Background: The genus Achromobacter was first described in 1981 and contains today 20 species. Species from this genus are considered as emerging nosocomial pathogens and as new species are described frequently, there is a growing need for robust identification systems with evolving databases such as MALDI-TOF. In order to update those databases, the characterization of the new species using the nrdA gene is necessary as 16S RNA sequencing is not discriminant enough. In this study, we describe the use of nrdA gene phylogeny combined with MALDI-TOF to identify Achromobacter species.
Materials/methods: A set of 110 strains representing 18 species of Achromobacter, from the bioMérieux collection, including type strains of new species and strains originally identified as A. denitrificans by phenotypic methods were sequenced on a polymorphic region of 760 bp of the nrdA gene. The phylogenetic trees were constructed using UPGMA and compared to dendrograms obtained using spectra acquired on the VITEK®MS platform in the 3000 to 17000 Da mass range.
Results: The analysis showed that 90.9% of the strains were identified to the genus Achromobacter. Among those 90.9%, 44.36% were identified as the expected species and 44.54% as a different one. 6.36% of the strains could not be classified at the species level and 2.74% have been reclassified in other genera (Bordetella and Cupriavidus). Concerning the historical species: 39 strains originally A. denitrificans, 5 strains A. xylosoxidans and 8 strains A. piechaudii were reclassified. The VITEK®MS spectra dendrograms were in accordance with the sequencing data.
Conclusions: The nrdA gene phylogeny and VITEK®MS spectra allow a good discrimination of Achromobacter species including species that were not discriminated by phenotypic techniques. We are waiting for strains from the two recently described species A. sediminum and A. aloeverae to complete the analysis. This study shows that phylogeny combined with MALDI-TOF is necessary to characterize species of the genus Achromobacter and to allow the identification systems to be more accurate.