Behavior of Shiga Toxigenic Escherichia coli Relevant to Lettuce Washing Processes and Consideration of Factors for Evaluating Washing Process Surrogates

Postharvest processes for fresh produce commonly include washing in water containing antimicrobial chemicals, such as chlorine; however, if the antimicrobials are not present in sufficient levels, washing can promote the spread of contamination that might be present. To understand cross-contamination risk during washing, we tested a collection of Shiga toxigenic Escherichia colt (STEC), including O157:H7 and other non-O157 strains, for certain traits during washing of fresh-cut lettuce, i.e., sensitivity to sublethal chlorine levels and ability to cross-contaminate (detach from and attach to) lettuce in the presence of sublethal chlorine levels. Nonpathogenic E. colt Nissle 1917 (EcN) and Pediococcus pentosaceus lactic acid bacterial species (LAB) were included as potential washing process validation surrogates. As measured by extension of the lag phase of growth in media containing 0.15 ppm of chlorine, chlorine sensitivity varied among the STECs. Cross-contamination was assessed by evaluating transfer of bacteria from inoculated to uninoculated leaves during washing. Without chlorine, similar transfer to wash water and uninoculated leaves was shown. In 1 ppm of chlorine, cross-contamination was not detected with most strains, except for the substantial transfer by a STEC 0111 strain and EcN in some replicates. Strain 0111 and EcN showed less inactivation in 0.25 ppm of chlorine water compared with O157 (P < 0.05). LAB showed similar transfer and similar chlorine inactivation to O157. Considering together the sublethal chlorine sensitivity and detachment/attachment traits, neither EcN nor LAB displayed optimal characteristics as washing process surrogates for the STEC strains, although further evaluation is needed. This work demonstrated a range of behaviors of STEC strains during lettuce washing and may be helpful in hazard characterization, identifying factors to consider for evaluating washing process efficacy, and identifying phenotypic traits to select surrogates to validate washing processes.