the hypothesis that physical contribution of indicator bacter ia from one environmental compartment to another, such as a storm drainage system to receiving coastal waters, can be a source of indicator bacteria. Specific Objectives of the Thesis Siesta Key Beach is located on the Gulf Co ast of Florida, south of Tampa, and is frequently mentioned among the top ten beaches in the US. In summer 2004, high levels of fecal coliforms and Enterococcus spp. caused health warnings to be posted by the Florida Department of Health. A stormwater drainage system was implicated as a possible source of microbial pollution (Fi gure 1). Stormwater flows through underground pipes to an underground concrete va ult, where it may be retained for many days. Overflow stormwater is de livered to an open retention pond located approximately 100 yards from the landward edge of the beac h. Rain events cause movement from this system to a ditch that empties into the Gulf of Mexico at Siesta Key Beach.
14 The specific objectives of this study were th reefold: 1) to asse ss and compare population diversity of E. coli and Enterococcus in the drainage system during a rain event and during dry conditions, 2) to observe similarity of the E. coli and Enterococcus populations between specific sites sampled th roughout the storm drainage system to the Gulf, and 3) to compare the Enterococcus populations of Siesta Key to that of sewage and of a pristine site (no known human impact, or urban stormwater runoff). These characteristics of the indicator bacteria popul ations were used to explore the hypothesis that the microbial contamination at Siesta Key Beach originated from the stormwater system.
15 MATERIALS AND METHODS Study Site and Sampling Strategy Siesta Key Beach is located on a barrier island on the west coast of Florida in Sarasota County. A stormwater conveyance sy stem runs parallel to the beach underneath a paved thoroughfare (Figure 1). The stormwater system receives runoff from an urban, residential area of approximately 60 acres. A portion of the stormwat er enters a canal on the east side of the road (northeast of th e beach), and the majority remains in the underground system, which runs southward to an underground concrete vault on the west side of the road, approximately 100 yards from the beach. Water may be retained in the vault for many days until a rain event causes overflow, which is pumped into an adjacent retention pond. Surface runoff from the road and overflow from the pond enter a ditch, which flows ~100 yards before it empties onto the beach. During heavy rain, the ditch outfall reaches the Gulf waters. Two sampling events were conducted during this study; one w ithin 48 hours of heavy rainfall (Figure 3), and one during a dr y period (Figure 4). Water and sediment samples were taken at various points, i.e ., access was obtained via a manhole to sample the stormpipe that feeds the vault, the va ult was sampled through a metal-covered access portal, and the ditch and its beach outfall we re sampled from the surface (Table 2). The land around the ditch and the ditch itself was heavily vegetated, and therefore shaded, with Brazilian pepper trees and mangroves. More surface sampling sites were added
16 (retention pond and Gulf of Mexico) for the second sampling (dry period) in order to obtain a more complete picture of the possi ble sources and sinks of microorganisms in the drainage system. For genetic diversity studies, Enterococcus spp. were also isolated from sewage and a pristine water site. Untreated sewage samp les were obtained from lift stations in the Florida counties of Duval and Wakulla. Water sa mples from a pristine site were collected at Deer Prairie Slough in the Myakka River, Myakka River State Park (Sarasota County; GPS N Latitude 27 10.543' and W Longitude 82 12.705'). This site was chosen due to the absence of known human impact and urban stormwater runoff. To examine variability in collection of subtypes during the sampling process, a study was conducted using repli cate water samples from a pond located on campus at the University of South Florida, Tampa campus (GPS N Latitude 28 03.704and W Longitude 82 25.060). One-liter grab samples (triplicate) were co llected in a one-meter 2 area, just below the water surf ace level, close to the shore. The pond covered ~ 3 acres, had little shade, and was inhabited by ducks. Isolation and Enumeration of Indicator Bacteria Water and sediment samples were collect ed in sterile containers, immediately placed on ice, and processed within 4 h of collection at the USF (Tampa, FL) laboratory. Water samples were collected in one-liter c ontainers (in duplicate) and filtered through sterile nitrocellulose membranes (0.45 m pore-size, 47 mm diameter) to enumerate fecal coliforms and Enterococcus spp. Sediment samples were colle cted (in duplicate) in 50 ml screw-cap conical tubes by sc ooping the top layer of sedime nt into the conical tube.
17 Twenty grams (wet weight) of sediment were added to 200 ml of sterile buffered water (0.0425 g L -1 KH 2 PO4 and 0.4055 g L -1 MgCl 2 ) and sonicated as previously described (6) to release bacteria from soil particles. A ra nge of sample volumes and dilutions for both water and sediment samples were filtered to allow for accurate enumeration of bacterial cells. Fecal coliforms were enumerated on mF C agar (Difco) and incubated for 24 h at 44.5 o C in a water bath (4). Blue colonies we re counted as fecal coliforms and then inoculated into microtiter plates containing EC broth amended with 4methylumbelliferyl-D-glucuronide (MUG) (50 g/ml) in order to determine the percentage of the colonies that were E. coli. After incubation for 24 h at 37 o C, the microtiter plates were exposed to ultraviolet (UV) light. Fluorescence indicated strains that had glucuronidase activity (MUG +), a characteristic of E. coli. For further confirmation, 25% of the MUG + isolates we re profiled biochemically using API 20E strips (BioMerieux), and 100% were identified as E. coli. MUG + fecal coliforms were therefore designated E. coli and fingerprinted by BOX-PCR for the similarity/diversity study. Enterococcus spp. were enumerated by USEP A Method 1600 (100), in which filters were incubated on mEI agar (base media from Difco; indoxyl -D glucoside from Sigma Aldrich) at 41 o C for 24 h. All resultant colonies w ith a blue halo were counted as Enterococcus spp. Plates with suitable colony nu mbers (10 100 CFU) were counted, and concentrations for each volume were calcu lated. If indicator bact eria concentrations were low, and no filtration volume contained more than 10 CFU/plate, plates with less than 10 CFU were counted. Concentra tions for all indicators were log 10 -transformed and recorded as CFU 100 ml -1 (water samples) or 100 g wet weight -1 (sediment samples).
