Genotyping of Pandemic Vibrio parahaemolyticus O3:K6 Still Open to Question

Vibrio parahaemolyticus is one of the major seafood-borne gastroenteritis-causing bacteria, frequently associated with consumption of raw or inappropriately cooked seafood. Thermostable direct hemolysin (TDH) and TDH-related hemolysin (TRH) are considered major virulence factors for the organism (7). V. parahaemolyticus can be classified into 13 O serotypes and 71 K serotypes (3). Although various serovars of the bacterium can cause infections, O3:K6 has been recognized as the predominant serovar responsible for most outbreaks worldwide since 1996 (5). The pandemic strains and other recently emerged serovars such as O4:K68 and O1:K untypeable showed almost identical pulsed-field gel electrophoresis (PFGE) patterns (1), suggesting that these strains are clonally related. Matsumoto et al. (5) reported that the pandemic strains exhibit a unique sequence within the toxRS operon, which encodes transmembrane proteins in the regulation of virulence-associated genes conserved in the genus Vibrio. On the other hand, Nasu et al. (6) isolated filamentous phage possessing a unique open reading frame, ORF8, from one of the pandemic strains, and Iida et al. (4) claimed that ORF8 was a useful genetic marker for identifying strains. It was thus necessary to evaluate the use of the toxRS sequence or ORF8 as a reliable genetic marker for the identification.

A total of 24 strains of V. parahaemolyticus that had been isolated from various sources with known serological identities were used in the present study and are listed in Table 1. These include 21 strains of O3:K6, consisting of 12 strains isolated before 1996 and 9 strains isolated since1996, and 3 strains of O4:K68 isolated since 1999. PFGE typing was performed on genomic DNAs of the O3:K6 and O4:K68 strains digested with the restriction enzyme SfiI, following a method described previously (1). Of 21 strains of O3:K6 subjected to the typing, 9 strains isolated since 1996 were classified as type A, 2 strains isolated in 1981 were classified as type B, and 5 strains isolated between 1981 and 1996 were classified as type C (Table 1). The 5 strains isolated between 1982 and 1988 showed PFGE fragment patterns different from any of the above PFGE types (Table 1).

The presence of TDH gene (tdh) and TRH gene (trh) was determined by PCR with a set of primers, 5′-GGTACTAAATGGCTGACATC-3′ and 5′-CCACTACCACTCTCATA-TGC-3′, and another set of primers, 5′-GGCTCAAAATGGTTAAGCG-3′ and 5′-CATTTCCG-CTCTCATATGC-3, respectively, following the protocols established by Tada et al. (7). All strains of PFGE types A and B were positive for tdh but negative for trh, whereas all strains of type C were negative for tdh, with four of them positive for trh (Table 1). The O3:K6 strains with the untypeable PFGE patterns were negative for both genes (Table 1).

PCRs using a method (5) designed to specifically detect the toxRS sequence of the new O3:K6 clone (toxRS/new) and that of the old O3:K6 clone (toxRS/old) were performed on the strains with primer set 5′-TAATGAGGTAGAAACA-3′ and 5′-ACGTAACGGGCCTACA-3′ and primer set 5′-TAATGAGGTAGAAACG-3′ and 5′-ACGTAACGGGCC-TACG-3′, respectively. All strains belonging to PFGE type A are found to possess toxRS/new, whereas all strains of type B and C strains possessed toxRS/old (Table 1). Interestingly, four strains of the PFGE untypeable O3:K6 strains were positive for the toxRS/new sequence (Table 1), suggesting that the sequence was not specific to the pandemic PFGE type (type A).

We also performed a PCR amplification (4) which was designed to amplify a partial DNA sequence of ORF8 in the genomic DNA, using primer set 5′-GTTCGCATACAGTTGAGG-3′ and 5′-AAGTACAGCAGGAGTGAG-3′. ORF8 was detected in all strains belonging to the PFGE type A but not in the rest of the strains tested (Table 1). The results suggest that ORF8 rather than toxRS/new is a more reliable genetic marker for identification of the pandemic strains. However, Iida et al. (4) pointed out that one of the O3:K6 strains belonging to the pandemic PFGE type was negative for ORF 8. More recently, Bhuiyan et al. (2) reported that eight of the O3:K6 clinical strains isolated between 1998 and 2000 were negative for ORF8, claiming that ORF8 was a poor genetic marker for the pandemic genotype. Nevertheless, the claim was made without any reference to the PFGE types of the strains used. Genotyping of the pandemic O3:K6 strains is thus still an open question. Further comprehensive investigation with a larger collection of the strains from various sources is necessary to draw any conclusion on this matter.

• American Society for Microbiology