Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Bower, P. A. Articles by McLellan, S. L. Search for Related Content PubMed PubMed Citation Articles by Bower, P. A. Articles by McLellan, S. L. Agricola Articles by Bower, P. A. Articles by McLellan, S. L.

 Previous Article  |  Next Article 

Applied and Environmental Microbiology, December 2005, p. 8305-8313, Vol. 71, No. 12
0099-2240/05/$08.00+0     doi:10.1128/AEM.71.12.8305-8313.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.

Detection of Genetic Markers of Fecal Indicator Bacteria in Lake Michigan and Determination of Their Relationship to Escherichia coli Densities Using Standard Microbiological Methods

Patricia A. Bower, Caitlin O. Scopel, Erika T. Jensen, Morgan M. Depas, and Sandra L. McLellan^*

Great Lakes WATER Institute, University of Wisconsin—Milwaukee, Milwaukee, Wisconsin 53204

Received 13 February 2005/ Accepted 11 August 2005

Top Abstract Introduction Materials and Methods Results Discussion References

 
Lake Michigan surface waters impacted by fecal pollution were assessed to determine the occurrence of genetic markers for Bacteroides and Escherichia coli. Initial experiments with sewage treatment plant influent demonstrated that total Bacteroides spp. could be detected by PCR in a 25- to 125-fold-higher dilution series than E. coli and human-specific Bacteroides spp., which were both found in similar dilution ranges. The limit of detection for the human-specific genetic marker ranged from 0.2 CFU/100 ml to 82 CFU/100 ml culturable E. coli for four wastewater treatment plants in urban and rural areas. The spatial and temporal distributions of these markers were assessed following major rain events that introduced urban storm water, agricultural runoff, and sewage overflows into Lake Michigan. Bacteroides spp. were detected in all of these samples by PCR, including those with <1 CFU/100 ml E. coli. Human-specific Bacteroides spp. were detected as far as 2 km into Lake Michigan during sewage overflow events, with variable detection 1 to 9 days postoverflow, whereas the cow-specific Bacteroides spp. were detected in only highly contaminated samples near the river outflow. Lake Michigan beaches were also assessed throughout the summer season for the same markers. Bacteroides spp. were detected in all beach samples, including 28 of the 74 samples that did not exceed 235 CFU/100 ml of E. coli. Human-specific Bacteroides spp. were detected at three of the seven beaches; one of the sites demonstrating positive results was sampled during a reported sewage overflow, but E. coli levels were below 235 CFU/100 ml. This study demonstrates the usefulness of non-culture-based microbial-source tracking approaches and the prevalence of these genetic markers in the Great Lakes, including freshwater coastal beaches.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Beach closings are a growing concern in both freshwater and marine coastal regions (14, 25). There is increasing awareness of the extent to which coastal recreational waters are impacted by fecal pollution, as beaches are monitored more frequently since the Federal BEACH Act was passed in 2000. Water quality advisories or closures occur when levels of fecal indicator organisms exceed standards set by individual state authorities (37). Since indicator bacteria, such as enterococci and Escherichia coli, are present in high abundance in the gastrointestinal tracts of most warm-blooded animals, the detection of these organisms implies the presence of fecal contamination (35, 36). While there is a correlation between indicator organisms and swimming-related illness (9, 27, 40), reliance on the indicator organisms may not be universally protective, as pathogens have been detected when culturable indicators do not exceed recommended standards (16, 19, 24).

Identifying sources of fecal pollution is important for assessing the public heath risk and devising management strategies (13). Human sources of fecal pollution constitute a serious health risk because of the high likelihood of the presence of human pathogens (21, 39). Zoonotic reservoirs (including cattle, swine, and chickens) of pathogens are also of high concern (24). There are a wide variety of microbial-source tracking approaches under development, and a single approach may not be adequate, since all these methods have limitations (30, 31, 34). Culture-independent, molecular methods detect genetic targets of organisms found in a specific host. These methods include detection of human enteroviruses or adenoviruses (16, 22), host-specific species of Bacteroides (1), or virulence genes in E. coli (6). Several of these approaches were shown to be useful in detecting human sources of fecal contamination but did not necessarily identify all of the contributing hosts when multiple sources were present (11).

Fecal anaerobes, such as Bacteroides and Bifidobacterium, have long been suggested as alternative indicators to the fecal-coliform group (5, 12). Fecal anaerobes make up the majority of fecal bacteria in the gastrointestinal tract of humans and may be present at 1,000-fold-higher densities than the fecal-coliform group, making these organisms highly sensitive indicators of fecal pollution (12, 15). The advent of molecular methods has made it more feasible to detect these organisms in contaminated waters (10). Culture-independent methods may more readily detect fecal indicator organisms than standard culture methods, since only intact cells are required, rather than cultivatable cells that need to be recovered on selective media. Recovery of organisms by either culture-based or molecular methods may be affected by survival characteristics of individual organisms, which have been cited as a complicating factor in utilizing indicators of fecal pollution (3, 5, 29, 33). Certain species of fecal anaerobes have been identified as host specific, and PCR assays based on the 16S rRNA genes of certain species, or closely related groups of species, allow the simultaneous detection of indicators and the source of the indicators (2).

