Mycobacterium tuberculosis (Mtb) is a causative agent of tuberculosis—an infectious disease responsible for ten million cases and over a million deaths every year (WHO, 2020). Despite the availability of antibiotics (Paulson, 2013) and vaccination with an attenuated form of Mycobacterium bovis, the Bacilli Calmette-Guérin (BCG), TB eventually kills 45% of HIV-negative people and nearly all HIV-positive individuals (WHO Key Facts, 2022).

Major histocompatibility complex receptor classes I and II (MHC-I and MHC-II) are two families of receptors involved in the recognition of “self” and surveillance of “foreign” antigens. While class II receptors evolved for the defense against pathogens and are present primarily in the immune cells, class I receptors are much more ubiquitous, also on MHC-II-negative cells. Ubiquitous presence may play a unique role in multiple-tissue surveillance, pathogen detection, and restriction of disease dissemination. MHC-I (or HLA class I) contributes both to innate immunity through the engagement of natural killer (NK) cells and to adaptive immunity through peptides presented to cytotoxic T cells (Jiang et al., 2019; Uzhachenko and Shanker, 2019). In contrast, MHC-II is solely dedicated to adaptive immunity (Leddon and Sant, 2010).

MHC-I molecules constitute a fundamental aspect of self in self-organizing networks of multicellular organisms starting from jawed vertebrates. Self is a process of continuous rebalancing of different biochemical and intercellular interactions rather than a constitutive feature of a single cell, as proposed initially by Turing (1990). MHC-I molecules present a snapshot of the intracellular proteome from all cell compartments by uploading self-peptides, byproducts of the reactions taking place in the cytosol. This proteome is recognized by T cells selected both in the thymus and periphery to have autoreactive cells eliminated or repressed (Perreault, 2010). MHC-I molecules can also present self-peptides that are mutated, as well as pathogen-derived mimicry peptides (Trost et al., 2010). The aim of the antigen presentation is the recurring recognition of cumulative signals up to the threshold where the “non-self” cell is destructed by adaptive immune cells, mainly CD8⁺ T cells, while protecting healthy cells from NK-mediated cytolysis. This requires coordinated action involving a series of brief encounters between antigen-presenting cells and their immune counterparts. As MHC-I molecules feature on every nucleated cell of the organism, they visualize pathological processes that occur in the organism at an early stage. Peptide presentation on MHC-I molecules, recognition by T-cell receptors (TCR), and immune cell activation are three stages in which cell interactions are inspected for affinity binding and initial signal strength, duration of stimulation, and decay (Segura et al., 2008; Pathni et al., 2022). Historically, the MHC-I system has been linked to transplant immunology and viral infections. MHC-I antigen presentation may contribute to the development of new vaccination strategies against chronic bacterial infections.

The major histocompatibility complex (MHC) locus, found on the short arm of chromosome 6 in humans, encodes genes from three classes of proteins (MHC-I, MHC-II, and MHC-III). There are three MHC-I groups, termed classical human leukocyte antigens A, B, and C (HLA), and several non-classical HLA groups, such as HLA-E, F, and G, a cluster of differentiation 1 (CD1) molecules, and MHC-I-related protein (MR1). Classical MHC-I molecules show a high degree of polymorphism, resulting in allele diversity predicted to encompass eight to nine million variants, although 80% of these occur only rarely and are represented by alleles differing by single base point mutations (Robinson et al., 2017). In humans, the most commonly found alleles are clustered into three major groups A–C, and 42 core classical types further grouped into 12 supertypes based on the similarities in peptide-binding motifs (Sette and Sidney, 1999). HLA class I supertypes have been linked to susceptibility and severity of tuberculosis (Balamurugan et al., 2004). Conversely, genes coding nonclassical HLA-E, HLA-F, and HLA-G are highly conserved as one or two alleles, often tissue-restricted, and involved in molecular mechanisms underpinning immune tolerance and fetus acceptance in pregnancy (Moscoso et al., 2006). Tissue expression of MHC-I molecules varies. In some cell subsets, such as postmitotic neurons, it is very limited (Neumann et al., 1997); in others, such as lymphocytes, MHC-I constitutive expression reaches up to 1% of the membrane protein content (Joly et al., 1991). It is also subject to transcriptional activation regulators TAF1, USF1/2, and CIITA (Howcroft et al., 2003).

