Update on Microbial Source Tracking from the ASM General Meeting in Orlando

            This year’s General Meeting of the American Society for Microbiology (ASM) featured two poster sessions devoted to “Indicators of Fecal Pollution”.  These sessions provided an excellent opportunity to get together with others in the field and to discuss recent developments and future directions.

            A number of different techniques were presented among the posters.  These included library-dependent and independent methods, bacterial, viral and chemical detection technologies, and targeted sampling strategies.  This diversity mirrored the many approaches to the problem of identifying the sources of fecal contamination and the wide range of goals in tracking microbial sources.

            One thing that was clear from the session was that the use of quantitative PCR (qPCR) coupled to host-specific primers has become increasingly popular.  Two of the specific approaches that were well represented in posters from both EPA and University labs were the use of the enterococcal surface protein (esp) and host-specific bacteroidetes probes.  I have included a brief summary of each approach along with a list of relevant posters.

Enterococcal Surface Protein

            Enterococcus faecium strains that contain a variant gene for esp have been shown to be unique to human fecal contamination.  The methods being developed by R. Haugland et al. at the USEPA (Q-205) involve using qPCR to determine the amount of human-specific contamination in recreational waters.  The same approach was used on posters by other researchers from the University of Georgia (Q-214), the University of Hawaii (Q-204) and the Massachusetts DEP (Q-219).  Haugland has compiled an impressive amount of data supporting this method and it is my impression that it is well on its way to becoming a “Standard Method” for recreational water monitoring.  One limitation of this method is that the target being used can only distinguish between human and non-human source contamination.  In order to distinguish between different non-human sources, other targets will need to be developed.

Selected Posters:

Q-204 Evaluation of the Enterococcal Surface Protein (esp) as an Alternative Indicator of Sewage Contamination of Hawaii’s Streams and Coastal Waters.

W. Q. Betancourt, R. S. Fujioka;
Univ. of Hawaii, Honolulu, HI.

Q- 205 Comparison of Real-Time PCR Fecal Bacteria Measurements in Recreational Waters Using Different Instruments and Reagent Systems.

R. Haugland, S. Siefring, M. Varma, E. Atikovic, L. Wymer;
U.S. Environmental Protection Agency, Cincinnati, OH.

Q-214 Identifying Sources of Fecal Contamination Inexpensively with Targeted Sampling and Bacterial Source Tracking.

J. L. McDonald, P. G. Hartel, L. C. Gentit, C. N. Belcher, K. W. Gates, K. Rodgers, J. A. Fisher, K. A. Smith, K. A. Payne;
Univ. of Georgia, Athens, GA.

Q-219 Assessment of Sewage Pollution in Massachusetts Rivers and Beaches Using a Sewage-Specific Marker PCR Assay Targeting a Putative Virulence Factor (esp Gene) in Enterococcus faecium.

R. J. Tang1, M. C. Waldron2, R. F. Breault2, P. K. Weiskel2, R. E. Stoner1, M. DiBara3, W. Dunn3, C. Duerring3, J. Beskenis3, R. F. Chase3, P. DiPietro4, T. Callaghan5, M. Celona6, D. J. Gray7, O. C. Pancorbo1;
1Massachusetts Department of Environmental Protection, Lawrence, MA, 2U.S. Geological Survey, Northborough, MA, 3Massachusetts Department of Environmental Protection, Worcester, MA, 4Massachusetts Department of Conservation and Recreation, Boston, MA, 5Massachusetts Office of Coastal Zone Management, Boston, MA, 6Massachusetts Department of Public Health, Boston, MA, 7U.S. Environmental Protection Agency, Boston, MA.

Bacteroidetes

            Bacteroidetes are a group of obligately anaerobic organisms that represent the dominant bacteria of the large intestine.  Their abundance in feces makes them an excellent indicator organism for direct detection by molecular methods.  In addition, there is a substantial amount of genetic diversity among different members of the group.  This diversity has allowed researchers to identify host-specific markers that can be used to distinguish between different sources of fecal contamination.  There were several new developments in identifying host-specific targets for bacteroidetes and applying these targets to study contamination on a watershed scale reported on at the conference.

