Abstract
Kazakhstan has one of the highest incidence rates of MDR-TB in the WHO European Region. However, data on the genomic diversity of circulating Mycobacterium tuberculosis strains in Kazakhstan remains limited. We performed whole-genome sequencing on 177 clinical M. tuberculosis isolates belonging to the L2/Beijing lineage, sourced from 15 regions across Kazakhstan. WGS data were analyzed using the computational genomics pipelines TB-Profiler, TB-Annotator, and MTBseq. We then carried out Bayesian analyses using the BEAST algorithm to predict the effective population size dynamics of M. tuberculosis in Kazakhstan over the last 50 years. Phenotypic drug susceptibility testing revealed a significant proportion of MDR (46.9%) and pre-XDR (14.7%) isolates. Phylogenetic analysis showed that 67.2% of M. tuberculosis L2/Beijing isolates belonged to the Central Asian outbreak (CAO) sublineage. Putative compensatory variants were detected in 89.6% of RIF-resistant CAO isolates, with rpoABC variants showing strong sublineage specificity. Furthermore, phylogenetic analysis contributed to the identification of six novel historical clusters, each characterized by distinct SNP signatures. Phylogenetic reconstruction indicated recent transmission and clonal expansion of CAO isolates. Meanwhile, Bayesian skyline analysis revealed a marked increase in the effective population size of Mtb. This study used genomic characterization to show that the CAO sublineage of M. tuberculosis lineage 2 has been highly successful during the MDR-TB epidemic in Kazakhstan, which has coincided with major economic fluctuations.