Classical MHC-I molecules are made with a heavy glycoprotein α chain, composed of three immunoglobulin-like domains α-1, α-2, and α-3, a transmembrane segment with a cytosolic tail, and a smaller noncovalently attached light β chain (β2-microglobulin). Processed antigen fragments in the form of 8 to 10 amino acid peptide chains are affinity bound with each amino acid contributing to the overall MHC affinity score (Lundegaard et al., 2010). The binding pocket is formed by two conformational domains of the heavy chain, α1 and α2, supported by the β2 chain and its own residues to prevent binding of the longer peptides that would need to bulge outside the pocket. There are usually two anchor residues within the octo- or nonameric peptide, one of which is predominantly at the C-terminus (Falk and Rötzschke, 1993). It is different in MHC-II class molecules where two similarly sized α and β chains form an open-ended binding site so that typically 13–18 amino acid long peptides can extend out (Germain, 1994). In both MHC-I and MHC-II molecules, there are pockets present in the binding groove that accommodate side chains of the peptides; physicochemical features of these pockets underpin allele-specific consensus motifs (Falk et al., 1991). In MHC-I molecules, the conformational stability of the whole molecule is equally dependent on MHC and the bound peptide; it disassociates with any of the components’ removal, leading to MHC-I recycling from the cell surface. The peptide-MHC (pMHC) complex is recognized by the T-cell receptors and its CD8⁺ co-receptor, which binds to the nonpolymorphic α3 domain in the heavy chain (Huppa et al., 2010).

Nonclassical MHC-like, MHC-Ib, or MHC-related molecules resemble the classical receptors with their heavy chains but do not always associate with β2 microglobulin, forming homodimers instead. They include MR1, HLA-E, HLA-F, HLA-G, and CD1 molecules. MR1 is an MHC-Ib molecule that uniquely binds small molecules of the microbial metabolome. Several isoforms of MR1 have been detected, albeit only one, MR1A bears a close resemblance to the classical MHC-I molecules and is fully functional (Riegert et al., 1998). Other isoforms are either nonfunctional or with uncertain properties, like MR1B, which forms homodimers that do not associate with the β2 light chain. Although these isoforms may not be functional in the antigen-presenting process, they can still play role in intracellular trafficking. HLA-E or Qa-1 binds leader sequences of other MHC-I molecules and, as such, is a sensor for their expression and checkpoint in antigen presentation to NK cells. In homeostasis, HLA-E binds only a restricted set of nonamers (O’Callaghan and Bell, 1998).

The CD1 proteins’ binding groove is deeper and lined almost entirely with nonpolar or hydrophobic amino acid chains. This groove can accommodate a long hydrophobic lipid tail inside its pockets and expose the hydrophilic part on the surface, where it directly contacts the T-cell antigen receptor. In contrast to MHC-I molecules, where the peptides are trimmed to 9–11Aa-long peptides, lipids bound by CD1 proteins are not cleaved but adapted within the hydrophobic clefts, which can accommodate up to C70–C80-long chains. In humans, there are four members of antigen-presenting CD proteins: CD1a, CD1b, CD1c, and CD1d, and CD1e, which is a soluble carrier in the endolysosomal network for other CD1-lipid complexes (Moody et al., 1999; Ly and Moody, 2014; Moody and Suliman, 2017).

