Reconstituting the genus Mycobacterium

Abstract

The definition of a genus has wide-ranging implications both in terms of binomial species names and also evolutionary relationships. In recent years, the definition of the genus Mycobacterium has been debated due to the proposed split of this genus into five new genera (Mycolicibacterium, Mycolicibacter, Mycolicibacillus, Mycobacteroides and an emended Mycobacterium). Since this group of species contains many important obligate and opportunistic pathogens, it is important that any renaming of species does not cause confusion in clinical treatment as outlined by the nomen periculosum rule (56a) of the Prokaryotic Code. In this study, we evaluated the proposed and original genus boundaries for the mycobacteria, to determine if the split into five genera was warranted. By combining multiple approaches for defining genus boundaries (16S rRNA gene similarity, amino acid identity index, average nucleotide identity, alignment fraction and percentage of conserved proteins) we show that the original genus Mycobacterium is strongly supported over the proposed five-way split. Thus, we propose that the original genus label be reapplied to all species within this group, with the proposed five genera potentially used as sub-genus complex names.

Keywords: Mycobacterium; genome; genus.

Fig. 1.

16S similarity scores between members of genera. Intra-genus scores are shown in red and inter-genus scores are shown in black. Histogram bin sizes are 0.5 %. The proposed genus boundary of 94.5 % is represented by the dashed line. Each genus is shown separately as follows: (a) Mycobacterium (original); (b) Mycobacterium (Gupta); (c) Mycobacteroides ; (d) Mycolicibacillus ; (e) Mycolicibacter ; (f) Mycolicibacterium . Note that the intra-genus proportion for Mycolicibacillus is 1.0 but the y-axis is cut at 0.4 for comparison between plots.

Fig. 2.

AAI similarity scores between members of genera. Intra-genus scores are shown in red and inter-genus scores are shown in black. Histogram bin sizes are 1 %. The proposed genus AAI boundaries of 65 % is represented by a dashed line. Each genus is shown separately as follows: (a) Mycobacterium (original) (b) Mycobacterium (Gupta); (c) Mycobacteroides ; (d) Mycolicibacillus ; (e) Mycolicibacter ; (f) Mycolicibacterium . Note that the intra-genus proportion for Mycolicibacillus is 1.0 but the y-axis is cut at 0.4 for comparison between plots.

Fig. 3.

POCP scores between members of genera. Intra-genus scores are shown in red and inter-genus scores are shown in black. Histogram bin sizes are 5 %. The proposed genus POCP boundary of 50 % is represented by the dashed line. Each genus is shown separately as follows: (a) Mycobacterium (original); (b) Mycobacterium (Gupta); (c) Mycobacteroides ; (d) Mycolicibacillus ; (e) Mycolicibacter ; (f) Mycolicibacterium . Note that the intra-genus proportion for Mycolicibacillus is 1.0 but the y-axis is cut at 0.6 for comparison between plots.

Fig. 4.

Alignment fraction (AF) and average nucleotide identity (ANI) between genomes and the type species for each genus. Intra-genus scores are shown in red and inter-genus scores are shown in black. The proposed demarcation points for the genus is shown with a dashed red line. The mean demarcation points based on Barco et al. [19] are shown with dashed black lines. Each genus is shown separately as follows: (a) Mycobacterium (original); (b) Mycobacterium (Gupta); (c) Mycobacteroides ; (d) Mycolicibacillus ; (e) Mycolicibacter ; (f) Mycolicibacterium .