The Mycobacterium tuberculosis genome at 25 years: lessons and lingering questions

First achieved in 1998 by Cole et al., the complete genome sequence of Mycobacterium tuberculosis continues to provide an invaluable resource to understand tuberculosis (TB), the leading cause of global infectious disease mortality. At the 25-year anniversary of this accomplishment, we describe how insights gleaned from the M. tuberculosis genome have led to vital tools for TB research, epidemiology, and clinical practice. The increasing accessibility of whole-genome sequencing across research and clinical settings has improved our ability to predict antibacterial susceptibility, to track epidemics at the level of individual outbreaks and wider historical trends, to query the efficacy of the bacille Calmette-Guérin (BCG) vaccine, and to uncover targets for novel antitubercular therapeutics. Likewise, we discuss several recent efforts to extract further discoveries from this powerful resource.

Not long after the first M. tuberculosis genome sequence was assembled, additional studies revealed the global diversity of M. tuberculosis genotypes. Previous methods of typing M. tuberculosis strains, including IS6110-RFLP (restriction fragment length polymorphism), spoligotyping (spacer oligonucleotide typing, which is recognized as using CRISPR array spacers), and MIRU-VNTR (mycobacterial interspersed repetitive unit variable number tandem repeat) typing, had found some success using variable numbers or positions of repetitive genomic elements, although they were limited by the low abundance and occasional non-uniqueness of these markers (40, 41). Indeed, previous attempts to predict clinical strain parameters by IS6110 copy number were found to be suboptimal, as the so-called low-copy and high-copy groups were later shown to be polyphyletic (42, 43). In contrast, comparing single-nucleotide polymorphisms (SNPs) across the genome allowed for unambiguous, higher-resolution assignment of M. tuberculosis strains into lineage clusters (Figure 1) (41–43). Strikingly, the broad lineages of M. tuberculosis also clustered with the birthplaces of the patients they infected, suggesting that the M. tuberculosis phylogeny and indeed the M. tuberculosis genome had captured geographic information (43, 44). Comparing M. tuberculosis strains from diverse lineages showed a paucity of large genomic deletions and rearrangements, evidence of an exceptionally stable genome among bacteria that preserved historical information (44). By the turn of the century, the M. tuberculosis genome was recognized as a powerful spatiotemporal resource to understand the epidemiology of the disease.

Figure 1

Phylogeny of M. tuberculosis lineage strains. Simplified maximum likelihood phylogeny of the 9 lineages of M. tuberculosis, as well as the related M. bovis strain and the M. canettii outgroup strain used as a root. Adapted with permission from Microbial Genomics (60).

M. tuberculosis has been classified into nine lineages mainly on the basis of large-scale genomic variations, variably described as regions of difference (RDs) or large sequence polymorphisms (LSPs) (45, 46). Lineages 2–4 (L2–L4) constitute a monophyletic group defined by the TbD1 deletion, a loss of the membrane proteins MmpS6 and MmpL6 that appears to confer enhanced resistance to oxidative and hypoxic stressors (47, 48). Collectively termed the “modern” lineages, L2–L4 cause the majority of globally distributed TB epidemics and hence most of the TB disease burden (Figure 2) (49). Among these, L4 has the broadest range, spanning throughout Africa, Asia, Europe, and the Americas. L2 causes the largest proportion of TB cases in East Asia as well as some cases in Central Asia and notably includes the hypervirulent Beijing strains, while L3 exists mainly in India, with additional presence in East Africa. The remaining “ancestral” lineages include L1, which predominantly occurs in Southeast Asia and India and has the widest distribution of the ancestral lineages. L5–L9 seemingly arise only in Africa, with L5–L6 (classically dubbed M. africanum) causing up to 40%–60% of TB cases there (50). Notably, the animal-adapted lineages of the broader M. tuberculosis complex, including M. bovis, share ancestry with L6 (51). Beyond these overarching categories, the modern lineages can be divided into sublineages that also correlate with human population geography (52, 53). These sublineages can be further classified into “generalist” sublineages with broader global distribution and greater variability in T cell epitopes than the more geographically confined “specialist” sublineages (53). Consistent with these observations, recent evidence argues for sympatric spread among specialist sublineages, suggesting that specialist strains have adapted to the human host genetics in their endemic regions (54). Recent methods expand beyond the use of RDs and incorporate SNPs into a highly granular sublineage classification schematic (55, 56).

Figure 2

Cartogram of global TB burden by M. tuberculosis lineage. Country areas are scaled to reflect TB incidence in 2021 (1) using the go-cart.io algorithm (172). Pie charts reflect distributions of the M. tuberculosis lineages L1–L9, as well as the animal-adapted M. bovis, M. caprae, and M. orygis, from clinical isolates by geographic region as previously described by Napier et al. (56).

By integrating the diversity of modern M. tuberculosis genomes, attempts have been made to determine the origin of human-adapted TB disease and follow its evolution with changing human migration and behavior. The discovery of Mycobacterium canettii in the Republic of Djibouti and its subsequent genomic characterization as an M. tuberculosis ancestor localized the origin of ancient TB to an origin around the Horn of Africa (57–59). Clinically, M. canettii strains are of relatively low virulence, and their genomes are generally devoid of the RDs/LSPs that define lineages L1–L9 (46, 60). Solidifying an emergence in East Africa, a new ancestral lineage with very deep phylogenetic branching, L7, was found in Ethiopia in 2012 and appears limited to residents of and immigrants from the region (61). In the past few years, two more analogously restricted East African lineages (L8–L9) were characterized (60, 62). While horizontal gene transfer mechanisms are not believed to occur in modern M. tuberculosis genomes, the ancient genome contains a mosaic of genetic material, likely from nonpathogenic bacteria (59). Earlier efforts to sequence large M. tuberculosis genomic regions identified a comparatively low rate of silent nucleotide mutations in comparison with other human pathogens, suggesting a population bottleneck with M. tuberculosis adaptation to the parasitic lifestyle (63). These findings collectively suggest a model in which environmental bacteria supplied genomic material to what would become the obligate human pathogen M. tuberculosis, replacing historical arguments for a zoonotic origin (59, 64). From East Africa, M. tuberculosis would have spread globally alongside its human host populations, and indeed the phylogeny of human mitochondrial DNA shows similar topology to that of the M. tuberculosis lineages (64). Like its human hosts, M. tuberculosis underwent several population bottlenecks during geographic spread (65). These events were followed by periods of diversification featuring many non-synonymous SNPs, notably in cell envelope proteins that may have facilitated bacterial virulence by adapting to the host immune system (65, 66). Alterations in human population demographics correlate well with M. tuberculosis evolution on both the ancient time scale, such as the dissemination of the L2 Beijing sublineage as an agricultural lifestyle spread from China across East Asia 3,000–5,000 years ago, and the nearer time scale, such as the spread of this sublineage to Afghanistan during recent wars and the rise in TB drug resistance in former Soviet states with the collapse of the USSR (64, 67). These findings demonstrate the impressive power of the M. tuberculosis genome to record and adapt to host changes throughout human history. Simultaneously, the recent data in particular highlight the tendency of M. tuberculosis to exploit periods of social instability and forced migration to escalate into a greater public health threat.

The non-random expansion of particular M. tuberculosis lineages into global pandemics suggests that these strains may have altered properties relevant to disease outcomes, a hypothesis that echoes the validation of the TbD1 deletion as a gain-of-virulence event based on animal model studies (48). Relatedly, differing lineages have been found to elicit variable immune responses in cellular and animal model systems, and there is some evidence of differential immune modulation by the modern lineages related to these properties (68–73). L2, for example, has been found to induce lower levels of inflammatory cytokines than L4 in some but not all studies (68, 69, 74). However, care must be taken in synthesizing results between such studies, particularly in attempting to generalize results from individual strains across entire lineages. As an example, virulence factors may exist only among certain subgroups within a lineage, as is the case with the cell wall phenolic glycolipid (PGL) that facilitates immune evasion by the subsets of the Beijing sublineage that express it (75, 76). Consistent with this explanation, a comparatively modern Beijing strain was found to cause reduced cytokine secretion in a macrophage model when compared with a more ancestral Beijing strain, despite similarities in bacterial burden and growth rate (71). While all M. tuberculosis strains isolated from patients with active TB are virulent by definition, the observed differences in global transmissibility suggest subtler differences in virulence by lineage due to genome variations, as has been suggested by high-throughput studies (73).

The growing antimicrobial resistance of M. tuberculosis presents an enormous clinical, financial, and public health challenge (1). Genomic sequencing has shown considerable promise in predicting M. tuberculosis susceptibility to TB drugs, and gene transfer had enabled the discovery of resistance genotypes, thereby elucidating previously unknown mechanisms of action (27, 77). In a study of over 10,000 clinical isolates, WGS informed by known resistance gene variants was successful in predicting susceptibility to isoniazid, rifampicin, ethambutol, and pyrazinamide, with sensitivities of 97.1%, 97.5%, 94.6%, and 91.3%, respectively (78). As a direct clinical diagnostic, WGS has been proposed for TB susceptibility testing in low-incidence, high-resource settings based primarily on its more rapid time to results than conventional culture-based testing (which typically requires about 2 weeks after initial cultivation), alongside a slightly cheaper cost (79–81). In 2017, the United Kingdom adopted a policy of routine WGS for taxonomic and drug susceptibility testing of positive mycobacterial cultures. Subsequent retrospective analyses found partial success, noting the potential benefits of WGS but also the substantial lag in turnaround time, which is of particular concern regarding isoniazid resistance (82). Both cost and turnaround time would be expected to decrease as sequencing technology improves and is better incorporated into the clinical workflow. Notably, proof-of-concept studies in the Kyrgyz Republic and South Africa have demonstrated a potential role for M. tuberculosis WGS for resistance testing in low- and middle-income, high-burden countries (9, 83). Settings with a known high burden of MDR- and XDR-TB in particular may benefit from using WGS to identify and discontinue ineffective drugs, which frequently carry high toxicities for patients (9).