Finally, for completeness, in humans, there are MHC class I chain-related molecules that are not involved in antigen presentation but are still involved in immune responses. These are MICA and EPCR proteins, with MICA being highly polymorphic glycoproteins that are expressed ubiquitously and serve as a “danger signal” for NK cells, γδ T cells, and CD8⁺ T cells via NKG2D receptors (Schrambach et al., 2007). EPCR, a 46-kDa protein, shares ~20% homology with the CD1d molecule. They are expressed in the vascular system (endothelium) as well as on various innate immune cells; they are ligands for TCR of Vδ2 γδ T cells. MICA and EPCR proteins, respectively, regulate inflammation and coagulation (Willcox et al., 2012).

Mtb is an intracellular microbe that can infect any tissue, but the lung is the main niche for its transmission (Torrelles and Schlesinger, 2017). The presence of the asymptomatic latent phase in a significant proportion of immunocompetent infected individuals makes this pathogen life cycle akin to other human pathobionts that cause harm only under host–pathogen disequilibrium (Hakansson et al., 2018). The typical Mtb life cycle includes a phase of the primary disease followed by a stage of long-term occult and persistent intracellular infection. During this “latent” period, bacilli survive in a non-vegetative dormant form characterized by a thick lipid-rich bacterial cell wall and lipid accumulations (Vázquez et al., 2014) protecting from the degrading activity of host autophagolysosomal enzymes ( Figure 1 ).

Alveolar macrophages serve as the first port of entry and primary host for Mtb, killing it and presenting antigens while also sheltering persistent bacilli (van Crevel et al., 2003). The process of phagosomal uptake involves a range of cell receptors, including cell surface pattern recognition receptors (PRR), which bind pathogen-associated molecular patterns (PAMPs). Toll-like receptors (TLRs)on the cell surface (TLR1, TLR2, TLR4, TLR6) and endosomal membranes (TLR3, TLR7, TLR8, TLR9) are key PRRs that are responsible for the induction of the signaling pathways downstream and the production of cytokines (Underhill et al., 1999). Interestingly, mycobacteria-led use of TLR3 appears to enhance IL-10 production (Bai et al., 2014), suggesting that the binding of Mtb RNA may influence cross-talk with other intracellular signaling pathways. It has also been shown that other receptors, like C-type lectin receptors, scavenger receptors, nucleotide-binding-oligomerization domain (NOD)-like receptors (NLRs), opsonin receptors, ficolin, C1q, complement receptors CR1, CR3, and CR4, and Fc receptors for bacilli opsonized with immunoglobulins, promote phagocytosis but not necessarily pathogen eradication (Hossain and Norazmi, 2013).

The arrest of phagosome maturation is an important stage in the Mtb life cycle ( Figure 1 ). Mycobacterium tuberculosis influences acidification of the phagosome in the early phase of infection by preventing tethering of V-ATPase to Mtb-containing vacuole and stabilizing its pH at 6.3–6.5 with tyrosine phosphatase PtpA (Wong et al., 2011). Early phagosomes undergo a series of encounters with other endocytic organelles to acquire various molecules, Rab5, EEA1, PI(3)P, and VPS 34 to name a few, before they progress to the late phagosome stage. Mtb prolongs that stage by hydrolyzing PI3P by the action of PI3P-specific acid phosphatase SapM (Vergne et al., 2005); mycobacterial ManLAM blocks the transport of acidic cargo from the trans-Golgi network via interference with early endosomal antigen EEA1-syntaxin 6 interactions (Fratti et al., 2003) and protein kinase G (pknG) induces continued accumulation of Rab5 and prevents Rab7 acquisition, further delaying phagosome maturation (Roberts et al., 2006). That temporary blocking effect is eventually overcome by the IFN-γ-activated macrophage. The immune response to mycobacterial early-conserved immunodominant epitopes of the Esx secretion system triggers pH to drop to 5.0 in the phagolysosome. However, the time gained allows Mtb to alter its ability to use carbon sources with the preference of glucose usage and storage of neutral lipids (triacylglycerols (TAG)) in droplets (Vázquez et al., 2014) followed by mycobacterial entry into the dormant phase.

