Tuberculosis (TB), the disease caused by the pathogen Mycobacterium tuberculosis (M.tb), is currently estimated to cause ~1.6 million deaths annually worldwide and is considered a leading cause of death due to a single infectious organism (WHO, 2022a). Recently, the World Health Organization (WHO) updated their definitions of drug-resistant (DR)-TB along with the newest approved treatment regimens being implemented. As defined, multi drug-resistant (MDR)-TB is caused by a M.tb strain with resistance to isoniazid (INH, H) and rifampicin (RIF, R). Pre-extensively drug-resistant (pre-XDR)-TB is caused by a M.tb strain resistant to RIF (may also be resistant to INH) and to at least one fluoroquinolone [levofloxacin (LEV) or moxifloxacin (MFX, M)]. Lastly, M.tb strains that cause extensively drug-resistant (XDR)-TB are resistant to RIF (may also be resistant to INH), at least one fluoroquinolone (LEV or MFX), and one “Group A” drug [bedaquiline (BDQ, B) or linezolid (LZD, L)] ( Table 1 ) (WHO, 2022c). From this point on, the one letter drug abbreviation will only be used when referring to a combined treatment.

Standard anti-TB treatment regimens for both drug-susceptible (DS)- and DR-TB were also recently updated by the WHO in May 2022. Previously, treatment for DS-TB included a 2-month intensive phase of treatment with INH, RIF, pyrazinamide (PZA, Z) and ethambutol (EMB, E), followed by a 4-month continuation phase with just INH and RIF (2HRZE/4HR) (WHO, 2022b). The new WHO recommendation for DS-TB treatment also includes implementation of a newer, 4-month daily treatment regimen consisting of rifapentine (RPT), INH, PZA, and MFX that is non-inferior to the 6-month DS-TB treatment regimens ( Table 2 ) (Carr et al., 2022, WHO, 2022b). The WHO updated recommendations for DR-TB treatment includes three separate treatment options for diagnosed patients. Choice of DR-TB treatment regimen is based on factors such as the M.tb drug-resistance profile, prior exposure to anti-TB drugs, and the extent of pulmonary TB disease among others (WHO, 2022c). Of the three possible treatment regimens, the newest and preferred DR-TB regimen for RIF-resistant (RR)/MDR-TB cases consists of BDQ, pretomanid (Pa), LZD, and MFX (BPaL+ M) as an all-oral regimen that lasts 6 months. For pre-XDR-TB cases, the same treatment regimen is recommended without MFX (Nyang'wa et al., 2022, WHO, 2022c). An additional recommended treatment for MDR/RR-TB is a 9-month all-oral regimen consisting of BDQ for 4-6 months in combination with a fluoroquinolone, ethionamide (ETH), EMB, INH (high dose), PZA, and clofazimine (CFZ), and treatment with a fluoroquinolone, CFZ, EMB, and PZA for the remaining 5 months (WHO, 2022c). Finally, a longer individualized treatment regimen is recommended for patients with DR-TB that have not had desirable outcomes with other regimens, are intolerant to specific components of the other regimens, or have XDR-TB (WHO, 2022c). These individualized regimens last at least 18 months and have poor adherence to due to the long duration and the increased likelihood of adverse side effects; thus, one of the other two regimens is preferred if possible.

Many anti-TB drugs target M.tb cell envelope biogenesis ( Table 3 ). However, the mechanisms used by M.tb to regulate its cell envelope composition and the outcomes in relation to drug resistance (and, consequently, how this regulation influences TB pathogenesis) are poorly understood. Various studies have examined the complex cell envelope of M.tb and show that there are differences in lipid composition that are strain specific (Torrelles et al., 2004; Torrelles et al., 2008; Torrelles et al., 2011; Howard et al., 2018). Thus, in order to advance the TB field, it is required that we increase our understanding of similarities and/or differences in the cell envelope composition of DR-M.tb strains and how these changes influence bacteria-host cell interactions and infection outcomes. In this review, we will present what is currently known about how the M.tb cell envelope changes in relation to drug resistance, examine the influence that drug resistance has on M.tb infection outcomes, and address some of the gaps in knowledge that remain to be elucidated.

