Method validation for norovirus detection in naturally contaminated irrigation water and fresh produce

Human enteric viruses are shed in extremely high numbers in the feces of infected individuals, becoming environmental contaminants and eventually leading to contamination of a variety of foodstuffs at the pre-harvest stage. Among these foods at risk is fresh produce and irrigation water is a major vehicle for crop contamination. In the present study, a standardized molecular method for quantitative virus assay has been evaluated in different types of fresh produce and in irrigation water for human norovirus (NoV) detection. Two different virus concentration procedures, polyethylene-glycol precipitation (PEG) and organic flocculation (OF), were employed. The procedures were initially assayed in spiked samples and later validated on naturally contaminated samples from the Nile Delta in Egypt. Overall, PEG provided significantly (p < 0.05) better virus recoveries than OF for both irrigation water and salad vegetable virus analysis. NOV GI and GII recoveries in spiked irrigation water ranged from 22.0% to 43.3% and from 12.6% to 16.4% with the PEG and OF methods, respectively. In experimentally contaminated salad vegetables, virus recoveries ranged from 28.0% to 48.0% and from 14.0% to 18.8% by PEG precipitation and OF, respectively. Using PEG precipitation, NoV was found in 31.9% of naturally contaminated irrigation water samples. Both NoV GI and GII were detected in these samples with genome copy numbers of around 10(2) per liter. Virus analyses performed in naturally contaminated fresh produce that included green onion, watercress, radish, leek, and lettuce show that NOV GI was present in 20.8%-34.0% of the samples with genome copy numbers of around 10(2) per gram. When OF was employed, NoV was found in 25.0% of the irrigation water samples. Both genogroups could be found in these samples with genome copy numbers of around 10 per liter. In fresh produce, GI was present in 16.0%-25.7% of the samples with genome copy numbers per gram of around 10. Surprisingly, Nov GII was not detected in any salad vegetable despite highly satisfactory virus/nucleic acid extraction and enzyme efficiencies reported in the assays. Available reliable standardized assays for virus detection in food matrices including appropriate quality assurance and quality control measures to assess the efficiency of critical steps in virus analysis open the possibility to produce consistent and accurate exposure data to be used in QMRA (quantitative microbial risk assessment) and at the same time may enable the formulation of guidelines to ensure the virological quality of selected commodities in specific scenarios to reduce the risk of foodborne virus infections. (C) 2013 Elsevier B.V. All rights reserved.