Genotypic diversity of escherichia coli isolates from environmental sources and the influence on transport behavior

Authors: Cook, Kimberly - Kim; Bolster, Carl; AYERS, KATI - Western Kentucky University; REYNOLDS, DALE - Kentucky Department For Environmental Protection

Submitted to: Kentucky Water Resources Research Institute Symposium
Publication Type: Proceedings
Publication Acceptance Date: 1/24/2011
Publication Date: 6/28/2011
Citation: Cook, K.L., Bolster, C.H., Ayers, K., Reynolds, D. 2011. Genotypic diversity of escherichia coli isolates from environmental sources and the influence on transport behavior. Kentucky Water Resources Research Institute Symposium. 53-55. http://www.uky.edu/WaterResources/2011_Proceedings.pdf

Interpretive Summary:

Technical Abstract: Escherichia coli (E. coli) is a dominant intestinal commensal organism, an important fecal indicator bacterium (FIB), a pathogen and a target for microbial source tracking (MST). Strain level differences (genotypic and phenotypic) in E. coli influence its fate and transport and therefore have important implications for its validity as an FIB and for MST. Strain survival and variation are regulated by environmental conditions and many of the factors that are important in virulence inside a host are also important when the organism is exposed to diverse and unpredictable environmental conditions outside of the host. Both environmental and cellular characteristics determine the fate and transport of microbial cells following deposition in soil or waters. Isolates that will be transported to water and sediments are the subset of the population that will be used for monitoring purposes. Therefore, the validity of the indictor paradigm and the feasibility of microbial source tracking can not be fully evaluated without a better understanding of the ecology of this important organism. The goals of this study were to (1) evaluate the diversity of E. coli in manures from livestock and stream-water samples taken following dry and wet weather events; (2) evaluate the effect of strain level differences on the attachment and transport of E. coli. and; (3) compare the concentration of E. coli present in water samples taken following wet or dry weather events to that of other indicator groups (Bacteroides, enterococci, clostridia). To evaluate diversity, 1346 E. coli isolates were obtained from poultry, swine and dairy manures and from seventeen stream-water samples taken from the Bacon Creek Watershed located in western Kentucky. Bacon Creek is on the EPA 303(d) list of impaired streams for pathogen presence. The predominant land use within the 90.5 square mile watershed is agricultural, but there are also surrounding rural communities with straight-pipes or septic systems. Samples were collected from the same locations following one dry weather event (n = 9; 72 hours without any rain) and one wet weather event (n = 8; 72 hrs of no rain followed by enough rainfall to cause runoff that reaches the stream). Slurry and litter samples (10 g or 10 mL) were plated onto selective media. Stream-water samples (0.5ml, 5.0ml, and 10.0ml) were filtered and placed onto selective media. Strain diversity among the 1346 E. coli isolates was evaluated by BOX-PCR analysis. E. coli source sub-group isolates were evaluated for the presence of genes associated with adhesion (afa/draBC, iha, agn43, eaeA and fimH), toxin production (hlyA, stx1, stx2), capsular polysaccharide synthesis (kpsMTII) and siderophores (iroNE.coli, chuA). Attachment efficiencies to quartz sand were calculated for 23 E. coli isolates following transport through saturated porous media. Concentrations of indicator groups were measured by quantitative, real-time PCR (qPCR). Richness of genotype profiles for livestock samples was relatively low (25, 12 and 11 for swine, poultry and dairy, respectively) compared to that of E. coli isolates from stream-water following dry or wet weather events (115 and 126, respectively). Genotype profiles for E. coli isolates from stream-water clustered with isolates from livestock species; however, over 34% of E. coli isolates from stream-water had genotype profiles that were distinct from those of the tested livestock species. Furthermore, only 18% of the 84 E. coli isolates from the wet and dry events clustered together, suggesting a high degree of temporal diversity. Genes associated with virulence (adhesions, toxins and siderophores) were present in E. coli isolates from all sources. The most commonly detected genes were the adhesions fimH (present in 80% to 95% of isolates) and agn43 (present in 40% to 100% of isolates). Bacterial attachment efficienci