All animal procedures were approved by the Ethics Committee on Animal Use of the Butantan Institute (CEUA/IB protocol 2012250722) and conducted in compliance with the Brazilian National Council for Animal Experimentation Control (CONCEA) guidelines. Female C57Bl/6 mice (4–8 weeks old) were obtained from the Butantan Institute’s Central Animal Facility. Knockout strains, Casp-1-/-, Nlpr3-/-, and Aim-2-/- (C57Bl/6 background) were kindly provided by Dr. Sérgio Costa Oliveira (University of São Paulo). AIRmax and AIRmin^TT mice are bred under standardized conditions at the Laboratório de Imunogenética (21). Mice were housed in the Laboratório de Desenvolvimento de Vacinas facility with ad libitum access to food and water. Environmental conditions were maintained at 20–24 °C, 40–70% relative humidity, and a 12-h light/dark cycle.

The recombinant rBCG-LTAK63 strain (ΔlysA::lysA-ltak63), generated from BCG Danish as previously described (14) and control strain (BCG Danish ATCC 35733, American Type Culture Collection, Rockville, MD, USA) were cultured in Middlebrook 7H9 (BD Difco™, Detroit, MI, USA) medium supplemented with 10% OADC (oleic acid–albumin–dextrose–catalase; BD BBL, Cockeysville, MD, USA), 0.05% Tween 80 (Sigma-Aldrich^®, Merck KGaA, St. Louis, MO, USA), and 0.5% glycerol (Sigma-Aldrich) at 37 °C with 5% CO₂ for 7 days. Cultures were expanded to 50 mL and grown to mid-log phase (OD₆₀₀ = 0.8). Bacterial cells were harvested by centrifugation at 3,180 × g, washed twice with ice-cold 10% glycerol, and resuspended in 1 mL of the same solution. Aliquots (100 µL) were cryopreserved at -80 °C for long-term storage. Post-preservation viability was determined by colony-forming unit (CFU) counts. Serial dilutions were plated on Middlebrook 7H10 (BD Difco™) agar supplemented with 10% OADC (BD BBL™) and incubated for 21 days at 37 °C with 5% CO₂.

The dataset is deposited in GEO under accession GSE278523, and the corresponding raw sequencing data are available in NCBI SRA under BioProject PRJNA1159865. BALB/c mice were immunized with rBCG-LTAK63 or BCG (10⁶ CFU, s.c.) and lymph node samples collected at 7 days post-immunization. Gene expression was analyzed using EdgeR (22). Raw gene counts were normalized using the trimmed mean of M-values (TMM) method to correct for library size variations. Differentially expressed genes (DEGs) were identified using a false discovery rate (FDR) threshold of < 0.05, with multiple testing correction applied via the Benjamini-Hochberg method. To investigate functional relationships among DEGs, protein-protein interaction (PPI) networks were constructed using the Search Tool for the Retrieval of Interacting Genes/Proteins (STRING) database v11.5, applying a medium confidence interaction score (≥ 0.4). Subsequently, KEGG pathway enrichment analysis was conducted to identify significantly overrepresented biological pathways within the DEG dataset, with statistical significance defined as FDR < 0.05.

C57Bl/6, Casp-1-/-, Nlrp3-/-, Aim-2-/-, AIRmax, and AIRmin^TT mice were euthanized with an overdose of anesthetic (300 mg/kg ketamine + 30 mg/kg xylazine) by intraperitoneal route, and tibias and femurs were aseptically collected. Bone marrow cells were flushed from the bones using RPMI-1640 medium (Gibco, Life Technologies, Paisley, UK) and a 10 mL disposable syringe (BD Plastipak™) with a 25G needle (BD PrecisionGlide™). The resulting cell suspension was centrifuged at 200 × g for 10 min at 4 °C. The pellet was resuspended in complete R10 medium (RPMI-1640 supplemented with 10% heat-inactivated fetal bovine serum (FBS; Sigma-Aldrich^®)), an antibiotic solution (penicillin 100 U/mL, streptomycin 100 µg/mL), and 30% L929-conditioned medium as a source of macrophage colony-stimulating factor (M-CSF). L929 cells were originally obtained from ATCC. Cells were seeded into 6-well plates (Corning^®, NY, USA) and incubated at 37 °C with 5% CO₂ for 6 days, with half of the medium replaced on day 4. On day 6, non-adherent cells were removed by washing twice with phosphate-buffered saline (PBS, 137 mM NaCl, 2.7 mM KCl, 8 mM Na2HPO4, and 1.5 mM KH2PO4, pH 7.2), and adherent cells were detached using a cell scraper (KASVI, Pinhais, PR, Brazil) in ice-cold RPMI-1640. Cells were centrifuged at 200 × g for 10 min and resuspended in complete R10 medium.

