Population Analyses Reveal Preenrichment Method and Selective Enrichment Media Affect Salmonella Serovars Detected on Broiler Carcasses.

Poultry is a major Salmonella reservoir, but conventional culture-based methods typically identify the most abundant serovars while those less abundant remain undetected. Choice of enrichment procedure also introduces bias, and for broiler carcasses, a 1-min rinse before preenrichment is insufficient to release all Salmonella present. The inability to assess serovar diversity means that serovars more often associated with human illness may be masked by more abundant Salmonella. CRISPR-SeroSeq (serotyping by sequencing clustered regularly interspaced short palindromic repeats), an amplicon-based, next-generation sequencing tool, allows detection of multiple serovars and maps the relative serovar frequencies in a sample. To address the preceding limitations, CRISPR-SeroSeq was used on broiler carcasses collected prechilled at a commercial plant. Standard carcass rinse aliquot preenrichments and whole carcass preenrichments that were enriched in Rappaport-Vassiliadis (RV) and tetrathionate (TT) broths were compared. On average, five serovars were observed per carcass, including nine on one carcass. CRISPR-SeroSeq detected serovars comprising as little as 0.005% of the population. CRISPR-SeroSeq data matched (28 of 32) standard culture analysis for abundant serovars. Salmonella serovars Kentucky, Typhimurium, and Schwarzengrund were found on each carcass. Overall, serovar diversity was higher in whole carcass preenrichments that were enriched in RV (P < 0.05). Serovar Schwarzengrund was present at higher frequencies in whole carcass preenrichments compared with rinse aliquot preenrichments (t test, P < 0.05), suggesting it adheres more strongly to the carcass. Salmonella serovar Enteritidis was enriched eightfold more in TT than in RV, and serovars Schwarzengrund and Reading were preferentially enriched in RV. Comparison of preenriched and enriched samples suggests that selective enrichment in RV or TT was inhibitory to some serovars. This article addresses limitations of Salmonella surveillance protocols and provides information related to Salmonella population dynamics.