|GRIFFING, SEAN - Centers For Disease Control And Prevention (CDC) - United States|
|MACCANNELL, DUNCAN - Centers For Disease Control And Prevention (CDC) - United States|
|SCHMIDTKE, AMBER - Centers For Disease Control And Prevention (CDC) - United States|
|FREEMAN, MOLLY - Centers For Disease Control And Prevention (CDC) - United States|
|HYYTIA-TREES, EIJA - Centers For Disease Control And Prevention (CDC) - United States|
|GERNER-SMIDT, PETER - Centers For Disease Control And Prevention (CDC) - United States|
|RIBOT, EFRAIN - Centers For Disease Control And Prevention (CDC) - United States|
|Bono, James - Jim|
Submitted to: PLOS ONE
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/11/2015
Publication Date: 7/1/2015
Publication URL: http://handle.nal.usda.gov/10113/61112
Citation: Griffing, S.M., Maccannell, D.R., Schmidtke, A.J., Freeman, M.M., Hyytia-Trees, E., Gerner-Smidt, P., Ribot, E.M., Bono, J.L. 2015. Canonical single nucleotide polymorphisms (SNPs) for high-resolution subtyping of Shiga-toxin producing Escherichia coli (STEC) O157:H7. PLoS One. 10(7):e0131967. DOI:10.137/journal.pone.0131967.
Interpretive Summary: Shiga toxin-producing Escherichia coli O157:H7 (STEC O157) recently emerged as a significant cause of diarrhea, hemorrhagic colitis, and hemolytic uremic syndrome throughout much of the world. Many outbreaks have been linked to the consumption of beef products or produce contaminated with the bacteria. Current outbreak investigational methods used to determine relatedness between E. coli cases take considerable time, human resources, and sometimes lack sufficient epidemiological resolving power. Single nucleotide polymorphisms (SNPs) are small DNA mutations that may facilitate better E. coli outbreak investigations by decreasing the time and manpower necessary to reach results, while simultaneously increasing their resolving power. We identified new SNPs and combined them with previously discovered SNPs to create a small, highly discriminatory panel for differentiating E. coli cases. We then compared this panel with current methods used in epidemiological laboratories. While the SNP assay had the same discriminatory power as current testing, it was unable to resolve some outbreak clusters identified by current methods. More E. coli SNPs need to be discovered in the unresolved outbreak clusters to reach or surpass the utility of existing methods.
Technical Abstract: The objective of this study was to develop a canonical SNP panel for subtyping of Shiga-toxin producing Escherichia coli (STEC). To this purpose, 906 putative SNPs were identified using resequencing tiling arrays. A subset of 391 SNPs was further screened using high-throughput TaqMan PCR against a diversity set of 177 STEC isolates. Another 30 SNPs identified from the literature were also screened. Validated SNPs were reduced to a minimum, parsimoniously-informative set with the goal to generate clustering consistent with epidemiological, PFGE, and MLVA data. Another round of SNP discovery was undertaken using comparative genomic resequencing of pooled DNA from clusters with inadequate resolution. There was considerable linkage disequilibrium at the population level between the 421 SNPs. A set of 32 SNPs excluded almost all of the 391 initial SNPs, the additional 30 SNPS, and the majority of SNPs reported by others and partitioned the diversity set in a manner consistent with epidemiological data, but with insufficient resolution in 10 of 47 clusters. The discriminatory power of the panel versus PFGE and MLVA were similar, but the number of subtypes varied widely between the methods. A second round of SNP discovery identified 4,051 potential SNPs in the 10 clusters. We concluded 1 of the 10 clusters was incorrectly grouped, reducing the SNP set to 2,891 in the remaining 9 clusters. Of these, 2,828 SNPs were previously unreported and 435 occurred across multiple clusters. Among highly clonal bacteria like STEC O157:H7,linkage disequilibrium greatly limits the number of parsimoniously informative SNPs. Our direct interrogation of 421 loci and indirect interrogation of numerous others reported in the literature suggests other SNPs should be vetted to establish a canonical panel sufficient for routine subtyping, including the 435 noted in our work.