ALTERNATIVE APPROACHES IN MOLECULAR CHARACTERIZATION OF FOODBORNE PATHOGENS: SHIGA TOXIN-PRODUCING Escherichia coli AND Salmonella SEROTYPES

Abstract

Shiga toxin-producing E. coli (STEC) and Salmonella enterica subspecies enterica (S. enterica) are two major foodborne pathogens. They cause almost 1.5 million of cases of disease each year in the US. Due to their public health impact, development of new methods for their detection and identification are top priority. This research focused on identifying alternative molecular methods and markers for the identification of STEC and Salmonella.

First, a suspension array was developed to simultaneously identify the seven most prevalent STEC (O26, O45, O103, O111, O121, O145, and O157) in the US. The panel targeted genes wzx or wzy and Shigatoxin genes. Testing and optimization employed four to eleven isolates of each serotype in the panel. STEC fluorescence values were 30 to >270 times greater than those of negative controls, demonstrating the method's effectiveness for the molecular serotyping of STEC.

STEC strains (n=194) of 43 serotypes were examined for clustered regularly interspaced short palindromic repeats (CRISPR) arrays to study relatedness among serotypes. A subset of strains (n=81) was analyzed for cas and virulence genes to determine a possible relationship. CRISPR spacer content correlated well with serotypes, although some strains with different serogroup but the same H type shared identical arrays (O26:H11, O103:H11, and O111:H11). cas and virulence genes were not associated, but strains with greater probability of causing outbreaks and disease showed fewer spacers than those less likely to cause them (p<0.05). Therefore, CRISPR array content correlated well with STEC serotype, and CRISPR-cas systems were inversely related to strain virulence potential.

Finally, the CRISPR arrays of 221 S. enterica of 53 serotypes were analyzed to define their relationship. CRISPR-cas systems of 50 S. enterica serotype Bareilly (S. Bareilly) were analyzed to resolve intra-serotype variations. CRISPR arrays correlated well with serotypes, although some serotypes displayed more than one type of array (e.g. S. Bareilly). Additionally, CRISPR-cas system elements reflected S. Bareilly phylogeny, but the array content was not linked to food vehicle or isolate's geographical origin. In conclusion, CRISPR array are useful for designing molecular serotyping assays, but a range of strains should be included to account for variation in S. enterica.