EpidemiologyNew Mycobacterium tuberculosis LAM sublineage with geographical specificity for the Old World revealed by phylogenetical and Bayesian analyses
Introduction
In a previous study, we explored the genetic specificities of Latin American Mediterranean (LAM) clade of Mycobacterium tuberculosis which is the most predominant genotypic lineage in the Americas [1]. Starting from a large dataset representing a total of 31 countries of the Americas extracted from the SITVIT2 database [2], we investigated the genetic structuration, evolutionary history and biogeographical pattern of the largest LAM sublineage – LAM9. By using a combination of phylogenetical and Bayesian analyses based on spoligotyping [3] and 12-loci MIRU-VNTRs (Mycobacterial Interspersed Repetitive Unit – Variable Number of Tandem Repeats) [4], this study revealed for the first time an obvious division of LAM9 isolates into two subpopulations named LAM9C1 and LAM9C2 [1]. Indeed, each subpopulation showed distinct phylogeographical specificity and genetic characteristics, briefly: (i) LAM9C2 sublineage showed older traces of expansion and expanded 20 times more than LAM9C1, (ii) LAM9C2 was characterized by higher allelic richness, and (iii) presented typical MIRU-VNTR signature of LAM-RDRio isolates. It was then proposed to consider these two clusters as distinct lineages. Subsequently, it was thought appropriate to extend these observations on an enlarged global sample to verify if this genetic structuration could be outlined equally out of Americas.
Section snippets
Global phylogenetic analysis of LAM sublineages
For the present study, we extracted anonymized data from the SITVIT2 proprietary database of Institut Pasteur de la Guadeloupe [2] which is an updated version of the previously released SITVITWEB database where lineages were subdivided into sublineages as described recently [5]. Global evolutionary relationships of LAM strains (n = 1923) from 35 countries were explored by combining spoligotyping and 12-loci MIRU-VNTR data using Bionumerics software 6.6 (Applied Maths, Sint-Martens-Latem,
Genetic structuration of LAM9 sublineage and phylogeography
To confirm putative subdivision of LAM9 isolates, a Bayesian model approach using STRUCTURE 2.3 software [10] was performed on the LAM9 dataset (n = 851 strains) using 12-loci MIRU-VNTRs with parameters set as in our recent study [1]. The results obtained (Supplemental Figure S2) confirmed the previously observed dichotomy in LAM9 population structure in the global dataset; indeed the majority of strains were grouped as LAM9C1 (n = 356) and LAM9C2 (n = 447) and only a minority were intermediate
Subdivision of LAM9 sublineage into distinct clusters
In order to test congruence of the clustering obtained by Bayesian analysis on STRUCTURE, and classical MST-based phylogenetic analysis within “out of Americas” setup, we performed a 2nd MST-based analysis in which all the LAM9 strains were prelabeled according to STRUCTURE-based groupings. The resulting tree (Figure 1B) corroborated the structuration for LAM9C1 but not for LAM9C2 since the later was distinctly subdivided into two groups tentatively named LAM9C2 (n = 117) and LAM9C3 (n = 97).
Genetic characterization and evolutionary history
Mean allelic richness was calculated on 12-loci MIRU-VNTR markers using a rarefaction procedure (compensating for sampling disparity) implemented in HP-RARE 1.0 software [11] when more than 19 samples were available (Table S1). This analysis was performed for the global dataset vs. out of Americas as well as at the country level in Belgium, France and Morocco. As summarized in Table S1, comparable results were found for these 2 sublineages irrespective of the analysis made globally or at
Conclusions
By combining spoligotyping and 12-loci MIRU-VNTR results on a global dataset of M. tuberculosis LAM9 isolates followed by phylogenetical and Bayesian analyses, we were able to describe a clear structuration into distinct clusters, tentatively named LAM9C1, C2 and C3. Although phylogenetically not flawless, LAM9C1 was pretty well clusterized. LAM9C2 appeared as loosely-knit group whose definition remained inconsistent. Lastly, LAM9C3 appeared as being remarkably tightly knit cluster exclusively
Acknowledgments
NR acknowledges help of David Couvin for the construction of the SITVIT2 database. Yann Reynaud was awarded a “Calmette and Yersin” postdoctoral fellowship by the Institut Pasteur.
References (16)
- et al.
SITVITWEB-a publicly available international multimarker database for studying Mycobacterium tuberculosis genetic diversity and molecular epidemiology
Infect Genet Evol
(2012) - et al.
Impact of homoplasy on variable numbers of tandem repeats and spoligotypes in Mycobacterium tuberculosis
Infect Genet Evol
(2012) - et al.
Latin-American-Mediterranean lineage of Mycobacterium tuberculosis: human traces across pathogen's phylogeography
Mol Phylogenet Evol
(2016) - et al.
Genetic structuration, demography and evolutionary history of Mycobacterium tuberculosis LAM9 sublineage in the Americas as two distinct subpopulations revealed by Bayesian analyses
PLoS One
(2015) - et al.
Phylogenetic associations with demographic, epidemiological and drug resistance characteristics of Mycobacterium tuberculosis lineages in the SITVIT2 database: macro- and micro-geographical cleavages and phylogeographical specificities
Int J Mycobact
(2015) - et al.
Simultaneous detection and strain differentiation of Mycobacterium tuberculosis for diagnosis and epidemiology
J Clin Microbiol
(1997) - et al.
Automated high-throughput genotyping for study of global epidemiology of Mycobacterium tuberculosis based on mycobacterial interspersed repetitive units
J Clin Microbiol
(2001) - et al.
Genotyping of genetically monomorphic bacteria: DNA sequencing in Mycobacterium tuberculosis highlights the limitations of current methodologies
PLoS One
(2009)