1. Autophagosome-to-cytosol pathway: small molecules. Various molecules are actively secreted by live Mtb cells. Nearly 30% of the Mtb proteome is composed of small proteins defined as <200Aa. Development of the strategies that disable these molecules facilitates processing and antigen presentation. If the autophagosome establishes direct membrane contact with ER via fusion (Guermonprez et al., 2003), smaller size molecules ( Table 1 ) can be transported out of the autophagosome either via ER-specific sec61 channel into ER, where post-translational modification are removed and then out to cytosol (Römisch, 2017). Retrograde transit facilitates entry of these molecules to the 19S unit of the proteasome and then back to the ER to form pMHC-I complexes via an antigen-processing transporter (TAP)-dependent pathway. Candidates could include Mtb 19-kDa glycosylated lipoprotein LpqH, fibronectin attachment protein Mpt32, and superoxide dismutase SodC (Mehaffy et al., 2019). Specific Mtb proteins lack fixed conformational structure, e.g., the PGRS part in PE-PGRS proteins or the PPE part in PE-PPE proteins. This results in their nonspecific interference with host signaling pathways by allosteric mimicry (Sharma et al., 2022). Cytosolic polyubiquitination of lysine residues in hydrophobic stretches of these proteins stabilizes their structure while directing them to the proteasome. Finally, the mycobacterial counterpart to the host sec61 channel is 8 membrane-bound Sec proteins that form translocation apparatus transporting many of Mtb-secreted proteins. In conjunctions with pore-forming type VII secretion system of esx1, esx3, and esx5-sec transport circumvent host ER sec-61 by directing small proteins out of Mtb and via punctured membrane into the cytosol (Divangahi et al., 2009).

2. Autophagosome-to-cytosol pathway: bigger molecules. Partially digested bacterial cells, e.g., BCG, molecular complexes, or even whole live bacilli, are transferred from the autophagosome to the cytosol in any conditions that significantly impair phagosome membrane stability, as described previously in Mycobacterium tuberculosis as an intracellular microbe: role of the phagosome. It was previously shown that in BCG-infected phagosomes, molecules as large as 70 kDa could access the cytosol (Teitelbaum et al., 1999), which may happen at a late stage when the phagolysosome membrane is partially degraded. Of note, both sec61 active transport and membrane disruption mode are not mutually exclusive. The repair patches of the plasma membrane are of ER origin and therefore sources of embedded molecules that may not be characteristic of the endosomal membrane itself.

3. Endosomal processing pathways. Mtb persistence in the phagosome (Mycobacterium tuberculosis as an intracellular microbe: role of the phagosome) is linked to the inhibition of its acidification. Less harsh conditions increase the likelihood of early antigens loading on classical MHC-I molecules. Conversely, loading on CD1 requires displacement of smaller self-lipids by bigger Mtb lipids in low pH, characteristic of late phagolysosomes (Ly and Moody, 2014), and is supported by CD1d and CD1b. Peptide-MHC-I loading is enhanced by endosomal TLR3, TLR7, an TLR9 signaling and any additional transport of surface recycled MHC-I vesicles or ER vesicles that may fuse with phagosomes/phagolysosomes and provide clusters of MHC-I pathway components. One of those is the TAP supplied in retrograde transport between endosome and ER (Harriff et al., 2013). In fact, in certain cases, ER vesicles with peptides processed in the cytosol can fuse to phagosomes for further antigen processing and MHC-I loading (Guermonprez et al., 2003; Gardiner et al., 2013). Late endosomes were also reported to be involved in MR1-specific molecule trafficking and release to the cell surface (Harriff et al., 2014).