Beyond its role in diagnostics, sequencing the M. tuberculosis genome catalyzed the development of tests for M. tuberculosis drug resistance, notably the Xpert platform with MTB/XDR cartridges. Using microfluidics, this instrument amplifies M. tuberculosis genomic regions of interest and employs 10 molecular beacon probes to detect characteristic shifts in hybridization melting temperature. The assay identifies known resistance mutations to isoniazid, fluoroquinolones, and select second-line drugs directly from patient sputum or other clinical samples (8). Additionally, the test distinguishes between low- and high-level resistance mutations to isoniazid and fluoroquinolones, as well as cross-resistance to multiple second-line drugs (8). These genomic targets were gleaned from sequencing of many resistant M. tuberculosis isolates to define a list of the most common mutations. Early studies of the Xpert MTB/XDR test for clinical evaluation showed promising results, with 98% to 100% specificity for each of isoniazid, fluoroquinolones, ethionamide, amikacin, kanamycin, and capreomycin (84, 85). Sensitivity was highest for isoniazid and fluoroquinolones but low for ethionamide; this was attributed to the comparatively poorly characterized ethA resistance mutations and highlighted that molecular tests are limited by the genomic knowledge that underpins them (81, 84).

Unique among diagnostic methods, WGS can not only examine known resistance-associated polymorphisms but also predict novel variants associated with phenotypic resistance. Using principles including positive selection, convergent evolution, and non-synonymous to synonymous SNP ratios, several groups have built lists of genes and genomic regions that may represent unknown M. tuberculosis resistance mechanisms or help compensate for the fitness costs of co-occurring ones (28–30). Intriguingly, the prevalence of particular resistance mutations differs between lineages, suggesting that local M. tuberculosis strain demographics must also be considered when cataloging resistance mechanisms (86). The Comprehensive Resistance Prediction for Tuberculosis: an International Consortium (CRyPTIC) collaboration recently published a compendium of phenotype-validated M. tuberculosis WGS resistance data, a vital resource that will aid future susceptibility testing (31). The developing role of WGS in M. tuberculosis resistance testing has been explored in further detail in other work (87–89).

The past decade has seen a renewed effort toward the development of new antitubercular drugs, with the successful implementation of bedaquiline, delamanid, and pretomanid for MDR-TB (10, 11). At all stages of this process, from initial laboratory characterization to optimization of clinical regimens, WGS has proven essential. In the case of bedaquiline, a consistent atpE resistance mutant in vitro revealed that this drug uniquely targeted ATP synthesis and suggested that the compound could kill even dormant bacteria (90, 91). This atpE mutant was identified by genome sequencing of the model organism M. smegmatis and confirmed to likewise arise in resistant M. tuberculosis (90, 92). Testing across mycobacterial species demonstrated broad activity against many non-tuberculosis mycobacteria, except for M. xenopi, which was found by WGS to have a preexisting polymorphism at an atpE position known to confer resistance (90, 93). Subsequently, mutations in Rv0678, a repressor of the MmpS5/MmpL5 efflux machinery, were identified by WGS to cause cross-resistance to both bedaquiline and clofazimine (94). While reports differ, some sequencing studies argue that the off-target Rv0678 polymorphisms are more frequent than atpE mutations in clinical isolates, at least in certain settings (95–97). An additional concern is the high prevalence of preexisting Rv0678 mutations conferring measurably increased minimal inhibitory concentrations for bedaquiline and clofazimine in populations with no known prior use of either drug, since it had been assumed that bedaquiline would be effective in virtually all M. tuberculosis isolates early in its use (98).

Unlike the mechanism of action of bedaquiline, that of the nitroimidazoles pretomanid and delamanid was not readily apparent upon selection for resistance mutations. Initial characterizations identified resistance mutations in an F420-dependent enzyme, fgd1; however, other resistance mutations were soon reported from genome sequencing data (99, 100). Furthermore, despite evidence that pretomanid treatment impaired synthesis of ketomycolic acids, this was unlikely to be the mechanism of M. tuberculosis killing, as the drug was bactericidal even in non-replicating culture conditions when the bacterium would have little need to produce a cell wall component (99). Subsequent work found that fgd1 is one half of a pair of redox enzymes that acts cyclically on the F420 redox cofactor, the other of which is ddn, a nitroreductase that activates pretomanid to eventually form bactericidal nitric oxide (101, 102). Resistance to pretomanid can thus occur via mutations in either fgd1 or ddn, or alternatively in the fbiABCD genes that synthesize F420 (95, 101, 103). Hence, the characterization by WGS of resistance polymorphisms to the nitroimidazole drugs led to the discovery of a mycobacterial metabolic pathway and a more plausible mechanism of killing: reactive nitrogen species production. In this way, genome sequencing can assign functions to unknown gene products. As with bedaquiline, mutations in the nitroimidazole-activating pathway genes have been found that predate drug use, some of which are predicted to confer resistance (104). Peculiarly, not all pretomanid-resistant mutants are cross-resistant to delamanid, so careful WGS of pretomanid-resistant, delamanid-susceptible isolates and vice versa is needed to fully characterize M. tuberculosis resistance patterns for the nitroimidazole class (103, 105).

Outbreak tracing. WGS can exploit polymorphisms in M. tuberculosis genomic isolates for robust epidemiological tracing regardless of whether the polymorphisms have a known functional role. On the local scale, genome sequencing can resolve TB outbreaks down to individual-level spreading events, enabling the identification of social and socioeconomic factors contributing to transmission (5–7). Older methods such as MIRU-VNTR typing may be sufficient for initial surveillance, though WGS allows for more precise tracing of spread through a community (5, 6). A systematic comparison of MIRU-VNTR and WGS found that MIRU-VNTR can help exclude transmission events on the basis of differences, but it is much less accurate for positively predicting transmission by strain relatedness (106). In addition, MIRU-VNTR predictive efficacy varies by M. tuberculosis lineage, with L2 in particular known to have poor discrimination due to evolutionary convergence in its VNTR regions (106, 107). Aggregate WGS surveillance data are more consistently accurate and can be systematically analyzed for variables such as M. tuberculosis sublineage and frequency of resistance polymorphisms, although currently the majority of M. tuberculosis isolates are not sequenced (49). To potentially decrease the computational load of screening the entire M. tuberculosis genome, a minimal set of SNPs has been proposed as a barcode to stratify M. tuberculosis sublineages as a rapid surveillance tool (55, 56). While widespread use of WGS for surveillance is limited by expense and expertise, particularly in high-burden settings, access may expand in the future as cost improves (108).

Recent versus remote transmission of TB. Prior to the 1990s, the prevailing belief held that the majority of active TB cases were the result of endogenous reactivation of latent TB infection (LTBI) that was acquired remotely in time (109). However, in the early 1990s, IS6110 molecular typing of large M. tuberculosis strain banks collected contemporaneously from active TB cases across US cities yielded the unexpected result that approximately 40% of strains matched another isolate in the collection (110–112). In some instances, it was possible to establish epidemiological links between individuals with matching strains (112). This suggested that recently transmitted TB could lead to active TB disease in a short period of time rather than requiring reactivation of LTBI acquired years or decades earlier. Such findings, based on M. tuberculosis genomic data, have prompted some to make the controversial argument that most infected individuals are cleared of bacteria after two years and are therefore no longer at risk for reactivation of LTBI (113, 114). The issue has important ramifications for how intensively public health organizations should seek out individuals with LTBI to provide secondary prevention therapy with isoniazid for 9 months or a shorter-course regimen (115, 116). Indeed, there has been a trend in the United States to limit LTBI testing by tuberculin skin test or interferon-γ release assay to select high-risk groups (117). Interestingly, recent work using WGS to address the matter of recent versus remote infection found that longer latency times do not necessarily correlate with SNP distance — a finding that calls into question the reliability of genomic determinants in estimating the recency of infection (118). Additional studies using high sequencing depth to query M. tuberculosis mutation rates in human patients, particularly during paucibacillary states, may refine the relationship between the rate of M. tuberculosis mutation acquisition and latency time to enhance epidemiological tracking tools.

