Four to five-week-old female C57BL/6 mice were purchased from DooYeol Biotech (Seoul, Korea). Mice were kept under standard environmental conditions and with ad libitum access to commercial food and tap water. All mouse studies were performed in accordance with the grant by the Laboratory Animal Welfare and Ethics Committee, Korea Disease Control and Prevention Agency (KDCA), in compliance with Institutional Animal Care and Use Committee guidelines for the care and use of animals (permit number: KCDC-124-19-2A). All processes complied with ARRIVE guidelines and American Veterinary Medical Association Guidelines for the Euthanasia of Animals.

The multi-antigenic recombinant antigen was constructed following the procedure described in a previous study (Zhou et al., 2018). Briefly, whole sequences of Ag85B, ESAT-6, a partial sequence of MPT64 (190–198 residues), and Rv2660c antigens of M. tuberculosis were obtained from Mycobrowser¹ and cloned into the pAd/CMV/V5-DEST vector (Invitrogen, Waltham, MA, United States) in order and linked using an Asp-Val-Ala and Gly-Ser-Gly linker to generate the candidate rAd-TB4 (Supplementary Figure S1A).

Subsequently, to confirm the construction of the adenovirus vector vaccine rAd-TB4, we analyzed its protein expression using western blotting. The culture supernatant and cell lysates obtained after infection were analyzed using a polyclonal-antibody specific to Ag85B, ESAT-6, MPT64 (Abcam, Cambridge, United Kingdom), and purified protein derivative (PPD; Mybiosource, San Diego, CA, United States). Briefly, HEK293A cells (Invitrogen) were grown in Dulbecco’s modified Eagle’s medium (DMEM, Gibco, Grand Island, NY, United States) containing 10% fetal bovine serum (FBS, Gibco) and 1% penicillin/streptavidin (P/S, Gibco) at 37°C for 24 h (Davis et al., 2001). Freshly grown HEK293A cells were then transfected with rAd-TB4 and cultured under DMEM containing 2% FBS at 37°C for 24 h. When approximately 80% of the cells displayed cytopathic effects, the infected cells and supernatants were collected to isolate the virus. The virus was purified and concentrated using a Virabind Adenovirus Purification Kit (Cell Biolabs, Inc., San Diego, CA, United States) following the manufacturer’s instruction, stored at -80°C, and titrated using the Adeno-X Rapid Titer Kit (Takara Bio, Japan). HEK293A cells were used as a negative control, and the Ag85B protein was used as a positive control for the Ag85B antibody.

M. tuberculosis Strains H37Rv and HN878, were obtained through BEI Resources, NIAID, NIH (NR-123 and NR-13647, respectively); BCG Pasteur 1173P2 was provided by the KDCA and grown in a medium containing Middlebrook’s 7H9 broth (Difco Laboratories, Detroit, MI, United States) supplemented with 10% Oleic Albumin Dextrose Catalase growth enrichment (Becton Dickinson, Sparks, MD, United States) and 0.2% glycerol at 37°C on a shaker at 200 rpm under aerobic conditions for 14–20 days (Cole et al., 1998). To prepare single-cell suspensions, mycobacterial cells were harvested via centrifugation at 10,000 × g for 20 min and washed thrice with PBS (pH 7.2). The pellets were passed through 40-, 20-, 10-, and finally, 8-μm filters (Millipore Corp., Burlington, MA, United States). The final stock was stored in small aliquots at −80°C until further use. Colony-forming units (CFUs) per milliliter of stock were measured using a counting assay on 7H10 agar plates.

Female mice (5–6 weeks old; n = 5) were vaccinated with BCG Pasteur 1173P2 (2 × 105 CFUs/mouse) subcutaneously (week 0), followed by immunization with rAd-TB4 (5 × 10⁷ infectious virus units/mouse) twice at 10 and 14 weeks after BCG priming via subcutaneous (SC) or IM injections. One week after the final immunization (at 15 weeks), the mice (n = 5) were euthanized via CO₂ inhalation to analyze immunogenicity. The measurement of immune response in the lung lymphocytes and splenocytes of rAd-TB4-administered mice was assessed by measuring ex vivo responses using the ELISpot assay, intracellular cytokine staining, and immunoglobulin G (IgG) titer measurements. Briefly, single lung lymphocyte and splenocyte cells were stimulated with different antigens—rAd-TB4 (Ag85B, Rv2660c, and ESAT-6) or PPD—and IFN-γ production per 1 × 10⁶ cells was determined using the ELISpot assay. For single-cell preparation, the spleens and lungs were aseptically collected and pooled for each group. Spleens were homogenized for immune assays using a gentleMACS μTissue Dissociator (Miltenyl Biotec, Germany), washed in RPMI-1640 medium (GenDEPOT, Baker, TX, United States) supplemented with 10% FBS and 1% P/S, rinsed in ammonium–chloride–potassium buffer to remove erythrocytes, and passed through a 40-μm cell strainer to generate single splenocytes. Lung tissues were incubated with DNase I (Roche, Swiss) and collagenase D (Roche) in a plain medium at 37°C for 1 h to isolate lymphocytes from mouse lungs. Lymphocytes were separated from 5 to 20 mL of lung cell suspension on a Lymphoprep gradient (STEMCELL Technologies, Canada) using density centrifugation, passed through a 40-μm cell strainer and resuspended in DMEM containing 10% FBS and 1% P/S.

