Contamination of beef by certain strains of Shiga toxin-producing Escherichia coli (STEC) called enterohemorrhagic E. coli (EHEC) can lead to outbreaks of severe disease. Therefore, accurate monitoring tests are needed to identify high risk beef products and divert them from consumers. Most EHEC testing focuses on the detection of their key virulence factors Shiga toxin (stx) and intimin (eae). However, these two factors can occur separately in lower risk nonpathogenic E. coli (STEC and enteropathogenic E. coli; EPEC) and confound testing if both are present. Accessory virulence factors like the Type III secreted effectors espK and espV may aid in increasing the specificity of EHEC testing. This work first evaluated collections of EHEC (n = 83), STEC (n = 100) and EPEC (n = 95), finding espK and/or espV in 100%, 0%, and 60% of each, respectively. Next, an inoculation study of beef trim samples (n = 118) examined the ability of including espK and espV in the monitoring test scheme to distinguish samples inoculated with EHEC from those inoculated with mixtures of STEC and EPEC (non-EHEC). Test accuracy was calculated as Area Under the Receiver Operating Characteristic curve (AUC) and found to be significantly (p < 0.05) different, increasing from 68.0% (stx/eae) to 76.8% by including espK and espV. Finally, 361 regulatory agency beef samples that had been identified as suspect for EHEC (stx+/eae+) were examined with the addition of espK and espV, and results compared to culture isolation. Culture isolation identified 42 EHEC, 82 STEC, and 67 EPEC isolates in 146 of the samples. In the case of these naturally contaminated samples, inclusion of espK and espV increased test accuracy compared to culture isolation from an AUC of 50.5% (random agreement) to 69.8% (good agreement). Results show that the inclusion of espK and espV can increase the specificity of identifying high risk EHEC contaminated beef and release beef contaminated with nonpathogenic or low risk E. coli. Further, use of espK and espV identified samples contaminated by common EHEC of serogroups O157, O26, and O103, as well as of less common serogroups O182, O177, and O5.
Read full abstract