Microbial Source Tracking for E. coli in 

Recreational and Wilderness Watersheds

(Mirror Lake, Seven Devils Wilderness Area, Idaho)
See more photos below

Background | Objectives | Location |  People | Sponsors | Methods | Data | Photos

Clear, sparkling water is not an accurate indicator that a water source in a natural area is free of bacterial contamination.  The impairment of water quality as a result of human activity, resulting in chemical or microbiological contamination, is one of the most significant threats to wild and natural areas.  The presence of enteric bacterial pathogens in drinking and recreational waters is a well known concern.  Less well known are the possible effects of introducing enteric strains of human and domestic animal origin into natural ecosystems.  In order to protect and improve aquatic resources in these areas, improved methods are needed to sample and asses water quality, to accurately identify sources of potential and actual microbial pollutants, and to develop management strategies to avoid future problems.  One of the most important and identifiable aspects of water quality is the presence of fecal coliform bacteria, especially E. coli.  Our project  focuses specifically on accurately identifying sources of fecal coliform bacteria (e.g., wild animal populations, vs. human and pack animal visitors), and the distribution of fecal coliforms in relation to recreational usage in natural areas.

Coliform bacteria include a number of bacterial genera; species within this group thrive in various soil and aquatic environments.  A subgroup, fecal coliform bacteria (e.g., Escherichia coli) reside in the intestines of all warm-blooded animals, and are excreted in waste materials.  The presence of fecal coliforms is specifically indicative of fecal contamination.  Sources of these bacteria may include indigenous mammals (e.g., bighorn sheep, deer, rodents, etc.), or ‘visitors’ (e.g., humans, pack animals, dogs).  The presence of high numbers of these bacteria in water samples traditionally has been used as an indicator of unsanitary conditions that may pose health concerns. 

Typically, it has been difficult to distinguish whether indigenous animals, pack animals, or people have been responsible for bacterial contamination of natural waters including lakes, streams and springs.  Recently, however, techniques of genetic fingerprinting have been found to be useful for matching bacterial strains from sampled sites with known source strains.  Microbial source tracking may allow the detection of the host origin of bacterial organisms, including fecal coliforms, from soil or water samples. 

Quantify populations of fecal coliform bacteria, at specific mapped locations relevant to trails, campsites, stream crossings and lakefronts at different alpine lakes in the Seven Devils Wilderness Area, and in vicinity of the U.S. Forest Service campground at Laird Park, Idaho.  At the Seven Devils sites, Sheep Lake is a lake which receives significant backpacker and horse-packer visitation, while Mirror Lake is relatively pristine, with low recreational usage and no horse access.

Construct and map genetic profiles of fecal coliform bacteria from the sampled sites and compare these profiles to known source species.  These profiles will be related spatially and temporally to give a series of “snapshots” of the fecal loading of each lake.

One testable hypothesis is that fecal coliform profiles (i.e., related to source animals) are significantly different in pristine lakes compared to high usage lakes.  In addition, there should be distinct, quantifiable spatial patterns to these profiles.  The quantity and modes of recreational and wildlife usage at a given site should directly affect the types of profiles found.

Chris LaPaglia, Scientific Aide

Louise-Marie Dandurand, Ph.D., Research Scientist

We visited the Seven Devils Wilderness in 2001 and 2002 to collect E. coli strains for our research, and obtained water samples from Seven Devils Lake, Sheep Lake and associated mud flats, and Mirror Lake.  Mirror Lake and the mud flats near Sheep Lake were judged inaccessible to horses, while Sheep Lake and Seven Devils Lake both had signs of recent visits by horses.  In addition to water samples, we isolated E. coli from all manure samples we could find in the study area.

The lakes were sampled in the late summer / early fall each year.  Water samples were plated onto TGY agar for total heterotroph counts and onto MacKonkeys Agar in order to count enteric bacteria (Clesceri, et al.1989).  Bacterial colonies, which grew red on MacKonkeys agar, were isolated and tested for citrate and sorbitol utilization.  Red colonies, which were citrate negative and sorbitol positive, were counted as E. coli (Difco Manual, 1984).  Colony counts were reported as colony forming units per 100 milliliters of lake water.  Representative results are shown below.

Isolates were grown overnight in Tryptic Soy Broth at 37^oC with shaking.  DNA from each isolate was prepared using a standard alkaline-lysis chromosomal mini-prep (Maniatis et al. 1982).  PCR fingerprinting/ribotyping of bacterial isolates followed an established procedure for restriction fragment length polymorphism using Hpy CH4 IV for restriction digests (Ibrahim, et al. 1996).  Initial results show clear differences in banding patterns for ruminant vs. human sources of E. coli, additional data will need to be collected and analyzed for discrimination at the species level.  

* Bacteria per 100 ml