Uptake of the antigenic molecules from apoptotic (apoptotic blebs, exosomes) or necrotic infected cells by dendritic cells, specifically CD8⁺, takes place via phagocytosis, receptor-mediated endocytosis, and micropinocytosis. Exosomes contain various cell wall and membrane byproducts, i.e., PIM, LAM, LM, lipoproteins (LpqH), trehalose dimycolate, and monoglycosylated PGL (Layre, 2020). Dendritic cells have an extensive network of specialized vesicular transport pathways (Montealegre and Van Endert, 2019). While this route is a critical source of Mtb pMHC-I complex presentation to CD8⁺ cells, it is less influenced by live intracellular bacilli, which disrupt phagosomal antigen processing and interfere with antigens present in the cytosol (Cruz et al., 2017).

Mtb survival strategy as an intracellular pathogen deploys various, often synergizing mechanisms to manipulate antigen presentation. Part of decoy strategies forms a tactic to “be recognized but not eradicated,” counteracting its host drive to “recognize and eliminate.” Cellular immunity is primarily involved in the local control of the infection via a cytokine-mediated feedback loop between antigen-presenting cells (APCs) and T cells. Sentinel lymph nodes, defined as first granuloma-draining lymph nodes (Nieweg et al., 2001), are the site for priming and expansion of the cognate T cells; they are also the main residence of various unconventional subsets of T cells that act as tissue-resident immunity. Lymph nodes can harbor pathogens themselves, transported here from lung parenchyma and directly infecting lymphatic endothelial cells (Lerner et al., 2016).

Cytotoxic T cells are the main responders to MHC-I presentation. The majority of them are classically (MHC-Ia) restricted; these are CD8⁺ T cells that recognize pMHC-I complex via TCR. Upon activation, they produce cytotoxic granules and discharge their content via direct contact with the target cell; they can also produce Th1-type cytokines IFN-γ, TNF-α, and IL-2. The functional profile of cytokine secretion differs between active and latent TB varying from poly- to monofunctional T cells (Rozot et al., 2013). CD8⁺ T cells preferentially recognize and destroy heavily infected macrophages and represent the sensor of the intracellular bacilli burden (Lewinsohn et al., 2003); or conversely, only heavily infected macrophages are able to stimulate cytotoxic T-cell granule exocytosis. Uniquely, CD8⁺ T cells recognize pMHC-I complexes on MHC-II-negative infected cells like some epithelia and can limit infection propagation in the local alveolar environment (Harriff et al., 2014), at least at an early stage. It is known that Mtb-infected cells, albeit recognized, do not elicit effective cytotoxic responses until late in infection, which hints at a possible decoy strategy of the bacilli, known to have many conserved T-cell epitopes (Comas et al., 2010). This appears to be the case for the TB10.4 (EsxH) antigen. It elicits a dominant CD8⁺ T-cell response which poorly recognizes Mtb-infected macrophages and is unable to lyse them (Yang et al., 2018). A substantial proportion of TB10.4-specific CD8⁺ T cells are directed to a single epitope, TB10.4_4-11, at the start of this 96aa small protein sequence, overshadowing other epitopes. This is an interesting example of a diversion strategy against CD8⁺ T cells. As intact TB10.4 or EsxH is reported to inhibit the endosomal sorting complex required for transport (ESCR) that processes antigens for MHC-II epitopes loading in phagosomes (Portal-Celhay et al., 2016), the gain is doubled.

A smaller population of unconventional cytotoxic T cells is MHC-Ib (HLA-E-H, CD1, MR1) restricted. HLA-E presents peptide sequences from other MHC-I molecules, i.e., VMAPRTLIL, VMAPRTLVL, VMAPRTLLL, VMAPRALLL, and VMAPRTLTL (O’Callaghan et al., 1998); it was reported that HLA-E may present a set of Mtb-derived peptides, i.e., VMATRRNVL, VLRPGGHFL, VMTTVLATL, and RLPAKAPLL (Caccamo et al., 2015), some with striking similarities to self-derived leader sequences. Among HLA-E-restricted nonclassical tolerogenic T cells, there has been also a smaller subset identified with Th2 cytokines (IL-4,5,13) secreting properties that activate B cells (Joosten et al., 2010). These cells, upon recognizing Mtb epitopes presented on HLA-E, are not only diverted from their cytolytic functions but induced secretion of cytokines is also acting to suppress other T cells in the vicinity (van Meijgaarden et al., 2015).