While progress has been made in identifying genetic markers that are useful for microbial-source tracking, few studies have evaluated how these markers vary temporally and spatially following contamination of surface waters (25), and even fewer studies have assessed their occurrence at beach sites (4, 23). We assessed the prevalence of E. coli and total and human- and cow-specific Bacteroides genetic markers in surface waters of Lake Michigan impacted by multiple pollution sources following major storm events. We also assessed nine freshwater beaches that encompass the western Lake Michigan shoreline, including a highly urbanized area, a rural community surrounded by agricultural land, and a resort/vacation area. All the beach sites are potentially impacted by different combinations of pollution sources, such as roosting waterfowl and shore birds, stormwater outfalls, river discharge carrying pollutants from the watershed, and sewage overflows. Detection of total, human-specific, and cow-specific Bacteroides spp. found in surface waters was linked to the detection of E. coli by PCR and the abundance of E. coli measured by culture-based methods.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Limits of detection and variability of fecal indicator markers in sewage.
The limits of detection and variability of four markers, the E. coli uidA gene, the Bacteroides sp. 16S rRNA gene, the cow-specific Bacteroides 16S rRNA gene, and the human-specific Bacteroides 16S rRNA gene were evaluated by PCR using influent from four different sewage treatment plants: Jones Island (Milwaukee), which services suburban and urban industrial and residential areas; South Shore (Milwaukee), which services primarily urban residential areas; one plant in Manitowoc County, which services a smaller urban area surrounded by agricultural land use; and one plant located in Door County, which services a small municipality in a resort/vacation area (Table 1). The samples were collected over a 9-month period. Each sample from the Jones Island and South Shore treatment plants was flow weighted over a 24-h period, providing good representation of the fecal indicator organism load entering the plant. For DNA analysis, the sewage samples were diluted 10-fold in sterile distilled water and filtered on a 0.45-µm nitrocellulose filter (47-mm nitrocellulose; Millipore, Bedford, MA). The filters were frozen at –80°C for storage until the DNA was extracted (see below). The isolated DNA was then serially diluted in fivefold steps to 10^–6 final dilution for analysis by PCR. At least two or three PCRs were conducted on each diluted DNA sample to determine the PCR variability for each dilution. Aliquots of each of the sewage influent samples were diluted 10^–4 to 10^–6 in sterile water to obtain E. coli cell counts.

View this table: [in this window] [in a new window]

TABLE 1. Relative abundances of E. coli, Bacteroides spp., and human-specific Bacteroides genetic markers in sewage treatment plant influent

Cow feedlot samples.
Seventy-five samples were collected from 15 dairy farms within the Lake Michigan drainage basin. Samples were obtained by collecting 5 to 10 g fecal material from feedlot detention basins in sterile 50-ml centrifuge tubes. Approximately three to five samples were collected from each site to obtain a representation of the animals on each farm. The material was transported on ice, and DNA was extracted from 200 mg of fecal material using a QIAamp stool DNA isolation kit (QIAGEN, Valencia, CA).

Nearshore Lake Michigan surface water samples.
A variety of environmental freshwater samples were collected from May 2004 until August 2004 in southern Lake Michigan, corresponding to latitude-longitude coordinates of 43.01°N, 87.55°W. The nearshore area surrounding the Milwaukee harbor was chosen for study, since the harbor is the primary discharge point for an 850-square-mile basin that includes agricultural, suburban, and urban land use, with no other major watershed inputs entering Lake Michigan within 10 km. The study area extended from sites immediately above the confluence of three major rivers to the channel leading to the harbor, the Milwaukee harbor, and outside of the harbor up to 8 km distance (Fig. 1). Nearshore transects were carried out 2 km from shore unless otherwise specified. In addition, sampling included sites 0.5 km off of two urban beaches in close proximity to the harbor, which were potentially the most susceptible to sewage overflows and agricultural runoff carried by river water.

View larger version (150K): [in this window] [in a new window]

FIG. 1. Aerial view of the study area. Sample regions included the Kinnickkinnic River, the channel, the Milwaukee harbor between the channel and the main gap in the breakwall, and nearshore Lake Michigan sites 4 km north and 8 km south of the harbor. The Milwaukee harbor is the discharge point of 850 square miles of watershed drainage from the Milwaukee River basin. Approximately 45 combined sewer outfalls are located on the lower reaches of the Milwaukee, Menomonee, and Kinnickinnic Rivers; CSO discharge from these outfalls mixes with river water that is potentially contaminated from upstream sources in the basin. (Aerial photograph provided by the Wisconsin Department of Natural Resources.)

Water samples were collected over a 90-day period following a series of major rain events totaling greater than 10 in. of precipitation in some areas of the watershed. During periods of the heaviest rain, the capacity of the combined sewer system was exceeded, resulting in a combined sewer overflow (CSO) that released a mixture of sanitary sewage and storm water from approximately 45 outfalls located on the three rivers within 0.5 to 3.0 km of the harbor. Only two CSO outfalls are located on Lake Michigan itself, both on the south end of the harbor, 1 km and 0.5 km from a beach site, respectively (beach site 2) (Fig. 1). Sanitary sewage overflows (SSO) also occurred during this time in the upper reaches of the rivers in areas that are serviced by a separated sewer system. SSO samples were collected immediately below outfalls at three sites within the Milwaukee River approximately 5 to 10 km from the harbor. Spatial and temporal surveys of the nearshore sites were conducted during CSO/SSO events (three CSOs and five SSOs), which occurred intermittently during a 15-day period in May 2004. Eighty nearshore samples were collected on 14 May 2004 after >4 in. of rainfall within 24 h; 32 of these were analyzed for genetic markers based on E. coli densities of >1,000 CFU/ml, along with samples that did not meet this criterion but that were collected within 0.5 km of beach site 2 (n = 5). Samples collected offshore of beach site 1, which may also be impacted by CSO contamination, were not analyzed for this date, since E. coli levels did not exceed 20 CFU/100 ml. Subsequent surveys were also carried out 1 to 3 days post-CSO, 7 days post-CSO, 9 days post-CSO, 23 days post-CSO, and >60 days post-CSO. These post-SSO/CSO surveys targeted four main regions—the channel, the harbor, Lake Michigan 1 to 2 km from the harbor, and Lake Michigan 3 to 8 km from the harbor; one to three samples were collected per region, and all samples were analyzed for genetic markers regardless of E. coli levels.

Lake Michigan beach water samples.
Nine coastal beaches spanning the Western shore of Lake Michigan were sampled from June to September 2004. The sites included three urban beaches in Milwaukee County, three beaches in Manitowoc County (one beach adjacent to agricultural land, one beach near a small municipality within the agricultural watershed, and the third adjacent to forested land), and three beaches in Door County, which is primarily a resort/vacation area. These sites are potentially influenced by localized sources, primarily storm water and shorebirds, as well as regional contamination, including urban storm water, agricultural runoff, and sewage overflows when they occur. At beaches with greater than 0.5 km of shoreline, two or three evenly spaced samples were taken; in all, 74 water samples were collected at the nine beach sites, including samples collected from outfalls or impervious surfaces that discharged directly to the beach.

