Clustering of Vibrio parahaemolyticus Isolates Using MLST and Whole-Genome Phylogenetics and Protein Motif Fingerprinting

Abstract

Vibrio parahaemolyticus is a ubiquitous and abundant member of native microbial assemblages in coastal waters and shellfish. Though V. parahaemolyticus is predominantly environmental, some strains have infected human hosts and caused outbreaks of seafood-related gastroenteritis. In order to understand differences among clinical and environmental V. parahaemolyticus strains, we used high quality DNA sequencing data to compare the genomes of V. parahaemolyticus isolates ( n = 43) from a variety of geographic locations and clinical and environmental sample matrices. We used phylogenetic trees inferred from multilocus sequence typing (MLST) and whole-genome (WG) alignments, as well as a novel classification and genome clustering approach that relies on protein motif fingerprints (MFs), to assess relationships between V. parahaemolyticus strains and identify novel molecular targets associated with virulence. Differences in strain clustering at more than one position were observed between the MLST and WG phylogenetic trees. The WG phylogeny had higher support values and strain resolution since isolates of the same sequence type could be differentiated. The MF analysis revealed groups of protein motifs that were associated with the pathogenic MLST type ST36 and a large group of clinical strains isolated from human stool. A subset of the stool and ST36-associated protein motifs were selected for further analysis and the motif sequences were found in genes with a variety of functions, including transposases, secretion system components and effectors, and hypothetical proteins. DNA sequences associated with these protein motifs are candidate targets for future molecular assays in order to improve surveys of pathogenic V. parahaemolyticus in the environment and seafood.

Phylogenetic and Biogeographic Patterns of Vibrio parahaemolyticus Strains from North America Inferred from Whole-Genome Sequence Data

“…As reported in a previous study using a subset of these isolates, ST36 and ST3 are the most frequent STs in this isolate collection. Both ST36 and ST3 were comprised only of clinical isolates collected on various dates and from multiple regions with O4:K12/O4:Kut and pandemic-related serotypes, respectively, as previously described for these STs (9,24). However, these data show high diversity regarding the residual STs of the isolates tested regardless of the method of analysis used (likelihood or genome distance).…”

“…However, recent studies have shown that some strains produce cytotoxic and enterotoxic effects independent of their hemolysin production (6,7). Also, some clinical isolates have been reported to lack one or both of the virulence markers tdh and trh, and phylogeny is independent of the genes (8)(9)(10). In light of this, type III secretion systems (T3SSs) have been investigated as potential pathogenicity factors for V. parahaemolyticus (11,12).…”

“…Previous investigations of V. parahaemolyticus utilizing multilocus sequence typing (MLST), pulsed-field gel electrophoresis (PFGE), and multiple-locus variable-number tandem-repeat analysis (MLVA) have indicated high genetic diversity (9,(21)(22)(23)(24). However, the availability of whole-genome sequencing (WGS), more powerful computing, and better algorithms has made phylogenetic analysis based on the data collected from the entire genome possible.…”

Human Health and Ocean Pollution

“…These changes will likely result in increased frequency of Vibrio infections in coming decades and possibly to appearance of Vibrio infections in previously unaffected areas [ 52 ]. There is some indication that after extreme weather events such as hurricanes, droughts, and tropical storms shifts occur in the composition of Vibrio species and that these shifts are driven by discharges of sewage and inorganic nutrients into coastal waters [ 410 ].…”

“…These models incorporate multiple layers of geocoded data and are designed to generate predictive forecasts [ 553 ]. New technologies such molecular and bioinformatics-based diagnostics [ 410 425 554 ], metabarcoding, “big data” mining and machine learning may be expected to contribute to further development of these efforts [ 40 555 556 ]. Implementation of real-time PCR-based approaches has already been shown to be a useful tool for diagnosing V. vulnificus wound infections [ 554 ].…”