Indonesia’s Tuberculosis Genomes Point to Fresher Drug Targets

Tuberculosis still kills more people than almost any other single infectious agent, and Indonesia sits at the center of the fight. The country ranks second worldwide in absolute tuberculosis case numbers, and multidrug-resistant disease continues to undermine national control efforts. Now, a comparative genomic study of 444 Mycobacterium tuberculosis isolates from across the Indonesian archipelago has delivered a finding with direct implications for drug discovery: the metabolic pathways that no current anti-tubercular drug engages are markedly more evolutionarily conserved than the classical resistance-associated loci that decades of chemotherapy have battered into mutational submission.

The research, published in Frontiers in Tropical Diseases, compiled publicly available whole-genome sequencing data from four Indonesian studies spanning Java and Papua. The isolates came from Bandung, the Jakarta metropolitan area, Banten, Jogjakarta, and Jayapura, giving the analysis a geographic spread that earlier Indonesian genomic studies, which were concentrated in Java and analyzed largely in isolation, had never achieved. Using the TBProfiler tool, the team assigned phylogenetic lineages and predicted genotypic drug resistance directly from sequencing reads, then reconstructed gene-level sequences for twenty drug-target genes across 435 isolates retained after quality filtering.

The resistance landscape the analysis revealed is sobering. Overall, 28.2 percent of isolates were genotypically multidrug-resistant, resistant to both rifampicin and isoniazid, and 5.4 percent were pre-extensively drug-resistant, carrying additional fluoroquinolone resistance. Predicted resistance was highest for rifampicin at 41.2 percent and isoniazid at 36.7 percent, followed by ethambutol, pyrazinamide, streptomycin, and fluoroquinolones. Notably, no isolate carried mutations associated with resistance to bedaquiline, linezolid, clofazimine, or delamanid, the newer and repurposed agents now anchoring drug-resistant tuberculosis regimens.

Lineage 2, the East-Asian lineage, dominated the cohort at 46.2 percent of isolates and carried a disproportionate resistance burden. Only 35.6 percent of Lineage 2 isolates were pan-susceptible, compared with 59.8 percent of Lineage 4 and 83.9 percent of Lineage 1 isolates. Lineage 2 accounted for 65.6 percent of all multidrug-resistant isolates and 70.8 percent of all pre-extensively drug-resistant isolates despite comprising less than half the cohort. A chi-square test confirmed the association between lineage and resistance (χ² = 49.07, df = 2, p < 0.001), and pairwise comparisons showed Lineage 2 isolates were nearly ten times more likely than Lineage 1 isolates to carry any genotypic resistance. The canonical resistance mutations rpoB Ser450Leu and katG Ser315Thr predominated, and 54 isolates carried both simultaneously, the single most common composite multidrug-resistant genotype.

The heart of the study, however, lies in its comparative architecture. The team pitted ten classical resistance-associated genes, including rpoB, katG, inhA, fabG1, embB, pncA, gyrA, gyrB, rpsL, and rrs, against ten emerging metabolic drug-target genes drawn from pathways that include de novo purine biosynthesis (guaA, guaB2), nitrogen assimilation (glnA1, gltB), the diaminopimelate pathway (dapB), folate biosynthesis (folP1), biotin biosynthesis (bioA), pantothenate and coenzyme A biosynthesis (panC), the shikimate pathway (aroG), and branched-chain amino acid biosynthesis (ilvB1). Several of these pathways have recently been shown to be essential for Mycobacterium tuberculosis survival during mammalian infection using genome-wide transposon mutagenesis, and many are absent from humans, offering plausible routes to selective, low-toxicity inhibition.

Across all twenty genes and 435 isolates, sequence conservation was uniformly high, ranging from 89.39 percent for rrs to 99.62 percent for bioA, with a total of 670 variable sites detected across 33,871 base pairs of aligned coding sequence. This narrow range is consistent with the highly clonal population structure of Mycobacterium tuberculosis, in which genome-wide diversity is low relative to most bacterial pathogens. But within that narrow range, the two panels separated clearly and consistently. Classical resistance-associated loci were significantly less conserved than emerging targets, with mean conservation of 96.47 percent versus 99.24 percent (p < 0.001), and they carried a significantly higher SNP density of 20.70 versus 6.58 SNPs per kilobase (p = 0.002). In absolute terms, the classical panel accounted for 537 of the 670 variable sites detected, four fifths of all variation, despite representing a comparable total length of coding sequence.