The M.tb cell envelope is a thick and complex structure that provides a unique barrier of protection to the bacterium in various environments (Garcia-Vilanova et al., 2019). It is comprised of multiple layers that each contribute to its complex composition (detailed in Figure 1 , adapted from (Jackson, 2014). The innermost layer is a plasma membrane similar to that of many other bacterial species (Jackson, 2014). Outside of the plasma membrane, a layer of peptidoglycan (PG) is covalently attached to the arabinogalactan (AG) polysaccharide. The non-reducing ends of AG are esterified to the long-chain (C₆₀-C₉₀) mycolic acids (MAcs) that are unique in nature to Mycobacterium spp and complete the ‘cell wall core’ of the cell envelope (Jackson, 2014; Garcia-Vilanova et al., 2019). Non-covalently bound lipoglycans, glycolipids, and lipids span the MAcs and create a peripheral lipid layer. On the top of this peripheral lipid layer the outer material ‘capsule’ is composed of α-glucan, arabinomannan, mannan, and several proteins among others. The peripheral lipid layer includes key lipid classes known to influence infection outcomes such as phthiocerol dimycoserosates (PDIMs), phenolic glycolipids (PGLs, present in some M.tb strains), trehalose monomycolate (TMM) and dimycolate (TDM), diacyl/triacyl/pentaacyl-trehaloses (DAT/TAT/PAT), sulfolipids (SL-1), and phosphatidyl-myo-inositol mannosides (PIMs) and their associated lipoglycans such as lipomannan (LM) and mannose-capped lipoarabinomannan (ManLAM) ( Figure 1 ) (Daffe and Etienne, 1999; Constant et al., 2002; Reed et al., 2004; Tsenova et al., 2005; Reed et al., 2007; Sinsimer et al., 2008; Jackson, 2014; Garcia-Vilanova et al., 2019). The importance of the peripheral lipid layer is underscored by the fact that it accounts for 40% of the M.tb cell envelope composition (Garcia-Vilanova et al., 2019).

The M.tb cell envelope is an evolving and adaptive component of the bacterium and is critical to its protection against adverse environmental conditions during infection (Garcia-Vilanova et al., 2019). Many anti-TB drugs such as INH and EMB target cell envelope biogenesis to break down this barrier and weaken the bacterium. Anti-TB drugs targeting the cell envelope can be bactericidal, allowing other drugs to go through. Importantly, many mutations that confer resistance to anti-TB drugs arise within these cell envelope biogenesis pathways, making the generation of new anti-TB drugs critical to combating the recent increase in DR-TB (Batt et al., 2020).

Overall, the M.tb cell envelope has many distinct layers that come together to give protection to M.tb in various environments and resistance to many different drugs (Jarlier and Nikaido, 1994; Jackson, 2014). These key lipids are described briefly below in the context of drug resistance, while a detailed description of the peripheral layer of the cell envelope is provided elsewhere (Garcia-Vilanova et al., 2019).

