Comparison of culture, PCR and immunoassays for detecting Escherichia coli O157 following enrichment culture and immunomagnetic separation performed on naturally contaminated raw meat products

Abstract

The aims of this study were (i) to evaluate the specificity and sensitivity of three previously described PCR assays for the detection of Escherichia coli O157 and, (ii) to compare PCR, culture, and two visual immunoassays (VIAs), BioSign and Path-Stik, for detecting E. coli O157 after enrichment culture and immunomagnetic separation (IMS) performed on various naturally contaminated raw beef, lamb and mixed meat products. Twelve sorbitol nonfermenting (SNF) verocytotoxin-producing (VT +) E. coli O157, 6 SNF VT - E. coli O157, 4 sorbitol fermenting (SF) VT + E. coli O157, 3 SF VT - E. coli O157, 23 E. coli belonging to 17 other serogroups and 12 organisms of other species were used to check the specificity of PCR reactions. Only one primer pair generated amplimers only with E. coli O157 and was used for all subsequent work. After enrichment culture and on inoculated minced beef samples, PCR was as sensitive as culture for detecting 9 of the 12 strains of E. coli O157, but up to 4 log10 more sensitive than culture for detecting three strains. Of the 120 samples of naturally contaminated meat products examined, 80 (67%) were positive by PCR, 70 (58%) were positive by BioSign, 69 (58%) were positive by culture and 67 (56%) were positive by Path-Stik. Although both VIAs lacked sensitivity when compared to PCR, both compared favourably with culture and both were extremely rapid and easy to perform, giving a result in less than 15 min. Eleven samples were positive by PCR and both VIAs, but negative by culture because culture plates were heavily overgrown with SF organisms, making detection of any E. coli O157 present impossible.