Many important discoveries in the innate response against Mtb (i.e., role of TNF-α, IFN-γ, IL-12, among others) have been obtained in the mouse model (61). Mice have also been used to study the most recently described role of micro-RNAs (like miR-223 and miR-155) (62, 63) and sensing pathways for bacterial nucleic acids in MTBC (64). Similarly, innate receptors such as TLR2, TLR9, the complement system, and other conserved immunological pathways in mammalians have been studied in the mouse model. Although they do not entirely mimic the human TB infection, mice provide similar infection routes to humans, cost-effective use of several biological replicates, and the availability to explore different genetic and immunological tools (65). Recent studies have shown that lower bacterial doses in mice, like the ultra-low dose (1-3 founding bacteria, measured using DNA-barcoded strains) can more successfully mimic human infection (61, 66).

Under the mouse background (BALB/c, C57BL/6, and C3HeB/FeJ, also known as Kramnik), transgenic and knock-out strains for specific genes have been used as models to evaluate different innate immune response mediators in the articles reviewed ( Table 1 ). These different genetic mice also reflect different degrees of susceptibility to Mtb infection. C57BL/6 mice are generally considered more resistant to Mtb infection than BALB/c mice, and both mice do not develop necrotic lung lesions, a pathology hallmark of human TB (67). On the contrary, C3HeB/FeJ mice are highly susceptible and fail in controlling Mtb infection, displaying a more severe disease progression with characteristics closer to human TB, including the development of large, caseous lung necrotic lesions (68–70).

Macaca mulatta (rhesus macaques, RM) and Macaca fascicularis (cynomolgus macaques, CM), were the second most used models to study TB, due to their ability to closely mimic human TB and immune responses. NHPs are the preferred model to investigate TB vaccines and pathogenesis. While NHP models are highly informative, they have limitations, including high costs, specialized infrastructure, and ethical concerns (66). In our review, five articles used RM (33, 52, 73, 85, 93) and two used CM (104, 111). There are notable differences between NHP species in their response to Mtb infection. Although both RM and CM are highly susceptible to Mtb, RM are generally more vulnerable to severe TB than CM (66, 112). This increased susceptibility in RM may be linked to differences in their innate immune responses. RM display an anti-inflammatory profile in peripheral monocytes even before infection, with reduced IFN-γ secretion, whereas CM show a stronger pro-inflammatory response, particularly the release of TNF-α, during early infection (113). Most of our reviewed articles using NHP focused on cellular responses and blood transcriptomic profiles, which we differentiated by NHP species ( Table 3 ).

Rabbits were reported by two articles that evaluated pathological features and transcriptional differences in innate-associated genes after external stimuli (74) or in response to Mtb strains with different virulence profiles (55). Kolloli et al. concluded that iron supplementation induces different transcriptomic changes, depending on the evaluated tissue, without affecting bacterial burden or disease pathology. For instance, a systemic downregulation of IFNG, TNFA, and IL1B and upregulation of IL6 and SMAD6 was reported four weeks p.i (w.p.i). In the lungs, IL6 is downregulated similarly to the transcriptional regulators SMAD6 and SMAD7, while IL10, NOS2, IFNG, and TNFA are upregulated simultaneously. SMAD6 and SMAD7 are involved in the expression of the hepcidin antimicrobial peptide (HAMP) (74).

Subbian et al. found transcriptional differences in 14 genes at 3 hours post-infection with greater recruitment of macrophages and polymorphonuclear leukocytes (PMN) (with higher myeloperoxidase activity) and cavitary disease more like human active TB in the lungs of rabbits infected with HN878 compared to those infected with CDC1551. Overall, pathways involved in macrophage activation, fMLP (N-formyl-Methionyl-Leucyl-Phenylalanine)-stimulation or PMN recruitment and activation were upregulated after HN878 versus CDC1551 infection ( Figure 3A ). Some of the genes with increased expression include TNF, CXCL10, STAT1, IL1A, SPP1, CCL4, CCL2, IRF5, CD38, while reduced expression of IL4R, CAV1, TGFB2, IL18, and CD36 were observed (66). That early signature profile was relatively constant at four w.p.i and proposed to determine long-term Mtb infection in rabbits (55).

