The role of molecular methods in the detection of pathogens in food

The research described here addresses several issues associated with testing food samples for the presence of pathogens. In the first study, we developed and evaluated a combined concentration/molecular detection method for the direct detection of foodborne pathogens (Listeria monocytogenes and Salmonella enterica serovar enteritidis) directly from complex dairy matrices, bypassing the need for cultural enrichment. PCR detection limits of 103 and 101 CFU per 11 g sample were achieved for L. monocytogenes and serovar. Enteritidis, respectively, in mild cheddar cheese and non-fat yogurt without prior cultural enrichment. In a related study, commercially available proprietary kits (MICROB Enrich® and MICROBExpress ®) for nucleic acid purification were evaluated to improve the direct detection of Listeria monocytogenes from a frankfurter matrix. PCR detection limits were 105 CFU/11g sample and RT-PCR detection limits were 103 CFU/11g. Detection limits were improved an additional 10-fold (to 102 CFU/11g) when extracted RNA was further purified using MICROBExpress®. In this study, amplification via RT-PCR yielded more sensitive detection that DNA based amplification and further purification of RNA extracts was beneficial for improving RT-PCR detection limits. Direct detection of pathogens from food matrices was achieved without cultural enrichment with detection limits approaching those of other “rapid” methods. Sample manipulations during sample preparation, nucleic acid extraction and amplification steps may aide in achieving desired detection limits. In the second study, automated ribotyping using the RiboPrinter® microbial characterization unit was evaluated for its ability to differentiate thirty-nine isolates of multi-drug resistant Salmonella enterica serovar Typhimurium. The strains could not be differentiated from one another using ribotyping, suggesting that in this case, phenotypic methods may be more discriminatory. Ribotyping data may best be used in conjunction with phenotypic and serological testing results in order to adequately differentiate isolates. In the final study, a formal decision analysis model was designed for a food-borne pathogen testing decision from a food processor's perspective. In general, the model indicated that testing for highly prevalent pathogens may provide an improvement in food safety but end product testing for pathogens of low prevalence should be carefully considered and may not be justified due to limited return on investment.