Institution: University of Georgia
Principal Investigator:
Nikki Shariat, Ph.D.
University of Georgia
Poultry Diagnostic and Research Center
Athens GA, 30602
Vaccine programs are a primary pre-harvest Salmonella mitigation strategy, and development of effective programs rely heavily on effective Salmonella serovar surveillance. Noteworthy hurdles to surveillance and vaccine program development includes the fact that traditional isolation identifies only the most abundant serovars in a population, while underlying serovars remain unknown. Further, there is a lack of understanding where in the supply chain samples should be taken to inform serovars present in the system.
This completed project encompassed two objectives. The first objective was to use deep serotyping by CRISPR-SeroSeq, a molecular tool that maps relative Salmonella serovar frequencies to identify all serovar populations present in surveillance samples taken throughout a turkey production chain. The second objective was to use CRISPR-SeroSeq data to resolve significant differences observed in serovar prevalence between ground turkey and mechanically separated turkey (MST) at processing.
Findings from objective 1 revealed that on-farm Salmonella did not correlate with prevalence at processing, with less than half of the flocks that were positive at the plant being positive on the farm. Salmonella prevalence was reduced 3-fold during pre-chill interventions. Higher prevalence in MST and ground turkey compared to pre-chill samples suggests the release of internalized Salmonella during grinding. This is supported by deep serotyping that showed that in most flocks there was at least one serovar found in product (MST or ground turkey) that was not found upstream.
Objective 2 results indicated that Salmonella prevalence was higher in MST than in ground turkey samples, but the difference was not statistically significant across all flocks. However, deep serotyping with CRISPR-SeroSeq showed that MST product contained a larger number of Salmonella serovars on a per sample basis than ground turkey. A third of MST samples contained multiple serovars, while less than one in 10 ground turkey samples contain more than one serovar.
This is the first high-resolution study of Salmonella in turkey that investigates serovar populations through the supply chain. Completion of this work highlighted the complexity of this pathogen in turkey production and processing. The study demonstrates flocks are likely contaminated with multiple serovars, and effective surveillance should include samples taken at processing where prevalence and diversity are highest (pre-scald, MST). Tracking whole Salmonella serotype populations from grow-out farms to finished products provided a better understanding of Salmonella diversity, but more importantly, has highlighted that, 1) relying solely on USDA finished product sampling is not sufficient to provide informative feedback to breeders/hatcheries regarding serovars in the supply chain, 2) product from different bird parts can harbor different serovars, thus multiple product types should be sampled for surveillance purposes, and, 3) farm booties collected at the end of grow-out are not an effective sample to understand serovars present in a flock prior to slaughter. There is a need for alternative sample types or sampling times to capture Salmonella, as the sampling performed here was not indicative of the Salmonella found down-stream.