Cattle infected with either M. bovis BCG or AF2122/97 were used as a model to study trained immunity (31) or MAIT cells (114), respectively. In the study by Guerra-Maupome et al., calves did not exhibit differences in the number of monocytes expressing CD14+, CD11b+, or TLR-4 at four weeks post-BCG-vaccination, compared to unvaccinated animals (31) as opposed to the increased trend observed in BCG-vaccinated humans (115, 116). Peripheral Blood Mononuclear Cells (PBMC) from the aerosol vaccinated compared to the unvaccinated cattle showed increased levels of IL-6, IL-1β, and TNFα, but only after ex-vivo stimulation with LPS or Pam3CSK4. The latter allowed the authors to conclude that the innate immune system of calves can be “trained” and could be used as an immunomodulatory strategy (31). However, they did not explore the trained immunity effect after the animals were re-infected with M. bovis.

The second article by Edmans et al. revealed a MAIT cell population in the blood that showed higher perforin expression two-weeks after M. bovis infection without changes in the number of these cells. This high perforin expression was seen mainly in animals with TB lesions (114).

Guinea pigs develop necrotic core granulomas, and their macrophages exhibit surface CD1b and CD1c, making them a beneficial small animal model for replicating human TB (67). This animal has been widely studied for host-pathogen interactions in metabolic co-morbidities (such as diabetes and malnutrition) and mainly adaptive responses against MTBC (117). However, only one article focused on innate responses in the guinea pig was found during our review, specifically studying the neutrophil extracellular traps (NETs) (chromatin webs). Although this was one of the earliest responses observed, it was not microbicidal (90).

Wild boars were orally vaccinated with a heat-inactivated M. bovis strain and then infected with the same (viable) strain. Beltrán-Beck et al. evaluated the transcriptomic responses post-vaccination in the PMN of the oral mucosa and serum and then after infection in vaccinated animals. Some differences were noted in the complement and inflammasome responses post-vaccination, but only C3 mRNA levels remained after M. bovis infection, which was associated with reduced pathology ( Figure 3B ) (37).

In recent years, invertebrate animal models such as the insect Galleria mellonella (99, 118, 119) and planarians (Dugesia japonica and Schmidtea mediterranea) (91) have been used to study TB. These smaller models are advantageous for high-throughput evaluation of innate immune responses and align with animal research’s 3Rs principles (Replacement, Reduction, and Refinement) (120). Invertebrates, which lack a well-developed adaptive immune system, rely primarily on robust innate immunity to defend against pathogens and other stressors, making them valuable for studying innate immune responses against bacteria, including MTBC (121). Insect models, for example, have contributed significantly to understanding innate immunity and facilitated the discovery of mediators, such as the Toll-like receptors (TLRs). Conserved immune responses across insects and mammals include phagocytosis, opsonization, oxidative stressors, and eicosanoid synthesis, among others yet to be characterized (121, 122).

One of the central and most frequently identified factors in these reviewed articles was the cytokine TNF-α that was reported in articles using mice (40, 43, 48, 50, 71, 72, 75–83, 131), NHP (73), and rabbits (74). This pathway is responsible for multiple biological functions, including inflammation, cell proliferation, differentiation, and apoptosis (132). TNF-α is a cytokine produced by immune and non-immune cells and has a crucial role in TB protection (9). Signaling involves different molecules, such as the transcriptional factor Nf-kB. Nf-kB was increased in animals treated with sphingolipid sphingosine-1-phosphate (S1-P) that exhibited protection to Mtb (84). Most of the trends reported for TNF-α, were similarly reported for IFN- γ, except for one study in which the Mif-/- animals were used ( Table 1 ). Significant trends identified in this pathway are described in the first section of Table 2 , while those obtained in KO mice strains and NHP are summarized in Tables 1 and 3 , respectively.