CD1 molecules are highly conserved and specialize in presenting to CD1-restricted T cells like double-negative (CD4⁻CD8⁻) and iNKT cells (Arora et al., 2013). CD1a receptors present Mtb mycoketides (Moody et al., 2004). CD1d is adapted for presenting phosphoglycolipids such as phosphatidylinositol mannosidase (PIM) (Fischer et al., 2004). Other CD molecules also participate in host cell responses to Mtb, with CD1c presenting lipids such as phosphodolichols, phosphomycoketides, and N-terminally acylated lipopeptides (Van Rhijn et al., 2009) and CD1b presenting mycobacterial mycolates and glycolipids, glucose-6-O-monomycolate, glycerol monomycolate, and sulfoglycolipids (Moody et al., 1997; Lopez et al., 2020). Invariant NKT cells’ recognition of mycobacterial fractions targets glycolipids via CD1d-restricted invariant TCRα chain paired with limited TCRβ chain; the percentage of these cells decreases in active TB as they express programmed death 1 (PD-1) molecule that marks their exhaustion. They are also present in local pleural effusion, where they can produce IL-21, taking part in the stimulation of humoral responses (Wu et al., 2015). Other subtypes of NKT cells produce cytokines like IFN-γ, IL-4, or IL-17a. iNKT cells can directly inhibit the intracellular growth of Mtb through the granulocyte-macrophage-colony stimulating factor (GM-CSF) (Rothchild et al., 2014; Di Carlo et al., 2022). Cytokines IL-12 and IL-18 stimulate GM-CSF production by iNKTs. GM-CSF is a potent cytokine involved in macrophage differentiation to M1 phenotype, upregulation of CD11c and MHC II markers and, intracellularly, shifting the balance from antioxidant responses to inflammasome processing and secretion IL-1β while protecting from DNA damages (Di Carlo et al., 2022; Vico et al., 2022). The role of IL-1β in responses to Mtb is complex: first in resistance to infection, then protection against cell death, and finally influencing the aforementioned death modality with the shift from necrosis to pyroptosis and the subsequent effect on antigen cross-presentation. Autophagosomal Mtb appears to actively counteract host cell inflammasomes via modulating its own antigen secretion and interference with host signaling pathways (Rastogi and Briken, 2022). Cytosolic virulent bacilli do the opposite (Beckwith et al., 2020).

Mucosa-associated invariant T cells (MAIT) are present in both the upper and lower respiratory tract and are likely to detect Mtb not only in macrophage host but also in infected epithelia while in transit (Harriff et al., 2014). MAIT cells express TCR receptors of restricted diversity and recognize small-molecule microbial metabolites (Chen et al., 2017) presented on MR1 receptors. MR1 ligands in Mtb infection are derivatives of riboflavin and folic acid synthesis pathway and possibly other transiently expressed molecules of mycobacterial metabolome like photolumazine I (Harriff et al., 2018). Mtb genome contains a family of RibA-H genes, indicating an essential requirement for riboflavin endogenous biosynthesis by the pathogen in the absence of a riboflavin transporter to acquire it or transport it back to the host (Long et al., 2010). The utilization of riboflavin derivatives intersects with iron and bacterial coenzyme F₄₂₀ redox metabolism. Microbial colonization of mucosal surfaces drives the expansion of MAIT cells; there is a proportionally higher abundance of MAIT cells in the jejunum than in lung tissue (Provine and Klenerman, 2020), and overexpression of MR1 ligands has been shown to provide higher protection against TB disease in preclinical models (Dey et al., 2022). The phenotype of MAIT cells is defined as CD161hiVa7.2+ T cells, predominantly CD8⁺. These cells, abundant in the blood, periphery (Greene et al., 2017), and nonlymphoid organs, respond locally and at an earlier stage than the adaptive CD8⁺ ones (Godfrey et al., 2019; Provine and Klenerman, 2020). MAIT cells, similar to classical CD8⁺ cells, produce both cytotoxic granules and cytokines—TNF-α, IFN-γ, IL-17A, IL-2, IL-22, and IL-13, depending on the local microenvironment stimuli, but unlike MHC-Ia-restricted CD8⁺ cells, they are capable of effector functions immediately after leaving the thymus. They have a primarily effector memory phenotype, contribute directly to mycobacterial burden reduction (Chua et al., 2012), and are involved in inflammatory responses to infection.