E. coli enumeration.
Each environmental water sample was filtered through a 0.45-µm-pore-size 47-mm nitrocellulose filter and placed on m-TEC (Difco, Sparkes, MD) or modified m-TEC (Difco, Sparkes, MD) agar according to the U.S. Environmental Protection Agency (EPA) method for E. coli enumeration (36). The volumes filtered varied according to expected contamination, where duplicates of 1-ml and 10-ml volumes were analyzed for river and selected beach samples, and duplicates of 10-ml and 100-ml volumes were analyzed for harbor, nearshore Lake Michigan, and the remaining beach samples. The plates were incubated at 37°C for 1 hour and then at 44.5°C for 18 h.

DNA extraction.
Overall, water sample volumes that were filtered for DNA extraction ranged from 200 to 300 ml for CSO samples and 300 to 500 ml for beach samples, with the exception of SSO samples collected below the outfall, where 50 ml was used; all samples were filtered onto a 0.45-µm nitrocellulose filter and frozen at –80°C for DNA extractions. The frozen filters were broken into small fragments with a metal spatula. The DNA was extracted using the Qbiogene FastDNA Spin Kit for Soil (Qbiogene, Inc., Irvine, CA) as specified in the manual, except that the cells were mechanically lysed using the MiniBeadBeater-8 Cell Disruptor (BioSpec Products, Bartlesville, OK) on the homogenization setting for 1.5 min at room temperature. The DNA was eluted in 75 or 100 µl of sterile distilled water. The number of culturable E. coli cells per filter was calculated from the cell counts on m-TEC or modified m-TEC medium. The DNA concentration was determined using the NanoDrop ND-1000 Spectrophotometer (NanaDrop Technologies, Wilmington, DE).

Primers and PCR analysis of genetic markers.
Bacteroides species (Bac32F/708R), Bacteroides human-specific (HF183F/708R), and Bacteroides cow-specific (CF128F/708R) primer sequences (Table 2) were based on the 16S rRNA gene as described previously (2). The Bacteroides cow-specific (CF193F/708R) primer was also used in reactions with feedlot manure samples but was not utilized for analysis of environmental samples due to variability and nonspecific amplification products with this reaction. E. coli-specific primers (uidA1318F/1698R) were designed to target the ß-glucuronidase gene using the uidA sequence of strain K-12 (NC 000913). All primers were synthesized by Sigma Genosys (The Woodlands, TX).

View this table: [in this window] [in a new window]

TABLE 2. PCR primers for amplification of fecal indicator bacterial genetic markers

PCR analysis was conducted with either 1 or 2 µl of undiluted DNA samples or samples diluted to a range of 10 to 40 ng/PCR. All samples were amplified with Bacteroides species, human-specific Bacteroides, cow-specific Bacteroides, and E. coli uidA primers. The 25-µl PCR mixtures consisted of the QIAGEN Taq PCR Master Mix Kit (2x-concentrated 5-U/µl Taq, 3 mM MgCl₂, and 400 µM of each deoxynucleoside triphosphate), and each primer was present at a final concentration of 0.35 µM. All PCRs were run in the MJ Research PTC-Quad thermal cycler (Watertown, MA). The PCR conditions were 1 cycle at 94°C for 4 minutes; 35 cycles at 94°C for 30 s; annealing temperature for 30 s; 72°C for 30 s; 1 cycle at 72°C for 6 min; and a 10°C hold. The annealing temperatures were as follows: Bac32F/Bac708R, 53°C; HF183F/Bac708R, 59°C; CF128F/Bac708R, 58°C; CF193F/Bac708R, 55°C; and uidA1318F/uidA1698R, 60°C. The PCR products were visualized on 2% agarose gels following staining with ethidium bromide and were compared with a 100-bp DNA size standard (Fisher Scientific, Pittsburgh, PA).

Cloning and sequencing of Bacteroides species 16S rRNA.
Two surface water samples that produced positive results for Bacteroides spp. in the absence of other evidence of fecal pollution were chosen for analysis to determine if the amplification product could be identified as Bacteroides spp. of fecal origin. The approximately 700-bp PCR products generated with the Bac32F and Bac708R primers were cloned into pCR2.1 using the TOPO TA cloning kit (Invitrogen, Carlsbad, CA). Plasmid DNA was prepared using a Bio-Rad miniplasmid prep kit (Hercules, CA), and sequencing was carried out from vector primer sites M13 Forward and M13 Reverse following the manufacturer's instructions on a Beckman Coulter CEQ8000 (Fullerton, CA). Sequences were analyzed with BLAST to determine the closest match and aligned with these and other published sequences obtained from GenBank using Vector NTI software (Invitrogen, Carlsbad, CA). Phylogenetic analysis was carried out on approximately 700 bp of the 16S rRNA sequence using ClustalW. Bootstrap values were obtained from a consensus of 1,000 neighbor-joining trees.

Nucleotide sequence accession numbers.
The partial 16S rRNA gene sequences have been deposited in the GenBank database under accession numbers DQ099450 to DQ099454.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Relative abundances of E. coli and Bacteroides genetic markers in sewage.
The limits of detection of E. coli and Bacteroides were determined by PCR in multiple sewage influent samples obtained from four wastewater treatment plants over a 9-month period. Jones Island treatment plant services the largest urban area and also receives residential and industrial waste. In seven samples from this wastewater treatment plant, Bacteroides sp. detection by PCR averaged 0.3-CFU/100 ml E. coli density, compared to E. coli and Bacteroides human-specific marker detection, which equated to 37- and 16-CFU/100 ml E. coli densities, respectively (Fig. 2). The mean dilution factor of sewage that corresponded to the limit of detection of the Bacteroides human-specific marker was 1:150,000. The other treatment plants produced positive results at similar dilution values, suggesting that the ratios of these organisms are fairly consistent in sewage. The levels of culturable E. coli that corresponded to the limits of detection for all four wastewater treatment plants are shown in Table 1. All samples were negative for the cow-specific Bacteroides species marker, with the exception of the sewage treatment plant located in an agricultural watershed, where this marker was positive in undiluted and 1:5-diluted samples. These positive samples corresponded to very high levels of culturable E. coli, e.g., >5.1 x 10⁴ CFU/100 ml.