Reactive oxygen species (ROS) and nitrogen species (RNS) are essential components of the antimicrobial system, produced mainly by macrophages (9, 133). iNOs was one of the most frequently reported proteins in this category. iNOs was higher in ATB in different mammals evaluated ( Table 2 ). Different articles described the role of iNOs during anti-mycobacterial defense, being stimulated by type-I IFN (86), S1-P (84), and cytokine-inducible SRC homology 2 (SH2) domain protein (CISH) (40). Conversely, overexpression of scaffold/matrix attachment region binding protein 1 (SMAR1) induced a significant reduction of iNOS expression in the lungs of infected mice at six w.p.i (87). In the same article, increased levels of IFN-γ and IL-12 in lung homogenates at four w.p.i were also reported, suggesting responses associated with adaptive immunity and potentially “sustained” innate immunity. These findings highlight the need to not only evaluate the kinetics of release, but also identify the cell where those innate responses are originating in vivo.

Neutrophil cytosolic factor 4 (NCF4) is a component of the NADPH oxidase complex, which is predicted to be involved in the production of O₂ ^-, via NADPH oxidase 2 (134). Increased NCF4 levels were related to reduced bacterial load and a protective response in animals immunized with an Ipr1-modified BCG vaccine (a BCG carrying the mouse gene Ipr1-intracellular pathogen resistance 1) (88). Other proteins and their trends are also summarized in Table 2 .

Lastly, the production of ROS by lung-residing myeloid cells (primarily PMN and AM) in mice was hampered by the early action of platelets that reduced the availability of these cells by forming aggregates in the lung by 21 d.p.i (independently of their canonical activation, by cyclooxygenase (COX)-1, glycoprotein IIb/IIIa or the ADP-receptor P2Y₁₂). This effect was confirmed by reduced bacterial load and lung pathology, as well as increased survival following platelet depletion around the onset of inflammation at seven d.p.i, but not before the infection (135). Increased early platelet response has been also reported in human patients that progress to ATB (136).

There are three types of IFNs: Type I (mainly IFN-α, -β, and others), Type II (IFN-γ), and Type III (IFN-λ). Reduced levels of IFN-γ were reported in diabetic mice as early as 24 h.p.i, that were exhibiting a dysregulated immune response with increased bacterial burden and inflammatory lesions ( Figure 3A ) (71, 72). In NHP, type I and type II IFN transcriptomic signatures were associated with ATB as early as two w.p.i ( Table 3 ) (104). Additional molecules and associated articles are summarized in Table 2 . Also, the interaction between this pathway and the neutrophil response was studied in the deficient IFNγR−/− mice ( Table 1 ).

IFN regulatory factors (IRF 1, 2, 3, 4, 5, and 7) showed a diverse response during Mtb infection. IRF7, among other innate transcriptional signatures from PMB cells in the blood, were correlated with the extent of disease and were strongly reduced in the vaccinated RM that exhibited Mtb control and reduced pathology ( Table 3 ) (93). In the article by Javed et al., IRF1 was highlighted among the early blood transcriptomic biomarkers in active TB that was also shared between human and macaque pathways as part of the type II IFN signaling, with sustained upregulation from the second through the 12th-w.p.i. Other IRFs, such as IRF2 and 4, were expressed differently, depending on the time point evaluated and the lineage of the infected animal (either Chinese or Mauritian macaque lines) ( Table 3 ) (104).

Neutrophils can internalize bacteria and activate inflammatory and effector responses, by the action of mediators like myeloperoxidase, elastase, and the production of NETs. As it was discussed earlier, neutrophils generate one of the earliest responses in the guinea pig ( Figure 3 , Table 2 ); however, they were ineffective at eliminating Mtb (90). Other publications concluded that the role of neutrophils is not necessarily bactericidal but perhaps in initiating inflammatory responses (29). Das et al. and Moreira-Texeira et al. concluded that neutrophil accumulation without a macrophage response leads to higher bacterial burden, increased disease severity, and greater lethality. This macrophage-reduced response was observed in MIF deficient mice (simultaneous to reduced macrophage innate cytokines: IL-6 TNF-α, IL-12, IL-10) (50, 54).