18 BOX-PCR of E. coli and Enterococcus spp. E. coli strains were grown overnight in micr ocentrifuge tubes containing 750 l of BHI broth (Becton Dickinson). After centr ifugation at 14,000 RPM for one minute, pellets were washed with sterile buffered wa ter two times and resuspended in 500 l of deionized sterile water. The cell suspension wa s boiled for 5 minutes to lyse the cells and then centrifuged again at 14,000 RPM for one minut e. One l of supernatant was used as template for each PCR reaction. BOX-PCR fingerprints were generated using the previously published BOXA1R primer (58), which has the following sequence: 5-CTA CGG CAA GGC GAC GCT GAC G3. Reagen ts and volumes for each 25 l reaction were: 2.5 l 10X Buffer B (Fishe r Scientific); 3.0 l 25mM MgCl 2 (Fisher Scientific); 1.0 l 10mM dNTPs (Fisher Scientific); 2.5 l 2% bovine serum albumin (Sigma); 1.3 l 10 M BOXA1R primer (IDT, Coralville, IA); 1.0 l Taq polymeras e (5000u/ml) (Fisher Scientific); and 12.7 l PCR-grade water (Fishe r Scientific). The thermocycler program contained three steps: 1) initial denaturation at 95C for 5 minutes; 2) 35 cycles of 94 C for 1 minute, 60 C for 1 minute, and 72 C for 1 minute; and 3) final extension at 72 C for 10 minutes. The preceding protocol wa s provided by correspondence with Dr. Cindy Nakatsu, Purdue University, West Lafayette, IN. Enterococcus spp. were grown overnight in mi crocentrifuge tubes containing 1.5 ml of BHI broth (Becton Dickinson). DNA was extracted using the DNeasy Tissue Kit (Qiagen, Valencia, CA) and the manufacturers protocol for Gram-positive bacteria. BOX-PCR fingerprints for enterococci were generated using the BOXA2R primer (58), which has the following sequence: 5-ACG TGG TTT GAA GAG ATT TTC G3. PCR reagents and conditions used were fro m previously published protocols with
19 modifications (65, 103). Each 25 l PCR reactio n contained: 5 l of 5X Gitschier Buffer (59); 2.5 l of 10% dimethyl sulfoxide; 0.4 l bovine se rum albumin(10mg/ml); 2.0 l 10mM dNTPs; 1.0 l Taq polymerase (5000u/ml ); 11.6 l PCR-grade water; 1.5 l 10M BOXA2R primer; and 1.0 l of DNA te mplate, containing between 30 to 100 ng l -1 The thermocycler program contained three st eps: 1) initial dena turation at 95C for 7 minutes; 2) 35 cycles of 90 C for 30 s econds, 40 C for 1 minute, and 65 C for 8 minute; and 3) final extension at 65 C for 16 minutes. Fragments were separated by electrophoresis through a 1.5% agarose gel for 4 hours at 90 volts ( E. coli fingerprints), or 6 hours at 60 volts ( Enterococcus spp. fingerprints). Gels were stained with ethidium bromide (1 % solution). Gels were digitally documented under UV light using a FOTO/Analyst Archiver (Fotodyne, Hartland, WI). Statistical Analysis Fingerprint patterns of E. coli and Enterococcus spp. subtypes generated by BOXPCR were analyzed with BioNumerics 4.0 software (Applied Maths, Belgium). Dendrograms were created using a densio metric curve-based algorithm (Pearson correlation coefficient, optimization 1%) and UPGMA to cluster patterns by similarity. Repeated runs of the control strains, ATCC 9637 for E. coli and ATCC 19433 ( E. faecalis ) for Enterococcus spp., were 86% and 93% simila r, respectively. Therefore, patterns showing the similarity value established by the control strains were considered identical. The relationship of patterns cons idered similar was confirmed by eye. The relationships of indicator bacteria populations at the various sites were determined by dendrograms constructed usi ng a population similarity coefficient (S p)
20 (Table 1), previously published by Kuhn et al (1991). The algorithm is based on the proportion of identical isolates between tw o populations; therefore, if two populations have no identical subtypes S p = 0, and as the number of identical subtypes increases between two populations, the S p increases to a maximum of 1.0 (60). The population similarity coefficient was used to compare E. coli and Enterococcus populations at Siesta Key during a rain event and during dr y conditions, and to further compare Enterococcus populations at Siesta Key to Enterococcus populations in sewage and in a sampled site on Myakka River. Accumulation curves and the Shannon-Wein er diversity index were calculated using EcoSim 7 software (Acquired Intelligen ce Inc. & Kesey-Bear, Jericho, VT). An accumulation curve measures the diversity of a sampled population by plotting new subtypes as a function of sampling effort. As the curve approaches an asymptote (slope = 0), the probability of obtaining new subtype s with additional sampling diminishes. The Shannon-Weiner index (H ) of diversity considers the frequency of the various subtypes in a population as well as the total number of subtypes (Table 1). Both the accumulation curve and the Shannon-Weiner index were used to compare the rela tive diversities of E. coli and Enterococcus populations during a rain event and dry conditions at Siesta Key, and to further compare Enterococcus populations at Siesta Key to Enterococcus populations in sewage and in a sampled si te on Myakka River. Paired t tests, nonparametric tests (Mann-Whitney), and ANOV A were used to determine significant difference in the comparisons. GraphPad Prism version 4.02 (GraphPad Software, San Diego, CA) was used for the statistical analyses.
Figure 1. Sampling locations within the stormwater system draining to Siesta Key Beach (light blue arrows indicate general direction of stormwater flow) StormwaterPipeStormwaterPipeStormwaterPipeVaultGulf of MexicoDitchBeach (ditch outfall)PondSiesta Key BeachGrand CanalBeach Road GulfIntracoastalWaterway 21
Figure 2. Ditch outfall at Siesta Key Beach Table 1. Equations for indicator population diversity and similarity Shannon-Weiner index (H) = -p i ln(p i ) p i = # isolates with pattern (i)/total isolates Population similarity coefficient (Sp) = (Sx + Sy)/2 Sx = qx i / Nx Sy = qy i / Ny Nx = total # isolates population x Ny = total # isolates population y qx i = proportion of isolates identical to isolate i in population x divided by proportion of isolates identical to isolate i in population y qy i = the proportion of isolates identical to isolate i in population y divided by the proportion of isolates identical to isolate i in population x 22
Figure 3. Rainfall (inches) during wet conditions sampling at Siesta Key Beach 00.20.40.60.8184.108.40.206.827/27/20047/28/20047/29/20047/30/20047/31/20048/1/20048/2/20048/3/2004Rainfall (in) Sampling Date Figure 4. Rainfall (inches) during dry conditions sampling at Siesta Key Beach 00.10.20.30.220.127.116.11/24/20048/25/20048/26/20048/27/20048/28/20048/29/20048/30/20048/31/2004Rainfall (in) Sampling Date 23
24 Table 2. Sites sampled at Siesta Key Beach afte r a rain event (08/03/04) and during dry conditions (08/31/04). Analyses conducted or not conducted Analyses Conducted Site Sample Date Fecal coliform concentration Enterococcus spp. concentration BOX-PCR E. coli BOX-PCR Enterococcus spp. 08/03/04 Stormpipe water 08/31/04 08/03/04 Stormpipe sediment 08/31/04 08/03/04 Vault water 08/31/04 08/03/04 Pond water 08/31/04 08/03/04 Pond sediment 08/31/04 08/03/04 Ditch water 08/31/04 08/03/04 Ditch sediment 08/31/04 08/03/04 Beach water 1 08/31/04 08/03/04 Beach sediment 08/31/04 08/03/04 Gulf water 2 08/31/04 08/03/04 Gulf sediment 08/31/04 1 Beach water and sediment were collected on the beach, within a few yards of the ditch 2 Gulf water and sediment were collected in the Gulf of Mexico Less than 10 isolates were recovered
25 RESULTS Enumeration of Indicator Bacteria Indicator bacteria were enumerated from water and sediment samples collected from the rain event (Figures 5 and 6) a nd during dry conditions (Figures 7 and 8). Indicator bacteria ( Enterococcus spp. and fecal coliform) concentrations exceeded the Florida standards for recreational waters during the rain event in all water samples, including tidal water on the beach (Figure 5). Enterococcus spp. concentrations were significantly higher than fecal coliforms ( P = 0.041, paired t test). The mean concentrations (log 10 -transformed) were 3.17 + 0.72 and 4.20 + 0.37 for fecal coliforms and Enterococcus spp., respectively. Indicator bacteria levels were also high in sediments collected during the rain event, at >10 3 CFU/100 g (Figure 6), although there are no regulatory standards for indicator concentrations in sediment. During dry conditions, fecal coliforms ex ceeded the standard only at the beach, where water pools from ditch outfall and/ or at high tide (Figures 2 and 7). Enterococcus spp. concentrations exceeded the standard in beach water, stormpipe water, and vault water (Figure 7). Water sampled from the re tention pond, ditch, and the Gulf was within the regulatory standard limits for recrea tional waters for both fecal coliforms and Enterococcus spp. (Figure 7). Indicator bacteria c oncentrations remained high in the sediments during dry conditions with stormpipe sediment the highest at >10 3.5 CFU/100 g for both fecal coliforms and Enterococcus spp. (Figure 8). Overall, Enterococcus spp.