View larger version (54K): [in this window] [in a new window]

FIG. 2. Detection of E. coli uidA (top), human-specific Bacteroides (middle), and total Bacteroides species (bottom) genetic markers in dilutions of DNA extracted from sewage. The dilutions are indicated at the top of each lane, where 0 corresponds to an initial 1:10 dilution, with lanes 1 to 10 representing serial 1:5 dilutions. E. coli levels were determined in the original sample; dilutions 5, 6, and 7 were calculated to have 77 CFU, 15 CFU, and 3 CFU/100 ml, respectively.

Detection of cow-specific Bacteroides spp.
The two previously described primers to detect cow-specific Bacteroides, CF128 and CF193, were tested in fecal samples collected from farm sites within the Lake Michigan drainage basin. Analysis of fecal samples from 15 different feedlots demonstrated that the CF128 primer could successfully detect cow-specific Bacteroides spp. in all the samples. Initial experiments using CF193 were positive for approximately 85% of the samples but produced nonspecific amplification products ranging from 100 bp to 1,500 bp in size in a number of the positive samples. The remaining samples that tested negative with the CF193 primer were from three different farm sites. The E. coli uidA marker was positive in all samples, and the human-specific Bacteroides marker was negative in all of the samples. These results demonstrate that the CF128 primer for cow-specific Bacteroides may be applicable for detecting fecal pollution in the Lake Michigan drainage basin.

Detection of fecal indicator genetic markers in Lake Michigan following contamination events.
Lake Michigan receives drainage from 850 square miles of agricultural, suburban, and urban land use via three rivers that discharge into the Milwaukee harbor and, subsequently, Lake Michigan (Fig. 1). The river flow volumes into Lake Michigan and the sewage overflow events are shown in Fig. 3 and represent the watershed drainage and sewage sources entering Lake Michigan during the sample times.

View larger version (19K): [in this window] [in a new window]

FIG. 3. Hydrograph of river discharge into Lake Michigan. Flow was measured at the channel downstream from the confluence of three major rivers that drain the Milwaukee River basin. The time spans of SSO and CSO events are indicated. The primary discharge points from sewage overflows are rivers prior to confluence and the harbor.

The detection of specific markers compared to culturable E. coli levels across sites over the 60-day sample period are shown in Table 3. Conductivity measurements approximate the amount of (contaminated) river water in relation to (uncontaminated) lake water, and the decrease in conductivity across sample locations indicates the amount of dilution. The rivers in this system, following precipitation events, have been found to range from 0.560 to 0.690 mS/cm, while Lake Michigan with no river water influences is typically measured at 0.281 to 0.223 mS/cm. All samples were positive for Bacteroides species genetic markers, even in the absence of other evidence of fecal pollution, e.g., detectable river water or culturable E. coli. Other genetic markers for fecal indicators were variable in relation to the amount of river water and culturable E. coli.

View this table: [in this window] [in a new window]

TABLE 3. Occurrence of host-specific markers in Lake Michigan following contamination with CSO, SSO, and river basin drainage containing urban and agricultural runoff

River water samples collected at three separate locations during an SSO event demonstrated the highest levels of E. coli; both Bacteroides spp. and human-specific Bacteroides were detected (Table 3). Interestingly, one site located in the suburban part of the watershed showed the presence of cow-specific Bacteroides; however, it could not be determined if this signal originated from the SSO discharge or from the river, which receives runoff from storm water outfalls located throughout the area.

During the SSO/CSO events, human-specific markers were detected at sites in nearshore Lake Michigan with >200 CFU/100 ml of E. coli. E. coli levels at distances of more than 2 km from the harbor contained <200 CFU/100 ml E. coli (data not shown); these samples were not analyzed for genetic markers unless the samples were collected near beach site 2. Notably, four beach samples collected during the SSO/CSO events (two at beach site 2 and two 100 m from the shore of beach site 2) (Table 4) and eight samples collected in the days that followed the SSO/CSO event (Table 3) were found to have E. coli levels below the EPA-recommended limit of 235 CFU/100 ml for recreational water but were positive for the human-specific Bacteroides genetic marker. Because the samples contained known sewage contamination, these results demonstrate that the human-specific genetic markers are more sensitive and/or persist longer than culturable E. coli.

View this table: [in this window] [in a new window]

TABLE 4. Occurrence of fecal indicator bacterial genetic markers in beach water samples at nine freshwater coastal beaches on Lake Michigan

Human-specific Bacteroides genetic markers were detected 9 days post-CSO, but not 7 days post-CSO; within this same time frame, 0.2 in. of rainfall occurred, contributing additional storm water to nearshore Lake Michigan. Likewise, the human-specific genetic marker was detected >60 days post-CSO following a rain event of 0.3 in. These findings suggest that human sources of fecal pollution may not be limited to reported sewage overflows, and unrecognized sanitary sewage inputs, e.g., leaking sewer lines or cross-connections, may also contribute to contamination in Lake Michigan.

During periods of heavy rain and subsequent SSO/CSO events, upstream river water from the agricultural land use in the basin is expected to be mixed with urban storm water and sewage discharge. Interestingly, the cow-specific genetic marker was detected following two major rain events in most samples within the harbor, but this marker was lost once water was discharged outside the harbor to Lake Michigan. Three of the 17 samples collected on 14 May 2004 during a CSO, which was expected to have large amounts of contaminated water from the drainage basin drainage (based on river flow into Lake Michigan) (Fig. 3), were negative for the cow-specific marker and corresponded to 940, 1,140, and 13,600 CFU/100 ml of E. coli.