Cell viability was assessed by Trypan Blue (Sigma-Aldrich^®) exclusion, and viable cells were counted using a Neubauer chamber. Macrophages were adjusted to 1 × 10⁶ cells/mL and seeded into 96-well plates (100 µL/well) for subsequent experiments. An aliquot was stained to confirm differentiation to BMDM by flow cytometry using anti-mouse CD11b conjugated to PE (BD Biosciences, San Diego, CA, USA) and anti-mouse F4/80 conjugated to Bv421 (BioLegend, San Diego, CA, USA).

BMDMs (1 × 10⁵ cells/well) were plated in 96-well plates (Corning^®) and either primed with LPS (500 ng/mL, Sigma-Aldrich^®) for 4 h or left unstimulated, followed by infection with BCG or rBCG-LTAK63 strains at a multiplicity of infection (MOI) 10:1 for 6 h. Appropriate positive controls were included for each experiment: Nigericin (20 μM, 40 min) or ATP (5 mM, 30 min; both from Sigma-Aldrich^®) for NLRP3 inflammasome activation. Methodologically, replacing Nigericin with ATP enhanced experimental consistency. Culture supernatants were collected and stored at -80 °C until use.

Female C57Bl/6 mice (5–7 weeks old) were randomly divided into three groups (n=5/group) receiving either: Saline (100 µL, subcutaneous (s.c.)), BCG Danish (100 µL, 10⁶ CFU, s.c.), or rBCG-LTAK63 (100 µL, 10⁶ CFU, s.c.). Thirty days later, spleens were aseptically collected for co-culture assays. Splenocytes were isolated by mechanical dissociation (Corning, Pyrex^® Ten Broeck Homogenizer), erythrocytes lysed by incubation with sterile ultrapure water for 10 s, and resuspended in complete R10 medium after centrifugation (200 × g, 10 min, 4 °C). Cell viability was assessed by trypan blue exclusion (Neubauer chamber), adjusting to 1 × 10⁶ cells/mL.

For co-culture, after differentiation BMDMs were seeded in 48-well culture plates and primed with LPS (500 ng/mL, 4 h) or left unprimed, then infected with BCG or rBCG-LTAK63 (MOI 10:1, 6 h). Controls included: unstimulated BMDMs (negative) and nigericin (48 µM, 40 min). After several PBS washes, BMDMs (2 × 10⁵/well) were co-cultured with CD28-stimulated splenocytes (2 × 10⁶/well; anti-CD28 1 µg/mL, 4 h) in 48-well plates (Corning^®). Monensin (3 µM) was added after 48 h, and the cultures were maintained for an additional 12 h prior to analysis.

Female AIRmax and AIRmin^TT mice (5–7 weeks old) were randomly divided into three groups receiving either: Saline (100 µL, subcutaneous (s.c.)), BCG Danish (100 µL, 10⁶ CFU, s.c.), or rBCG-LTAK63 (100 µL, 10⁶ CFU, s.c.). Thirty days later, aliquots of Mtb (H37Rv) maintained at –80 °C were thawed, and the concentration adjusted to 1.25 x 10⁴ CFU/mL using saline. Mice under mild anesthesia were instilled with 40 µL (500 CFU) into the nostrils with aid of a micropipette. Thirty days after challenge, mice were euthanized and lungs collected. For CFU counts, the cranial and median lobes of the lungs were homogenized, diluted in PBS and plated on Middlebrook 7H10 agar (BD Difco™), supplemented with 0.5% glycerol (Sigma-Aldrich^®), 10% OADC (BD BBL™), 5 µg/ml of TCH (2-thiophenecarboxylic acid hydrazide, Sigma-Aldrich^®, a BCG growth inhibitor) and incubated at 37 °C and 5% CO2 for up to 3 weeks to evaluate the bacterial load.

Samples were thawed and assayed for IL-1β by ELISA using the DuoSet^® Mouse IL-1β/IL-1F2 (R&D Systems, Inc., Minneapolis, MN, USA). For inflammation-related cytokines (IL-6, IL-10, MCP-1, IFN-γ, TNF-α, and IL-12p70) a Cytometric Beads Array (CBA mouse inflammatory kit, BD Biosciences) was used according to the manufacturer’s instructions. The assay lower limits of detection were between 2.5 and 17.5 pg/mL, depending on the cytokine, and the higher limit was 5,000 pg/mL. Data were acquired using FACS Canto II equipment and analyzed using FCAP Array Software (BD Biosciences).