26 concentrations in sediments were significantly higher than fecal co liform concentrations by a paired t-test ( P = 0.020) during dry conditions. The mean concentrations (log 10 transformed) were 1.70 + 1.38 and 3.21 + 1.11 for fecal coliforms and Enterococcus spp., respectively. Indicator organism concentra tions in water samples at sites sampled on both dates (i.e., stormpipe water, vault water, ditch water, and beach water; Table 2) were compared. Mean indicator organism concentr ations were significantly higher during the rain event than during dry conditions as assessed by a nonparametric, Mann-Whitney t test. Differences in mean log 10 -transformed concentrations were statistically significant for Enterococcus spp. ( P = 0.028) and nearly significant for fecal coliforms ( P = 0.057) at the = 0.05 level. The mean fecal coliform concentration (log 10 -transformed) on 8/3/04 (rain event) was 3.17 + 0.72, while it was 1.57 + 0.91 on 8/31/04 (dry conditions). Corresponding means for Enterococcus spp. were 4.20 + 0.37 on 8/3/04 and 2.55 + 0.68 on 8/31/04.
Figure 5. Fecal coliform and Enterococcus spp. concentrations from water samples collected during the rain event (log 10 CFU/100ml) 00.511.522.533.544.55StormpipewaterVault waterDitch waterBeachwaterSitesLog10 CFU/100 ml Fecal Coliforms Enterococci FC regulatory Ent regulatory Figure 6. Fecal coliform and Enterococcus spp. concentrations from sediment samples collected during the rain event (log 10 CFU/100g) 00.511.522.533.544.55Ditch SedimentBeach SedimentSitesLog10 CFU/100g Fecal Coliforms Enterococci 27
Figure 7. Fecal coliform and Enterococcus spp. concentrations from water samples collected during dry conditions (log 10 CFU/100ml) 00.511.522.533.544.55Stormpipewater Vault waterPond waterDitch waterBeachwaterGulf waterSitesLog 10 CFU/100 ml Fecal Coliforms Enterococci FC regulatory Ent regulatory Figure 8. Fecal coliform and Enterococcus spp. concentrations from sediment samples collected during dry conditions (log 10 CFU/100g) 00.511.522.533.544.55StormpipeSedimentPondSedimentDitchSedimentBeachSedimentGulfSedimentSitesLog10 CFU/100g Fecal Coliforms Enterococci 28
29 Diversity Measured by Accumulation Curves and the Shannon-Weiner Index Accumulation curves for the E. coli populations (Figure 9) and the Enterococcus populations (Figure 10) sampled during the rain event do not reach an asymptote, clearly showing that the population diversity for thes e sites was not completely captured by the sampling effort. For dry conditions, the accumulation curve for the E. coli population sampled from the ditch water and vault water (Figure 11) and the Enterococcus population sampled from the Gulf water (Figur e 12) reached an asymptote, showing that the population diversity was captured by the sampling effort. Overall, accumulation curves indicated a trend in lower diversity during dry conditions for E. coli and Enterococcus populations. Averaged accumulation curves were constructed for Enterococcus populations for the rain event (n = 4), dry conditions (n = 4), sewage samples (n = 3), and samples collected at Myakka River (n = 3) (Figure 13). A higher diversity of Enterococcus populations in the rain event and sewage and a lower diversity of Enterococcus populations in dry conditions and Myakka River reflect th e differences in the ShannonWeiner index for these four groups (see below, and Table 4). Sites that were sampled for both the rain event and dry condi tions and had 14 to 20 isolates per site were chosen for compar ison of diversity by using the Shannon-Weiner index ( H ). In comparing population diversity of E. coli versus Enterococcus spp., there was no significant difference in either the ra in event or dry cond itions; however, there was a significant difference in the population diversity of E. coli when comparing the rain event versus dry conditions ( P = 0.047, Table 3), and a sign ificant difference in the
30 population diversity of Enterococcus spp. when comparing the rain event versus dry conditions ( P = 0.008, Table 3). Enterococcus populations from sewage samples and from Myakka River (pristine site) samples were measured for diversity using the Shannon-Weiner index. There was a significant difference in the populatio n diversity between sewage (mean H = 2.69) and Myakka River (mean H = 1.96) with the diversity being higher in the sewage samples than in the samples collected from Myakka River ( P = 0.024). Furthermore, a one-way analysis of variance showed significant difference when comparing the population diversities of the rain event, dry conditi ons, sewage, and Myakka River (Table 4).
Figure 9. Accumulation curves for E. coli populations during the rain event. Subtypes are fingerprint patterns of E. coli isolates by BOX-PCR 024681012141618135791113151719Sampling effortSubtypes Beach Water Beach Sediment Vault Water Ditch Water Ditch Sediment Stormpipe Water Figure 10. Accumulation curves for Enterococcus populations during the rain event. Subtypes are fingerprint patterns of Enterococcus spp. isolates by BOX-PCR 02468101214161820135791113151719Sampling effortSubtypes Beach Water Beach Sediment Vault Water Ditch Water Ditch Sediment Stormpipe Water 31
Figure 11. Accumulation curves for E. coli populations during dry conditions. Subtypes are fingerprint patterns of E. coli isolates by BOX-PCR 024681012141618135791113151719Sampling effortSubtypes Beach Water Vault Water Ditch Water Stormpipe Sediment Gulf Water Figure 12. Accumulation curves for Enterococcus populations during dry conditions. Subtypes are fingerprint patterns of Enterococcus spp. isolates by BOX-PCR 0246810121416182013579111315171921Sampling effortSubtypes Beach Water Vault Water Ditch Water Stormpipe Water Stormpipe Sediment Gulf Water 32
Figure 13. Averaged accumulation curves for Enterococcus populations. Sites included for both rain and dry conditions: beach water, ditch water, vault water, and stormpipe water. Subtypes are fingerprint patterns of Enterococcus spp. isolates by BOX-PCR 024681012141618135791113151719Sampling effortSubtypes Rain Event Dry Event Myakka River Sewage Table 3. Comparison of the population diversity of E. coli and Enterococcus spp. during the rain event versus dry conditions. Paired t test, (=0.05, + standard deviation) Indicator (sites) Mean H P value E. coli (beach water, ditch water, vault water) Rain event = 2.39 + 0.22 Dry conditions =1.12 + 0.34 P = 0.047 Enterococcus spp. (beach water, ditch water, stormpipe water, vault water) Rain event = 2.65 + 0.13 Dry conditions =1.88 + 0.28 P = 0.008 Table 4. Comparison of the population diversity of Enterococcus spp. in sewage, Myakka River, rain event, and dry conditions. Values that share the same letter within columns are not significantly different. ANOVA, (P = 0.0001, =0.05, + standard deviation) Sample events Mean H Sewage 2.69 + 0.09 (a) Myakka 1.96 + 0.22 (b) Dry conditions 1.88 + 0.45 (b) Rain event 2.66 + 0.13 (a) 33
34 Similarity Measured by the Popul ation Similarity Coefficient An estimate of the similarity of Enterococcus populations isolated from three replicate samples on the same day was carried out. This experiment was meant to provide a benchmark for population similarity in sa mples in which the population structure was expected to be very similar, and to ex amine the variability in observed population structure of Enterococcus spp. contributed at the level of replicate samples. The water samples were collected from a pond located on the campus at the University of South Florida. The Enterococcus spp. concentration was 40 CFU100 ml -1 therefore 400 possible (culturable) subtypes were in each one-liter sample. Approximately twenty isolates from each replicate were fingerprint ed by BOX-PCR and compared for similarity by using the population similar ity coefficient (see Materi als and Methods). Among the 58 isolates subtyped from the three replicate sa mples, only five diffe rent BOX-PCR patterns were observed. Samples A and B were 88% simi lar, while sample C was 52% similar to samples A and B (Figure 14). All three samples (A, B, and C) shared two patterns out of five total patterns. Samples A and B shared one pattern and samples B and C shared another pattern. Sample C had one pattern th at was not shared with any other sample. Fingerprint patterns of indicator isolates ( E. coli or Enterococcus spp.) for sampled sites (rain event and dry conditions) were compared to each other to determine similarity between site populations. The population simila rity was calculated by using the population similarity coefficient. Sa mpled sites included for comparison of E. coli populations during the rain event were: beach sediment, stormpipe water, beach water, ditch sediment, ditch water and vault water (Figure 15). During dry conditions, sampled sites for E. coli population comparisons were: beach water, ditch water, stormpipe