            One of the most significant developments was a poster by Orin Shanks et al. (Q-226) that reported on an efficient method for identifying new host-specific markers.  The specific target identified in this study was for cattle-source contamination which expands on the previous targets of human and general ruminant.  Two other posters (Q-215, Q218), both with US-EPA participants, used host-specific bacteroidetes markers to track fecal pollution in two different watersheds.  The level of detail in identifying sources by using these markers along with comprehensive sampling plans was quite impressive. 

Selected Posters:

Q-215 Tracking Fecal Contamination with Bacteroidales Molecular Markers: An Analysis of the Dynamics of Fecal Contamination in the Tillamook Basin, Oregon.

O. C. Shanks1, C. Nietch1, M. Simonich2, D. Reynolds3, K. G. Field2;
1US EPA, Cincinnati, OH, 2Oregon State Univ., Corvallis, OR, 3Tillamook County Performance Partnership, Garibaldi, OR.

Q-218 Assessment of Fecal Pollution Sources in Plum Creek Watershed Using Bacteroidetes 16S rDNA-Based Assays.

R. Lamendella1, J. W. Santo Domingo2, D. Oerther1, J. Vogel3, D. Stoeckel4;
1Univ. of Cincinnati, Cincinnati, OH, 2Environmental Protection Agency, Cincinnati, OH, 3U.S. Geological Survey, Lincoln, NE, 4U.S. Geological Survey, Columbus, OH.

Q-226 Competitive Metagenomic DNA Hybridization Identifies Host-Specific Genetic Markers in Cattle Fecal Samples.

O. C. Shanks1, J. Santo Domingo1, R. Lamendella1, C. A. Kelty1, J. E. Graham2;
1US EPA, Cincinnati, OH, 2Univ. of Louisville, Louisville, KY.

Poster Abstracts:

Q-204. Evaluation of the Enterococcal Surface Protein (esp) as an Alternative Indicator of Sewage Contamination of Hawaii’s Streams and Coastal Waters

W. Q. Betancourt, R. S. Fujioka;
University of Hawaii, Honolulu, HI.

Enterococci are routinely evaluated in recreational waters in order to determine the occurrence of fecal pollution and possible presence of enteric pathogens. In tropical areas, the occurrence of environmentally-adapted types of these bacteria can inaccurately indicate risk of fecal contamination. Hence, alternative indicators of fecal contamination are being investigated in order to make better determinations of the sanitary quality of tropical recreational waters. The adhesin-type enterococcal surface protein (esp) identified in clinical enterococcal isolates has been used as an indicator of human fecal pollution in the environment. The objective of this study was to evaluate the occurrence of the esp gene as a marker of fecal pollution in enterococcal isolates from sewage effluents and culturable enterococci obtained from an ocean sewage outfall and stream and coastal waters not known to be contaminated with sewage. DNA obtained from single isolates and partially enriched membrane filters along with standard PCR were used for detection of the esp gene using primers and cycling conditions described by Scott et al. 2005 and Shankar et al. 1999. The results revealed that when using both molecular methods ten of ten (100%) of sewage outfall sites and two of ten (20%) stream samples were positive for the esp marker, while five of the coastal waters analyzed were devoid of the marker. Nineteen enterococcal isolates from sewage effluents were identified as E. faecalis by PCR and thirteen (68%) of these isolates carried the esp gene using the primers of Shankar et al. which target a fragment of the gene present in both E. faecalis and E. faecium. None of the single isolates gave a positive PCR signal with the primers of Scott et al. designed to amplify a fragment of the gene in E. faecium. The method required a minimum of 20 enterococci CFU for PCR detection. The results of this investigation demonstrate that further analysis of the esp marker in enterococci recovered from water can be use to confirm the presence of sewage (point source contamination) and to differentiate between samples contaminated by non point source (non-sewage) contamination.