Understanding the M.tb cell envelope is critical to recognizing the complexity of the anti-TB drug treatment regimens required to kill DS- and DR-M.tb ( Table 3 ). For 2HRZE/4HR treatment regimen, INH is a prodrug and its activation is associated with reduction of M.tb ferric catalase-peroxidase KatG by hydrazine to form ferrous KatG. Then, ferrous KatG reacts with oxygen to form oxyferrous KatG to activate INH (Wang et al., 1998). Once activated, INH inhibits InhA, the M.tb enoyl reductase, by forming a covalent adduct with the NAD cofactor. This INH-NAD adduct acts as a slow, tight-binding competitive inhibitor of InhA to interfere with MAc biosynthesis (Wilson et al., 1995; Chen et al., 2002; Broussy et al., 2003; Schroeder et al., 2005; Vilchèze and Jacobs, 2007). RIF inhibits M.tb DNA-dependent RNA polymerase, leading to a suppression of RNA synthesis and bacterial death (Campbell et al., 2001; Tupin et al., 2010). PZA diffuses into active M.tb expressing the pyrazinamidase enzyme that converts PZA to the active form pyrazinoic acid, where it interferes with fatty acid synthase (FAS) I (Ngo et al., 2007; Zimhony et al., 2007). This hinders M.tb’s ability to synthesize new fatty acids required for bacterial growth and replication (Ngo et al., 2007). Accumulation of pyrazinoic acid also disrupts membrane potential by interfering with ATP production as well as binding to the ribosomal protein S1 (RpsA), inhibiting trans-translation and blocking M.tb proteins being synthesized with high fidelity. (Shi et al., 2011) EMB directly inhibits M.tb arabinosyltransferases (EmbA, EmbB, and EmbC), interfering with the biosynthesis of arabinose-containing M.tb cell envelope components such as AG and LAM, as well as preventing bacterial division (Escuyer et al., 2001; Zhang et al., 2003; Amin et al., 2008; Goude et al., 2009; Zhang et al., 2020; Lee and Nguyen, 2023). EMB is thought to have a major impact on the cell envelope structure and permeability, as the arabinan domain of AG sustains the MAc barrier (Goude et al., 2009). Thus, less or smaller/truncated arabinan domain in AG could decrease the binding sites for MAcs, leading to a thinner MAc barrier and the accumulation of free MAcs, TMM, and TDM (Takayama et al., 1979). Further, M.tb surface exposed LAM interacts with host cells; thus, EMB-derived truncated or reduced levels of LAM on the M.tb cell envelope may decrease M.tb interactions with phagocytes (Zhang et al., 2003; Torrelles et al., 2004; Kang et al., 2005; Torrelles et al., 2008). Overall, the 2HRZE/4HR treatment for DS-TB is designed to increase permeability by dismantling the M.tb cell envelope (INH, EMB, PZA), followed by a direct inhibition of RNA-polymerases (RIF) driving bacterial death.

In the case of the new WHO treatment regimen recommendation for DS-TB of 4-month daily treatment regimen including RPT, INH, PZA, and MFX, the substitution of RIF for RPT adds value in multiple ways. Although RPT also inhibits M.tb DNA-dependent RNA polymerase activity, it has lower minimum inhibitory concentration (MIC), a longer half-life, and higher protein binding affinity, which altogether results in shorter therapies with reduced cytotoxicity (Alfarisi et al., 2017; Zheng et al., 2017). Further, replacing EMB with MFX is motivated by the fact that MFX inhibits M.tb topoisomerase II (DNA gyrase) (Naidoo et al., 2017). This enzyme is essential, as DNA gyrase is involved in the replication, transcription and repair of M.tb DNA. Thus, adding RPT and MFX while keeping the bactericidal effects of INH and PZA on the M.tb cell envelope is intended to maximize the effects of RPT and MFX on the bacterium.

