R-054. Use of a Genotyping DNA Microarray Representing Diverse Pathotypes of E.coli and Shigella sp for Strain Identification and Discrimination Between and Within Closely Related Species

S. A. Jackson, I. Patel, M. K. Mammel, T. Mays, J. E. LeClerc, T. A. Cebula;
US FDA, Laurel, MD.

Background: Illnesses associated with the consumption of foods contaminated with the pathogens Escherichia coli and Shigella result in thousands of hospitalizations and hundreds of deaths annually throughout the world. The ability of these pathogens to rapidly adapt to novel environmental niches necessitates highly parallel analysis methods in order to accurately assess their genomic diversity and genomic content. An important issue is the degree of diversity present among the individual clinical and environmental strains associated with a foodborne outbreak. Previous microarray approaches have been limited in their ability to measure genome content and accurately represent the genomic diversity due to limited availability of sequence information from any given species. Methods: A custom Affymetrix microarray, representing 32 diverse E. coli and Shigella whole genome sequences as well as 46 related plasmid sequences, was developed and validated by analysis of sequenced strains. Using this array, we interrogated a collection of E.coli and Shigella strains that were associated with temporally and geographically diverse outbreaks, including the two recent outbreaks of E. coli O157:H7 linked to spinach and lettuce. Additionally, the ECOR collection was analyzed in order to compare and contrast the diversity of outbreak strains to the global diversity of this species. Results/Conclusions: This array design allowed us to accurately detect each gene target represented on the array. We were able to differentiate closely related strains within a pathotype. While most of the isolates of E. coli O157:H7 associated with the spinach outbreak were indistinguishable, several clinical isolates contained apparent chromosomal rearrangements. These results demonstrated the potential of this array for use in evolutionary, epidemiological, and forensics studies for strain identification and discrimination.