WGS to determine mutation rates in latent versus active TB. In addition to studying the emergence of particular variants of interest across broad geographies, WGS offers the capability of examining mutational changes in the M. tuberculosis population within a single individual. Studies of M. tuberculosis genomes from patients receiving TB treatment have consistently found heterogeneity in bacterial populations, with drug resistance driven by mutants that are initially detectable only at low levels (32–34). Most population heterogeneity was driven by SNPs with less than 20% frequency, and indeed a population prevalence of merely 19% was estimated to be the threshold required for a drug resistance variant to proceed toward fixation (33, 34). The most common variation at the nucleotide level was the GC > AT transition, predicted to be a product of oxidative damage faced by the bacterium in the host macrophage, which was first noted in prior nonhuman primate data (34, 119). This latter model also found that M. tuberculosis mutation rates remained similar in active, latent, and reactivated TB disease, a striking finding given that LTBI was previously believed to reflect low numbers of bacilli with little or no replication and thus a negligible risk of mutation. Such findings raise the concern that the use of isoniazid monotherapy for 9 months as secondary prevention for LTBI would potentially risk isoniazid resistance (119–121). Consistent with the threat of resistance due to inadequate treatment, M. tuberculosis WGS data during active TB treatment showed a risk of excessive drug resistance mutations in patients receiving fewer than 4 effective drugs (33). These results collectively underscore the high threshold for adequate antimicrobial pressure, below which even rare M. tuberculosis resistance mutants can ultimately reach fixation within a patient. Additionally, there is some evidence linking high intra-host M. tuberculosis diversity to worse TB severity metrics (122). It remains to be seen whether the use of WGS to query M. tuberculosis heterogeneity within a patient has a role as a disease prognostic marker or for monitoring the development of resistance.

Epidemiological studies tracking M. tuberculosis strains and lineages are greatly aided, if not outright enabled, by the slow mutation rate of the organism. In vitro, the M. tuberculosis genome is remarkably stable, with mutations mainly at the level of individual bases; polymorphisms due to polymerase errors in repetitive microsatellite regions are comparatively rarer than would be expected by chance, and large-scale chromosomal rearrangements are almost entirely absent (44, 123). Complicating the earlier finding of a low, constant mutation rate in nonhuman primates, subsequent studies using deeper sequencing in human patients found substantial population heterogeneity, suggesting that the in vivo mutation rate may be higher than that observed in vitro (32–34, 119). Additionally, the Beijing sublineage of M. tuberculosis has been proposed to acquire antimicrobial resistance more rapidly owing to a faster mutation rate, though this remains controversial (123–127).

Molecular epidemiology of drug-resistance emergence. Public health interventions that target the emergence of drug-resistant TB are severely needed. Successful treatment outcomes are achieved in just under 60% of patients with MDR-TB globally and only 40% of patients with XDR-TB, and some cohorts experience much less success in the latter case (1, 128, 129). Algorithms that calculate the most parsimonious path to a common ancestor strain enable so-called molecular clock analyses that predict the order in which SNPs, including those conferring drug resistance, arise within a pool of clinical isolates. When applied to WGS data of contemporary isolates from the KwaZulu-Natal province of South Africa, these tools predicted that the evolution of current XDR strains in South Africa began in the mid-1950s, with isoniazid mono-resistance occurring first, followed by sequential accumulation of additional drug resistance mutations over time (35). In addition, this work predicted a rapid acquisition of rifampicin resistance polymorphisms once M. tuberculosis acquired isoniazid mono-resistance and found both preexisting and novel mutations likely to compensate for the fitness costs of acquiring rifampicin resistance (35). Broadening this WGS approach to encompass a global range of M. tuberculosis isolates revealed a similar finding: that the Ser315Thr mutation in katG conferring isoniazid resistance occurred prior to rifampicin resistance in over 90% of measured instances, regardless of time period or geography (36). Notably, isoniazid mono-resistance is not detected by the Xpert MTB/RIF cartridge, a broadly used molecular test to both diagnose TB and screen for common rifampicin resistance mutations (130). As a result, TB cases found to be rifampicin-susceptible by Xpert receive standard TB therapy using only two drugs — isoniazid and rifampicin — in the 4-month continuation phase (120, 121). Thus, undetected isoniazid–mono-resistant cases may receive the equivalent of unprotected rifampicin monotherapy, thereby risking rapid loss of rifampicin susceptibility.

The only currently approved and effective TB vaccine is BCG, a live, attenuated derivative of M. bovis developed by serial passage in vitro over a 13-year period (131). This vaccine strain was rapidly distributed at global scale and maintained as numerous daughter strains (Figure 3) that underwent decades of further passages prior to the advent of bacterial storage technology. Consequently, these various BCG strains contain overlapping but distinct subsets of RDs from the parental M. bovis. Chief among these, RD1 is absent from all BCG strains but present in virulent M. bovis isolates (132). RD1 deletions in three distinct M. tuberculosis strains and M. bovis caused avirulence, which was restored by complementation, validating RD1 as the primary attenuating mutation in BCG (25). Beyond RD1, however, the BCG daughter strains are not uniform for the remaining RDs and possess additional polymorphisms beyond the RD deletions (12, 15, 134). Numerous studies have examined the TB protection conferred by the global BCG repertoire in animal models and consistently found measurable differences between BCG strains (12–14, 135). Additionally, BCG strains differ in their predicted T cell epitopes, gene expression, and virulence toward immune-deficient mice, among other properties (12, 14, 136, 137). This variability extends to comparative studies of differing BCG strains in infants, as assessed by immune metrics (138–141). It is not clear at present whether these findings should influence the choice of BCG vaccine strain in different settings, particularly because of the infrastructural and ethical challenges associated with such a decision. As genome sequencing uncovers higher-resolution information regarding the global diversity of BCG strains, it is challenging to identify a consistent impact of these polymorphisms on clinical efficacy. However, this body of work clearly justifies additional consideration in BCG strain selection for laboratory studies, particularly if the aim is to generate an improved TB vaccine (142).

Figure 3

Genealogy of BCG daughter strains. Evolutionary trajectory depicting selected regions of difference (RDs) that define the modern BCG vaccine strains in relation to the parental virulent M. bovis. Strains are clustered by the presence of tandem duplications (DU1 and DU2). Adapted with permission from Vaccines (173).

There remain additional opportunities to make further use of WGS related to the BCG vaccine. Surveillance of BCG seed lots using deep sequencing has uncovered heterogeneity between and even within lots (143, 144). In select instances, variants were identified in known virulence-promoting genes or in secretory loci that may alter the antigenic profile of the vaccine (144). BCG diversity extends far beyond the large genomic deletions classified as RDs, and many variants similarly impact cell wall and secretory factors predicted to interact with the host (12, 137). Additional data regarding BCG protective efficacy and adverse outcomes may uncover determinants of mycobacterial immunogenicity if these vaccine properties can be consistently correlated to BCG genotype.

While most proteins in the M. tuberculosis genome could be classified into known or putative functions once their sequences were obtained, there remained 16% without similarity to previously known proteins (4). Among these are the PE and PPE families, named for the conserved Pro-Glu and Pro-Pro-Glu amino acids near their N-termini, which were uniquely discovered by M. tuberculosis WGS (4). Phylogenetic analysis of these families revealed a massive duplication of these proteins in pathogenic mycobacteria, especially the PE_PGRS (polymorphic GC-rich repeat sequence) and the PPE-MPTR (major polymorphic tandem repeat) subfamilies (145). Notably, this expansion was predated by the duplication of the ESX-2 secretion system to yield ESX-5, the most recently evolved type VII secretion system in M. tuberculosis (145, 146). The ESX-5 system was subsequently found to secrete PE/PPE proteins via a conserved secretion signal on the C-terminus of the PE and PPE domains, potentially rendering ESX-5 responsible for the export of approximately 150 members of the largely uncharacterized PE/PPE families (147, 148). Saturating transposon mutagenesis demonstrated that the vast majority of PE/PPE proteins are nonessential for M. tuberculosis growth in vitro, suggesting instead a role during infection (149–151). Indeed, an M. tuberculosis strain deficient for ESX-5 showed attenuation in mice, and PE/PPE proteins are highly secreted in M. tuberculosis animal models and human infections (152–156). Intriguingly, mutation at the ppe38 locus impairs secretion of many PE_PGRS and PPE-MPTR proteins but increases virulence, which is notable because the hypervirulent Beijing sublineage possesses this mutation (157, 158). As a class, PE/PPE genes show frequent polymorphisms and recombination events across clinical strains, which may impact their efficacy as antigens and putative vaccine candidates; the PPE18 component of the M72/AS01E vaccine has over 60 known non-synonymous SNPs, for example (159, 160). Clear functions have been identified for several family members, but most remain weakly characterized (161, 162). Relatedly, reads encompassing these genes are often omitted from M. tuberculosis WGS and similar studies owing to the high similarity between family members. Other reviews document the growing understanding of the roles of PE/PPE proteins during M. tuberculosis infection (163–165).

With the immense global need for novel antitubercular drugs, the M. tuberculosis genome was probed to uncover targetable pathways almost as soon as it was reported in 1998. Transposon sequencing (TnSeq) uses mobile genetic elements to create loss-of-function insertions across the bacterial genome. Large pooled libraries of M. tuberculosis strains with single transposon insertions are then evaluated in different environmental conditions to identify mutants that decrease in abundance over time, suggesting fitness defects conferred by disruption of individual genes. This approach identified genes essential for growth in vitro, a candidate list for potential therapeutic intervention (149–151). To identify M. tuberculosis genes essential in vivo and determine the subset of these factors that interact with host genotype, an M. tuberculosis transposon library was infected into the Collaborative Cross panel of outbred mice (39, 166). Notably, expanding the range of mouse genotypes more than doubled the number of M. tuberculosis genes implicated in infecting at least one host background (39). This study made exceptional use of WGS technology in a rigorous attempt to model interindividual diversity during M. tuberculosis infection.

