RESUMEN
Enteric fever is caused by three Salmonella enterica serovars: Typhi, Paratyphi A, and Paratyphi B sensu stricto Although vaccines against two of these serovars are licensed (Typhi) or in clinical development (Paratyphi A), as yet there are no candidates for S. Paratyphi B. To gain genomic insight into these serovars, we sequenced 38 enteric fever-associated strains from Chile and compared these with reference genomes. Each of the serovars was separated genomically based on the core genome. Genomic comparisons identified loci that were aberrant between serovars Paratyphi B sensu stricto and Paratyphi B Java, which is typically associated with gastroenteritis; however, the majority of these were annotated as hypothetical or phage related and thus were not ideal vaccine candidates. With the genomic information in hand, we engineered a live attenuated S. Paratyphi B sensu stricto vaccine strain, CVD 2005, which was capable of protecting mice from both homologous challenge and heterologous challenge with S. Paratyphi B Java. These findings extend our understanding of S. Paratyphi B and provide a viable vaccine option for inclusion in a trivalent live attenuated enteric fever vaccine formulation.IMPORTANCE We developed a live attenuated Salmonella enterica serovar Paratyphi B vaccine that conferred protection in mice against challenge with S Paratyphi B sensu stricto and S Paratyphi B Java, which are the causes of enteric fever and gastroenteritis, respectively. Currently, the incidence of invasive S. Paratyphi B sensu stricto infections is low; however, the development of new conjugate vaccines against other enteric fever serovars could lead to the emergence of S. Paratyphi B to fill the niche left by these other pathogens. As such, an effective S. Paratyphi B vaccine would be a useful tool in the armamentarium against Salmonella infections. Comparative genomics confirmed the serovar-specific groupings of these isolates and revealed that there are a limited number of genetic differences between the sensu stricto and Java strains, which are mostly hypothetical and phage-encoded proteins. The observed level of genomic similarity likely explains why we observe some cross-protection.