Genome-wide host-pathogen analyses reveal genetic interaction points in tuberculosis disease

The genetics underlying tuberculosis (TB) pathophysiology are poorly understood. Human genome-wide association studies have failed so far to reveal reproducible susceptibility loci, attributed in part to the influence of the underlying Mycobacterium tuberculosis (Mtb) bacterial genotype on the outcome of the infection. Several studies have found associations of human genetic polymorphisms with Mtb phylo-lineages, but studies analysing genome-genome interactions are needed. By implementing a phylogenetic tree-based Mtb-to-human analysis for 714 TB patients from Thailand, we identify eight putative genetic interaction points (P < 5 × 10-8) including human loci DAP and RIMS3, both linked to the IFNγ cytokine and host immune system, as well as FSTL5, previously associated with susceptibility to TB. Many of the corresponding Mtb markers are lineage specific. The genome-to-genome analysis reveals a complex interactome picture, supports host-pathogen adaptation and co-evolution in TB, and has potential applications to large-scale studies across many TB endemic populations matched for host-pathogen genomic diversity.

Indo-Oceanic Mycobacterium tuberculosis strains from Thailand associated with higher mortality

SETTING: This study was conducted among tuberculosis (TB) patients in a highly endemic Thai province.OBJECTIVE: To evaluate the association between different Mycobacterium tuberculosis lineages and clinical characteristics, especially mortality.DESIGN: We enrolled 1,304 TB patients registered from 2002-2011 with culture isolates whose lineages were identified by specific regions of deletion. Data on mortality within 1 year of follow-up were extracted from the registration system and hospital records. Mortality-associated risk factors, including bacterial lineages, as independent variables were analysed using Cox regression models.RESULTS: Of 1,304 isolates, 521 (40.0%) and 582 (44.6%) belonged to Indo-Oceanic and East-Asian lineages, respectively. Indo-Oceanic strains significantly increased the mortality risk compared with East-Asian strains (adjusted hazard ratio [aHR] 1.42, 95%CI 1.02-1.99) or modern lineages (aHR 1.49, 95%CI 1.08-2.06) in the 172 patients who died within 1 year after TB diagnosis. The former also caused significantly higher mortality than modern lineages among patients who died within 6 months after TB diagnosis (aHR 1.62, 95%CI 1.12-2.35). No significant association was found between drug resistance and death.CONCLUSION: In Thailand, the Indo-Oceanic lineage of M. tuberculosis increased mortality risk compared with modern lineages or the East-Asian lineage, the latter being considered highly virulent in previous studies.

Evidence for Host-Bacterial Co-evolution via Genome Sequence Analysis of 480 Thai Mycobacterium tuberculosis Lineage 1 Isolates

Tuberculosis presents a global health challenge. Mycobacterium tuberculosis is divided into several lineages, each with a different geographical distribution. M. tuberculosis lineage 1 (L1) is common in the high-burden areas in East Africa and Southeast Asia. Although the founder effect contributes significantly to the phylogeographic profile, co-evolution between the host and M. tuberculosis may also play a role. Here, we reported the genomic analysis of 480 L1 isolates from patients in northern Thailand. The studied bacterial population was genetically diverse, allowing the identification of a total of 18 sublineages distributed into three major clades. The majority of isolates belonged to L1.1 followed by L1.2.1 and L1.2.2. Comparison of the single nucleotide variant (SNV) phylogenetic tree and the clades defined by spoligotyping revealed some monophyletic clades representing EAI2_MNL, EAI2_NTM and EAI6_BGD1 spoligotypes. Our work demonstrates that ambiguity in spoligotype assignment could be partially resolved if the entire DR region is investigated. Using the information to map L1 diversity across Southeast Asia highlighted differences in the dominant strain-types in each individual country, despite extensive interactions between populations over time. This finding supported the hypothesis that there is co-evolution between the bacteria and the host, and have implications for tuberculosis disease control.

Pathogen lineage-based genome-wide association study identified CD53 as susceptible locus in tuberculosis

Tuberculosis (TB) is known to be affected by host genetic factors. We reported a specific genetic risk factor through a genome-wide association study (GWAS) that focused on young age onset TB. In this study, we further focused on the heterogeneity of Mycobacterium tuberculosis (M. tb) lineages and assessed its possible interaction with age at onset on host genetic factors. We identified the pathogen lineage in 686 Thai TB cases and GWAS stratified by both infected pathogen lineage information and age at onset revealed a genome-wide significant association of one single-nucleotide polymorphism (SNP) on chromosome 1p13, which was specifically associated with non-Beijing lineage-infected old age onset cases (P=2.54E-08, OR=1.74 (95% CI=1.43–2.12)), when we compared them to the population-matched healthy controls. This SNP locates near the CD53 gene, which encodes a leukocyte surface glycoprotein. Interestingly, the expression of CD53 was also correlated with the patients’ active TB status. This is the first report of a pathogen lineage-based genome-wide association study. The results suggested that host genetic risk in TB is depended upon the pathogen genetic background and demonstrate the importance of analyzing the interaction between host and pathogen genomes in TB.

Polymorphisms in the TMEM132D region are associated with panic disorder in HLA-DRB1*13:02-negative individuals of a Japanese population

We herein report an association between TMEM132D and panic disorder (PD) in a Japanese population, evaluating the effects of HLA-DRB1*13:02, which we previously reported as a susceptibility genetic factor for PD. SNPs in TMEM132D showed significant associations with PD in subjects without HLA-DRB1*13:02 (rs4759997; P=5.02×10⁻⁶, odds ratio=1.50) but not in those with the HLA allele. TMEM132D might have a role in the development of PD in subjects without HLA-DRB1*13:02.