A New Phylogenetic Framework for the Animal-Adapted Mycobacterium tuberculosis Complex

Abstract

Tuberculosis (TB) affects humans and other animals and is caused by bacteria from the Mycobacterium tuberculosis complex (MTBC). Previous studies have shown that there are at least nine members of the MTBC infecting animals other than humans; these have also been referred to as ecotypes. However, the ecology and the evolution of these animal-adapted MTBC ecotypes are poorly understood. Here we screened 12,886 publicly available MTBC genomes and newly sequenced 17 animal-adapted MTBC strains, gathering a total of 529 genomes of animal-adapted MTBC strains. Phylogenomic and comparative analyses confirm that the animal-adapted MTBC members are paraphyletic with some members more closely related to the human-adapted Mycobacterium africanum Lineage 6 than to other animal-adapted strains. Furthermore, we identified four main animal-adapted MTBC clades that might correspond to four main host shifts; two of these clades are hypothesized to reflect independent cattle domestication events. Contrary to what would be expected from an obligate pathogen, MTBC nucleotide diversity was not positively correlated with host phylogenetic distances, suggesting that host tropism in the animal-adapted MTBC seems to be driven by contact rates and demographic aspects of the host population rather by than host relatedness. By combining phylogenomics with ecological data, we propose an evolutionary scenario in which the ancestor of Lineage 6 and all animal-adapted MTBC ecotypes was a generalist pathogen that subsequently adapted to different host species. This study provides a new phylogenetic framework to better understand the evolution of the different ecotypes of the MTBC and guide future work aimed at elucidating the molecular mechanisms underlying host range.

Keywords: genetic diversity; host range; host–pathogen interactions; specificity; whole-genome sequencing.

FIGURE 1

Maximum Likelihood topology of 322 human-adapted and 529 animal-adapted MTBC members. Branch lengths are proportional to nucleotide substitutions and the topology is rooted with Mycobacterium canettii. Support values correspond to bootstrap values. Main large deletions defining the animal-adapted MTBC are indicated by red arrow heads. The asterisk marks host range expansion within the MTBC.

FIGURE 2

Topology showed in Figure 1 after collapsing all human-adapted branches. Branch lengths are proportional to nucleotide substitutions and the topology is rooted with Mycobacterium canettii. Support values are those of Figure 1. Main large deletions discussed in the text are indicated by red arrow heads and RD specific nomenclatures are indicated when available (Brodin et al., 2002; Mostowy et al., 2004; Alexander et al., 2010; Parsons et al., 2013; Dippenaar et al., 2015). Deletions which are polymorphic in terms of their presence or absence within main clades are indicated in italics.

FIGURE 3

Pair-wise SNP distances within lineage and ecotype from human and animal-adapted MTBC, respectively. Each box corresponds to the 25 and 75% quantiles, the black line represents the median and the whiskers extend to 1.5 times the interquartile range.

FIGURE 4

Principal component analysis (PCA) derived from whole-genome SNPs. The two first principal components are shown.

FIGURE 5

Pair-wise SNP distances between one randomly chosen representative of each human adapted MTBC lineage (A) and animal-adapted MTBC (B). Darker and lighter blue indicate higher and lower genetic distances respectively.

FIGURE 6

Schematic illustration of the putative evolutionary history of the animal-adapted MTBC. The length of the branch is not proportional to genetic distances.