TnSeq is valuable in building breadth of understanding about the M. tuberculosis genome, but an ideal ranking scheme would quantitatively prioritize candidate essential genes where pharmacological inhibition would be most lethal for the pathogen. This is the basis of “vulnerability” studies, which have demonstrated that certain bacterial pathways are highly susceptible to low levels of depletion while others are durable even to near-total knockdown (167, 168). Recently, this concept has incorporated CRISPR interference (CRISPRi), which uses a deactivated version of the Cas machinery that acts as a sequence-specific transcriptional repressor. This targeted approach allows the tuning of expression levels across almost all essential genes of M. tuberculosis (37). Adapting this approach to a chemical-genetic design demonstrated known and new resistance pathways for first- and second-line TB drugs, including cell envelope and stress response pathways that caused broad cross-resistance to many compounds (38). Most excitingly, this study identified a new whiB7 polymorphism that caused macrolide hypersensitivity in a particular M. tuberculosis L1 sublineage and further showed that mice infected with an L1 strain responded well to clarithromycin therapy in a short-term model. Addition of a macrolide to standard TB therapy might accelerate the treatment time for TB patients in Southeast Asia who are infected by L1 strains with this polymorphism (38). This recently reported example of unexpected macrolide susceptibility in an appreciable number of TB strains in Asia represents a promising example of how M. tuberculosis genomics can yield clinical solutions.

The M. tuberculosis genome sequence and subsequent M. tuberculosis WGS efforts have impacted nearly every aspect of the TB field. Many therapeutic innovations, including the newest generation of antitubercular drugs, rely on this technology for an improved ability to identify therapeutic targets and track resistance mechanisms (81, 95, 96). Diagnostic advancements, such as the Xpert MTB/XDR assay, likewise hinge on the growing power of M. tuberculosis genomics (8, 84, 85). Epidemiological surveillance of TB has been extensively informed by WGS both to track spread within communities and to follow M. tuberculosis lineages across broader patterns of geographic and temporal spread (6, 36, 49). Innovations stemming from the M. tuberculosis genome sequence will likely further broaden in the coming years, as findings from patient-level investigations and innovative whole-genome bacteriological designs are pursued for translational application (32–34, 37, 38).

While the genome sequence is a valuable road map, it is important to note that causality of phenotypes cannot be concluded from DNA sequences but rather requires gene transfer technologies based on plasmids, phage elements, and gene recombination tools (16–27). The many understudied variants seen in M. tuberculosis clinical isolates may provide additional insight into bacterial factors that promote host infection (31, 86, 159, 160). Relatedly, there is growing interest in associating bacterial and human genome variations to demonstrate bidirectional selective pressure exerted by the pathogen and the host immune system on one another (53, 54, 68, 71–74). We note as well several examples in which WGS has identified preexisting drug resistance or susceptibility in populations, though the appropriate use of this technology to inform region-specific regimen selection must be further considered (38, 98, 104). In addition, the adoption of a persistence phenotype by M. tuberculosis, which allows the organism to survive in the host for extreme durations, remains an area of great interest. Understanding how gene expression regulates the persistence phenotype may lead to novel approaches to allow drugs and the immune response to sterilize M. tuberculosis infections (77, 169–171).

The support of NIH grants AI152688, AI155346, AI167750, and AI155602 (to BNK and WRB) and AI026170, AI134650, and AI156853 (to WRJ) is gratefully acknowledged.

Address correspondence to: William R. Bishai, Department of Medicine, Johns Hopkins University School of Medicine, 1550 Orleans Street, CRB2 Room 108, Baltimore, Maryland 21231, USA. Phone: 410.955.3507; Email: wbishai1@jhmi.edu.

Conflict of interest: WRB and WRJ hold the following patents on recombinant mycobacteria: US6372478, US7758874, US10130663, US10842828, and US11590177.

Copyright: © 2023, Koleske et al. This is an open access article published under the terms of the Creative Commons Attribution 4.0 International License.

Reference information: J Clin Invest. 2023;133(19):e173156. https://doi.org/10.1172/JCI173156.

