SOLiD sequencing of four Vibrio vulnificus genomes enables comparative genomic analysis and identification of candidate clade-specific virulence genes

Abstract

Background: Vibrio vulnificus is the leading cause of reported death from consumption of seafood in the United States. Despite several decades of research on molecular pathogenesis, much remains to be learned about the mechanisms of virulence of this opportunistic bacterial pathogen. The two complete and annotated genomic DNA sequences of V. vulnificus belong to strains of clade 2, which is the predominant clade among clinical strains. Clade 2 strains generally possess higher virulence potential in animal models of disease compared with clade 1, which predominates among environmental strains. SOLiD sequencing of four V. vulnificus strains representing different clades (1 and 2) and biotypes (1 and 2) was used for comparative genomic analysis.

Results: Greater than 4,100,000 bases were sequenced of each strain, yielding approximately 100-fold coverage for each of the four genomes. Although the read lengths of SOLiD genomic sequencing were only 35 nt, we were able to make significant conclusions about the unique and shared sequences among the genomes, including identification of single nucleotide polymorphisms. Comparative analysis of the newly sequenced genomes to the existing reference genomes enabled the identification of 3,459 core V. vulnificus genes shared among all six strains and 80 clade 2-specific genes. We identified 523,161 SNPs among the six genomes.

Conclusions: We were able to glean much information about the genomic content of each strain using next generation sequencing. Flp pili, GGDEF proteins, and genomic island XII were identified as possible virulence factors because of their presence in virulent sequenced strains. Genomic comparisons also point toward the involvement of sialic acid catabolism in pathogenesis.

Figure 1

Graphical representation of coverage of the reference genome components by sequences of each of the four newly sequenced genomes. The depth of coverage (number of matched 35-nt reads per 100-nt window of the reference genomes) is plotted for both chromosomes of the reference CMCP6 and YJ016 genomes and the YJ016 plasmid. The source strain for the reads being matched are as follows: M06 - M06-24/O, B5 - 99-738 DP-B5, B8 - 99-520 DP-B8, ATCC - ATCC 33149. It should be noted that coverage of the reference genomes is not as continuous as it appears in the figures.

Figure 2

Distribution of SNPs Relative to the CMCP6 Genome and Subsets of Genes. Box and whisker plots of the SNPs/base for each of the subsets of annotated genes are shown.