Detection of fecal indicator genetic markers at Lake Michigan beaches.
All nine beaches exceeded the EPA-recommended limit for E. coli of 235 CFU/100 ml on at least one of the days tested. Of samples that exceeded the recommended limits, all were positive for E. coli by PCR detection methods (Table 4). In addition, approximately 80% of the beach water samples that contained >30 CFU but <235 CFU/100 ml were positive for E. coli by PCR.

Human-specific Bacteroides spp. were found at three of the nine beach sites tested. At beach site 1, beach water and the storm water outfalls that drain directly across the beach were both positive for the human-specific genetic marker on the same day, and the outfalls were positive 3 days later. Samples collected at later dates were negative for the human marker at the beach and outfall; however, E. coli levels in theses samples were found to be 5,800 CFU/100 ml and 2,200 CFU E. coli/100 ml for the beach water and outfall, respectively. Results from samples collected at nearshore sites (0.5 km from shore) on that same day had 8 and 13 CFU E. coli/100 ml. The beach water and outfall samples were positive for E. coli and Bacteroides spp. by PCR, while only Bacteroides spp. were detected at the two sites 0.5 km from shore (Table 4).

Beach site 2 is located 0.5 km from a CSO outfall that drains directly into Lake Michigan and is in close proximity to the Milwaukee harbor; therefore, this site is most likely to be influenced by sewage overflows when they occur. Two samples collected at beach site 2 during a CSO did not exceed the EPA-recommended limit for E. coli but were positive for human-specific Bacteroides, demonstrating the high sensitivity of this genetic marker. Notably, 20 days post-CSO, a nearby storm water outfall was positive for the human-specific genetic marker; samples collected 100 m from shore and beach samples were also positive for the human-specific marker on the same day (Table 4).

The third beach site positive for the human-specific marker, beach site 6, was found to have two of seven samples positive; however, detection of the PCR product was weak in one of the samples. This beach is surrounded by forested land with no known urban or agricultural influences; the park itself is serviced by septic drain fields rather than contained septic systems. Five additional samples that exceeded 235 CFU/100 ml did not display evidence of human sources.

Surface runoff may contribute large loads of fecal indicator bacteria to adjacent beach waters. Storm water runoff discharged from outfalls in the absence of sewage overflows was found to have E. coli levels as high as 15,900 CFU/100 ml. Runoff collected directly from the parking lot surfaces was found to be as high as 50,000 CFU/100 ml. The parking lot surface runoff was evaluated for fecal indicator genetic markers to ensure that the human-specific primers would not cross-react with organisms found in surface runoff, which was expected to originate from urban wildlife. In these samples, Bacteroides spp. and E. coli were positive, but the human-specific genetic marker was negative (Table 4).

The remaining six beaches did not demonstrate evidence of human sources of fecal pollution. In addition, none of the nine beaches was positive for the cow-specific genetic markers. These six beaches exceeded the EPA limit for more than 40% of the samples, which were primarily collected following rain events, when contamination is expected to be highest. These findings suggest that non-point-source runoff, which can be delivered from storm water outfalls, impervious surfaces, and the beach sand itself, is a major influence on beach water quality at these sites.

Sequencing of Bacteroides spp. from Lake Michigan samples with no other evidence of fecal pollution.
The Bacteroides spp. that were detected at the two most distant sites in Lake Michigan were compared to previously published sequences from host and environmental samples (Fig. 4). One of the samples was collected 1 day post-CSO 8 km from the harbor and contained <10 CFU/100 ml. Sequencing of multiple clones revealed two unique sequences: one sequence most closely matched an uncultured Bacteroides sp. isolated from human feces (99% identity; GenBank accession number AY985132), and the other sequence most closely matched an uncultured Bacteroides sp. isolated from water 20 m upstream of a manure pile (99% identity; GenBank AY212541). The second sample chosen for sequencing was collected 7.5 km from the shore in Green Bay, which was the control site used for comparison of beach water samples; this sample contained <1 CFU E. coli/100 ml. Three different amplified sequences were found; the first sequence was identical to an uncultured Bacteroides sp. from human feces (100% identity; GenBank AY597205), the second sequence most closely matched an uncultured Bacteroides sp. cloned from a biopsy sample of a human intestine, and the third sequence most closely matched a Prevotella sp. (99% identity; GenBank AY581270).

View larger version (21K): [in this window] [in a new window]

FIG. 4. Phylogenetic tree derived from partial 16S RNA gene sequence data obtained from Bacteroides spp. amplified from Lake Michigan water samples. Clones 257 were obtained from DNA isolated from samples collected 2 km from shore and 8 km from the Milwaukee harbor, and clones designated 245 were obtained from DNA isolated from samples collected 7.5 km from shore in Green Bay. Previously published sequences are designated by the accession number or species name. Bootstrap values are shown as percentages of 1,000 trees; values less than 70% are omitted. The scale refers to the number of nucleotide substitutions per nucleotide position.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Information as to the source of pollution ideally should be linked to standard fecal indicators (11), particularly in the case of beach-monitoring programs. Since fecal indicator densities are generally the sole parameter used to assess beach water quality, identifying the sources of the indicator organisms would allow beach managers to minimize beach closings by identifying and addressing problems that result in elevated fecal indicator levels. Further, identifying the presence of human and zoonotic sources that are reservoirs for human disease, even if they are not major contributors to overall fecal indicator levels, would be useful for resolving the disparity between detectable indicator organisms and the occurrence of pathogens (19).

This work demonstrated that the human-specific Bacteroides genetic marker is a sensitive measure of sewage contamination. The detection limit and relative abundance of the genetic markers in relation to culturable E. coli were consistent for several wastewater treatment plants servicing different regions of the Lake Michigan drainage basin, supporting the reliability of these markers. Similarly, Bernhard et al. found that the human-specific genetic marker was widely distributed and detectable in wastewater treatment plants separated by 100 miles (1). Overall, the human-specific genetic marker could be detected when culturable E. coli levels were as low as 30 to 105 CFU/100 ml in nearshore Lake Michigan and 110 to 170 CFU/100 ml at beach site 2 following a CSO. Two previous field studies utilizing the human-specific genetic marker also reported detection when culturable E. coli levels were low; however, these were primarily conducted in a marine estuary and coastal bay, where E. coli survival might be affected by salinity (1, 4).