For immunophenotyping of co-culture cells, the following extracellular markers were used: BV510-CD3 (clone: 145-2C11, BD Horizon™), FITC-CD4 (clone: GK1.5, BD Pharmingen™), PerCP-CD69 (clone: H1.2F3, BD Pharmingen™), and APC-Cy7-CD44 (clone: IM7, BioLegend). Cells were incubated with antibodies for 30 min, washed with PBS, and fixed/permeabilized using Fix/Perm Buffer (BD Biosciences™). For intracellular staining, the following markers were used: PE-IFN-γ (clone: XMG1.2, BD Pharmingen™), and BV421-IL-17 (clone: TC11-18H10, BD Horizon™). Cells were incubated with intracellular markers for 30 min, followed by washes with Perm Wash buffer (BD Biosciences™). All analyses were performed by acquiring 100,000 events on a FACSCanto II flow cytometer. Data were analyzed using FlowJo v10. Lymphocyte selection was based on size (FSC) and granularity (SSC) characteristics. CD4+ were gated to evaluate T cell activation markers (CD44+ and CD69+) and Th1/Th17-related intracellular cytokines (IFN-γ and IL-17). Gating strategy is depicted in Supplementary Figure S1.

Data was analyzed using GraphPad Prism 8.0 (GraphPad Software) and expressed as mean ± standard deviation (SD). Comparisons between two groups were performed using an unpaired two-tailed Student’s t-test. Multiple group comparisons were conducted using one-way analysis of variance (ANOVA), followed by Tukey’s post hoc test. p values < 0.05 were considered statistically significant. The experimental design figures were created with http://biorender.com.

We analyzed the transcriptomic profile of differentially expressed genes (DEGs) from the draining lymph nodes of mice at seven days post-immunization, comparing rBCG-LTAK63 to BCG using STRING analysis. A cluster of upregulated genes was revealed, indicating the transcriptional signature linked to early immune activation (Fosb, Fos, Atf3, Ptgs2), cellular stress (Hspa1a, Hspa1b, Dnajb1) and leukocyte recruitment (Cxcl1, Cxcl2) (Figure 1A). The protein-protein interaction map (Figure 1B) shows two main hubs. A chaperone-centered hub was revealed, which indicates an active and highly stressed environment, involving core (Hspa1a, Hspa1b, Hspa8), assistant co-chaperones (Dnaja4, Dnajb13, Dnajb4, Dnajb1) and other important members (Hsp90a1, Hsph1). The other hub reinforces this active immune environment linking transcription factors (Fos, Jun, Junb, Klf4, Klf6), chemokines (Cxcl1, Cxcl2), stress (Atf4, Hspa1a, Hspa1b) and inflammation resolution (Dusp1, Atf3) (Figure 1B).

The cell specific enrichment analysis of these genes (using human orthologues) points to expression of these genes primarily in macrophages (Figure 1C). The lack of direct up-regulation of IL-1 or core inflammasome genes may reflect the cellular composition of the lymph nodes, mainly T and B cells or even the time-point used (7 days after immunization). Pathway enrichment analysis (Figure 1D) demonstrated strong activation of key immune and inflammatory pathways, including IL-17 (mmu04657), TNF (mmu04668) and NK-κB signaling (mmu04064), antigen processing and presentation (mmu04612) and NOD-like receptor signaling (mmu04621). This molecular signature suggests a coordinated engagement of innate and adaptive immune responses, supported by an intense cellular activity demonstrated by the enrichment of Protein processing in endoplasmic reticulum (mmu04141), MAPK signaling (mmu04010) and apoptosis (mmu04210) pathways. Interestingly, the induction of pathogen-response pathways (mmu05145: Toxoplasmosis, mmu05134: Legionellosis, mmu05162: Measles) indicates broad-spectrum immune priming against intracellular pathogens.

Collectively the transcriptional analysis indicates that rBCG-LTAK63 promotes a potent and early immune response in comparison to BCG and points to macrophages as the main responsible for this response.

To characterize the contribution of inflammasome-associated pathways to the immune response induced by rBCG-LTAK63, we combined in vitro assays using murine macrophages with in vivo infection models. These analyses were performed in wild-type mice, in knockout strains lacking key inflammasome components, and in AIRmax and AIRmin^TT mouse lines, selected for maximal and minimal inflammatory responsiveness including inflammasome-associated responses, respectively. Bacterial burden was assessed in these models through CFU counts (Figure 2).

To evaluate inflammasome activation by rBCG-LTAK63, we measured IL-1β concentration in the supernatants of BMDMs according to standard protocols (23). Our results demonstrate that only rBCG-LTAK63 was able to induce detectable levels of IL-1β in the supernatant of non-primed cells (Figure 3A). On the other hand, when the BMDMs were primed with LPS, BCG induced IL-1β expression as compared to unstimulated cells, while rBCG-LTAK63 induced even higher levels (Figure 3B).

