ABSTRACT
Three methods to enumerate Escherichia coli in food were compared. They were based on AOAC methods using lauryl tryptose broth (LST) medium, LST-4-methylumbelliferyl-beta-D-glucuronide (MUG) medium, and a proposed method using regular LST in combination with E. coli (EC)-MUG medium. An efficacious and cost-effective method is needed that can detect E. coli and does not produce false presumptive positives. We tested 170 cheeses, 40 frozen processed seafood samples, 210 tree nuts, and 40 other samples. The method of choice for enumerating E. coli depends on the commodity itself. For a product, such as hard cheese or processed seafood, with a history of being negative for E. coli and other lactose-fermenting organisms, the proposed method using regular LST/EC-MUG is a good choice. These samples were seldom presumptive positive in the primary LST medium. If gas was produced, EC-MUG was an effective secondary medium. No false positives (fluorescence) or negatives were detected in EC-MUG medium. For a product with a history of being positive for E. coli and/or other lactose fermenting organisms, such as tree nutmeats or cheeses that are ripened by bacteria or mold, the method using LST-MUG is the method of choice. A presumptive positive in the LST-MUG medium was highly correlative with the biochemical tests that confirmed a sample contain E. coli. For samples spiked with E. coli, the results from each of these 3 methods were identical, and were consistent in enumerating E. coli.
