BackgroundVibrio parahaemolyticus is frequently isolated from environmental and seafood samples and associated with gastroenteritis outbreakes in American, European, Asian and African countries. To distinguish between different lineages of V. parahaemolyticus various genotyping techniques have been used, incl. multilocus sequence typing (MLST). Even though some studies have already applied MLST analysis to characterize V. parahaemolyticus strain sets, these studies have been restricted to specific geographical areas (e.g. U.S. coast, Thailand and Peru), have focused exclusively on pandemic or non-pandemic pathogenic isolates or have been based on a limited strain number.ResultsTo generate a global picture of V. parahaemolyticus genotype distribution, a collection of 130 environmental and seafood related V. parahaemolyticus isolates of different geographical origins (Sri Lanka, Ecuador, North Sea and Baltic Sea as well as German retail) was subjected to MLST analysis after modification of gyrB and recA PCRs. The V. parahaemolyticus population was composed of 82 unique Sequence Types (STs), of which 68 (82.9%) were new to the pubMLST database. After translating the in-frame nucleotide sequences into amino acid sequences, less diversity was detectable: a total of 31 different peptide Sequence Types (pSTs) with 19 (61.3%) new pSTs were generated from the analyzed isolates. Most STs did not show a global dissemination, but some were supra-regionally distributed and clusters of STs were dependent on geographical origin. On peptide level no general clustering of strains from specific geographical regions was observed, thereby the most common pSTs were found on all continents (Asia, South America and Europe) and rare pSTs were restricted to distinct countries or even geographical regions. One lineage of pSTs associated only with strains from North and Baltic Sea strains was identified.ConclusionsOur study reveals a high genetic diversity in the analyzed V. parahaemolyticus strain set as well as for geographical strain subsets, with a high proportion of newly discovered alleles and STs. Differences between the subsets were identified. Our data support the postulated population structure of V. parahaemolyticus which follows the ‘epidemic’ model of clonal expansion. Application of peptide based AA-MLST allowed the identification of reliable relationships between strains.