In order to identify the inflammasome pathway activated by rBCG-LTAK63 we used inflammasome-deficient macrophages. In Casp-1-/- or Nlrp3-/- macrophages, neither BCG nor the positive control nigericin could induce IL-1β secretion. In contrast, rBCG-LTAK63 stimulation of either Casp-1-/- or Nlrp3-/- macrophages induced IL-1β (Figures 4A, B). The production of IL-1β in Aim-2-deficient macrophages was maintained by both stimulus (Figure 4C). Comparative analyses across genotypes (e.g., wild-type vs. knockout) showed that Casp-1-/- or Nlrp3-/- cells exhibited an ~10-fold reduction in IL-1β compared to wild-type cells, whereas IL-1β production in Aim-2-/- macrophages was comparable to wild-type levels (Figures 4D–F).

Together, these results indicate that rBCG-LTAK63 induces IL-1β secretion predominantly through the canonical NLRP3/caspase-1 inflammasome pathway, while also engaging caspase-1 and NLRP3-independent mechanisms that partially compensate for the absence of these components.

To investigate the crosstalk between innate and adaptive immunity, we established a co-culture system combining inflammasome-activated BMDMs with naïve splenocytes or splenocytes from immunized mice (Figure 2B).

When co-cultured with naïve splenocytes, macrophages stimulated with either BCG or rBCG-LTAK63 increased the percentage of memory-phenotype (CD44⁺, Figure 5A) and activated (CD69⁺, Figure 5B) CD4⁺ T cells, and promoted direction toward a Th1 phenotype (IFN-γ⁺, Figure 5C). However, no significant differences were observed between the BCG and rBCG-LTAK63 groups in these subsets, including Th17 (IL-17⁺) cells (Figure 5D).

In contrast, when the co-culture was performed using splenocytes from immunized mice (Figure 2B) a more pronounced effect was observed. rBCG-LTAK63-stimulated macrophages induced a significantly higher percentage of both activated (CD69⁺) and memory-phenotype (CD44⁺) CD4⁺ T cells compared to its own basal control (Supplementary Figures S2D, E). Under these conditions, rBCG-LTAK63 stimulation also led to a robust increase in IFN-γ⁺ (Th1) and IL-17⁺ (Th17) CD4⁺ T cells relative to its basal group (Supplementary Figures S2F, G). These findings demonstrate that pre-existing immunity unleashes rBCG-LTAK63’s unique capacity to drive a robust and polyfunctional adaptive immune response.

We investigated the specific involvement of innate inflammatory responses using macrophages (BMDMs) from AIRmax and AIRmin mouse strains. These strains were genetically selected for high and low acute inflammatory responses, respectively. The AIRmin^TT strain is a subline with a characterized loss-of-function mutation in ASC (24).

In AIRmax macrophages, rBCG-LTAK63 triggered significantly higher IL-1β secretion than BCG or unstimulated cells (Figure 6A). In AIRmin^TT macrophages, the IL-1β response was completely abolished for all stimuli including the positive control ATP (Figures 6B, C). rBCG-LTAK63 was statistically different from BCG group but not in comparison to unstimulated cells (Figure 6B).

We also investigated whether rBCG-LTAK63 would also stimulate a cytokine response beyond IL-1β. The results showed a significant reduction in the production of IL-6 (Figure 6D), TNF-α (Figure 6E), and MCP-1 (Figure 6F) was observed in AIRmin^TT in comparison to AIRmax macrophages. Collectively, these data demonstrate that rBCG-LTAK63 can induce IL-1β secretion in both macrophage lineages but with a reduced efficiency in AIRmin BMDMs.

rBCG-LTAK63 has consistently demonstrated increased protection of mice against Mtb challenge including in different BCG background strains (Moreau or Danish), expression systems (auxotrophic complementation or antibiotic-maintained), challenge models (intratracheal or intranasal) and Mtb strains (H37Rv or Beijing) (14, 15). Here, we used AIRmax and AIRmin mouse models to assess whether inflammasome-associated innate immune competence influences vaccine-induced protection in vivo. AIRmax and AIRmin^TT mice were immunized with either BCG or rBCG-LTAK63, or received saline, and subsequently challenged with Mtb H37Rv. The pulmonary bacterial burden was quantified 30 days post-infection (Figure 7A).

Non-immunized mice (Saline group) from both AIRmax and AIRmin^TT genotypes exhibited a similarly high bacterial load in the lungs of challenged mice (Figure 7B). In AIRmin^TT mice, BCG did not induce protection, but in AIRmax mice, protection was similar to that obtained in naïve mice. More importantly, rBCG-LTAK63 induced higher protection as compared to BCG in both genotypes. Comparison between genotypes demonstrates that AIRmax mice exhibit significantly lower CFU counts in comparison to AIRmin^TT mice in both BCG and rBCG-LTAK63 groups. Nevertheless, the ability of rBCG-LTAK63 to confer protection in AIRminTT mice indicates that this recombinant vaccine can partially overcome limitations imposed by a genetically attenuated inflammatory background.