Feasibility of Using Multiple-Locus Variable-Number Tandem-Repeat Analysis for Epidemiology Study of Vibrio parahaemolyticus Infections

Vibrio parahaemolyticus causes foodborne gastroenteritis, which is often associated with the consumption of raw or undercooked shellfish. Molecular typing can provide critical information for detecting outbreaks and for source attribution. In this study, we describe the development and evaluation of an optimized multiple-locus variable-number tandem-repeat (VNTR) analysis (MLVA) for the characterization of V. parahaemolyticus isolates. The discriminatory power of MLVA was compared to that of pulsed-field gel electrophoresis (PFGE) by typing 73 sporadic isolates. Epidemiologic concordance was evaluated by typing 23 isolates from five epidemiologically well-characterized outbreaks. The optimized MLVA was applied in early warning, epidemiological surveillance, and source tracking for V. parahaemolyticus infections. There was no significant difference in the discriminatory power of PFGE and MLVA with six or eight VNTR loci for the sporadic isolates. All isolates within an outbreak were indistinguishable by MLVA with six loci, except for one outbreak. Typically, the epidemiological survey could be initiated according to PFGE clusters. We applied MLVA with six loci on 22 isolates in two PFGE clusters. Isolates in one PFGE cluster were distinguished by MLVA. Although a follow-up investigation showed that both clusters had no epidemiological concordance, MLVA decreased the frequency of initiation of epidemiological surveys, thereby reducing labor costs. The ability of MLVA to trace the source of infection was evaluated by isolates from two outbreaks and shrimp samples. The isolates from one of outbreaks and a shrimp had the same MLVA type, suggesting that an epidemiological survey was initiated. Data from the epidemiological investigation subsequently indicated that contaminated shrimp from a nearby city (Dongguan) might be the source of the outbreak. In conclusion, these results indicate that the optimized MLVA may be a promising tool for early warning and epidemiological surveillance of V. parahaemolyticus infections.