Other proteins also associated with this pathway are described in Table 2 . Among those, proteolytic matrix metalloproteinases (MMP) participate in the degradation of the lung extracellular matrix components, with the subsequent Mtb spread. These proteins have been proposed as biomarkers of ATB disease (137). During our review, we found enhanced MMP-1, 8, and 9 levels in blood and lungs during active TB that was responsible for excessive inflammation and tissue damage in mouse-susceptible strains (54). MMPs were also highly expressed during neutrophil activation and recruitment and induced neutrophil responses such as the NETs (43, 54, 85).

The dual role of neutrophil described in the reviewed articles is represented by excessive inflammation in acute phase after infection (probably bacterial dose/strain-dependent) (56) and protection in “sustained” or trained immunity after BCG vaccination in mice (29). In humans, neutrophil degranulation and NETs markers are also increased early in patients that progress to ATB (136). In RM, a higher expression of neutrophil degranulation makers was seen before exposure to Mtb that resulted in reduced lung pathology and bacterial load after infection, and therefore categorized as a protection marker (pre-infection) ( Table 3 ) (93). This protective role of neutrophils may be driven by a specific subpopulation that could be induced since early developmental stages in the bone marrow after BCG immunization (29, 35).

The inflammasome is a complex consisting of a sensor (such as a cytosolic pattern recognition receptor or PRR), an adaptor, and an effector (pro-caspase-1) proteins that, when activated, cleaves pro-inflammatory cytokines essential for MTBC control (138, 139). We found articles describing two types of inflammasome receptors: “NOD-like” receptor (NLR) pyrindomain-containing protein 3 (NLRP3) and the DNA cytosolic sensor “absent in melanoma 2” (AIM2) ( Table 2 ). In mice, NLRP3 interacts with the Mtb Esat-6 protein and influences the expression of IL-18 and IFN-γ (94). In addition to mice (94, 95), the inflammasome complex NLRP3/caspase-1/IL-1β was reported in wild boars (37). In that model, mRNA levels of these inflammasome-associated molecules were increased in the PBMC from oral mucosa after oral vaccination with BCG ( Figure 3B ) (37). On the other hand, NLRP3 was shown to be mitigated by miR-20b binding to the 3’-UTR of NLRP3 mRNA, inducing M2 macrophage polarization in the mouse ( Figure 4 ) (95).

Another inflammasome-related PRR is the AIM2-like receptor. Kuptz et al. demonstrated that NLRP3 rather than AIM2 was involved in the activation of IL-18 in mice (94). In NHP (RM), AIM2 was increased in animals with active TB and reduced in vaccinated animals that showed Mtb control (no signs of granulomatous disease) ( Table 3 ) (93).

More than 30 autophagy-related genes (ATGs) and other players, such as the microtubule-associated protein 1A/1B-light chain 3 (LC3), perform this innate protein degradation and regulatory mechanism in mammals. The activation of these proteins results in the membrane invagination of the bacteria, its products, or even damaged self-organelles. There is also a non-canonical autophagy known as LC3-associated phagocytosis (LAP), with the participation of PRR, independently of autophagosome formation (144). Deficient ATG5 mice exhibit more severe inflammation and succumb earlier to the infection ( Table 1 ) (53). More recently, Kinsella et al. reported that ATG5 was required in CD11c⁺ cells (lung macrophages and DCs) to control the production of pro-inflammatory cytokines and chemokines and avoid neutrophil recruitment (81).

LC3 participates in the autophagosome maturation and has two recognized isoforms: a cytosolic (LC-I) and a membrane-associated isoform (LC3-II, that is LC-3 conjugated to phosphatidylethanolamine) (145). LC3-II, regularly measured as marker of autophagic activity, was found increased in M. bovis infected mice after treatment with nilotinib (also discussed in section 3.2.5) (92). LC3 was evaluated also in planarians. LC3 co-localization with Mtb in mature phagolysosomes and LAP was fundamental for Mtb elimination in planarians. Finally, the Membrane Occupation and Recognition Nexus repeat-containing-2 (MORN2) transcript promoted both LC3 processes, and Mtb phagocytosis in planarians and humans (91).

