PCR 16S rRNA gene analyses identified 18 strains as V. parahaemolyticus with 100% identity, but yielded uncertain identification for 14 isolates. Twenty-one strains were confirmed as V. parahaemolyticus by PCR assays to detect species-specific targets (in Fig. 1 an example of ToxR PCR detection is shown); three strains Stem Cell Compound Library order were trh positive. The comparison of biochemical and molecular results (Table 1) showed that, among the 21 V. parahaemolyticus strains, 19 were identified by one or both API systems, but only two of them yielded coherent responses with biochemical features reported by Alsina’s scheme; in particular, API 20E yielded only one false positive (Table 2) and six false negatives,

while API 20NE yielded no false-positive results, but eight false negatives. The results obtained in the present work contribute to the debate about the problematic phenotypic identification of environmental V. parahaemolyticus strains. TCBS agar is the only proven selective medium for Vibrio spp. isolation,

but a large number of marine microorganisms may also grow (Thompson et al., 2004). In this study, the screening phase selected 58% of the analyzed strains as belonging to genus Vibrio. Our results confirm those of Croci et al. (2001), who evidenced how strains isolated from seawater and mussels on TCBS agar were principally vibrios (about 50%) while the remaining were Aeromonas, Pseudomonas, Flavobacterium, Pasteurella and Agrobacterium. API systems and Alsina’s scheme (Alsina & Blanch, 1994a, b) are the most extensively used techniques KU-60019 by Italian Laboratories to screen the diversity Vorinostat in vitro of Vibrio spp. strains associated with marine organisms and their habitats (Croci et al., 2007). However, several authors reported that V. parahaemolyticus phenotypic identification is difficult because of the huge variability of diagnostic features among the species (O’Hara et al., 2003; Thompson et al., 2004 and references therein; Croci et al., 2007) and the molecular analyses considered necessary, either for additional confirmatory testing or for a certain identification method. In our study, the

amplification of the 16S rRNA gene produced misidentifications because of the strictly genetic similarity between V. parahaemolyticus and Vibrio alginolyticus, Vibrio campbelli, Vibrio carchariae and Vibrio harveyi (Dorsch et al., 1992). Molecular confirmation performed through PCR assays for toxR and tlh genes produced the same results in contrast to that reported by Croci et al. (2007), who reported that tlh gene detection yields false-positive identifications. Although different studies highlighted the inadequacy of API systems for Vibrio identification (Dalsgaard et al., 1996; Colodner et al., 2004; Croci et al., 2007), in the research, the use of both API 20E and API 20NE, using bacterial suspensions with a slight modification of the salinity from 0.