For DR-TB, the newest oral treatment regimen approved is BPaL + MFX (for MDR/RR-TB) or BPaL alone for pre-XDR TB. The overall effects of the combination of these drugs on the cell envelope is unknown. However, BDQ is a diarylquinoline that inhibits the activity of mycobacterial ATP synthase that is essential for the generation of energy in M.tb (Sarathy et al., 2019). Pretomanid (Pa) inhibits M.tb MAc biosynthesis and causes respiratory poisoning of M.tb through the nitrosylating effects of nitroimidazole, causing the release of reactive nitrogen species under anaerobic conditions (Singh et al., 2008; Thompson et al., 2017, 2019). Pa molecular mechanisms of action may also involve effects on fasI and fasII (involved in mycobacterial fatty acid synthesis), as well as efpA (encoding the efflux pump EfpA implicated in the efflux of INH), the cyd operon (encoding cytochrome bd oxidase involved in the mycobacterial electron transport chain), and the iniBAC operon (INH induced operon involved in mycobacterial cell envelope stress responses) (Manjunatha et al., 2009). Finally, LZD interferes with bacterial protein translation by preventing the formation of the functional bacterial 70S ribosomal initiation complex, which is essential for M.tb proliferation (Diekema and Jones, 2000). Thus, the BPaL treatment regimen alters the M.tb MAc layer (biosynthesis), ATP production, and bacterial proliferation. These effects, together with the effects of MFX on M.tb DNA replication, transcription, and repair, initially makes BPaL + MFX a suitable treatment regimen for MDR/RR-TB. In this regard, other drugs such as ETH, delamanid (DLM), CFZ, and high doses of INH could also be used for the treatment of MDR/RR-TB. Increasing INH critical concentration aids in maintaining disruption of MAc biosynthesis. Further, ETH, like PZA, is a nicotinic acid derivative related to INH and thus, it acts similarly to INH where the covalent ETH-NAD adduct formed can act as a slow, tight-binding competitive inhibitor of InhA (Banerjee et al., 1994). DLM similarly inhibits the synthesis of methoxy- and keto-MAcs by disrupting the mycobacteria coenzyme F420 system and generating nitrous oxide through the nitrosylating effects of its nitroimidazole metabolites (Singh et al., 2008; Khoshnood et al., 2021). In the case of CFZ, its activity appears to be membrane-directed, interacting with M.tb membrane phospholipids to generate antimicrobial lysophospholipids which destabilizes the membrane, interfering with K⁺ uptake and ultimately ATP production (Cholo et al., 2012). However, the use of CFZ could be a double-edged sword, as it is also linked to local inhibition of host memory and effector T cell responses.

In summary, all anti-TB drug treatment regimens are developed to target M.tb cell envelope permeability, energy production, and DNA/RNA/protein synthesis ( Table 3 ). The use of a combination of drugs in treatment regimens increases the permeability of the M.tb cell envelope, while simultaneously targeting key processes for survival. These are also necessary to prevent the emergence of drug resistance and shorten treatments. Treatment length and drug cytotoxicity drives drug misuse and provides M.tb with an ideal scenario to evolve mechanisms of drug resistance, potentially changing its cell envelope properties to its own benefit.

The M.tb cell envelope has evolved over time to increase M.tb hydrophobicity, transmissibility, and pathogenesis during infection (Jankute et al., 2017). Indeed, some reports suggest that modern M.tb strains evolved from M. canetti, increasing their hydrophobicity and ability to be transmitted by increasing the amount of nonpolar lipids in comparison to polar lipids on the cell envelope (Fabre et al., 2004; Jankute et al., 2017; Briquet et al., 2019). However, while studies have examined the evolution of the M.tb cell envelope over time, there is still a large gap in knowledge related to how the M.tb cell envelope changes in relation to drug resistance. Here we summarize what is known about DS vs. DR-M.tb cell envelope changes and how they influence M.tb-host interactions.

Fitness has been defined as a comprehensive measure of the bacteria’s ability to survive, reproduce, and undergo transmission, and could represent virulence as well (Zhan et al., 2020). It is true that many mechanistically related forms of drug resistance are associated with reduced bacterial growth (Gagneux et al., 2006b; Andersson and Hughes, 2010; Bhatter and Mistry, 2013; Zhan et al., 2020). This fact in the TB field is debatable; indeed, studies show that INH-resistant M.tb strains with mutations in katG have decreased virulence, pathology, and fitness in the mouse model (Nieto et al., 2016). However, other studies report that DR-M.tb strains range in virulence and that certain drug resistance conferring mutations are at “non-cost” and do not decrease fitness (Ordway et al., 1995). This could be explained by the existence of secondary mutations compensating for fitness cost. These secondary mutations could be an indirect, subsequent result of the primary mutations driving drug resistance. For example, an epidemiological study found that 47% of M.tb clinical isolates with a rpoB mutation conferring RIF resistance also had fitness-compensatory mutations in rpoA or rpoC (Casali et al., 2014).