Increased antimicrobial peptides were reported in the early response to MTBC infection, which was more prominent when a hypervirulent strain infected the host (96, 99). Cecropins increased 48 hours post BCG infection in the insect model (99), while hypervirulent bacteria induced higher cathelicidin-related antimicrobial peptide (Cramp) and defensins response in the mouse model as soon as one d.p.i compared to the levels induced by H37Rv ( Figure 3A ) (96). In other cases, the antimicrobial activity of these peptides was evaluated in the mouse model as a potential therapeutic alternative (79, 97). For instance, Ramos-Espinosa et al. evaluated the use of recombinant adenovirus as a delivery strategy for antimicrobial peptides [Human β-defensin 3 (HβD3) or cathelicidin (LL37)]. This treatment strategy resulted in reduced bacterial load and pneumonia with higher expression of proinflammatory cytokines and a synergistic effect with anti-TB drugs (97). In line with this, KO mice for Cramp (LL37 mouse homolog) were unable to control Mtb and succumbed earlier after infection (44) ( Table 1 ).

In the rabbit, the antimicrobial peptide HAMP was studied, which is also known to regulate systemic iron. In this study, HAMP levels in the lung were upregulated after 8 weeks post-Mtb infection in animals supplemented with iron without affecting bacterial load (74). Although antimicrobial peptides are mainly produced by epithelial cells and macrophages (96), none of the reviewed articles evaluated the epithelial cells in vivo during an MTBC infection. Lastly, lower expression of the gene RegIII γ, involved in defense peptide production in the intestine, was found in mice with antibiotic induced-intestinal dysbiosis. This trend was more intense after the mice were infected with H37Rv and in the infected animals that were treated with Isoniazid (INH) ( Table 2 ) (76).

These groups of PRR were represented in the reviewed articles mainly by the Macrophage inducible C-type lectin receptor (MINCLE) and macrophage C-type lectin (MCL). Macrophages and DCs express MCL that recognize Mtb TDM (Trehalose 6,6’-dimycolate). Reduced MCL levels were linked to increased inflammation, bacterial burdens, and mortality in the mouse model ( Table 1 ) (41). These receptors exert non-redundant anti-Mtb function, facilitating the uptake of the bacteria and the signaling towards proinflammatory responses (41, 131). Gut microbiota dysbiosis reduces the expression of MINCLE in DCs that induces adaptive T cell CD4⁺ responses, demonstrating a strong relationship between gut microbiota and DCs with subsequent Mtb elimination (146). Gut microbiota disruption also reduces the expression of other innate receptors, such as NOD2, MHC-II, and TLR2 in pulmonary DCs that participate in anti-Mtb responses in the mouse model (76, 100, 146).

We found three TLRs in our reviewed articles: TLR2, 4, and 9. Significant findings related to these TLRs and other molecules participating in this innate pathway are described in Table 2 . The activation of TLR4 can be dependent or independent of myeloid differentiation primary-response protein 88 (MyD88). MyD88 is fundamental to clear MTBC infection. The reactivation of MyD88 signaling in myeloid cells (macrophages and DCs) during M. bovis BCG infection is sufficient to control pathogen growth and reinstate local inflammatory cytokine production (IL-12p40, IFN-γ, and IL-1β in lungs) in mice (103). Before Mtb infection, Myd88 and the Toll/IL-1R domain-containing adaptor-inducing IFN-β (TRIF) transcripts were higher in vaccinated wild boars than in unvaccinated wild boars, as a sign of activation of the innate response and reduced pathology ( Figure 3B ) (37). Another mediator in TLR4-mediated responses is the tumor necrosis factor receptor-associated factor 6 (TRAF6). TRAF6 was associated with TB signs and tissue damage in the TCRβ deficient mice ( Table 1 ) (43).