Q-205. Comparison of Real-Time PCR Fecal Bacteria Measurements in Recreational Waters Using Different Instruments and Reagent Systems

R. Haugland, S. Siefring, M. Varma, E. Atikovic, L. Wymer;
U.S. Environmental Protection Agency, Cincinnati, OH.

U.S. EPA guidance on the safety of surface waters for recreational use is currently based on concentrations of culturable fecal indicator bacteria. Attention is now shifting to more rapid molecular monitoring methods. A multi-year epidemiological study is in progress to determine the relationship between illness rates in bathers and concentrations of fecal bacteria in recreational water as determined by real-time PCR analysis. Analyses in the first 2 years of the study were limited to one PCR reagent and type of instrument. Acceptance of this technology will be aided by the availability of choices in instruments and newer PCR reagents that offer even shorter analysis times. DNA extracts of 50 ml beach water filtrates (N = 396) collected from Biloxi, MS during the 2005 study were analyzed on 3 instruments: Applied Biosystems model 7700 (analysis time ~ 2 hr); Cepheid Smart Cycler (analysis time ~ 30 min with TaqMan probe & ~ 45 min with Scorpion probe); and Applied Biosystems fast block model 7900 (analysis time ~ 35 min), using customized PCR reagents and primer/probe sets for each instrument. Mean Enterococcus and Bacteroidetes calibrator cell equivalents/filtrate determined from all analyses were 37 and 388 (CV between instrument means: 0.26 and 0.24), based on an amplification efficiency of 0.92 for both assays. Significant differences (P < 0.05) were found in the means for both groups of organisms between each of the systems with exception of the Bacteroidetes results on the models 7700 and 7900. Normalization of results using reference control analyses made these differences non-significant in all comparisons except those involving the model 7700. Failure to show comparability between model 7700 and other systems results may be related to the use of a different reference assay. Variance in replicate analyses was highest on the model 7700; this system gave a significantly lower percentage of non-detects. Our results suggest that different instrument/reagent systems can give comparable results but further analyses are needed to demonstrate relationships between epidemiological results and fecal bacteria measurements with the newer systems.
Notice: Although this work was reviewed by EPA and approved for publication, it may not necessarily reflect official Agency policy.

Q-214. Identifying Sources of Fecal Contamination Inexpensively with Targeted Sampling and Bacterial Source Tracking

J. L. McDonald, P. G. Hartel, L. C. Gentit, C. N. Belcher, K. W. Gates, K. Rodgers, J. A. Fisher, K. A. Smith, K. A. Payne;
University of Georgia, Athens, GA.

Most bacterial source tracking (BST) methods are too expensive for most communities to afford. We developed targeted sampling as a prelude to BST to reduce these costs. We combined targeted sampling with three inexpensive BST methods, Enterococcus speciation, detection of the esp gene, and fluorometry, to confirm the sources of fecal contamination to beaches on Georgia's Jekyll and Sea Islands during calm and stormy weather conditions. For Jekyll Island, the most likely source of contamination was bird feces because the percentage of Ent. faecalis was high (30%) and the esp gene was not detected. For the Sea Island beach during calm conditions, the most likely sources of fecal contamination were leaking sewer lines and wildlife feces. The leaking sewer lines were confirmed with fluorometry and detection of the esp gene. For the Sea Island beach during stormflow conditions, the most likely sources of fecal contamination were wildlife feces and runoff discharging from two county-maintained pipes. For the pipes, the most likely source of contamination was bird feces because the percentage of Ent. faecalis was high (30%) and the esp gene was not detected. Sediments were also a reservoir of fecal enterococci for both Jekyll and Sea Islands. Combining targeted sampling with two or more BST methods identified sources of fecal contamination quickly, easily, and inexpensively. This combination was the first time targeted sampling was conducted during stormy conditions, and the first time targeted sampling was combined with enterococcal speciation, detection of the esp gene, and fluorometry.

Q-215. Tracking Fecal Contamination with Bacteroidales Molecular Markers: An Analysis of the Dynamics of Fecal Contamination in the Tillamook Basin, Oregon

O. C. Shanks1, C. Nietch1, M. Simonich2, D. Reynolds3, K. G. Field2;
1US EPA, Cincinnati, OH, 2Oregon State Univ, Corvallis, OR, 3Tillamook County Performance Partnership, Garibaldi, OR.

