Evaluation of Molecular Methods for Serotyping Shigella flexneri
Gentle A., Ashton PM., Dallman TJ., Jenkins C.
<jats:p><jats:named-content content-type="genus-species">Shigella flexneri</jats:named-content>can be phenotypically serotyped using antisera raised to type-specific somatic antigens and group factor antigens and genotypically serotyped using PCR targeting O-antigen synthesis or modification genes. The aim of this study was to evaluate a real-time PCR for serotyping<jats:named-content content-type="genus-species">S. flexneri</jats:named-content>and to use whole-genome sequencing (WGS) to investigate the phenotypic and genotypic serotype identifications. Of the 244 cultures tested retrospectively, 226 (92.6%) had concordant results between phenotypic serotyping and PCR. Seventy of the 244 isolates (including 15 of the 18 isolates where a serotype-PCR mismatch was identified) were whole-genome sequenced, and the serotype was derived from the genome. Discrepant results between the phenotypic and genotypic tests were attributed to insertions/deletions or point mutations identified in O-antigen synthesis or modification genes, rendering them dysfunctional; inconclusive serotyping results due to nonspecific cross-reactions; or novel genotypes. Phylogenetic analysis of the WGS data indicated that the serotype, regardless of whether it was phenotypically or genotypically determined, was a weak predictor of phylogenetic relationships between strains of<jats:named-content content-type="genus-species">S. flexneri</jats:named-content>. WGS data provided both genome-derived serotyping, thus supporting backward compatibility with historical data and facilitating data exchange in the community, and more robust and discriminatory typing at the single-nucleotide-polymorphism level.</jats:p>