Many of Minnesota's rivers and streams do not achieve the Clean Water Act "swimmable" goal due to elevated numbers of fecal coliform bacteria. Sources of fecal coliform bacteria include runoff from feedlots and manure-amended agricultural land, wildlife, inadequate septic systems, urban runoff, and sewage discharges. High levels of fecal bacteria in Minnesota's rivers, lakes and streams threaten the use of these water resources for swimming and other forms of water recreation.

The ability to distinguish between human and animal sources of fecal contamination is an important assessment tool. From a public health perspective, fecal contamination originating from human sources poses a different human health risk than that originating from animal sources. From a water quality perspective the ability to narrow the source of fecal contamination among the many potential sources will facilitate more tailored and cost effective pollution abatement efforts.

Standard microbiological methods for enumerating fecal coliform bacteria do not differentiate between sources of fecal pollution giving rise to elevated coliform counts. Currently, several investigators are exploring various methods to determine the source of water-borne fecal contamination. These methods generally fall into two categories: DNA fingerprinting and antibiotic resistance profiles. The approach used for both methods is to create a database of DNA fingerprints or antibiotic resistance profiles generated from fecal bacteria isolated from known human and animal sources. The known-source database is then applied to fecal bacteria isolated from impacted waterways. Organisms yielding similar DNA banding patterns (or antibiotic resistance profiles) can be regarded as being identical or near-identical, and as such, define the source of the fecal contamination.

In our laboratory we have utilized two types of DNA fingerprinting for source tracking. Originally, we worked with rep-PCR DNA fingerprinting, a PCR-based method that exploits naturally occurring, highly conserved, repetitive DNA sequences, present in multiple copies in bacterial genomes. More recently we have switched to horizontal, fluorophore-enhanced, rep-PCR (HFERP). HFERP improves overall fingerprint resolution by including a ROX ladder with each fluoresceine labelled DNA fingerprint. This allows every single fingerprint in the database to be normalized. The rep-PCR and HFERP techniques have been shown to provide the necessary sensitivity and resolving power to differentiate between strains of fecal coliform bacteria originating from different human and animal sources. Of the various genetic fingerprinting strategies, rep-PCR and HFERP are relatively simple and cost-effective technique that can be adapted for high throughput applications.

Project Hypotheses

• E. coli strains form various ecotypes that are differentially adapted to inhabit the GI tracts of different animals
• The rep-PCR and HFERP techniques have the necessary resolving power and robustness to differentiate between different E. coli ecotypes of human and animal origin
• A database of DNA fingerprints produced from a large number of E. coli strains isolated from different known fecal sources can be used to identify the source of unknown E. coli strains isolated from water

Project Goals

Create a "known source" E. coli rep-PCR DNA fingerprint database to include about 2400 isolates obtained from 13 sources: cows, pigs, sheep, goats, turkeys, chickens, ducks, geese, deer, horses, dogs, cats, and humans.

Apply the database to E. coli isolated from four impacted Minnesota waterways (~300 E. coli per watershed).