Sanitary sewage overflow samples taken in the suburban part of the watershed showed the presence of cow-specific Bacteroides. This area has a high deer density, which may account for detection of the cow-specific genetic marker, since the cow-specific primers do not differentiate between types of ruminants, i.e., elk, deer, and cows (2). In addition, the cow-specific marker was found in sewage samples from the wastewater treatment plant in an agricultural watershed, though at a much lower abundance than the human-specific marker, and was detected only when E. coli levels were >50,000 CFU. These findings demonstrate that there are complications in using this marker as a sole indicator of agricultural inputs, and interpretations should be made in light of these factors.

In this study, we were interested in the dynamics of distribution and subsequent detection of the fecal indicator genetic markers and in how these markers correspond to culturable E. coli levels. Strict anaerobes are not expected to survive in the environment (5, 18, 28); however, the ability to detect these organisms in surface waters over time and the relationship to E. coli survival and recovery by standard culture methods are unknown. Because the major input of contamination is via three rivers that discharge into the harbor, measurements of the river flow (Fig. 3) and conductivity allowed the contamination plume to be tracked in Lake Michigan. The SSO/CSO events provided the ideal model to test how these markers might fluctuate in the nearshore during a known sewage contamination event and for several days post-sewage contamination. Comparing E. coli detection by PCR with the Bacteroides PCR gave some context to the Bacteroides genetic targets compared to culturable E. coli. The simultaneous disappearance of the E. coli and human-specific genetic markers in Lake Michigan as contamination dispersed suggests that these markers have similar longevities in terms of intact cells that can be recovered for analysis.

Notably, Bacteroides spp. were detected by PCR when there was no other conclusive evidence of fecal pollution; E. coli was absent or at very low levels, e.g., 0 to 10 CFU/100 ml. Previous studies indicated that Bacteroides spp. may be one of the most sensitive fecal indicator genetic markers, since the primers are homologous to multiple Bacteroides species that would be present in fecal pollution at a much higher abundance than fecal coliforms (1, 17, 41). The Bacteroides species genetic marker may be useful for tracking low levels of fecal pollution, particularly in a system such as Lake Michigan, where dilution makes it difficult to track the distribution and fate of pollution inputs from watershed drainage. In this study, the Bacteroides sp. genetic marker was found in all samples that were analyzed. Sequencing of the amplification product from two samples with no other evidence of fecal pollution showed the presence of Bacteroides spp. identical or nearly identical (99% homology) to Bacteroides spp. from human sources. A recent study has shown 97 to 99% identity in the same region of the 16S rRNA gene of Bacteroides-Prevotella recovered from humans, cats, dogs, and gulls (7); therefore, the origin of the Bacteroides spp. detected in Lake Michigan cannot be inferred as being from human sources without additional testing. Two other sequences closely matched a Bacteroides sp. and a Prevotella sp. that were previously recovered from water samples with unknown contamination histories, making it difficult to determine if these sequences represent species of fecal origin or species found naturally in surface waters. Overall, the sequence information supports the notion that Bacteroides spp. of fecal origin were detected in apparently "uncontaminated" waters.

While the Bacteroides genetic markers have been shown to provide useful information in field studies (1, 3, 4), their applicability to the beach environment in freshwater systems has not been evaluated. In this study, multiple beaches along Lake Michigan were assessed for E. coli and Bacteroides genetic markers by PCR, and the markers were compared to standard microbiological monitoring methods. Because the sewage treatment plant influent and the sewage overflow samples were all positive when expected, the absence of the human-specific markers at beaches with elevated E. coli levels would support the notion that the beach was contaminated by nonhuman sources. Importantly, the human-specific genetic marker was detected at beaches during a sewage overflow when E. coli was at 110 to 170 CFU/100 ml, below the recommended limit for E. coli in recreational water. This illustrates how information about the source of pollution might protect public health more effectively than reliance on standard measures of fecal indicators.

In the absence of reported sewage overflow events, two of the three urban beaches were positive for the human-specific genetic marker. Undetected sanitary sewage infiltration into the storm sewer system might be a source of sanitary sewage contamination (26). Many beaches experience closing due to nonpoint runoff (20, 24, 38); however, epidemiology studies aimed at correlating indicator organism densities with illness are based on point source exposure (24). Differentiation of these sources might be an important first-tier approach (4) to understanding locations of fecal pollution loading, as well as the contributing host sources.

Nonculture molecular approaches to detect genetic markers of Bacteroides spp. appear promising, since these markers are a sensitive measure of fecal pollution, are relatively simple to detect, and have been shown to be host specific (10, 11). Further work is necessary to assess the informational value of detecting these markers in the environment and to determine their association with pollution sources (8, 11, 24, 32, 34). Extensive field testing will be necessary to determine the geographical distribution of source-specific genetic markers and their correlation with standard fecal indicators (2). Most importantly, future studies determining the relationship of genetic markers to pathogens subjected to the same physical and ecological influences (24) will provide invaluable information.

 
We thank the Manitowoc Soil and Water Conservation Department for providing feedlot manure samples. We also thank Sara Goetz for invaluable assistance in editing the manuscript.

Funding for this work was provided by the University of Wisconsin Sea Grant Program.

 
* Corresponding author. Mailing address: 600 E. Greenfield Ave., Milwaukee, WI 53204. Phone: (414) 382-1700. Fax: (414) 382-1705. E-mail: mclellan{at}uwm.edu.