35 sediment, vault water, and Gulf water (Figur e 16). During the rain event, the highest E. coli population similarity was between ditch water and ditch sediment, followed by similarity between stormpipe water and b each sediment. During dry conditions, ditch water and vault water had the highest similarity, followed by beach water and stormpipe sediment. Overall, there was higher similar ity between sites during dry conditions when compared to the rain event. The Gulf water population, which was only sampled during dry conditions, had no similarity to any other sites. Sampled sites included for comparison of Enterococcus populations during the rain event were: beach sediment, stormpipe water, beach water, ditch sediment, ditch water and vault water (Figure 17). Du ring dry conditions, sampled sites for Enterococcus population comparisons were: be ach water, ditch water, st ormpipe sediment, stormpipe water, vault water, and Gulf water (Figure 18). During the rain ev ent, beach water and ditch sediment had the highest similarit y. During dry conditions, stormpipe water and vault water had the highest similarity followe d by beach water and ditch water. The Gulf water population, which was only sampled during dry conditions, had no similarity to any other sites. Overall, pop ulation similarities were highe r during dry conditions than during the rain event. Sites that were compared for similarity for the rain event and for dry conditions were grouped together and labeled Sampling 1 and Sampling 2, respectively. The two sampling dates were then compared for similarity to sewage and Myakka River (Figure 19). The two populations with the hi ghest similarity were Myakka River and sampling 2, and the population w ith the least similarity to all other groups was sewage.
Figure 14. Similarity of Enterococcus populations from three replicate water samples collected from a pond, based on BOX-PCR fingerprints 100 95 90 85 80 75 70 65 60 55 A BC Figure 15. Similarity of E. coli populations by site during the rain event, based on BOX-PCR fingerprints 100 90 80 70 60 50 40 30 20 10 beach sedimentstorm pipe waterbeach waterditch sedimentditch watervault water Figure 16. Similarity of E. coli populations by site during dry conditions, based on BOX-PCR fingerprints 100 90 80 70 60 50 40 30 20 10 0 ditch watervault waterbeach waterstorm pipe sedimentgulf water 36
Figure 17. Similarity of Enterococcus populations by site during the rain event, based on BOX-PCR fingerprints 100 90 80 70 60 50 40 30 20 10 ditch waterstorm pipe waterbeach waterditch sedimentbeach sedimentvault water Figure 18. Similarity of Enterococcus populations by site during dry conditions, based on BOX-PCR fingerprints 100 90 80 70 60 50 40 30 20 10 0 beach waterditch waterstorm pipe sedimentstorm pipe watervault watergulf water Figure 19. Similarity of Enterococcus populations sampled during the rain event (sampling 1), dry conditions (sampling 2), from sewage, and from Myakka River, based on BOX-PCR fingerprints 100 90 80 70 60 50 40 30 20 10 Myakkasampling 2sampling 1sewage 37
38 DISCUSSION The stormwater drainage system at Siesta Key Beach was sampled within 48 hours of a rain event (~2 in ches) and during dry conditions (no precipitation 6 days prior). During the rain event, stormwater flowed through an underground stormpipe to an underground vault. Because of the high volume during the rain event, the stormwater was pumped into a retention pond. Stormwater from the retention pond and surface runoff from the main road flowed through a ditch to the beach, where it flowed into the Gulf of Mexico. At that time (rain even t), levels of indicator bacteria were above the regulatory standards at all sites sampled throughout the drainage system (stormpipe to beach). In contrast, during dry conditions, no water was observed flowing through the system except for a trickle from the ditch to the beach, where the water pooled and did not reach the Gulf. During this time, levels of indicator b acteria were much lower in the water column samples; however, the stormpipe, vault, and beach sites still exceeded the regulatory standards. The stormpipe and vault are enclos ed structures that co uld provide protection to indicator bacteria from stressors (discussed below) and the water pooled onto the beach could be directly impacted by anothe r source such as seagulls (Figure 2). The stormwater and surface runoff are fr om an approximately 60-acre area of the residential community of Siesta Key. High levels of indicator bacteria in the stormwater drainage system initially suggested a possible sewage influence. However, prior to the study, the wastewater collection system was examined for any leaks into the stormwater
39 conveyance system by the Siesta Key Utilities Au thority. Furthermore, as part of this study, another laboratory (Biol ogical Consulting Services of North Florida) conducted tests for human polyomaviruses and the enterococcal surface protein gene ( esp ) for Enterococcus faecium Both tests have previously been used to determine the presence of human sewage in environmental waters (74, 89) and produced negative results for this study, which suggests that human se wage input was not involved. Previous studies have shown that stormwat er runoff can elevate levels of indicator bacteria (1, 27, 51, 81, 86). A study conducted in a coastal ur ban watershed in southern California (2004) observed that duri ng dry conditions, total coliforms, E. coli and Enterococcus spp. were highly concentrated in runoff from forebays (underground storage tanks), and that indicat or bacteria concentrations we re higher in residential runoff when compared to other land-uses, incl uding channels, parks, agricultural, and commercial (86). Underground storage of urban runoff ma y well provide favorable conditions for bacterial persistence, allowing it to act as a source of indicator bacteria. Two conditions known to affect the survival of E. coli and Enterococcus spp. are temperature and sunlight. Increased die-off rates were obser ved with an increase in temperature (5, 30, 80) and exposure to sunlight (19, 34, 93) The underground system provides protection from these abiotic influences, and supplies nu trients such as nitr ogen and phosphate from residential fertilizers, promoting survival and possible regrowth. High concentrations of both fecal coliforms and Enterococcus spp. were found in sediments for both sampling events. Enterococcus spp. concentrations remained high during dry conditions even when the overlay ing water column (retention pond, ditch, and