Background: Although PCR of source-specific molecular markers from Bacteroidales fecal bacteria identifies several sources of fecal contamination in water, it is unclear how this technique relates to fecal indicator measurements in natural waters. Objectives of this study were to elucidate spatial and temporal dynamics of source-specific Bacteroidales 16S rRNA markers across a watershed; to compare these to fecal indicator counts, measurements of water quality, and climatic forces; and to identify sites of intense exposure to specific sources of contamination. Methods: Biweekly sampling occurred over a two-year period in the Tillamook basin situated on the north central coast of Oregon, at 30 sites located along five river tributaries and in the Tillamook Bay. We performed Bacteroidales PCR assays with ruminant and human source-specific primers for fecal source identification. We measured E. coli MPN, temperature, turbidity, and 5-day precipitation. Results and Conclusions: Fecal sources were statistically more linked to ruminants than humans; there was a 40% greater probability of detecting a ruminant source marker than a human one across the basin. At individual sites, fecal source tracking data linked elevated fecal indicator bacteria loads to specific point and non-point sources of fecal pollution in the basin. In upstream sites, ruminant contamination was highest in fall, lower in winter, and lowest in summer. This pattern suggests non-point sources, and fits rainfall-runoff models, in which pollutants build up on the landscape between rain events, and are washed off during subsequent rain events. At sites associated with point sources, however, both ruminant and human marker probabilities were positively correlated to rainfall alone. Inconsistencies in E. coli and host-specific marker trends in the rivers and estuary suggested that factors that control the quantity of fecal indicators in the water column are different than those influencing presence of Bacteroidales markers at specific times of the year. This is important if fecal indicator counts are used as a criterion for source loading potential in receiving waters.

Q-218. Assessment of Fecal Pollution Sources in Plum Creek Watershed Using Bacteroidetes 16S rDNA-Based Assays

R. Lamendella1, J. W. Santo Domingo2, D. Oerther1, J. Vogel3, D. Stoeckel4;
1University of Cincinnati, Cincinnati, OH, 2Environmental Protection Agency, Cincinnati, OH, 3U.S. Geological Survey, Lincoln, NE, 4U.S. Geological Survey, Columbus, OH.

Recently, 16S rDNA Bacteroidetes-targeted PCR assays were developed to discriminate between ruminant and human fecal pollution. These assays are rapid and relatively inexpensive but have been used in a limited number of studies. In this study, we evaluated the efficacy of human and ruminant-specific 16S rDNA Bacteroidetes assays in determining the primary sources of fecal pollution in Plum Creek (NE) watershed. The sensitivity and specificity, as well as the temporal and spatial application of these assays were challenged against feces from different animals, water, and sediment samples from this watershed. Phylogenetic analyses of 981 fecal and environmental 16S rDNA clones were also performed to study the diversity of Bacteroidetes in this watershed. On average, the host specific assays indicated that ruminant feces were present in more than one-third of the water samples and in all sampling seasons, with increasing frequency in downstream sites along Plum Creek. The human-targeted assay indicated that only 5% of the water samples were positive for human fecal signals, although a higher percentage of human-associated signals (24%) were detected in sediment samples. Phylogenetic analysis of Bacteroidetes 16S rDNA sequences derived from fecal and environmental samples demonstrated the presence of a high level of sequence diversity, with nearly half of the recovered sequences showing less than 97% identity to sequences found in publicly available databases. Approximately, 60% of all clones clustered with yet to be cultured Bacteroidetes species associated with sequences obtained from ruminant feces, further supporting the prevalence of ruminant contamination in this watershed. The sequencing data also indicated that other sources, including humans, sediments, and wildlife may be significantly contributing to the molecular diversity of fecal Bacteroidetes in this watershed. Since several clusters contained sequences from multiple sources, future studies must take into consideration the potential cosmopolitan nature of some of latter bacterial populations when assessing fecal pollution sources using Bacteroidetes markers.