Top Abstract Introduction Materials and Methods Results Discussion References

 

1
1. Bernhard, A. E., and K. G. Field. 2000. Identification of nonpoint sources of fecal pollution in coastal waters by using host-specific 16S ribosomal DNA genetic markers from fecal anaerobes. Appl. Environ. Microbiol. 66:1587-1594.[Abstract/Free Full Text]
2
2. Bernhard, A. E., and K. G. Field. 2000. A PCR assay to discriminate human and ruminant feces on the basis of host differences in Bacteroides-Prevotella genes encoding 16S rRNA. Appl. Environ. Microbiol. 66:4571-4574.[Abstract/Free Full Text]
3
3. Bernhard, A. E., T. Goyard, M. T. Simonich, and K. G. Field. 2003. Application of a rapid method for identifying fecal pollution sources in a multi-use estuary. Water Res. 37:909-913.[Medline]
4
4. Boehm, A. B., J. A. Fuhrman, R. D. Morse, 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 Island, California. Environ. Sci. Technol. 37:673-680.[Medline]
5
5. Carrillo, M., E. Estrada, and T. C. Hazen. 1985. Survival and enumeration of the fecal indicators Bifidobacterium adolescentis and Escherichia coli in a tropical rain forest watershed. Appl. Environ. Microbiol. 50:468-476.[Abstract/Free Full Text]
6
6. Chern, E. C., Y. L. Tsai, and B. H. Olson. 2004. Occurrence of genes associated with enterotoxigenic and enterohemorrhagic Escherichia coli in agricultural waste lagoons. Appl. Environ. Microbiol. 70:356-362.[Abstract/Free Full Text]
7
7. Dick, L. K., A. E. Bernhard, T. J. Brodeur, J. W. Santo Domingo, J. M. Simpson, S. P. Walters, and K. G. Field. 2005. Host distributions of uncultivated fecal Bacteroidales bacteria reveal genetic markers for fecal source identification. Appl. Environ. Microbiol. 71:3184-3191.[Abstract/Free Full Text]
8
8. Dick, L. K., and K. G. Field. 2004. Rapid estimation of numbers of fecal Bacteroidetes by use of a quantitative PCR assay for 16S rRNA genes. Appl. Environ. Microbiol. 70:5695-5697.[Abstract/Free Full Text]
9
9. Dufour, A. P. 1984. Bacterial indicators of recreational water quality. Can. J. Public Health 75:49-56.[Medline]
10
10. Field, K. G., A. E. Bernhard, and T. J. Brodeur. 2003. Molecular approaches to microbiological monitoring: fecal source detection. Environ. Monit. Assess. 81:313-326.[CrossRef][Medline]
11
11. Field, K. G., E. C. Chern, L. K. Dick, J. Fuhrman, J. Griffith, P. A. Holden, M. G. LaMontagne, J. Le, B. Olson, and M. T. Simonich. 2003. A comparative study of culture-independent, library-independent genotypic methods of fecal source tracking. J. Water Health 1:181-194.[Medline]
12
12. Fiksdal, L., J. S. Maki, S. J. LaCroix, and J. T. Staley. 1985. Survival and detection of Bacteroides spp., prospective indicator bacteria. Appl. Environ. Microbiol. 49:148-150.[Abstract/Free Full Text]
13
13. Griffith, J. F., S. B. Weisberg, and C. D. McGee. 2003. Evaluation of microbial source tracking methods using mixed fecal sources in aqueous test samples. J. Water Health 1:141-151.[Medline]
14
14. Henrickson, S. E., T. Wong, P. Allen, T. Ford, and P. R. Epstein. 2001. Marine swimming-related illness: implications for monitoring and environmental policy. Environ. Health Perspect. 109:645-650.[Medline]
15
15. Howard, B. J., J. Klaas, S. J. Rubin, A. S. Weissfeld, and R. C. Tilton. 1987. Clinical and pathogenic microbiology. C.V. Mosby Company, Saint Louis, Mo.
16
16. Jiang, S., R. Noble, and W. Chu. 2001. Human adenoviruses and coliphages in urban runoff-impacted coastal waters of Southern California. Appl. Environ. Microbiol. 67:179-184.[Abstract/Free Full Text]
17
17. Kreader, C. A. 1995. Design and evaluation of Bacteroides DNA probes for the specific detection of human fecal pollution. Appl. Environ. Microbiol. 61:1171-1179.[Abstract]
18
18. Kreader, C. A. 1998. Persistence of PCR-detectable Bacteroides distasonis from human feces in river water. Appl. Environ. Microbiol. 64:4103-4105.[Abstract/Free Full Text]
19
19. Lipp, E. K., S. A. Farrah, and J. B. Rose. 2001. Assessment and impact of microbial fecal pollution and human enteric pathogens in a coastal community. Mar. Pollut. Bull. 42:286-293.[CrossRef][Medline]
20
20. McLellan, S. L., and A. K. Salmore. 2003. Evidence for localized bacterial loading as the cause of chronic beach closings in a freshwater marina. Water Res. 37:2700-2708.[Medline]
21
21. Moe, C. L. 2002. Waterborne transmission of infectious agents, p. 184-204. In J. H. Hurst (ed.), Manual of environmental microbiology, 2nd ed. ASM Press, Washington, D.C.
22
22. Noble, R. T., S. M. Allen, A. D. Blackwood, W. Chu, S. C. Jiang, G. L. Lovelace, M. D. Sobsey, J. R. Stewart, and D. A. Wait. 2003. Use of viral pathogens and indicators to differentiate between human and non-human fecal contamination in a microbial source tracking comparison study. J. Water Health 1:195-207.[Medline]
23
23. 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 California shoreline. J. Water Health 1:23-31.[Medline]
24
24. National Research Council of the National Academies. 2004. Indicators for waterborne pathogens. National Research Council of the National Academies, Washington, D.C.
25
25. Natural Resources Defense Council. 19 February 2004, posting date. Swimming in sewage: the growing problem of sewage pollution and how the Bush administration is putting our health and environment at risk. [Online.] Natural Resources Defense Council, New York, N.Y. http://www.hrdc.org/water/pollution/sewage/contents@asp.
26
26. O'Shea, M. L., and R. Field. 1992. Detection and disinfection of pathogens in storm-generated flows. Can. J. Microbiol. 38:267-276.[Medline]
27
27. Pruss, A. 1998. Review of epidemiological studies on health effects from exposure to recreational water. Int. J. Epidemiol. 27:1-9.[Abstract/Free Full Text]
28
28. Resnick, I. G., and M. A. Levin. 1981. Assessment of bifidobacteria as indicators of human fecal pollution. Appl. Environ. Microbiol. 42:433-438.[Abstract/Free Full Text]
29
29. Roll, B. M., and R. S. Fujioka. 1997. Sources of faecal indicator bacteria in a brackish tropical stream and their impact on recreational water quality. Water Sci. Technol. 35:179-186.
30
30. Scott, T. M., J. B. Rose, T. M. Jenkins, S. R. Farrah, and J. Lukasik. 2002. Microbial source tracking: current methodology and future directions. Appl. Environ. Microbiol. 68:5796-5803.[Free Full Text]
31
31. Simpson, J. M., J. W. Santo Domingo, and D. J. Reasoner. 2002. Microbial source tracking: state of the science. Environ. Sci. Technol. 36:5279-5288.[Medline]
32
32. Simpson, J. M., J. W. Santo Domingo, and D. J. Reasoner. 2004. Assessment of equine fecal contamination: the search for alternative bacterial source-tracking targets. FEMS Microbiol. Lett. 47:65-75.
33
33. 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. 66:230-237.[Abstract/Free Full Text]
34
34. Stewart, J. R., R. D. Ellender, J. A. Gooch, S. Jiang, S. P. Myoda, and S. B. Weisberg. 2003. Recommendations for microbial source tracking: lessons from a methods comparison study. J. Water Health 1:225-231.[Medline]
35
35. Toranzos, G. A., and G. A. McFeters. 1997. Detection of indicator microorganisms in environmental fresh waters and drinking waters, p. 184-194. In C. J. Hurst, G. R. Knudsen, M. J. McInerney, L. D. Stetzenbach, and M. V. Walter (ed.), Manual of environmental microbiology. American Society for Microbiology, Washington, D.C.
36
36. U.S. Environmental Protection Agency. 2000. Improved enumeration methods for recreational water quality indicators: enterococci and Escherichia coli. EPA/821/R-97/004. U.S. Environmental Protection Agency, Washington, D.C.
37
37. U.S. Environmental Protection Agency. 2003. Bacterial water quality standards for recreational waters. EPA/823/R-03/008. U.S. Environmental Protection Agency, Washington, D.C.
38
38. U.S. Environmental Protection Agency. 2003. EPA's BEACH Watch Program: 2002 swimming season. EPA/823/F-03/007. U.S. Environmental Protection Agency, Washington, D.C.
39
39. U.S. Environmental Protection Agency. 2004. Impacts and Control of CSOs and SSOs. EPA/833/R-04/001. U.S. Environmental Protection Agency, Washington, D.C.
40
40. Wade, T. J., N. Pai, J. N. Eisenberg, and J. M. Colford, Jr. 2003. Do U.S. Environmental Protection Agency water quality guidelines for recreational waters prevent gastrointestinal illness? A systematic review and meta-analysis. Environ. Health Perspect. 111:1102-1109.[Medline]
41
41. Wang, R. F., W. W. Cao, and C. E. Cerniglia. 1996. PCR detection and quantitation of predominant anaerobic bacteria in human and animal fecal samples. Appl. Environ. Microbiol. 62:1242-1247.[Abstract]