40 Gulf) had concentrations below the regulatory standard. This implies that the dynamics of indicator populations differ between the water column and sediments. Both E. coli and Enterococcus spp. are known to persist in a cultura ble state in sediments (6, 17, 20, 51). Studies conducted have shown lower decay rates of indicator bacteria in sediment than in water (6, 48, 92), indicating that sediments provide protection from harmful stressors (e.g. high temperatures and sun light). Two studies (17, 20) suggest that soil contains the nutrients needed for regrowth of indicator bacteria. Byappanahalli and Fujioka (1998) observed an increase in fecal coliforms and E. coli when adding sewage to cobaltirradiated soil, and Desmarais et al (2002) observed an increase in E. coli and Enterococcus spp. after adding sterile sediment to river water. This supports the premise that sediments are a possible reservoir for i ndicator organisms once introduced into the environment. In comparing the Siesta Key indicator b acteria populations originating from the rain event and from dry conditions, not only were the levels of indicator bacteria different, but also the genotypic makeup of th e indicator bacteria populations. Increased population diversity for E. coli and Enterococcus spp. during the rain event indicates a trend for greater diversity during conditions th at result in stormwater influence on surface water quality. Higher diversity would implicate r ecent inputs, possibly from multiple sources. The diversity of Enterococcus populations during the ra in event and during dry conditions was compared to the diversity of Enterococcus populations found in sewage and in water samples collected from Myakka Ri ver, considered to be a pristine site with no known human input or urban runoff. Similar diversity levels were observed in Siesta Key Enterococcus populations during the rain event and Enterococcus populations in
41 sewage samples collected from lift stations in two Florida counties. Previous studies have shown that the Enterococcus population in domestic sewage has a higher diversity when compared to river water (105), and animal f eces (61, 67). In contrast, significantly lower diversity was found in Enterococcus populations during dry conditions and at a pristine site (Myakka River). This suggests that st ormwater and urban runoff can influence the diversity of indicator bacteria populations in the environment to mimic that of sewage input, although the subtypes re presented in these two environments were dissimilar (see below). Increased population similarity for E. coli and Enterococcus spp. during dry conditions suggests that a subs tantial portion of the population is composed of survivor isolates (6). Both a diversity decrease a nd a similarity increase were observed in the stormpipe and vault Enterococcus populations as well as the beach and ditch under dry conditions compared to wet conditions. For E. coli, a diversity decrease and a similarity increase were observed in the vault, ditch, and beach populations. These populations also shared similarity with the stormpip e sediment. During dry conditions, both E. coli and Enterococcus populations had similarity between a ll sites with the exception of Gulf water. During this time, the water from th e ditch pooled onto the beach and did not reach the Gulf. Studies on the population similarity of an indicator bacterium in environmental waters are relatively rare in the literatur e (60, 72, 105). To demonstrate similarity in indicator bacteria populations considered to be similar, three water samples were collected from the same pond on the same day. Enterococcus spp. from each sample were typed by BOX-PCR and compared, show ing high similarity among samples. The
42 number of isolates per sample was 19 or 20 and the total number of subtypes was 5. This data represents one end of the spectrum w ith low diversity and high similarity from samples collected at one site. When compari ng populations with a much higher diversity and a broader area of sample collection, such as the Enterococcus population at Siesta Key in the rain event, percent similarity is greatly reduced. An inverse relationship was observed during dry conditions; as the popul ation diversity decr eased, the population similarity increased. Sewage isolates, which displayed th e highest diversity, was the group least related to populations isolated du ring the rain event, dry conditions, and from the Myakka River. The fate of the two indicator groups in the environmental habitat probably contributed to observed differences in their population similarity, in that E. coli populations displayed greater popul ation similarity than the Enterococcus spp. populations. Since concentrations of Enterococcus spp. were higher than fecal coliforms, this could be a contributing factor and has b een previously reported in estuarine sites (27, 51, 81), suggesting that Enterococcus spp. are better survivors in estuarine-type waters. Moreover, it is plausible that Enterococcus spp. as a genetic group provides more variability and possible candida tes for survival when compared to the available genetic variability of the one Escherichia species. Even though human sewage input is not evident at Siesta Key Beach, it cannot be definitively stated that there is less risk to human health when indicator bacteria concentrations exceed the regulatory standard. The health risks associated with exposure to recreational waters impacted by stormwater runoff have not been as well studied as the risks associated with sewage impacted wate rs. In one study, Haile et al (42) observed
43 that respiratory and gastrointestinal symp toms increased as the distance decreased between swimmers and a stormwater outlet in Santa Monica Bay, CA. Dwight et al (24) observed that during an El Ni no year, surfers in Orange County, CA reported twice as many symptoms as surfers in Santa Cruz County, considered to be less impacted by urban runoff. These studies show that a dverse heath outcomes are associated with stormwater impact of recreational waters. The population dynamics of indicator bacter ia in the storm drainage system at Siesta Key Beach are evidently affected by ra in events. A change in concentrations and diversity, as well as similarity, of the popul ations extending from the stormpipe to the Gulf was observed. The transport of urban runoff coll ecting for days in the stormpipe and vault, and the persistence of survivor isol ates in the sediments, suggests a reservoir for indicator bacteria that can be flushed through the system to the Gulf during a rain event, causing high levels of indicator bacteria. Such e nvironmental reservoirs of indicator bacteria further complicate the already questionable relationship between indicator organisms and human pathogens (6, 17 ) and call for a bett er understanding of the ecology, fate and persistence of indicator bacteria in water.
44 References 1. Ahn, J. H., S. B. Grant, C. Q. Surbeck, P. M. DiGiacomo, N. P. Nezlin, and S. Jiang. 2005. Coastal water quality impact of stormwater runoff from an urban watershed in southern Calif ornia. Environ Sci Technol 39: 5940-53. 2. Altekruse, S. F., J. M. Hunt, L. K. Tollefson, and J. M. Madden. 1994. Food and animal sources of human Campylobacter jejuni infection. J Am Vet Med Assoc 204: 57-61. 3. American Public Health Association. 1905. Standard Methods of Water Analysis, 1st ed. American Public Health Association, Washington, D.C. 4. American Public Health Association. 1998. Standards for the Examination of Water and Wastewater, 20th ed. Amer ican Public Health Association, Washington, D.C. 5. Anderson, I. C., M. W. Rhodes, and H. I. Kator. 1983. Seasonal variation in survival of Escherichia coli exposed in situ in membrane diffusion chambers containing filtered and nonfiltered estuarine water. Appl Environ Microbiol 45: 1877-83. 6. Anderson, K. L., J. E. Whitlock, and V. J. Harwood. 2005. Persistence and differential survival of fecal indicato r bacteria in subtropical waters and sediments. Appl Environ Microbiol 71: 3041-8. 7. Anderson, M. A. 2003. Frequency distributions of Escherichia coli subtypes in various fecal sources over time and geographical space: application to bacterial source tracking methods. University of South Florida, Tampa. 8. Aschengrau, A., and G. R. Seage, III. 2003. Experimental Studies, p. 163-164, 200, Essentials of Epidemiology in Public Health. Jones and Bartlett Publishers, Sudbury, Massachusetts. 9. Atlas, R. M., and R. Bartha. 1998. Microbial Ecology: Fundamentals and Applications, p. 191, Fourth ed. Benjam in/Cummings Science Publishing, Menlo Park. 10. Behzad-Behbahani, A., P. E. Klapper, P. J. Vallely, G. M. Cleator, and S. H. Khoo. 2004. Detection of BK virus and JC virus DNA in urine samples from immunocompromised (HIV-infected) and immunocompetent (H IV-non-infected) patients using polymerase chain reacti on and microplate hybridisation. J Clin Virol 29: 224-9. 11. Bernhard, A. E., T. Goyard, M. T. Simonich, and K. G. Field. 2003. Application of a rapid method for iden tifying fecal pollution sources in a multiuse estuary. Water Res 37: 909-13.
