Evaluation of Microbial Water Quality Indicators in a Forested and Agricultural Watershed

Authors: Shelton, Daniel; Karns, Jeffrey; Sadeghi, Ali; Coppock, Cary; Pachepsky, Yakov

Submitted to: Journal of Food Protection
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 10/2/2011
Publication Date: 1/15/2011
Citation: Shelton, D.R., Karns, J.S., Sadeghi, A.M., Coppock, C.R., Pachepsky, Y.A. 2011.Relationship between eae and stx virulence genes and Escherichia coli in an agricultural watershed: Implications for irrigation water standards and leafy green commodities. Journal of Food Protection. 74(1):18-23.

Interpretive Summary: The U.S. Environmental Protection Agency (EPA) and states are engaged in an extensive effort to assess and improve surface water quality, including decreasing risks to public health from water-borne pathogens. Since many pathogens originate from feces (either direct deposition or land application), indicators of fecal contamination are utilized to assess water quality. However, the relationship(s) between indicators and pathogens, and their fate and survival in watersheds are poorly understood. We conducted a monitoring study in a small rural watershed with inputs from wildlife and grazing cattle to evaluate the usefulness of microbial water indicators, such as Escherichia coli or virulence factors typically associated with pathogenic E. coli, in assessing water quality. E. coli concentrations were substantially higher in agricultural than in forested sites indicative of the much higher fecal inputs from grazing cattle vs. wildlife. A general decrease was observed in E. coli concentrations from summer through fall/winter. This decrease was partially due to decreased wildlife activity and cattle densities. However, an additional factor was likely “flushing” of E. coli caused by high water flow (due to high rainfall) beginning in late fall. Virulence factors associated with pathogenic E. coli (O157 serogroup, eae gene, and stx1/2 genes) were prevalent throughout the watershed. However, no correlation was observed between concentrations of generic E. coli and any virulence factor. No definitive conclusions could be drawn regarding the presence or absence of specific pathogenic E. coli strains due to the distribution of these virulence factors among other E. coli strains and other fecal bacteria. Our results are consistent with the well established principle that fecal runoff and/or deposition are the predominant source of water-borne E. coli contamination. However, they also illustrate the difficulty associated with utilizing E. coli data to predict water-borne pathogen contamination; hence, public health risk.

Technical Abstract: The U.S. Environmental Protection Agency (EPA) and European Union (EU) are engaged in an extensive effort to assess and improve surface water quality, including decreasing risks to public health from water-borne pathogens. In the absence of data for specific pathogens, indicators of fecal contamination such as Escherichia coli are utilized to assess water quality. However, the relationship(s) between indicators and pathogens, and their population dynamics in watersheds are poorly understood. We undertook this monitoring study in a small rural watershed with inputs from wildlife and grazing cattle to (i) evaluate fluctuations in E. coli populations and (ii) assess the use of virulence factors typically associated with pathogenic E. coli as indicators of water quality. Generic E. coli concentrations were substantially higher in agricultural than in forested sites indicative of the much higher fecal inputs from grazing cattle vs. wildlife. However, high E. coli concentrations found in sediments suggest that these may be relatively stable habitats for E. coli growth/survival and be responsible for some portion of the downstream contamination. A general decrease was observed in E. coli concentrations from summer through fall/winter. This decrease was partially due to decreased wildlife activity and cattle densities. However, an additional factor was likely “flushing” of sediment-borne E. coli caused by high discharge levels (due to high rainfall) beginning in late fall. Virulence factors associated with pathogenic E. coli (O157 serogroup, eae gene, and stx1/2 genes) were prevalent throughout the watershed; population dynamics were similar to generic E. coli. However, no definitive conclusions could be drawn regarding the presence or absence of specific pathogenic E. coli strains due to the distribution of these virulence factors among other E. coli strains and other enteric bacteria. Also, no correlation was observed between concentrations of generic E. coli and the eae gene at agricultural sites, suggesting that generic E. coli data cannot be used to predict the risk of pathogen exposure. Although our results are consistent with the well established principle that fecal runoff and/or deposition are the predominant source of water-borne E. coli contamination, they also illustrate the difficulty associated with the interpretation of water-borne E. coli data. Watershed water quality models should account for E. coli growth and survival in indigenous habitats and “flushing” of sediment-borne E. coli from watersheds, as well as for fecal runoff.