1. World Health Organization. Global Tuberculosis Report 2021. Geneva: World Health Organization; 2021. https://www.who.int/publications/i/item/9789240037021 Accessed August 9, 2023.
2. Sekaggya-Wiltshire C, et al. Anti-TB drug concentrations and drug-associated toxicities among TB/HIV-coinfected patients. J Antimicrob Chemother. 2017;72(4):1172–1177.
   View this article via: PubMed Google Scholar
3. CDC. Tuberculosis (TB) Incidence in the United States, 1953–2021. https://www.cdc.gov/tb/statistics/tbcases.htm Updated November 9, 2022. Accessed August 9, 2023.
4. Cole ST, et al. Deciphering the biology of Mycobacterium tuberculosis from the complete genome sequence. Nature. 1998;393(6685):537–544.
   View this article via: CrossRef PubMed Google Scholar
5. Black AT, et al. Tracking and responding to an outbreak of tuberculosis using MIRU-VNTR genotyping and whole genome sequencing as epidemiological tools. J Public Health (Oxf). 2018;40(2):66–73.
   View this article via: CrossRef PubMed Google Scholar
6. Gardy JL, et al. Whole-genome sequencing and social-network analysis of a tuberculosis outbreak. N Engl J Med. 2011;364(8):730–739.
   View this article via: CrossRef PubMed Google Scholar
7. Bryant JM, et al. Inferring patient to patient transmission of Mycobacterium tuberculosis from whole genome sequencing data. BMC Infectious Diseases. 2013;13(1):11–21.
   View this article via: CrossRef PubMed Google Scholar
8. Cao Y, et al. Xpert MTB/XDR: a 10-color reflex assay suitable for point-of-care settings to detect isoniazid, fluoroquinolone, and second-line-injectable-drug resistance directly from Mycobacterium tuberculosis-positive sputum. J Clin Microbiol. 2021;59(3):e02314-20.
   View this article via: CrossRef PubMed Google Scholar
9. Cox H, et al. Whole-genome sequencing has the potential to improve treatment for rifampicin-resistant tuberculosis in high-burden settings: a retrospective cohort study. J Clin Microbiol. 2022;60(3):e0236221.
   View this article via: CrossRef PubMed Google Scholar
10. Conradie F, et al. Treatment of highly drug-resistant pulmonary tuberculosis. N Engl J Med. 2020;382(10):893–902.
    View this article via: CrossRef PubMed Google Scholar
11. Gler MT, et al. Delamanid for multidrug-resistant pulmonary tuberculosis. N Engl J Med. 2012;366(23):2151–2160.
    View this article via: CrossRef PubMed Google Scholar
12. Leung AS, et al. Novel genome polymorphisms in BCG vaccine strains and impact on efficacy. BMC Genomics. 2008;9:413.
    View this article via: CrossRef PubMed Google Scholar
13. Castillo-Rodal AI, et al. Mycobacterium bovis BCG substrains confer different levels of protection against Mycobacterium tuberculosis infection in a BALB/c model of progressive pulmonary tuberculosis. Infect Immun. 2006;74(3):1718–1724.
    View this article via: CrossRef PubMed Google Scholar
14. Zhang L, et al. Variable virulence and efficacy of BCG vaccine strains in mice and correlation with genome polymorphisms. Mol Ther. 2016;24(2):398–405.
    View this article via: CrossRef PubMed Google Scholar
15. Behr MA, et al. Comparative genomics of BCG vaccines by whole-genome DNA microarray. Science. 1999;284(5419):1520–1523.
    View this article via: CrossRef PubMed Google Scholar
16. Heym B, et al. Implications of multidrug resistance for the future of short-course chemotherapy of tuberculosis: a molecular study. Lancet. 1994;344(8918):293–298.
    View this article via: CrossRef PubMed Google Scholar
17. Banerjee A, et al. inhA, a gene encoding a target for isoniazid and ethionamide in Mycobacterium tuberculosis. Science. 1994;263(5144):227–230.
    View this article via: CrossRef PubMed Google Scholar
18. Al-Sayyed B, et al. Monoclonal antibodies to Mycobacterium tuberculosis CDC 1551 reveal subcellular localization of MPT51. Tuberculosis (Edinb). 2007;87(6):489–497.
    View this article via: CrossRef PubMed Google Scholar
19. Jacobs WR, et al. Introduction of foreign DNA into mycobacteria using a shuttle phasmid. Nature. 1987;327(6122):532–535.
    View this article via: CrossRef PubMed Google Scholar
20. Snapper SB, et al. Lysogeny and transformation in mycobacteria: stable expression of foreign genes. Proc Natl Acad Sci U S A. 1988;85(18):6987–6991.
    View this article via: CrossRef PubMed Google Scholar
21. Lee MH, et al. Site-specific integration of mycobacteriophage L5: integration-proficient vectors for Mycobacterium smegmatis, Mycobacterium tuberculosis, and bacille Calmette-Guérin. Proc Natl Acad Sci U S A. 1991;88(8):3111–3115.
    View this article via: CrossRef PubMed Google Scholar
22. Stover CK, et al. New use of BCG for recombinant vaccines. Nature. 1991;351(6326):456–460.
    View this article via: CrossRef PubMed Google Scholar
23. Telenti A, et al. The emb operon, a gene cluster of Mycobacterium tuberculosis involved in resistance to ethambutol. Nat Med. 1997;3(5):567–570.
    View this article via: CrossRef PubMed Google Scholar
24. Larsen MH, et al. Overexpression of inhA, but not kasA, confers resistance to isoniazid and ethionamide in Mycobacterium smegmatis, M. bovis BCG and M. tuberculosis. Mol Microbiol. 2002;46(2):453–466.
    View this article via: CrossRef PubMed Google Scholar
25. Hsu T, et al. The primary mechanism of attenuation of bacillus Calmette-Guerin is a loss of secreted lytic function required for invasion of lung interstitial tissue. Proc Natl Acad Sci U S A. 2003;100(21):12420–12425.
    View this article via: CrossRef PubMed Google Scholar
26. Vilchèze C, et al. Transfer of a point mutation in Mycobacterium tuberculosisinhA resolves the target of isoniazid. Nat Med. 2006;12(9):1027–1029.
    View this article via: CrossRef PubMed Google Scholar
27. Vilchèze C, Jacobs WR. The mechanism of isoniazid killing: clarity through the scope of genetics. Annu Rev Microbiol. 2007;61:35–50.
    View this article via: CrossRef PubMed Google Scholar
28. Osório NS, et al. Evidence for diversifying selection in a set of Mycobacterium tuberculosis genes in response to antibiotic- and nonantibiotic-related pressure. Mol Biol Evol. 2013;30(6):1326–1336.
    View this article via: CrossRef PubMed Google Scholar
29. Farhat MR, et al. Genomic analysis identifies targets of convergent positive selection in drug-resistant Mycobacterium tuberculosis. Nat Genet. 2013;45(10):1183–1189.
    View this article via: CrossRef PubMed Google Scholar
30. Zhang H, et al. Genome sequencing of 161 Mycobacterium tuberculosis isolates from China identifies genes and intergenic regions associated with drug resistance. Nat Genet. 2013;45(10):1255–1260.
    View this article via: CrossRef PubMed Google Scholar
31. Brankin A, et al. A data compendium of Mycobacterium tuberculosis antibiotic resistance [preprint]. https://doi.org/10.1101/2021.09.14.460274 Posted on bioRxiv March 29, 2022.
32. Sun G, et al. Dynamic population changes in Mycobacterium tuberculosis during acquisition and fixation of drug resistance in patients. J Infect Dis. 2012;206(11):1724–1733.
    View this article via: CrossRef PubMed Google Scholar
33. Trauner A, et al. The within-host population dynamics of Mycobacterium tuberculosis vary with treatment efficacy. Genome Biol. 2017;18(1):71.
    View this article via: CrossRef PubMed Google Scholar
34. Vargas R, et al. In-host population dynamics of Mycobacterium tuberculosis complex during active disease. Elife. 2021;10:e61805.
    View this article via: CrossRef PubMed Google Scholar
35. Cohen KA, et al. Evolution of extensively drug-resistant tuberculosis over four decades: whole genome sequencing and dating analysis of Mycobacterium tuberculosis isolates from KwaZulu-Natal. PLoS Med. 2015;12(9):e1001880.
    View this article via: CrossRef PubMed Google Scholar
36. Manson AL, et al. Genomic analysis of globally diverse Mycobacterium tuberculosis strains provides insights into the emergence and spread of multidrug resistance. Nat Genet. 2017;49(3):395–402.
    View this article via: CrossRef PubMed Google Scholar
37. Bosch B, et al. Genome-wide gene expression tuning reveals diverse vulnerabilities of M. tuberculosis. Cell. 2021;184(17):4579–4592.
    View this article via: CrossRef PubMed Google Scholar
38. Li S, et al. CRISPRi chemical genetics and comparative genomics identify genes mediating drug potency in Mycobacterium tuberculosis. Nat Microbiol. 2022;7(6):766–779.
    View this article via: CrossRef PubMed Google Scholar
39. Smith CM, et al. Host-pathogen genetic interactions underlie tuberculosis susceptibility in genetically diverse mice. Elife. 2022;11:e74419.
    View this article via: CrossRef PubMed Google Scholar
40. Goyal M, et al. Differentiation of Mycobacterium tuberculosis isolates by spoligotyping and IS6110 restriction fragment length polymorphism. J Clin Microbiol. 1997;35(3):647–651.
    View this article via: CrossRef PubMed Google Scholar
41. Gutacker MM, et al. Genome-wide analysis of synonymous single nucleotide polymorphisms in Mycobacterium tuberculosis complex organisms: resolution of genetic relationships among closely related microbial strains. Genetics. 2002;162(4):1533–1543.
    View this article via: CrossRef PubMed Google Scholar
42. Gutacker MM, et al. Single-nucleotide polymorphism-based population genetic analysis of Mycobacterium tuberculosis strains from 4 geographic sites. J Infect Dis. 2006;193(1):121–128.
    View this article via: CrossRef PubMed Google Scholar
43. Filliol I, et al. Global phylogeny of Mycobacterium tuberculosis based on single nucleotide polymorphism (SNP) analysis: insights into tuberculosis evolution, phylogenetic accuracy of other DNA fingerprinting systems, and recommendations for a minimal standard SNP set. J Bacteriol. 2006;188(2):759–772.
    View this article via: CrossRef PubMed Google Scholar