45 12. Boehm, A. B., J. A. Fuhrman, R. D. Mrse, and S. B. Grant. 2003. Tiered approach for identification of a human fecal pollution source at a recreational beach: case study at Avalon Bay, Catalina Is land, California. Environ Sci Technol 37: 673-80. 13. Bofill-Mas, S., M. Formiga-Cruz, P. Cl emente-Casares, F. Calafell, and R. Girones. 2001. Potential transmission of human polyomaviruses through the gastrointestinal tract after exposure to virions or viral DNA. J Virol 75: 10290-9. 14. Bofill-Mas, S., S. Pina, and R. Girones. 2000. Documenting the epidemiologic patterns of polyomaviruses in human popul ations by studying their presence in urban sewage. Appl Environ Microbiol 66: 238-45. 15. Bohannan, B. J., and J. Hughes. 2003. New approaches to analyzing microbial biodiversity data. Curr Opin Microbiol 6: 282-7. 16. Byappanahalli, M., and R. Fujioka. 2004. Indigenous soil bacteria and low moisture may limit but allow faecal bact eria to multiply and become a minor population in tropical soils. Water Sci Technol 50: 27-32. 17. Byappanahalli, M. N. and R. S. Fujioka. 1998. Evidence that tropical soil environment can support the growth of Escherichia coli Water Science and Technology 38: 171-174. 18. Carson, C. A., B. L. Shear, M. R. Ellersieck, and J. D. Schnell. 2003. Comparison of ribotyping and repetit ive extragenic palindromic-PCR for identification of fecal Escherichia coli from humans and an imals. Appl Environ Microbiol 69: 1836-9. 19. Davies, C. M., and L. M. Evison. 1991. Sunlight and the survival of enteric bacteria in natural waters. J Appl Bacteriol 70: 265-74. 20. Desmarais, T. R., H. M. Solo-Gabriele, and C. J. Palmer. 2002. Influence of soil on fecal indicator organisms in a tida lly influenced subtro pical environment. Appl Environ Microbiol 68: 1165-72. 21. 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: 2572-7. 22. Dubnau, D. 1999. DNA uptake in bacteria. Annu Rev Microbiol 53: 217-44. 23. Dunne, W. M., Jr., and W. Wang. 1997. Clonal dissemination and colony morphotype variation of vancomycin-resistant Enterococcus faecium isolates in metropolitan Detroit, Michigan. J Clin Microbiol 35: 388-92. 24. Dwight, R. H., D. B. Baker, J. C. Semenza, and B. H. Olson. 2004. Health effects associated with recreational co astal water use: urban versus rural California. Am J Public Health 94: 565-7. 25. Elshahed, M. S., J. M. Senko, F. Z. Naja r, S. M. Kenton, B. A. Roe, T. A. Dewers, J. R. Spear, and L. R. Krumholz. 2003. Bacterial diversity and sulfur cycling in a mesophilic sulfide-ri ch spring. Appl Environ Microbiol 69: 5609-21. 26. Fayer, R., and B. L. Ungar. 1986. Cryptosporidium spp. and cryptosporidiosis. Microbiol Rev 50: 458-83. 27. Ferguson, D. M., D. F. Moore, M. A. Getrich, and M. H. Zhowandai. 2005. Enumeration and speciation of enterococci found in marine and intertidal sediments and coastal water in southern California. J Appl Microbiol 99: 598-608.
46 28. Fleisher, J. M., D. Kay, R. L. Salmon, F. Jones, M. D. Wyer, and A. F. Godfree. 1996. Marine waters contaminated with domestic sewage: nonenteric illnesses associated with bather exposure in the United Kingdom. Am J Public Health 86: 1228-34. 29. Fleisher, J. M., D. Kay, M. D. Wyer, and A. F. Godfree. 1998. Estimates of the severity of illnesses asso ciated with bathing in ma rine recreational waters contaminated with domestic sewage. Int J Epidemiol 27: 722-6. 30. Flint, K. P. 1987. The long-term survival of Escherichia coli in river water. J Appl Bacteriol 63: 261-70. 31. Florida Department of Environmental Protection. 1996. Criteria for Surface Water Quality Classifications Chapte r 62-302.530, Florida Administrative Code. 32. Fong, T. T., D. W. Griffin, and E. K. Lipp. 2005. Molecular assays for targeting human and bovine enteric viruses in coas tal waters and their application for library-independent source tracki ng. Appl Environ Microbiol 71: 2070-8. 33. Fujioka, R. S. 2001. Monitoring coastal marine wate rs for spore-forming bacteria of faecal and soil origin to determine point from non-point source pollution. Water Sci Technol 44: 181-8. 34. Fujioka, R. S., H. H. Hashimoto, E. B. Siwak, and R. H. Young. 1981. Effect of sunlight on survival of indicator bact eria in seawater. Appl Environ Microbiol 41: 690-6. 35. Gaffield, S. J., R. L. Goo, L. A. Richards, and R. J. Jackson. 2003. Public health effects of inadequately managed stormwater runoff. Am J Public Health 93: 1527-33. 36. Gauthier, F., and F. Archibald. 2001. The ecology of "fecal indicator" bacteria commonly found in pulp and paper mill water systems. Water Res 35: 2207-18. 37. Geldreich, E. E., B. A. Kenner, and P. W. Kabler. 1964. Occurrence of coliforms, fecal coliforms, and streptococci on vegetation and insects. Appl Microbiol 12: 63-9. 38. Griffin, D. W., K. A. Donaldson, J. H. Paul, and J. B. Rose. 2003. Pathogenic human viruses in coastal wa ters. Clin Microbiol Rev 16: 129-43. 39. Griffin, D. W., E.K. Lipp, M.R. McLaughlin, and J.B. Rose. 2001. Marine Recreation and Public Health Microbio logy: Quest for the Ideal Indicator. BioScience 51. 40. Hagedorn, C., J. B. Crozier, K. A. Mentz A. M. Booth, A. K. Graves, N. J. Nelson, and R. B. Reneau, Jr. 2003. Carbon source utiliz ation profiles as a method to identify sources of faecal po llution in water. J Appl Microbiol 94: 7929. 41. Hagedorn, C., S. L. Robinson, J. R. Filtz, S. M. Grubbs, T. A. Angier, and R. B. Reneau, Jr. 1999. Determining sources of feca l pollution in a rural Virginia watershed with antibiotic resistance patte rns in fecal streptococci. Appl Environ Microbiol 65: 5522-31.
47 42. Haile, R. W., J. S. Witte, M. Gold, R. Cressey, C. McGee, R. C. Millikan, A. Glasser, N. Harawa, C. Ervin, P. Harmon, J. Harper, J. Dermand, J. Alamillo, K. Barrett, M. Nides, and G. Wang. 1999. The health effects of swimming in ocean water contaminated by storm drain runoff. Epidemiology 10: 355-63. 43. Hammerum, A. M., and L. B. Jensen. 2002. Prevalence of esp, encoding the enterococcal surface protein, in Enterococcus faecalis and Enterococcus faecium isolates from hospital patients, poultry, and pigs in Denmark. J Clin Microbiol 40: 4396. 44. Hancock, D. D., T. E. Besser, D. H. Rice, E. D. Ebel, D. E. Herriott, and L. V. Carpenter. 1998. Multiple sources of Escherichia coli O157 in feedlots and dairy farms in the northwestern USA. Prev Vet Med 35: 11-9. 45. Harwood, V. J., J. Whitlock, and V. Withington. 2000. Classification of antibiotic resistance patterns of indicator bacteria by discriminant analysis: use in predicting the source of fecal contaminati on in subtropical waters. Appl Environ Microbiol 66: 3698-704. 46. Hassan, W. M., S. Y. Wang, and R. D. Ellender. 2005. Methods to increase fidelity of repetitive extragenic palindromic PCR fingerprint-based bacterial source tracking efforts. Appl Environ Microbiol 71: 512-8. 47. Havelaar, A. H., W. M. Pot-Hogeboom, K. Furuse, R. Pot, and M. P. Hormann. 1990. F-specific RNA b acteriophages and sensitive host strains in faeces and wastewater of human and animal origin. J Appl Bacteriol 69: 30-7. 48. Howell, J. M., M. S. Coyne, and P. L. Cornelius. 1996. Effect of sediment particle size and temperature on fecal b acteria mortality rates and the fecal coliform/fecal streptococci ratio. J Environ Qual. 25: 1216-1220. 49. Hsu, F. C., Y. S. Shieh, J. van Duin, M. J. Beekwilder, and M. D. Sobsey. 1995. Genotyping male-specific RNA coliphages by hybridization with oligonucleotide probes. A ppl Environ Microbiol 61: 3960-6. 50. Hughes, J. B., J. J. Hellmann, T. H. Ricketts, and B. J. Bohannan. 2001. Counting the uncountable: statistical appro aches to estimating microbial diversity. Appl Environ Microbiol 67: 4399-406. 51. Jeng, H. C., A. J. England, and H. B. Bradford. 2005. Indicator organisms associated with stormwater suspended particles and estuarine sediment. J Environ Sci Health A Tox Hazard Subst Environ Eng 40: 779-91. 52. Jiang, S., R. Noble, and W. Chu. 2001. Human adenoviruses and coliphages in urban runoff-impacted coastal waters of Southern California. Appl Environ Microbiol 67: 179-84. 53. 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 from different animals influence accuracy of determin ing sources of fecal pollution. Appl Environ Microbiol 70: 4478-85.