Additional factors that can influence infection outcomes include M.tb lineage and host genetics and epigenetics that can cause variabilities in the immune response during infection (Saelens et al., 2019; Carey et al., 2022). Previous epidemiological studies suggest that lineage 2 “Beijing” strains are associated with increased virulence, transmissibility, and drug resistance compared to other lineages (Kong et al., 2007; Cowley et al., 2008; de Jong et al., 2008; Thwaites et al., 2008; Niemann et al., 2010; Casali et al., 2014; Carey et al., 2022). Recent reports on DR-TB patients also show cytokine signatures associated with increased TB disease severity and hyper-inflammation, indicating that some DR-M.tb strains may have increased virulence (Sampath et al., 2023). However, this conclusion needs to be handled with caution, as it seems possible that DR-M.tb appears more virulent in patients simply because it is drug resistant.

Another study assessed the adaptation of M.tb and found that strains from multiple lineages have adapted to specific human populations, and that these adaptations are consistent with the geographic spread of each lineage in various parts of the world (Gagneux et al., 2006a). Thus, it is plausible that DR-M.tb strains’ primary and secondary mutations, lineage, and host specific genetic and epigenetic factors may account for the vast amount of variability in disease severity and the immune response seen in DR-TB cases. These factors may allow for a small subset of DR-M.tb strains with little to no fitness cost to be preferentially transmitted to new hosts, ultimately favoring an increase in DR-TB over time ( Figure 2 ) (Zhan et al., 2020).

M.tb persisters also complicate our understanding of DR-TB and infection outcomes. These persisters are a subpopulation of M.tb that are genotypically the same as the rest of the population, but phenotypically drug tolerant and do not replicate in the presence of drugs (Torrey et al., 2016). Thus, bactericidal drugs do not kill these. Indeed, studies show that clinical isolates with a high-persister phenotype have greater ex vivo survival than those with a low-persister phenotype, with specific mutations found in genes related to PDIM biosynthesis in the high persisters (Torrey et al., 2016). However, while persisters can act as a source for drug resistant mutations (Prabowo et al., 2013), the association between drug resistance (not simply drug tolerance) and persister M.tb needs to be studied further.

Aside from active TB and latent M.tb infection (LTBI), incipient TB (likely to progress to active TB without causing detectable abnormalities) and subclinical TB (does not cause clinical TB-related symptoms, but does cause abnormalities that are detectable using radiologic or microbiologic assays) are also forms of the disease that need to be considered (Drain et al., 2018). In this regard, in 2021 it was estimated that approximately 7 million people were living with subclinical TB, and while these people do not exhibit classic symptoms of active TB disease, they are still able to transmit the infection to others (Kendall et al., 2021). Unfortunately, little is known about incipient and subclinical cases of TB, and even less is known about how drug resistance influences the likelihood of developing these TB stages in comparison to LTBI or active TB. What is clear, however, is that the prevalence of DR-TB is likely to increase if patients presenting subclinical TB are indeed infected with DR-M.tb and able to transmit it to others. Thus, understanding how drug resistance influences M.tb infection stages and outcomes is an important avenue for further exploration as we work to end TB.

Most studies examining DR-TB are in the context of pulmonary TB. However, there is also evidence of extrapulmonary TB resistance to both first- and second-line anti-TB drugs. Several studies involving extrapulmonary TB indicate that the rate of DR-TB in these patients goes between 4.4% to 20% (Dusthackeer et al., 2015). Similar to extrapulmonary TB, cases of disseminated DR-TB are reported around the world and can also complicate the already difficult-to-manage problem of DR-TB (Jung et al., 2016; Kaplan et al., 2018).