44. Hirsh AE, et al. Stable association between strains of Mycobacterium tuberculosis and their human host populations. Proc Natl Acad Sci U S A. 2004;101(14):4871–4876.
    View this article via: CrossRef PubMed Google Scholar
45. Coscolla M, Gagneux S. Consequences of genomic diversity in Mycobacterium tuberculosis. Semin Immunol. 2014;26(6):431–444.
    View this article via: CrossRef PubMed Google Scholar
46. Bespiatykh D, et al. A comprehensive map of Mycobacterium tuberculosis complex regions of difference. mSphere. 2021;6(4):e0053521.
    View this article via: CrossRef PubMed Google Scholar
47. Brosch R, et al. A new evolutionary scenario for the Mycobacterium tuberculosis complex. Proc Natl Acad Sci U S A. 2002;99(6):3684–3689.
    View this article via: CrossRef PubMed Google Scholar
48. Bottai D, et al. TbD1 deletion as a driver of the evolutionary success of modern epidemic Mycobacterium tuberculosis lineages. Nat Commun. 2020;11(1):684.
    View this article via: CrossRef PubMed Google Scholar
49. Żukowska L, et al. An overview of tuberculosis outbreaks reported in the years 2011-2020. BMC Infect Dis. 2023;23(1):253.
    View this article via: CrossRef PubMed Google Scholar
50. de Jong BC, et al. Mycobacterium africanum — review of an important cause of human tuberculosis in West Africa. PLoS Negl Trop Dis. 2010;4(9):e744.
    View this article via: CrossRef PubMed Google Scholar
51. Brites D, et al. A new phylogenetic framework for the animal-adapted Mycobacterium tuberculosis complex. Front Microbiol. 2018;9:2820.
    View this article via: CrossRef PubMed Google Scholar
52. Brynildsrud OB, et al. Global expansion of Mycobacterium tuberculosis lineage 4 shaped by colonial migration and local adaptation. Sci Adv. 2018;4(10):eaat5869.
    View this article via: CrossRef PubMed Google Scholar
53. Stucki D, et al. Mycobacterium tuberculosis lineage 4 comprises globally distributed and geographically restricted sublineages. Nat Genet. 2016;48(12):1535–1543.
    View this article via: CrossRef PubMed Google Scholar
54. Gröschel MIP-L, et al. Host-pathogen sympatry and differential transmissibility of Mycobacterium tuberculosis complex [preprint]. https://doi.org/10.1101/2022.08.04.22278337 Posted on medRxiv February 15, 2023.
55. Coll F, et al. A robust SNP barcode for typing Mycobacterium tuberculosis complex strains. Nat Commun. 2014;5:4812.
    View this article via: CrossRef PubMed Google Scholar
56. Napier G, et al. Robust barcoding and identification of Mycobacterium tuberculosis lineages for epidemiological and clinical studies. Genome Med. 2020;12(1):114.
    View this article via: CrossRef PubMed Google Scholar
57. van Soolingen D, et al. A novel pathogenic taxon of the Mycobacterium tuberculosis complex, Canetti: characterization of an exceptional isolate from Africa. Int J Syst Bacteriol. 1997;47(4):1236–1245.
    View this article via: CrossRef PubMed Google Scholar
58. Fabre M, et al. High genetic diversity revealed by variable-number tandem repeat genotyping and analysis of hsp65 gene polymorphism in a large collection of “Mycobacterium canettii” strains indicates that the M. tuberculosis complex is a recently emerged clone of “M. canettii”. J Clin Microbiol. 2004;42(7):3248–3255.
    View this article via: CrossRef PubMed Google Scholar
59. Gutierrez MC, et al. Ancient origin and gene mosaicism of the progenitor of Mycobacterium tuberculosis. PLoS Pathog. 2005;1(1):e5.
    View this article via: CrossRef PubMed Google Scholar
60. Coscolla M, et al. Phylogenomics of Mycobacterium africanum reveals a new lineage and a complex evolutionary history. Microb Genom. 2021;7(2):1–14.
    View this article via: PubMed Google Scholar
61. Blouin Y, et al. Significance of the identification in the Horn of Africa of an exceptionally deep branching Mycobacterium tuberculosis clade. PLoS One. 2012;7(12):e52841.
    View this article via: CrossRef PubMed Google Scholar
62. Ngabonziza JCS, et al. A sister lineage of the Mycobacterium tuberculosis complex discovered in the African Great Lakes region. Nat Commun. 2020;11(1):2917.
    View this article via: CrossRef PubMed Google Scholar
63. Sreevatsan S, et al. Restricted structural gene polymorphism in the Mycobacterium tuberculosis complex indicates evolutionarily recent global dissemination. PNAS. 1997;94(18):9869–9874.
    View this article via: CrossRef PubMed Google Scholar
64. Comas I, et al. Out-of-Africa migration and Neolithic coexpansion of Mycobacterium tuberculosis with modern humans. Nat Genet. 2013;45(10):1176–1182.
    View this article via: CrossRef PubMed Google Scholar
65. Hershberg R, et al. High functional diversity in Mycobacterium tuberculosis driven by genetic drift and human demography. PLoS Biol. 2008;6(12):e311.
    View this article via: CrossRef PubMed Google Scholar
66. Namouchi A, et al. After the bottleneck: genome-wide diversification of the Mycobacterium tuberculosis complex by mutation, recombination, and natural selection. Genome Res. 2012;22(4):721–734.
    View this article via: CrossRef PubMed Google Scholar
67. Eldholm V, et al. Armed conflict and population displacement as drivers of the evolution and dispersal of Mycobacterium tuberculosis. Proc Natl Acad Sci U S A. 2016;113(48):13881–13886.
    View this article via: CrossRef PubMed Google Scholar
68. Sarkar R, et al. Modern lineages of Mycobacterium tuberculosis exhibit lineage-specific patterns of growth and cytokine induction in human monocyte-derived macrophages. PLoS One. 2012;7(8):e43170.
    View this article via: CrossRef PubMed Google Scholar
69. Reiling N, et al. Clade-specific virulence patterns of Mycobacterium tuberculosis complex strains in human primary macrophages and aerogenically infected mice. mBio. 2013;4(4):e00250-13.
    View this article via: CrossRef PubMed Google Scholar
70. Romagnoli A, et al. Clinical isolates of the modern Mycobacterium tuberculosis lineage 4 evade host defense in human macrophages through eluding IL-1β-induced autophagy. Cell Death Dis. 2018;9(6):624.
    View this article via: CrossRef PubMed Google Scholar
71. Chen YY, et al. The pattern of cytokine production in vitro induced by ancient and modern Beijing Mycobacterium tuberculosis strains. PLoS One. 2014;9(4):e94296.
    View this article via: CrossRef PubMed Google Scholar
72. Krishnan N, et al. Mycobacterium tuberculosis lineage influences innate immune response and virulence and is associated with distinct cell envelope lipid profiles. PLoS One. 2011;6(9):e23870.
    View this article via: CrossRef PubMed Google Scholar
73. Yimer SA, et al. Lineage-specific proteomic signatures in the Mycobacterium tuberculosis complex reveal differential abundance of proteins involved in virulence, DNA repair, CRISPR-Cas, bioenergetics and lipid metabolism. Front Microbiol. 2020;11:550760.
    View this article via: CrossRef PubMed Google Scholar
74. Chacón-Salinas R, et al. Differential pattern of cytokine expression by macrophages infected in vitro with different Mycobacterium tuberculosis genotypes. Clin Exp Immunol. 2005;140(3):443–449.
    View this article via: CrossRef PubMed Google Scholar
75. Reed MB, et al. A glycolipid of hypervirulent tuberculosis strains that inhibits the innate immune response. Nature. 2004;431(7004):84–87.
    View this article via: CrossRef PubMed Google Scholar
76. Reed MB, et al. The W-Beijing lineage of Mycobacterium tuberculosis overproduces triglycerides and has the DosR dormancy regulon constitutively upregulated. J Bacteriol. 2007;189(7):2583–2589.
    View this article via: CrossRef PubMed Google Scholar
77. Vilchèze C, Jacobs WR. Resistance to isoniazid and ethionamide in Mycobacterium tuberculosis: genes, mutations, and causalities. Microbiol Spectr. 2014;2(4):MGM2-0014-2013.
    View this article via: CrossRef PubMed Google Scholar
78. Allix-Béguec C, et al. Prediction of susceptibility to first-line tuberculosis drugs by DNA sequencing. N Engl J Med. 2018;379(15):1403–1415.
    View this article via: CrossRef PubMed Google Scholar
79. Pankhurst LJ, et al. Rapid, comprehensive, and affordable mycobacterial diagnosis with whole-genome sequencing: a prospective study. Lancet Respir Med. 2016;4(1):49–58.
    View this article via: CrossRef PubMed Google Scholar
80. Witney AA, et al. Clinical use of whole genome sequencing for Mycobacterium tuberculosis. BMC Med. 2016;14:46.
    View this article via: CrossRef PubMed Google Scholar
81. Enkirch T, et al. Systematic review of whole-genome sequencing data to predict phenotypic drug resistance and susceptibility in Swedish Mycobacterium tuberculosis isolates, 2016 to 2018. Antimicrob Agents Chemother. 2020;64(5):e02550-19.
    View this article via: CrossRef PubMed Google Scholar
82. Park M, et al. Evaluating the clinical impact of routine whole genome sequencing in tuberculosis treatment decisions and the issue of isoniazid mono-resistance. BMC Infect Dis. 2022;22(1):349.
    View this article via: CrossRef PubMed Google Scholar
83. Vogel M, et al. Implementation of whole genome sequencing for tuberculosis diagnostics in a low-middle income, high MDR-TB burden country. Sci Rep. 2021;11(1):15333.
    View this article via: CrossRef PubMed Google Scholar
84. Penn-Nicholson A, et al. Detection of isoniazid, fluoroquinolone, ethionamide, amikacin, kanamycin, and capreomycin resistance by the Xpert MTB/XDR assay: a cross-sectional multicentre diagnostic accuracy study. Lancet Infect Dis. 2022;22(2):242–249.
    View this article via: CrossRef PubMed Google Scholar
85. Pillay S, et al. Xpert MTB/XDR for detection of pulmonary tuberculosis and resistance to isoniazid, fluoroquinolones, ethionamide, and amikacin. Cochrane Database Syst Rev. 2022;5(5):CD014841.
    View this article via: PubMed Google Scholar