---

Applied and Environmental Microbiology, December 2005, p. 8305-8313, Vol. 71, No. 12
0099-2240/05/$08.00+0     doi:10.1128/AEM.71.12.8305-8313.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.

This article has been cited by other articles:

• McLaughlin, M. R., Brooks, J. P., Adeli, A. (2009). Characterization of Selected Nutrients and Bacteria from Anaerobic Swine Manure Lagoons on Sow, Nursery, and Finisher Farms in the Mid-South USA. J. Environ. Qual. 38: 2422-2430 [Abstract] [Full Text]  
• Brooks, J. P., McLaughlin, M. R. (2009). Antibiotic Resistant Bacterial Profiles of Anaerobic Swine Lagoon Effluent. J. Environ. Qual. 38: 2431-2437 [Abstract] [Full Text]  
• Jeter, S. N., McDermott, C. M., Bower, P. A., Kinzelman, J. L., Bootsma, M. J., Goetz, G. W., McLellan, S. L. (2009). Bacteroidales Diversity in Ring-Billed Gulls (Laurus delawarensis) Residing at Lake Michigan Beaches. Appl. Environ. Microbiol. 75: 1525-1533 [Abstract] [Full Text]  
• Lee, Y.-J., Molina, M., Santo Domingo, J. W., Willis, J. D., Cyterski, M., Endale, D. M., Shanks, O. C. (2008). Temporal Assessment of the Impact of Exposure to Cow Feces in Two Watersheds by Multiple Host-Specific PCR Assays. Appl. Environ. Microbiol. 74: 6839-6847 [Abstract] [Full Text]  
• Gourmelon, M., Caprais, M. P., Segura, R., Le Mennec, C., Lozach, S., Piriou, J. Y., Rince, A. (2007). Evaluation of Two Library-Independent Microbial Source Tracking Methods To Identify Sources of Fecal Contamination in French Estuaries. Appl. Environ. Microbiol. 73: 4857-4866 [Abstract] [Full Text]  
• Stoeckel, D. M., Harwood, V. J. (2007). Performance, Design, and Analysis in Microbial Source Tracking Studies. Appl. Environ. Microbiol. 73: 2405-2415 [Full Text]  
• McQuaig, S. M., Scott, T. M., Harwood, V. J., Farrah, S. R., Lukasik, J. O. (2006). Detection of Human-Derived Fecal Pollution in Environmental Waters by Use of a PCR-Based Human Polyomavirus Assay. Appl. Environ. Microbiol. 72: 7567-7574 [Abstract] [Full Text]  

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Bower, P. A. Articles by McLellan, S. L. Search for Related Content PubMed PubMed Citation Articles by Bower, P. A. Articles by McLellan, S. L. Agricola Articles by Bower, P. A. Articles by McLellan, S. L.