The most striking signal came from substitution-pattern analysis. The non-synonymous-to-synonymous substitution ratio, a proxy for the direction and intensity of selective pressure, was significantly higher among classical loci (median 1.67 versus 0.96; p = 0.033). But the distribution was strongly bimodal, and the excess was concentrated precisely where pharmacology predicts. The two prodrug-activating enzymes showed by far the highest ratios in the entire panel: pncA, which activates pyrazinamide, showed a ratio of 49.0, and katG, which activates isoniazid, showed 10.4. For both drugs, resistance arises through loss of function of the activating enzyme, so amino-acid-altering and disruptive changes are directly selected. Both genes also carried loss-of-function variants, which are otherwise almost absent from the panel. By contrast, classical genes that are themselves direct enzymatic targets, such as gyrB, fabG1, and embB, fell within the emerging-panel range, consistent with strong purifying selection on essential catalytic function where resistance arises through a small number of specific substitutions rather than a gene-wide shift.

The emerging metabolic targets, by contrast, formed a tight cluster around parity in their substitution ratios, with no gene showing the strong non-synonymous excess characteristic of the drug-activating loci. Absolute variable-site counts were uniformly low, with four genes carrying fewer than eight variable sites across 435 isolates, and the most variable emerging gene, gltB, carried 47 variable sites across 4,584 base pairs, a density still below that of eight of the ten classical loci. The team had originally intended to prioritize individual candidate genes by combining above-median conservation with below-median SNP density, but they concluded that gene-level discrimination is not statistically supportable with these data, because the absolute counts are too small and the ranking too sensitive to the addition or removal of individual variants. They instead interpret the emerging panel collectively as a low-variability target space.

The authors are careful about what their metrics can and cannot claim. The non-synonymous-to-synonymous ratio is a simple variant-count ratio rather than a codon-normalized dN/dS estimate, and they read it as descriptive evidence of relative non-synonymous enrichment rather than a formal test of positive selection. They also note that conservation percentages derived from consensus sequences without a minimum-depth mask should be regarded as upper-bound estimates in one respect, since low-coverage regions inherit the reference base and appear conserved. The TBProfiler mutation library used dates from November 2021 and predates the current World Health Organization catalogue of resistance mutations, and phenotypic drug susceptibility testing was not available to cross-validate genotypic predictions.

What the study ultimately establishes is a quantitative framework for target-space prioritization in high-burden tuberculosis settings. A candidate drawn from the emerging metabolic pathways begins from a position of substantially lower standing sequence variation, and therefore lower pre-existing potential for target-site resistance, than the classical target set that current chemotherapy engages. The authors suggest that extending this comparative framework to additional Indonesian provinces and linking it prospectively to phenotypic drug susceptibility and treatment outcome data would strengthen its value for both drug target discovery and routine programmatic surveillance of multidrug-resistant tuberculosis. In a country that bears a disproportionate share of the global drug-resistant tuberculosis caseload, that combination of drug discovery guidance and scalable genomic surveillance may prove to be the study’s most consequential legacy.

Subject of Research: Comparative genomic analysis of established and emerging anti-tubercular drug targets reveals evolutionary conservation and antimicrobial resistance patterns in Mycobacterium tuberculosis isolates from Indonesia

Article Title: Comparative genomic analysis of established and emerging anti-tubercular drug targets reveals evolutionary conservation and antimicrobial resistance patterns in Mycobacterium tuberculosis isolates from Indonesia

Article References: Puspawati, N. M., Suprihatin, I. E., Suthar, A., Vansia, A., & Patel, R. (2026). Comparative genomic analysis of established and emerging anti-tubercular drug targets reveals evolutionary conservation and antimicrobial resistance patterns in Mycobacterium tuberculosis isolates from Indonesia. Frontiers in Tropical Diseases, 7, Article 1953716. https://doi.org/10.3389/fitd.2026.1953716

Image Credits: AI Generated

DOI: 10.3389/fitd.2026.1953716

Keywords: Comparative, genomic, analysis, established, emerging, anti-tubercular, drug, targets, reveals, evolutionary, conservation, antimicrobial