86. Shanmugam SK, et al. Mycobacterium tuberculosis lineages associated with mutations and drug resistance in isolates from India. Microbiol Spectr. 2022;10(3):e0159421.
    View this article via: CrossRef PubMed Google Scholar
87. Walker TM, et al. Whole genome sequencing for M/XDR tuberculosis surveillance and for resistance testing. Clin Microbiol Infect. 2017;23(3):161–166.
    View this article via: CrossRef PubMed Google Scholar
88. Cohen KA, et al. Deciphering drug resistance in Mycobacterium tuberculosis using whole-genome sequencing: progress, promise, and challenges. Genome Med. 2019;11(1):45.
    View this article via: CrossRef PubMed Google Scholar
89. Finci I, et al. Investigating resistance in clinical Mycobacterium tuberculosis complex isolates with genomic and phenotypic antimicrobial susceptibility testing: a multicentre observational study. Lancet Microbe. 2022;3(9):e672–e682.
    View this article via: CrossRef PubMed Google Scholar
90. Andries K, et al. A diarylquinoline drug active on the ATP synthase of Mycobacterium tuberculosis. Science. 2005;307(5707):223–227.
    View this article via: CrossRef PubMed Google Scholar
91. Koul A, et al. Diarylquinolines are bactericidal for dormant mycobacteria as a result of disturbed ATP homeostasis. J Biol Chem. 2008;283(37):25273–25280.
    View this article via: CrossRef PubMed Google Scholar
92. Sparks IL, et al. Mycobacterium smegmatis: the vanguard of mycobacterial research. J Bacteriol. 2023;205(1):e0033722.
    View this article via: CrossRef PubMed Google Scholar
93. Petrella S, et al. Genetic basis for natural and acquired resistance to the diarylquinoline R207910 in mycobacteria. Antimicrob Agents Chemother. 2006;50(8):2853–2856.
    View this article via: CrossRef PubMed Google Scholar
94. Andries K, et al. Acquired resistance of Mycobacterium tuberculosis to bedaquiline. PLoS One. 2014;9(7):e102135.
    View this article via: CrossRef PubMed Google Scholar
95. Kadura S, et al. Systematic review of mutations associated with resistance to the new and repurposed Mycobacterium tuberculosis drugs bedaquiline, clofazimine, linezolid, delamanid and pretomanid. J Antimicrob Chemother. 2020;75(8):2031–2043.
    View this article via: CrossRef PubMed Google Scholar
96. Nieto Ramirez LM, et al. Whole genome sequencing for the analysis of drug resistant strains of Mycobacterium tuberculosis: a systematic review for bedaquiline and delamanid. Antibiotics (Basel). 2020;9(3):133.
    View this article via: CrossRef PubMed Google Scholar
97. Omar SV, et al. Bedaquiline-resistant tuberculosis associated with Rv0678 mutations. N Engl J Med. 2022;386(1):93–94.
    View this article via: CrossRef PubMed Google Scholar
98. Villellas C, et al. Unexpected high prevalence of resistance-associated Rv0678 variants in MDR-TB patients without documented prior use of clofazimine or bedaquiline. J Antimicrob Chemother. 2017;72(3):684–690.
    View this article via: PubMed Google Scholar
99. Stover CK, et al. A small-molecule nitroimidazopyran drug candidate for the treatment of tuberculosis. Nature. 2000;405(6789):962–966.
    View this article via: CrossRef PubMed Google Scholar
100. Matsumoto M, et al. OPC-67683, a nitro-dihydro-imidazooxazole derivative with promising action against tuberculosis in vitro and in mice. PLoS Med. 2006;3(11):e466.
     View this article via: CrossRef PubMed Google Scholar
101. Manjunatha UH, et al. Identification of a nitroimidazo-oxazine-specific protein involved in PA-824 resistance in Mycobacterium tuberculosis. Proc Natl Acad Sci U S A. 2006;103(2):431–436.
     View this article via: CrossRef PubMed Google Scholar
102. Singh R, et al. PA-824 kills nonreplicating Mycobacterium tuberculosis by intracellular NO release. Science. 2008;322(5906):1392–1395.
     View this article via: CrossRef PubMed Google Scholar
103. Rifat D, et al. Mutations in fbiD (Rv2983) as a novel determinant of resistance to pretomanid and delamanid in Mycobacterium tuberculosis. Antimicrob Agents Chemother. 2020;65(1):e01948-20.
     View this article via: CrossRef PubMed Google Scholar
104. Gómez-González PJ, et al. Genetic diversity of candidate loci linked to Mycobacterium tuberculosis resistance to bedaquiline, delamanid and pretomanid. Sci Rep. 2021;11(1):19431.
     View this article via: CrossRef PubMed Google Scholar
105. Fujiwara M, et al. Mechanisms of resistance to delamanid, a drug for Mycobacterium tuberculosis. Tuberculosis (edinb). 2018;108:186–194.
     View this article via: CrossRef PubMed Google Scholar
106. Wyllie DH, et al. A quantitative evaluation of MIRU-VNTR typing against whole-genome sequencing for identifying Mycobacterium tuberculosis transmission: a prospective observational cohort study. EBioMedicine. 2018;34:122–130.
     View this article via: CrossRef PubMed Google Scholar
107. Luo T, et al. Combination of single nucleotide polymorphism and variable-number tandem repeats for genotyping a homogenous population of Mycobacterium tuberculosis Beijing strains in China. J Clin Microbiol. 2012;50(3):633–639.
     View this article via: CrossRef PubMed Google Scholar
108. Asare P, et al. The relevance of genomic epidemiology for control of tuberculosis in west Africa. Front Public Health. 2021;9:706651.
     View this article via: CrossRef PubMed Google Scholar
109. Rich AR, ed. The Pathogenesis of Tuberculosis. 2nd ed. Charles C. Thomas, Publisher; 1951.
110. Alland D, et al. Transmission of tuberculosis in New York City. An analysis by DNA fingerprinting and conventional epidemiologic methods. N Engl J Med. 1994;330(24):1710–1716.
     View this article via: CrossRef PubMed Google Scholar
111. Small PM, et al. The epidemiology of tuberculosis in San Francisco — a population-based study using conventional and molecular methods. N Engl J Med. 1994;330(24):1703–1709.
     View this article via: CrossRef PubMed Google Scholar
112. Bishai WR, et al. Molecular and geographic patterns of tuberculosis transmission after 15 years of directly observed therapy. JAMA. 1998;280(19):1679–1684.
     View this article via: CrossRef PubMed Google Scholar
113. Behr M, et al. Is Mycobacterium tuberculosis infection life long? BMJ. 2019;367:l5770.
     View this article via: CrossRef PubMed Google Scholar
114. Behr MA, et al. Latent tuberculosis: two centuries of confusion. Am J Respir Crit Care Med. 2021;204(2):142–148.
     View this article via: CrossRef PubMed Google Scholar
115. Sterling TR, et al. Three months of rifapentine and isoniazid for latent tuberculosis infection. N Engl J Med. 2011;365(23):2155–2166.
     View this article via: CrossRef PubMed Google Scholar
116. Swindells S, et al. One month of rifapentine plus isoniazid to prevent HIV-related tuberculosis. N Engl J Med. 2019;380(11):1001–1011.
     View this article via: CrossRef PubMed Google Scholar
117. Mangione CM, et al. Screening for latent tuberculosis infection in adults: US preventive services task force recommendation statement. JAMA. 2023;329(17):1487–1494.
     View this article via: CrossRef PubMed Google Scholar
118. Nelson KN, et al. Mutation of Mycobacterium tuberculosis and implications for using whole-genome sequencing for investigating recent tuberculosis transmission. Front Public Health. 2021;9:790544.
     View this article via: CrossRef PubMed Google Scholar
119. Ford CB, et al. Use of whole genome sequencing to estimate the mutation rate of Mycobacterium tuberculosis during latent infection. Nat Genet. 2011;43(5):482–486.
     View this article via: CrossRef PubMed Google Scholar
120. Balcells ME, et al. Isoniazid preventive therapy and risk for resistant tuberculosis. Emerg Infect Dis. 2006;12(5):744–751.
     View this article via: CrossRef PubMed Google Scholar
121. Cattamanchi A, et al. Clinical characteristics and treatment outcomes of patients with isoniazid-monoresistant tuberculosis. Clin Infect Dis. 2009;48(2):179–185.
     View this article via: CrossRef PubMed Google Scholar
122. Genestet C, et al. Within-host genetic micro-diversity of Mycobacterium tuberculosis and the link with tuberculosis disease features [preprint]. https://doi.org/10.1101/2021.04.07.438754 Posted on bioRxiv April 8, 2021.
123. McGrath M, et al. Mutation rate and the emergence of drug resistance in Mycobacterium tuberculosis. J Antimicrob Chemother. 2014;69(2):292–302.
     View this article via: CrossRef PubMed Google Scholar
124. Kremer K, et al. Definition of the Beijing/W lineage of Mycobacterium tuberculosis on the basis of genetic markers. J Clin Microbiol. 2004;42(9):4040–4049.
     View this article via: CrossRef PubMed Google Scholar
125. Ford CB, et al. Mycobacterium tuberculosis mutation rate estimates from different lineages predict substantial differences in the emergence of drug-resistant tuberculosis. Nat Genet. 2013;45(7):784–790.
     View this article via: CrossRef PubMed Google Scholar
126. Hakamata M, et al. Higher genome mutation rates of Beijing lineage of Mycobacterium tuberculosis during human infection. Sci Rep. 2020;10(1):17997.
     View this article via: CrossRef PubMed Google Scholar
127. Werngren J, Hoffner SE. Drug-susceptible Mycobacterium tuberculosis Beijing genotype does not develop mutation-conferred resistance to rifampin at an elevated rate. J Clin Microbiol. 2003;41(4):1520–1524.
     View this article via: CrossRef PubMed Google Scholar
128. Abubakar M, et al. Treatment outcomes of extensively drug-resistant tuberculosis in Pakistan: a countrywide retrospective record review. Front Pharmacol. 2021;12:640555.
     View this article via: CrossRef PubMed Google Scholar
129. Pietersen E, et al. Long-term outcomes of patients with extensively drug-resistant tuberculosis in South Africa: a cohort study. Lancet. 2014;383(9924):1230–1239.
     View this article via: CrossRef PubMed Google Scholar