This phagocytic receptor binds to the Fc portion of immunoglobulin G (IgG). One article evaluated the neonatal Fc receptor (FcRn) in a KO mutant mouse strain ( Table 1 ). Mtb-infected mice lacking FcRn had a reduced neutrophil infiltration in Mtb-infected lungs, concomitant to reduced bacterial burden and pathology. Because of the absence of FcRn reduced the capacity of CD103⁺ DCs to eliminate bacteria, other phagocytic cells may be driving the Mtb elimination and contributing to the low pathology profile observed (49).

In this pathway, we have grouped the different IL observed in four mammalian species ( Table 4 ) and the triggering receptors expressed on myeloid cells (TREM)-1. A high frequency of articles reporting IL and other cytokines and chemokines could be explained by their relevance in the inflammatory responses generated after MTBC infection, the availability of analytical methods (ELISA, Luminex^®, flow cytometry), and reagents that allow the multiplexed study of these molecules, particularly in the mouse model. In the remaining animal models, IL and other cytokines were primarily studied using gene expression analysis, either by using qRT-PCR (40, 87), whole transcriptomic studies, or a combination of both (55, 104). The increase of acute-phase cytokines such as IL-1β, IL-6, or anti-inflammatory IL-10 in serum or different tissues of mammals depended mainly on the specific factor being studied (KO mouse strains, ATB vs LTBI, etc.) and how those induce early or “sustained” changes in innate responses ( Tables 1 , 3 ; Figure 3 , section 3.2.2). Some of the articles included in this review evaluated these innate cytokines in response to different factors, such as the use of hypervirulent strains (55, 147). For instance, IL-17 is required for early protective immunity against the hypervirulent Mtb HN878, but not much against Mtb H37Rv or CDC1551 (147).

Besides myeloid cells, TREM-1 is highly expressed in Vδ2 T cells of human patients with ATB and is involved in inflammation (153). In our review, increased TREM-1 signature-associated transcripts were associated with higher inflammation and disease severity in NHP at the time of diagnosis (90-180 d.p.i), similarly observed in human studies (111).

Eicosanoids are lipid mediators derived from arachidonic acid, produced by cyclooxygenases (COXs) and lipoxygenases (LOs), among other enzymes. These mediators include prostaglandins, resolvins, lipoxins, and leukotrienes (154–159). Eicosanoids have protective functions in innate and adaptive responses after Mtb infection (9, 155, 160). The most frequently reported eicosanoid, prostaglandin E2 (PGE2), generated by COX1, limits tissue damage and bacteria growth in the mouse lungs at 15 d.p.i (105). Mayer-Barber et al. similarly showed that PGE2 induced anti-Mtb activity and was stimulated by IL-1 during infection in mice, resulting in reduced CFU in the lung four w.p.i (86). During TB progression in susceptible mice, increased serum levels of leukotriene B4 (LTB4) and PGE2, as well as decreased levels of lipoxin A4 (LXA4) were seen in the first 30 d.p.i. LTB4 increases susceptibility to Mtb. Therefore, a balanced PGE2/LTB4 response influences the severity of Mtb infection (82). The other eicosanoid-producing enzyme, 5-LO, generates LTB4 and LXA4. In mice, 5-LO is negatively associated with protection against Mtb, since its absence led to higher levels of PGE2, low bacterial counts, and rearrangement of the profile of cells recruited to the lung (favoring CD11c⁺, CD19⁺, and CD3⁺/CD4⁺ cells) after Mtb infection (105).

This review explores the role of various metabolites that influence the early immune response to MTBC infection. Vitamin B5 has shown promise by reducing bacterial load, promoting macrophage maturation, and increasing pro-inflammatory cytokine levels (TNF-α, IFN-γ, IL-17) in mice as early as one-week postinfection (83). In contrast, vitamin B6 did not appear to affect the innate immune response in mice (149), and iron supplementation did not significantly alter bacterial burden or disease progression in rabbits (74). Additionally, S1-P induced M1 macrophage polarization, iNOS expression in the lungs and protection against infection (84). This review summarizes other factors influencing M1 and M2 macrophage polarization, shown in Figure 4 . Recent findings have demonstrated that the classification into M1 and M2 macrophages is rather simplistic in explaining the events happening during infection in vivo; however, it is still helpful to describe an inflammatory or anti-inflammatory profile (174).