Q-219. Assessment of Sewage Pollution in Massachusetts Rivers and Beaches Using a Sewage-Specific Marker PCR Assay Targeting a Putative Virulence Factor (esp Gene) in Enterococcus faecium

R. J. Tang1, M. C. Waldron2, R. F. Breault2, P. K. Weiskel2, R. E. Stoner1, M. DiBara3, W. Dunn3, C. Duerring3, J. Beskenis3, R. F. Chase3, P. DiPietro4, T. Callaghan5, M. Celona6, D. J. Gray7, O. C. Pancorbo1;
1Massachusetts Department of Environmental Protection, Lawrence, MA, 2U.S. Geological Survey, Northborough, MA, 3Massachusetts Department of Environmental Protection, Worcester, MA, 4Massachusetts Department of Conservation and Recreation, Boston, MA, 5Massachusetts Office of Coastal Zone Management, Boston, MA, 6Massachusetts Department of Public Health, Boston, MA, 7U.S. Environmental Protection Agency, Boston, MA.

Fecal contamination, as demonstrated by elevated concentrations of fecal indicator bacteria in the water column, is a leading cause of watershed impairment in Massachusetts. Confirmation of the source of fecal pollution in a watershed segment as human (i.e., illicit sewage source) is the logical first step in the development of an appropriate corrective action plan. Using MA raw municipal sewage and individual human/animal fecal samples, our laboratory demonstrated that the human-specific Enterococcus faecium PCR assay targeting the putative virulent esp gene is highly specific for sewage samples (i.e., generally positive for sewage and negative for individual human/animal fecal samples). We subsequently applied this assay to both dry and wet weather samples collected in 2005 from MA fresh and saline waters without permitted sewage discharges, and detected the sewage marker in an average of 14% of samples from the lower Charles River (3 of 14 samples), Shawsheen River (13 of 104), Little River (2 of 17), Wollaston Beach/Quincy Bay (6 of 36), and Sandy Beach/Upper Mystic Lake (1 of 9). The marker was not detected in 18 and 23 samples from Carson Beach/Boston Harbor and Salem Beach, respectively. The marker was detected in samples with total enterococci concentrations that met as well as in samples that exceeded the corresponding U.S. EPA recreational water quality criterion. Also, the marker was more frequently detected in wet weather than in dry weather samples from the Shawsheen and Little River, Wollaston Beach, and Sandy Beach. In dry weather samples with high levels of fecal indicator bacteria taken from a Charles River tributary thought to be impacted by raw sewage, we detected the esp marker along with four sewage-specific chemical indicators - fluorescent whitening agent # 1, acetaminophen, caffeine, and 1,7-dimethylxanthine. Neither the esp marker nor any of the sewage-specific chemicals were detected at an upstream control station with no known sewage sources. Along with selected sewage-specific chemical indicators, the E. faecium sewage marker PCR assay is a promising tool for identifying illicit sewage sources in watersheds.

Q-226. Competitive Metagenomic DNA Hybridization Identifies Host-Specific Genetic Markers in Cattle Fecal Samples

O. C. Shanks1, J. Santo Domingo1, R. Lamendella1, C. A. Kelty1, J. E. Graham2;
1US EPA, Cincinnati, OH, 2University of Louisville, Louisville, KY.

Several PCR methods have recently been developed to identify sources of fecal contamination in surface waters. In all cases, researchers have relied on one gene or one microorganism for selection of host specific markers. Here, we describe the application of a genome fragment enrichment (GFE) method to identify host-specific genetic markers from fecal microbial community DNA. As a proof of concept, cattle fecal DNA was challenged against a pig fecal DNA background to select for cattle specific DNA sequences. Bioinformatic analyses of 380 cattle enriched metagenomic sequences indicated a preponderance of Bacteroidales-like regions predicted to encode membrane-associated and secreted proteins. Oligonucleotide primers capable of annealing to select Bacteroidales-like cattle GFE sequences exhibited extremely high specificity (> 99%) in PCR assays with total fecal DNA from 279 different animal sources. These primers also demonstrated a broad distribution of corresponding genetic markers (81% positive) among 148 different cattle sources. These data demonstrate that direct metagenomic DNA analysis using the competitive solution hybridization approach described is an efficient method for identifying potentially useful host specific fecal genetic markers, and for characterizing differences between environmental microbial communities.