48 54. Kay, D., J. M. Fleisher, R. L. Salmon, F. Jones, M. D. Wyer, A. F. Godfree, Z. Zelenauch-Jacquotte, and R. Shore. 1994. Predicting likelihood of gastroenteritis from sea bathing: results from randomized exposure. Lancet 344: 905-9. 55. Kidd, M., J. C. Atherton, A. J. Lastovica, and J. A. Louw. 2001. Clustering of South African Helicobacter pylori isolates from peptic ul cer disease patients is demonstrated by repetitive extragenic palindromic-PCR fingerprinting. J Clin Microbiol 39: 1833-9. 56. Kinzelman, J., S. L. McLellan, A. D. Daniels, S. Cashin, A. Singh, S. Gradus, and R. Bagley. 2004. Non-point source pollution: determination of replication versus persistence of Escherichia coli in surface water and sediments with correlation of levels to readily measurable environmental parameters. J Water Health 2: 103-14. 57. Knowles, W. A., P. Pipkin, N. Andrews, A. Vyse, P. Minor, D. W. Brown, and E. Miller. 2003. Population-based study of antibody to the human polyomaviruses BKV and JCV and the simian polyomavirus SV40. J Med Virol 71: 115-23. 58. Koeuth, T., J. Versalovic, and J. R. Lupski. 1995. Differential subsequence conservation of inte rspersed repetitive Streptococcus pneumoniae BOX elements in diverse bacter ia. Genome Res 5: 408-18. 59. Kogan, S. C., M. Doherty, and J. Gitschier. 1987. An improved method for prenatal diagnosis of genetic diseases by analysis of amplified DNA sequences. Application to hemophilia A. N Engl J Med 317: 985-90. 60. Kuhn, I., G. Allestam, T. A. Stenstrom, and R. Mollby. 1991. Biochemical fingerprinting of water coliform bacter ia, a new method for measuring phenotypic diversity and for comparing different bacterial populati ons. Appl Environ Microbiol 57: 3171-7. 61. Kuhn, I., A. Iversen, L. G. Burman, B. Olsson-Liljequist, A. Franklin, M. Finn, F. Aarestrup, A. M. Seyfarth, A. R. Blanch, X. Vilanova, H. Taylor, J. Caplin, M. A. Moreno, L. Dominguez, I. A. Herrero, and R. Mollby. 2003. Comparison of enterococcal populations in animals, humans, and the environment--a European study. Int J Food Microbiol 88: 133-45. 62. Ley, V., J. Higgins, and R. Fayer. 2002. Bovine enteroviruses as indicators of fecal contamination. Appl Environ Microbiol 68: 3455-61. 63. Lim, D. 1998. Microbiology, p. 562-563, Second ed. WCB/McGraw-Hill, Boston. 64. Madigan, M. T., J.M Martinko and J. Parker. 2000. Brock Biology of Microorganisms, p. 446-450, Ninth ed. Pren tice Hall International, Inc., Upper Saddle River, New Jersey. 65. Malathum, K., K. V. Singh, G. M. Weinstock, and B. E. Murray. 1998. Repetitive sequence-based PCR versus pul sed-field gel electrophoresis for typing of Enterococcus faecalis at the subspecies level. J Clin Microbiol 36: 211-5. 66. Maluquer de Motes, C., P. Clemente-Casares, A. Hundesa, M. Martin, and R. Girones. 2004. Detection of bovine and porci ne adenoviruses for tracing the source of fecal contamination. Appl Environ Microbiol 70: 1448-54.
49 67. Manero, A., X. Vilanova, M. Cerda-Cuellar, and A. R. Blanch. 2002. Characterization of sewage waters by bi ochemical fingerprinting of enterococci. Water Res 36: 2831-5. 68. Martin, B., O. Humbert, M. Camara, E. Guenzi, J. Walker, T. Mitchell, P. Andrew, M. Prudhomme, G. Alloing, and R. Hakenbeck. 1992. A highly conserved repeated DNA element located in the chromosome of Streptococcus pneumoniae Nucleic Acids Res 20: 3479-83. 69. Maugeri, T. L., M. Carbone, M. T. Fera, G. P. Irrera, and C. Gugliandolo. 2004. Distribution of potentially pathogeni c bacteria as free living and plankton associated in a marine coas tal zone. J Appl Microbiol 97: 354-61. 70. Mayer, E. 1999. Systematics and the Origin of Species from the Viewpoint of a Zoologist. Havard University Press, Cambridge. 71. McCaig, A. E., L. A. Glover, and J. I. Prosser. 1999. Molecular analysis of bacterial community structure and dive rsity in unimproved and improved upland grass pastures. Appl Environ Microbiol 65: 1721-30. 72. McLellan, S. L. 2004. Genetic diversity of Escherichia coli isolated from urban rivers and beach water. Appl Environ Microbiol 70: 4658-65. 73. McLellan, S. L., A. D. Daniels, and A. K. Salmore. 2003. Genetic characterization of Escherichia coli populations from host sources of fecal pollution by using DNA fingerprinting. Appl Environ Microbiol 69: 2587-94. 74. McQuaig, S. M., T. M. Scott, J.O. Lukasik, and S.R. Farrah. 2005. Detection of human polyomaviruses in environmen tal waters as an indication of human fecal contamination. Presented at the Am erican Society for Microbiology, 105th General Meeting, Atlanta, GA. 75. Moore, D. F., V. J. Harwood, D. M. Ferguson, J. Lukasik, P. Hannah, M. Getrich, and M. Brownell. 2005. Evaluation of antibioti c resistance analysis and ribotyping for identification of faecal po llution sources in an urban watershed. J Appl Microbiol 99: 618-28. 76. Muller, T., A. Ulrich, E. M. Ott, and M. Muller. 2001. Identification of plantassociated enterococci. J Appl Microbiol 91: 268-78. 77. Myoda, S. P., C. A. Carson, J. J. Fuhrmann, B. K. Hahm, P. G. Hartel, H. Yampara-Lquise, L. Johnson, R. L. Kunt z, C. H. Nakatsu, M. J. Sadowsky, and M. Samadpour. 2003. Comparison of genotypic-based microbial source tracking methods requiring a host or igin database. J Water Health 1: 167-80. 78. Nielsen, E. M., J. Engberg, and M. Madsen. 1997. Distribution of serotypes of Campylobacter jejuni and C. coli from Danish patients, poultry, cattle and swine. FEMS Immunol Med Microbiol 19: 47-56. 79. Noble, R. T., S. M. Allen, A. D. Bl ackwood, 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 different iate between human and non-human fecal contamination in a microbial source tr acking comparison study. J Water Health 1: 195-207. 80. Noble, R. T., I. M. Lee, and K. C. Schiff. 2004. Inactivation of indicator microorganisms from various sources of faecal contamination in seawater and freshwater. J Appl Microbiol 96: 464-72.
