Detection of enteric viruses in East Tennessee public ground water systems

by 1979- Johnson, Trisha Baldwin

A two-part study was conducted by University of Tennessee-Knoxville, the U.S. Geological Survey, and the U.S. Environmental Protection Agency-National Exposure Research Laboratory to (1) develop, validate, and test a real-time reverse transcriptionpolymerase chain reaction (real-time RT-PCR) assay for enteroviruses in ground water samples and to (2) perform the first survey of enteric viral occurrence in the karst aquifers of East Tennessee. Karst aquifers are expected to have a high susceptibility to viral contamination because of the rapid flow (100’s of m/day) and frequent occurrence of fecal indicator bacteria typically observed in these systems. Real-time RT-PCR primers and probes specific for enteroviruses were developed and tested at the University of Tennessee’s Center for Environmental Biotechnology (UTCEB). The real-time RT-PCR assay was validated using co-standards: attenuated poliovirus and a DNA plasmid constructed at UTCEB (cDNA to the attenuated poliovirus). The assay was confirmed to have good PCR efficiency, reproducibility, and sensitivity. The real-time RT-PCR assay was quantitative over 6 orders of magnitude and had low minimum detection limits (0.5 plaque forming units (PFU) of the attenuated poliovirus per reaction and 10 copies of the DNA plasmid per reaction). In the field study, eight wells and springs used as raw water sources for East Tennessee public ground water systems were sampled between March and August of 2004. The wells and springs were sampled one to two times under baseflow conditions. The ground water samples were tested for enteroviruses and reoviruses by cell culture methods (total culturable viruses), enteroviruses and reoviruses by conventional RT-PCR, enteroviruses by the real-time RT-PCR assay developed at UTCEB, fecal indicator bacteria (E. coli and Bacteroides), total coliforms, and physical and chemical waterquality parameters. The wells and springs were chosen on the basis of prior monitoring of E. coli and geochemical parameters, their hydrogeologic settings, and the presence or absence of likely input sources of fecal contamination to the ground water supplies. Four sites were designated as “high risk” for fecal contamination and four sites were designated as “low risk” for fecal contamination. “High risk” sites were expected to have higher occurrences and concentrations of enteric viruses as well as other indicators of fecal contamination, such as Bacteroides and E. coli, than “low risk” sites. The major results of the field study were: (1) 88% of the wells and springs sampled were positive for culturable viruses (concentrations ranged from 2 MPN/100 mL to 156 MPN/100 mL), (2) 75% of the wells and springs were positive for at least one of the indicator organisms, (3) None of the wells or springs were positive for enteric viruses using the conventional RT-PCR or real-time RT-PCR methods, and (4) “High risk” sites had more frequent detections of enteric viruses and indicator bacteria than “low risk” sites. However, only total coliform concentrations were statistically different (higher) between “high risk” and “low risk” sites. A statistically significant positive correlation was found between total culturable virus concentrations and total coliform concentrations. Of the fecal indicators, Bacteroides had the highest co-occurrence with enteric viruses. iv