γδ T cells are a subset of lymphocytes that enrich epithelial tissues. Their major blood subset of Vγ9Vδ2 cells can directly recognize microbial phosphoantigens in a non-MHC-dependent fashion; Vγ9Vδ2 T cells recognize not only molecules belonging to the family of butyrophilin but also mycobacterial 6-O-methyl-glucose containing lipopolysaccharides and phosphomonoester molecules. These are known as phosphoantigens, byproducts of the Mtb mevalonate metabolic pathway. The smaller periphery-bound subset of Vδ1 T cells bind to antigens displayed by several subtypes of MHC-I-like molecules, i.e., MHC-I chain-related gene A (MICA), via the NKG2D receptor shared also with CD8⁺ and NK cells, CD1, and endothelial protein C receptor (EPCR) (Witherden and Havran, 2012). Like other potentially highly cytotoxic cells, γδ T cells express CD94/NKG2A that inhibits MHC-I cells’ destruction by Vγ9Vδ2 ones. All γδ T cells are endowed with innate immune functions, allowing them to directly lyse infected cells as well as produce cytokines that stimulate αβ T cells.

Although not the subject of this review, it is worth remarking that CD4⁺ T cells can also display cytotoxic properties, and the plasticity of the immune response can include smaller populations of either double-negative T cells or MHC-II-restricted CD8⁺ T cells. Linked via their cytotoxicity properties are also NK cells, which contribute to overall immune responses to Mtb. Leader signal peptides presented via HLA-E downregulate NK cells via CD94/NKG2A receptors when processed via TAP antigen processing machinery. In cases of TAP inhibition, HLA-E is more likely to present exogenous antigens.

Despite considerable efforts in constructing a vaccine prototype that would effectively stimulate cytotoxic T-cell responses, the positive results have been moderate so far (Behar et al., 2007) and usually more pronounced in preclinical testing than as an outcome in immunogenicity testing of clinical trial samples (Rodo et al., 2019). Although it gets more accepted that protective immunity to TB may include other than CD4⁺ cell subsets, aggregated CD8⁺ responses constitute, so far, the main readout for current TB vaccine candidates, as summarized in Table 2 .

Mycobacterium tuberculosis’ adaptation to its host has been refined through thousands of years of coevolution. The renewed interest in intracellular antigen processing and presentation on MHC-I molecules has arisen in an attempt to better define immune correlates of protection against TB infection. Although classical CD8⁺ cells are still considered the main effector for peptides presented by MHC-I molecules and one of the main outcomes for immunogenicity assays, growing interest in donor unrestricted T cells (DURTs) may soon change that readout (Gela et al., 2022). In the majority of already completed trials, BCG vaccination is either a comparator or an inclusion criterion for patients’ eligibility. It is also a frequent Mtb surrogate for any in vitro and preclinical experiments. Although BCG-induced immune responses set a high testing threshold of efficacy for any new vaccine candidate entering clinical testing, our incomplete understanding of how these immune responses are induced and why they are insufficient in many TB-endemic countries outlines questions yet to be answered both scientifically and therapeutically. Together with new technological developments in cytometry, forthcoming clinical trials indicate a dynamic landscape in TB vaccinology with new, yet unexplored directions ahead.

The author confirms being the sole contributor of this work and has approved it for publication.