50 81. Noble, R. T., S. B. Weisberg, M. K. Leecaster, C. D. McGee, J. H. Dorsey, P. Vainik, and V. Orozco-Borbon. 2003. Storm effects on regional beach water quality along the southern Califor nia shoreline. J Water Health 1: 23-31. 82. Nubel, U., F. Garcia-Pichel, M. Kuhl, and G. Muyzer. 1999. Quantifying microbial diversity: morphotypes, 16S rRNA genes, and carotenoids of oxygenic phototrophs in microbial mats Appl Environ Microbiol 65: 422-30. 83. Osawa, S., K. Furuse, and I. Watanabe. 1981. Distribution of ribonucleic acid coliphages in animals. Appl Environ Microbiol 41: 164-8. 84. Pina, S., M. Puig, F. Lucena, J. Jofre, and R. Girones. 1998. Viral pollution in the environment and in shellfish: human adenovirus detection by PCR as an index of human viruses. A ppl Environ Microbiol 64: 3376-82. 85. Pruss, A. 1998. Review of epidemiological studies on health effects from exposure to recreational water. Int J Epidemiol 27: 1-9. 86. Reeves, R. L., S. B. Grant, R. D. Mrse, C. M. Copil Oancea, B. F. Sanders, and A. B. Boehm. 2004. Scaling and management of fecal indicator bacteria in runoff from a coastal urban watershed in southern California. Environ Sci Technol 38: 2637-48. 87. Rivera, S. C., T. C. Hazen, and G. A. Toranzos. 1988. Isolation of fecal coliforms from pristine sites in a tropi cal rain forest. Appl Environ Microbiol 54: 513-7. 88. Rodgers, P., C. Soulsby, C. Hunter, and J. Petry. 2003. Spatial and temporal bacterial quality of a lowland agricultura l stream in northeast Scotland. Sci Total Environ 314-316: 289-302. 89. Scott, T. M., T. M. Jenkins, J. Lukasik, and J. B. Rose. 2005. Potential use of a host associated molecular marker in Enterococcus faecium as an index of human fecal pollution. Environ Sci Technol 39: 283-7. 90. Seurinck, S., T. Defoirdt, W. Verstraete, and S. D. Siciliano. 2005. Detection and quantification of the human-specif ic HF183 Bacteroides 16S rRNA genetic marker with real-time PCR for asse ssment of human faecal pollution in freshwater. Environ Microbiol 7: 249-59. 91. Shankar, V., A. S. Baghdayan, M. M. Huycke, G. Lindahl, and M. S. Gilmore. 1999. Infection-derived Enterococcus faecalis strains are enriched in esp, a gene encoding a novel surface protein. Infect Immun 67: 193-200. 92. Sherer, B. M., J. R. Miner, J. A. Moore, and J. C. Buckhouse. 1992. Indicator bacterial survival in stream sediments. J Environ Qual. 21: 101-115. 93. Sinton, L. W., C. H. Hall, P. A. Lynch, and R. J. Davies-Colley. 2002. Sunlight inactivation of fecal i ndicator bacteria and bacteriophages from waste stabilization pond effluent in fresh and saline waters. Appl Environ Microbiol 68: 1122-31. 94. Solo-Gabriele, H. M., M. A. Wolfert T. R. Desmarais, and C. J. Palmer. 2000. Sources of Escherichia coli in a coastal subtropical environment. Appl Environ Microbiol 66: 230-7. 95. Solomon, E. P., L.R. Berg, D.W. Martin, and C. Villee. 1996. Biology, Fourth ed. Saunders College Publishing, Fort Worth.
51 96. Spratt, B. G., and M. C. Maiden. 1999. Bacterial population genetics, evolution and epidemiology. Philos Trans R Soc Lond B Biol Sci 354: 701-10. 97. Stoeckel, D. M., M. V. Mathes, K. E. Hyer, C. Hagedorn, H. Kator, J. Lukasik, T. L. O'Brien, T. W. Fenger, M. Samadpour, K. M. Strickler, and B. A. Wiggins. 2004. Comparison of seven protocols to identify fecal contamination sources using Escherichia coli Environ Sci Technol 38: 6109-17. 98. USEPA. 1983. Health Effects Criteria for Marine Recreational Waters EPA600/1-80-031. US Environmental Protection Agency. 99. USEPA. 2002. Implementation Guidance for Ambient Water Quality Criteria for Bacteria EPA-823-B-02-003. US Environmental Protection Agency. 100. USEPA. 2000. Improved Enumeration Methods for the Recreational Water Quality Indicators: Enterococci and Escherichia coli EPA/821/R-97/004. US Environmental Protection Agency. 101. USEPA. 2005. Microbial Source Tracking Guide Document EPA/600-R-05-064. US Environmental Protection Agency. 102. van Belkum, A., M. Sluijuter, R. de Groot, H. Verbrugh, and P. W. Hermans. 1996. Novel BOX repeat PCR assa y for high-resolution typing of Streptococcus pneumoniae strains. J Clin Microbiol 34: 1176-9. 103. Versalovic, J., T. Koeuth, and J. R. Lupski. 1991. Distribution of repetitive DNA sequences in eubacteria and applic ation to fingerprint ing of bacterial genomes. Nucleic Acids Res 19: 6823-31. 104. Vilanova, X., A. Manero, M. Cerda-Cuellar, and A. R. Blanch. 2004. The composition and persistence of faecal colif orms and enterococcal populations in sewage treatment plants. J Appl Microbiol 96: 279-88. 105. Vilanova, X., A. Manero, M. Cerda-Cuellar, and A. R. Blanch. 2002. The effect of a sewage treatment plant effluent on the faecal coliforms and enterococci populations of the reception rive r waters. J Appl Microbiol 92: 210-4. 106. 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: 27-32. 107. Whitlock, J. E., D. T. Jones, and V. J. Harwood. 2002. Identification of the sources of fecal coliforms in an urba n watershed using antibiotic resistance analysis. Water Res 36: 4273-82. 108. 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 wate r and beach sand of Lake Michigan. Appl Environ Microbiol 69: 4714-9. 109. Wiggins, B. A., R. W. Andrews, R. A. Co nway, C. L. Corr, E. J. Dobratz, D. P. Dougherty, J. R. Eppard, S. R. Knupp, M. C. Limjoco, J. M. Mettenburg, J. M. Rinehardt, J. Sonsino, R. L. Torrijos, and M. E. Zimmerman. 1999. Use of antibiotic resistance analysis to identify nonpoint sources of fecal pollution. Appl Environ Microbiol 65: 3483-6. 110. Woese, C. R. 1987. Bacterial evolu tion. Microbiol Rev 51: 221-71. 111. Wolf, H. W. 1972. The Coliform Count as a Measure of Water Quality. In R. Mitchell (ed.), Water Pollution Microb iology. Wiley-Interscience, New York.