AUTHOR=Gao Ruimin , Duceppe Marc-Olivier , Chattaway Marie Anne , Goodridge Lawrence , Ogunremi Dele 

TITLE=Application of prophage sequence analysis to investigate a disease outbreak involving Salmonella Adjame, a rare serovar and implications for the population structure

JOURNAL=Frontiers in Microbiology

VOLUME=Volume 14 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/microbiology/articles/10.3389/fmicb.2023.1086198

DOI=10.3389/fmicb.2023.1086198

ISSN=1664-302X

ABSTRACT=The outbreak investigation of foodborne salmonellosis is hindered when the source is contaminated by multiple strains of Salmonella, creating difficulties matching an incriminated organism recovered from patients with the specific strain in the suspect food.  An outbreak of the rare Salmonella Adjame was found to be caused by multiple strains of the organism as demonstrated by variation in the single nucleotide polymorphism (SNP) profile.  For this reason, the outbreak was deemed atypical and investigation was further hindered by the failure to recover the organism from potential food sources.  To improve the investigation of foodborne salmonellosis by ensuring accurate strain characterization that can be applied to both food and clinical sources and enable precise matching, we applied a previously described highly discriminatory, prophage sequence typing (PST) pipeline to evaluate the strain composition of the atypical S. Adjame outbreak. The PST analysis of the S. Adjame isolates showed a high degree of strain heterogeneity. We observed small clusters made up of 2-6 isolates (n = 27) and singletons (n = 11) in stark contrast with the three clusters observed by SNP analysis.  In total, we detected 24 prophages of which only four were highly prevalent among the S. Adjame strains, namely Salmon_SEN34 (36/38 strains), Entero_p88 (35/38 strains), Burkho_phiE255 (33/38 strains) and Edward_GF (28/38 strains). Despite the marked strain diversity seen with prophage analysis, the distribution of the four most common prophages matched the clustering observed using core genome.  A single nucleotide change in the conserved Entero-p88 and Burkho-phiE255 prophages correlated with the two clusters as defined by their core genomes, shedding light on the evolutionary relationships among the related Adjame strains and illustrating a convergence of variation in both the core and accessory genomes. We conclude that complementary core and accessory genomes analyses should be adopted in foodborne bacteria outbreak investigations to provide a more accurate strain description to facilitate the reliable matching of isolates from patients and incriminated food sources.  The outcomes can be expected to translate to a better understanding of the microbial population structure and an improved source attribution in foodborne illnesses.