The M.tb cell envelope is a unique and complex structure that can adapt to its environment. Key peripheral lipids on the M.tb cell envelope contribute to the hydrophobicity of the bacterium and act as virulence factors to influence M.tb-host interactions. Although MDR/RR-, pre-XDR-, and XDR-TB have emerged in recent decades, how the cell envelope of DR-M.tb changes in comparison to DS-M.tb remains poorly understood. Recent studies have started to address genotypic changes naturally occurring in DR-M.tb strains that influence the efficacy of anti-TB drugs (Bellerose et al., 2020), and what is currently known about these evolutionary processes and adaptation of DR-M.tb are reviewed in detail elsewhere (Allué-Guardia et al., 2021). Examination of other superbugs such as methicillin-resistant Staphylococcus aureus (MRSA) showed that MRSA strains can alter their cell envelope to strategically evade the immune response (Gerlach et al., 2018), and that the glycosylation of cell envelope teichoic acids is required for methicillin resistance (Brown et al., 2012). It is possible that we can apply knowledge gained from other superbugs to DR-TB with respect to changes in the cell envelope composition of MDR/RR-, pre-XDR-, and XDR-M.tb strains and how these changes impact TB clinical presentation, host responses, and subsequent pathogenesis.

Studies in recent years suggest that the cell envelope of DR-M.tb strains thicken, and PDIMs may be playing a determinant role in controlling permeability to hydrophobic drugs. If this is the case, infection with a DR-M.tb strain may result in increased virulence, as the host immune system could skew towards a more pro-inflammatory response and increased necrosis of infected cells due to the greater amount of hydrophobic lipids on the cell envelope ( Figure 3 ). Another possibility is that, due to the increase in the amount of hydrophobic vs. hydrophilic lipids and their associated lipoglycans (e.g. PIMs/LM/ManLAM), there is missed opportunity for host cells to recognize hydrophilic lipids that favorably modulate the host immune response towards decreasing inflammation. One outstanding question is whether host-signaling patterns are altered by DR-M.tb due to cell envelope changes and, if so, do these alterations influence immune cell recruitment to and granuloma formation at the site of infection? Recent studies have begun to address the genotypic changes in DS- vs. MDR-M.tb, focusing on mutations that alter the efficacy of anti-TB drugs in vivo (Bellerose et al., 2020). However, it will be critical in future studies to also examine pre-XDR- and XDR-M.tb strains to assess their natural genetic variations that may influence drug efficacy, if any of these genetic mutations are associated with changes in the M.tb cell envelope composition, and how these changes ultimately influence TB pathogenesis.

Another outstanding question related to DR-TB is whether increased amounts of nonpolar lipids (and thus, increased hydrophobicity) on the cell envelope of DR-M.tb strains contributes to enhanced aerosolization and transmissibility of DR-M.tb, which was previously demonstrated in a study showing that increased hydrophobicity enhances aerosol transmissibility in non-tuberculous mycobacteria (Falkinham, 2003; Jankute et al., 2017). If so, it is plausible that increased hydrophobicity would ultimately allow these strains to be easily transmissible and increase their prevalence over time, thus associating drug resistance with an overall increase in fitness, TB pathogenesis, and impaired host immune responses. Addressing these questions will ultimately aid the TB field in designing novel anti-TB drugs in response to these changes and improving diagnosis of DR-TB in POC settings to reduce transmission and improve public health worldwide.

AS: Conceptualization, Funding acquisition, Investigation, Writing – original draft, Writing – review & editing. MI: Conceptualization, Writing – review & editing. JB: Conceptualization, Writing – review & editing. JT: Conceptualization, Funding acquisition, Supervision, Writing – original draft, Writing – review & editing.