Detection methods for human enteric viruses in representative foods

Although viral foodborne disease is a significant problem, foods are rarely tested for viral contamination, and when done, testing is limited to shell fish commodities. In this work, we report a method to extract and detect hu man enteric viruses from alternative food commodities using an elution-conc entration approach followed by detection using reverse transcription-polyme rase chain reaction (RT-PCR). Fifty-gram lettuce or hamburger samples were artificially inoculated with poliovirus type 1 (PV1), hepatitis A virus (HA V), or the Norwalk virus and processed by the sequential steps of homogeniz ation, filtration, Freon extraction (hamburger), and polyethylene glycol (P EG) precipitation. To reduce volumes further and remove RT-PCR inhibitors, a secondary PEG precipitation was necessary, resulting in an overall 10- to 20-fold sample size reduction from 50 g to 3 to 5 ml. Virus recoveries in secondary PEG concentrates ranged from 10 to 70% for PV1 and 2 to 4% for HA V as evaluated by mammalian cell culture infectivity assay. Total RNA from PEG concentrates was extracted to a small volume (30 to 40 mu1) and subject ed to RT-PCR amplification of viral RNA sequences. Detection limit studies indicated that viral RNA was consistently detected by RT-PCR at initial ino culum levels greater than or equal to 10(2) PFU/50-g food sample for PV1 an d greater than or equal to 10(3) PFU/50-g food sample for HAV. In similar s tudies with the Norwalk virus, detection at inoculum levels greater than or equal to1.5 x 10(3) PCR-amplifiable units/50-g sample for both food produc ts was possible. All RT-PCR amplicons were confirmed by subsequent Southern hybridization. The procedure reported represents progress toward the devel opment of methods to detect human enteric viral contamination in foods othe r than shellfish.