Abstract
The ad hoc committee for the re-evaluation of the species definition in bacteriology encouraged researchers to base a species description on more than a single strain and to apply DNA profiling (i.e. genotyping) as a method of discrimination at the subspecies level (Stackebrandt et al., 2002). Genotyping avoids a species being defined on the basis of isolates that could belong to the same strain. This may be critical for isolates recovered from the same source and/or site, where different strains or multiple isolates of the same strains may co-exist. This letter aims to support the committee's recommendation by highlighting the need for the use of genotyping techniques in the description of novel taxa. To illustrate this, we investigated the three most recently described species of Aeromonas. These are: Aeromonas culicicola, described on the basis of the type strain MTCC 3249T, isolated from the midgut of the mosquito Culex quinquefasciatus, and strains SH and SLH, both from the mosquito Aedes aegyptii (Pidiyar et al., 2002); Aeromonas simiae, described on the basis of two strains (CIP 107798T and CIP 107797) recovered from the faeces of the monkey Macaca fascicularis (Harf-Monteil et al., 2004); and Aeromonas molluscorum, described on the basis of five strains recovered from bivalve molluscs (Miñana-Galbis et al., 2004). In the studies describing A. culicicola and A. simiae, genotyping techniques were not applied, while they were in the study of A. molluscorum.
We genotyped all available strains of the species mentioned above using enterobacterial repetitive intergenic consensus sequence (ERIC)-PCR, a method that has been shown previously to be highly discriminatory for Aeromonas (Sechi et al., 2002; Soler et al., 2003; Szczuka & Kaznowski, 2004; Aguilera-Arreola et al., 2005). The primers and conditions were those of Soler et al. (2003). This method proved that the two strains SH and SLH of A. culicicola showed the same pattern (Fig. 1a), as was also the case for the two strains of A. simiae (Fig. 1b). Random amplification of polymorphic DNA (RAPD) analysis and pulsed-field gel electrophoresis (PFGE) (data not shown) further confirmed this. Furthermore, a re-evaluation of the hydrolysis of aesculin showed that the two strains of A. simiae are indistinguishable, contradicting the findings of Harf-Monteil et al. (2004). In contrast, the five strains of A. molluscorum showed individual ERIC-PCR patterns (Fig. 1c), which confirmed previous amplified fragment length polymorphism (AFLP) results (Miñana-Galbis et al., 2004). In conclusion, in the cases of A. culicicola and A. simiae, where no typing methods were included in the original publications (Pidiyar et al., 2002; Harf-Monteil et al., 2004), two isolates of a single strain were included in the species description.