Another metabolite examined is 5-OP-RU ((5-(2-oxopropylideneamino)-6-D-ribitylaminouracil), an intermediate in bacterial riboflavin biosynthesis, which has demonstrated contrasting effects in vitro and in vivo. In vitro, 5-OP-RU activates MAIT cells, but in vivo, it requires co-stimulation with the TLR2/6 agonist Pam2Cys to promote pulmonary MAIT cell expansion in mice (89). In NHP infected with Mtb, treatment with 5-OP-RU activated MAIT cells, but did not expand them, nor did it help control the infection. In some cases, animals treated with 5-OP-RU developed acute respiratory symptoms, prompting the authors to reconsider its therapeutic potential (175). Effects of metabolites (including eicosanoids) in innate immunity should be evaluated as a sole factor, but also in combination with others (metabolites and proteins), due to the reactive nature of these molecules, their half-life and their co-stimulatory and antagonistic effect observed with other molecules. For example, the co-stimulatory activity described between 5-OP-RU and Pam2Cys (89) and the antagonistic effect observed for PGE2 and LTB4 (82). Moreover, while 5-OP-RU showed minimal impact on the intestinal microbiota in NHPs (175), its effects on the respiratory tract microbiota remain an area for further investigation.

We found several drug treatments with the potential for modulating the innate immune response and inhibiting Mtb or M. bovis growth, all of them evaluated in the mouse model. Biapenem and tubastatin A activated DCs and macrophages, reducing Mtb growth and suggesting new therapeutic avenues that stimulate the innate immune response (75, 176). Additionally, non-steroidal anti-inflammatory drugs (NSAIDs) like aspirin and ibuprofen, particularly when administered in later infection stages, reduced pro-inflammatory cytokines, alleviated lung pathology, and decreased bacterial load (80, 82).

Hussain et al. also reported that nilotinib induced both parkin and LC3-II proteins in the lung of M. bovis infected animals at 63 d.p.i. In a natural infection, M. bovis leads to the overexpression of the Abelson tyrosine kinase (Abl), that inhibits parkin, which is crucial to promote ubiquitin accumulation around the bacilli for its elimination (92). Nilotinib is a tyrosine kinase inhibitor that promotes parking activity by inhibiting Abl (177). A protective role of parkin was also described in the KO mouse model ( Table 1 ). The LC3-II increased after Nilotinib treatment also suggests the induction of autophagy, an antibacterial process described in section 3.2.4 (145). Targeting these pathways with Nilotinib resulted in reduced pathology, bacterial counts in lung and spleen, and higher survival of M. bovis-infected animals, highlighting its potential therapeutic benefits (92, 178).

Pharmacological inhibition of COX-2, administering Celecoxib shortly before and daily after infection, significantly reduced bacterial load in the lungs of infected mice (105). These findings highlight the potential of various pharmacological approaches to combat MTBC and possibly other non-tuberculous mycobacteria, warranting further evaluation in clinical settings.

Pretreatment with anti-TB drugs like INH and pyrazinamide (but not rifampicin) disrupted the mouse microbiota, leading to an increase in the Firmicutes phylum and reduced MHC-II expression in AM at five d.p.i, which worsened lung and extrapulmonary pathology (100). Additionally, pretreatment with a broad-spectrum antibiotic cocktail exacerbated this effect, decreasing MHC-II and CD86 expression in lung myeloid dendritic cells and likely impairing antigen presentation. Disruption of gut microbiota also resulted in higher levels of pro-inflammatory cytokines (IL-6, IL-1β, IL-12) and lower levels of IL-10, further hindering the antibacterial efficacy of the subsequent INH treatment and leading to higher bacterial loads and more severe lung pathology compared to control mice (76).