Genotypic characterization of bovine Shiga toxin-producing Escherichia coli (STEC) strains with persistent or sporadic colonization types

Question: Certain STEC strains not yet identified as human pathogenic EHEC strains can persist in cattle at herd or even at single animal level over several months and may serve as gene reservoir during the genesis of highly virulent zoonotic pathogens. Reduction of human risk for acquiring EHEC infections thus particularly requires strategies to control STEC strains persisting in cattle. We therefore aimed at identifying gene patterns associated with the STEC colonization types in the bovine host. Methods: We monitored 4 cattle farms over 28 months and isolated 85 persistent (pSTEC, shedding ≥ 4 months) and 72 sporadically colonizing STEC (sSTEC, shedding ≤ 2 months) strains. Additionally, 16 STEC from sampling a cohort of calves were included. Genoserotype and MLST were determined for all 173 strains by whole genome sequencing and the strains were probed with the „E. coli Genotyping“ DNA microarray (Alere Technologies GmbH, Germany). Results: While all pSTEC belonged to only 4 genoserotypes (O26:H11, O156:H25, O165:H25, O182:H25), 28 genoserotypes were present in sSTEC with O157:H7 being the most prevalent. The microarray analysis identified 135 unique gene patterns based on the occurrence or absence of 104 virulence-associated genes (VAGs) or VAG-variants. The VAG-patterns clustered with the genoserotypes and MLST types of the strains. Comparing 122 strains with known colonization type and only one representative of each VAG-pattern, pSTEC possessed significantly more often the genes astA, stx1, eae, lpfA, efa-1/lifA, espB, espJ, nleA, nleB, and nleC (Chi², p ≤ 0.01), sSTEC more often cdtB, stx2, and toxB (Chi², p < 0.01). Conclusions: Microarray analysis including only known VAGs did not allow identification of gene profiles characteristic for a persistent colonization type in bovine STEC. Analysis of the sequencing data is currently underway to differentiate the genomic backgrounds more clearly. This study is funded by the German Research Foundation (DFG).