130. Boehme CC, et al. Rapid molecular detection of tuberculosis and rifampin resistance. N Engl J Med. 2010;363(11):1005–1015.
     View this article via: CrossRef PubMed Google Scholar
131. Calmette A, Guérin C. Nouvelles recherches expérimentales sur la vaccination des bovidés contre la tuberculose. Ann Inst Pasteur. 1920;34:553–561.
132. Mahairas GG, et al. Molecular analysis of genetic differences between Mycobacterium bovis BCG and virulent M. bovis. J Bacteriol. 1996;178(5):1274–1282.
     View this article via: CrossRef PubMed Google Scholar
133. Pym AS, et al. Recombinant BCG exporting ESAT-6 confers enhanced protection against tuberculosis. Nat Med. 2003;9(5):533–539.
     View this article via: CrossRef PubMed Google Scholar
134. Brosch R, et al. Genome plasticity of BCG and impact on vaccine efficacy. Proc Natl Acad Sci U S A. 2007;104(13):5596–5601.
     View this article via: CrossRef PubMed Google Scholar
135. Ritz N, et al. Influence of BCG vaccine strain on the immune response and protection against tuberculosis. FEMS Microbiol Rev. 2008;32(5):821–841.
     View this article via: CrossRef PubMed Google Scholar
136. Zhang W, et al. Genome sequencing and analysis of BCG vaccine strains. PLoS One. 2013;8(8):e71243.
     View this article via: CrossRef PubMed Google Scholar
137. Abdallah AM, et al. Genomic expression catalogue of a global collection of BCG vaccine strains show evidence for highly diverged metabolic and cell-wall adaptations. Sci Rep. 2015;5:15443.
     View this article via: CrossRef PubMed Google Scholar
138. Davids V, et al. The effect of bacille Calmette-Guérin vaccine strain and route of administration on induced immune responses in vaccinated infants. J Infect Dis. 2006;193(4):531–536.
     View this article via: CrossRef PubMed Google Scholar
139. Anderson EJ, et al. The influence of BCG vaccine strain on mycobacteria-specific and non-specific immune responses in a prospective cohort of infants in Uganda. Vaccine. 2012;30(12):2083–2089.
     View this article via: CrossRef PubMed Google Scholar
140. Kiravu A, et al. Bacille Calmette-Guérin vaccine strain modulates the ontogeny of both mycobacterial-specific and heterologous T cell immunity to vaccination in infants. Front Immunol. 2019;10:2307.
     View this article via: CrossRef PubMed Google Scholar
141. Subbian S, et al. BCG vaccination of infants confers Mycobacterium tuberculosis strain-specific immune responses by leukocytes. ACS Infect Dis. 2020;6(12):3141–3146.
     View this article via: CrossRef PubMed Google Scholar
142. da Costa AC, et al. Recombinant BCG: innovations on an old vaccine. Scope of BCG strains and strategies to improve long-lasting memory. Front Immunol. 2014;5:152.
     View this article via: PubMed Google Scholar
143. Wada T, et al. Deep sequencing analysis of the heterogeneity of seed and commercial lots of the bacillus Calmette-Guérin (BCG) tuberculosis vaccine substrain Tokyo-172. Sci Rep. 2015;5:17827.
     View this article via: CrossRef PubMed Google Scholar
144. Narvskaya O, et al. First insight into the whole-genome sequence variations in Mycobacterium bovis BCG-1 (Russia) vaccine seed lots and their progeny clinical isolates from children with BCG-induced adverse events. BMC Genomics. 2020;21(1):567.
     View this article via: CrossRef PubMed Google Scholar
145. Gey van Pittius NC, et al. Evolution and expansion of the Mycobacterium tuberculosis PE and PPE multigene families and their association with the duplication of the ESAT-6 (esx) gene cluster regions. BMC Evol Biol. 2006;6:95.
     View this article via: CrossRef PubMed Google Scholar
146. Gey Van Pittius NC, et al. The ESAT-6 gene cluster of Mycobacterium tuberculosis and other high G+C Gram-positive bacteria. Genome Biol. 2001;2(10):RESEARCH0044.
     View this article via: PubMed Google Scholar
147. Abdallah AM, et al. PPE and PE_PGRS proteins of Mycobacterium marinum are transported via the type VII secretion system ESX-5. Mol Microbiol. 2009;73(3):329–340.
     View this article via: CrossRef PubMed Google Scholar
148. Daleke MH, et al. General secretion signal for the mycobacterial type VII secretion pathway. Proc Natl Acad Sci U S A. 2012;109(28):11342–11347.
     View this article via: CrossRef PubMed Google Scholar
149. Sassetti CM, et al. Genes required for mycobacterial growth defined by high density mutagenesis. Mol Microbiol. 2003;48(1):77–84.
     View this article via: CrossRef PubMed Google Scholar
150. Griffin JE, et al. High-resolution phenotypic profiling defines genes essential for mycobacterial growth and cholesterol catabolism. PLoS Pathog. 2011;7(9):e1002251.
     View this article via: CrossRef PubMed Google Scholar
151. DeJesus MA, et al. Comprehensive essentiality analysis of the Mycobacterium tuberculosis genome via saturating transposon mutagenesis. mBio. 2017;8(1):e02133-16.
     View this article via: CrossRef PubMed Google Scholar
152. Tiwari S, et al. BCG-Prime and boost with Esx-5 secretion system deletion mutant leads to better protection against clinical strains of Mycobacterium tuberculosis. Vaccine. 2020;38(45):7156–7165.
     View this article via: CrossRef PubMed Google Scholar
153. Kruh NA, et al. Portrait of a pathogen: the Mycobacterium tuberculosis proteome in vivo. PLoS One. 2010;5(11):e13938.
     View this article via: CrossRef PubMed Google Scholar
154. Soldini S, et al. PPE_MPTR genes are differentially expressed by Mycobacterium tuberculosis in vivo. Tuberculosis (Edinb). 2011;91(6):563–568.
     View this article via: CrossRef PubMed Google Scholar
155. Copin R, et al. Sequence diversity in the pe_pgrs genes of Mycobacterium tuberculosis is independent of human T cell recognition. mBio. 2014;5(1):e00960-e00913.
     View this article via: CrossRef PubMed Google Scholar
156. Vita R, et al. The Immune Epitope Database (IEDB): 2018 update. Nucleic Acids Res. 2019;47(d1):D339–D343.
     View this article via: CrossRef PubMed Google Scholar
157. McEvoy CR, et al. Evidence for a rapid rate of molecular evolution at the hypervariable and immunogenic Mycobacterium tuberculosis PPE38 gene region. BMC Evol Biol. 2009;9:237.
     View this article via: CrossRef PubMed Google Scholar
158. Ates LS, et al. Mutations in ppe38 block PE_PGRS secretion and increase virulence of Mycobacterium tuberculosis. Nat Microbiol. 2018;3(2):181–188.
     View this article via: CrossRef PubMed Google Scholar
159. Phelan JE, et al. Recombination in pe/ppe genes contributes to genetic variation in Mycobacterium tuberculosis lineages. BMC Genomics. 2016;17:151.
     View this article via: CrossRef PubMed Google Scholar
160. Homolka S, et al. High sequence variability of the ppE18 gene of clinical Mycobacterium tuberculosis complex strains potentially impacts effectivity of vaccine candidate M72/AS01E. PLoS One. 2016;11(3):e0152200.
     View this article via: CrossRef PubMed Google Scholar
161. Deb C, et al. A novel lipase belonging to the hormone-sensitive lipase family induced under starvation to utilize stored triacylglycerol in Mycobacterium tuberculosis. J Biol Chem. 2006;281(7):3866–3875.
     View this article via: CrossRef PubMed Google Scholar
162. Wang Q, et al. PE/PPE proteins mediate nutrient transport across the outer membrane of Mycobacterium tuberculosis. Science. 2020;367(6482):1147–1151.
     View this article via: CrossRef PubMed Google Scholar
163. Fishbein S, et al. Phylogeny to function: PE/PPE protein evolution and impact on Mycobacterium tuberculosis pathogenicity. Mol Microbiol. 2015;96(5):901–916.
     View this article via: CrossRef PubMed Google Scholar
164. Ates LS. New insights into the mycobacterial PE and PPE proteins provide a framework for future research. Mol Microbiol. 2020;113(1):4–21.
     View this article via: CrossRef PubMed Google Scholar
165. De Maio F, et al. PE_PGRS proteins of Mycobacterium tuberculosis: a specialized molecular task force at the forefront of host-pathogen interaction. Virulence. 2020;11(1):898–915.
     View this article via: CrossRef PubMed Google Scholar
166. Churchill GA, et al. The Collaborative Cross, a community resource for the genetic analysis of complex traits. Nat Genet. 2004;36(11):1133–1137.
     View this article via: CrossRef PubMed Google Scholar
167. Wei JR, et al. Depletion of antibiotic targets has widely varying effects on growth. Proc Natl Acad Sci U S A. 2011;108(10):4176–4181.
     View this article via: CrossRef PubMed Google Scholar
168. Carroll P, et al. Identifying vulnerable pathways in Mycobacterium tuberculosis by using a knockdown approach. Appl Environ Microbiol. 2011;77(14):5040–5043.
     View this article via: CrossRef PubMed Google Scholar
169. Jain P, et al. Dual-reporter mycobacteriophages (Φ2DRMs) reveal preexisting Mycobacterium tuberculosis persistent cells in human sputum. mBio. 2016;7(5):e01023-16.
     View this article via: CrossRef PubMed Google Scholar
170. Vilchèze C, Jacobs WR. The isoniazid paradigm of killing, resistance, and persistence in Mycobacterium tuberculosis. J Mol Biol. 2019;431(18):3450–3461.
     View this article via: CrossRef PubMed Google Scholar
171. Vilchèze C, et al. Sterilization by adaptive immunity of a conditionally persistent mutant of Mycobacterium tuberculosis. mBio. 2021;12(1):e02391-20.
     View this article via: CrossRef PubMed Google Scholar
172. Gastner MT, et al. Fast flow-based algorithm for creating density-equalizing map projections. Proc Natl Acad Sci U S A. 2018;115(10):E2156–E2164.
     View this article via: CrossRef PubMed Google Scholar
173. Panaiotov S, et al. Complete genome sequence, genome stability and phylogeny of the vaccine strain Mycobacterium bovis BCG SL222 Sofia. Vaccines (Basel). 2021;9(3):237.
     View this article via: CrossRef PubMed Google Scholar