Three weeks after the final immunization (at 17 weeks), another batch of mice was challenged with the M. tuberculosis H37Rv strain (experiment 1: n = 5; experiment 2: n = 3) or HN878 (n = 6) strain using a Glas-Col aerosol generator (Glas-Col LLC., Terre Hautre, IN, United States). The infection conditions were calibrated to expose each mouse to approximately 200 CFU following the procedures described in a previous study (Kwon et al., 2019). To estimate antigen specific IgG and subtypes, whole blood from the mice n the experiment 1 batch was collected eight weeks post-challenge and centrifuged to obtain serum.

For multifunctional T cell analysis, single cells (2 × 10⁵) were stimulated with PPD (NIBSC, Blanche Lane, UK), ESAT-6, Ag85B, or Rv2660c peptide (100 ng/mL; JPT Peptide Technologies GmbH, Berlin, Germany) for 5 h at 37 °C in the presence of GolgiPlug and GolgiStop (BD Biosciences, East Rutherford, NJ, United States). Stimulated cells were washed with PBS containing 3% bovine serum albumin (BSA; Sigma Aldrich, Saint Louis, MO, United States) and stained with FITC-conjugated anti-MHC II, V450-conjugated anti-CD4, and PerCP-Cy5.5-conjugated anti-CD8 antibody for 30 min at 4°C. The cells were then permeabilized with the Fixation/Permeabilization Solution Kit (BD Biosciences) for 30 min at 4°C, intracellularly stained with APC-conjugated anti-IFN-γ, PE-Cy7-conjugated anti-TNF-α, and PE-conjugated anti-IL-2 antibody for 30 min at 4°C, fixed with 4% paraformaldehyde, and resuspended in PBS containing 3% BSA. Analyses were performed using FACSVerse (BD Biosciences) and FlowJo software v9 (BD Biosciences) to gate triple-, double-, or single-positive T-cell populations (Supplementary Figure S2). Data are expressed as pie charts with the percentage of positively gated T-cells among CD4- or CD8-positive (CD4⁺ or CD8⁺) populations. For the detection of memory T cells and activated phenotypes, lymphocytes were stained with BV605-conjugated anti-CD3 (BD Biosciences), PerCP-Cy5.5-conjugated anti-CD4 (BD Biosciences), PE-conjugated anti-CD44 (eBioscience, San Diego, CA, United States), FITC-conjugated anti-CD62L (eBioscience), and APC-conjugated anti-CD127 antibody (eBioscience) for 30 min at 4°C, fixed, and resuspended in PBS containing 3% BSA. Analyses were performed using identical equipment and software.

The ELISpot assay was performed using an IFN-γ secretion ELISpot kit. Briefly, a single-cell suspension (5 × 10⁵ cells) was stimulated with PPD, Ag85B, Rv2660c, and ESAT-6 peptide (100 ng/mL) for 36 h at 37°C in anti-IFN-γ antibody-coated filter plates. Subsequently, biotinylated anti-IFN-γ antibody, streptavidin-horseradish peroxidase (HRP) conjugate, and 3-amino-9-ethylcarbazole were added as substrates to develop secreted cell spots, which were quantified using an Immunospot S6 analyzer (Cellular Immunospot Limited, Shaker Heights, OH, United States). The results are presented as mean values of triplicate wells for each group. All substrates and the ELISpot kits were purchased from BD Biosciences.

A single-cell suspension (5 × 10⁵ cells) was stimulated with Ag85B peptide for 36 h at 37°C, and the supernatant was collected to measure cytokine expression. Supernatants were diluted in a 1:2 ratio with complete RPMI medium and assayed in triplicate wells of each group using the Bio-Plex Pro Mouse Cytokine 10-plex, which was customized to include IFN-γ, TNF-α, IL-2, Th17-related cytokine (IL-17A), GM-CSF, IL-6, IL-12p40, IL-12p70, and IL-10 (Bio-Rad Laboratories, Hercules, CA, United States), following the manufacturer’s instructions. The results were acquired using a Bio-Plex MAGPIX reader (Bio-Rad Laboratories), and the mean values were used to plot graphs.

We coated 96-well flat-bottom Immuno Plates (Thermo Fisher Scientific, Waltham MA, United States) with 100 ng/mL PPD, Ag85B, Rv2660c, or ESAT-6 for 18 h at 4°C. After washing with PBS containing 10% tween-20, each well was blocked with PBS containing 3% BSA for 2 h at 37°C. Serum samples were diluted at 1:200 with PBS containing 3% BSA and incubated for 2 h at 37 °C. After washing, a 1:2,000 dilution of goat anti-mouse IgG-HRP (Thermo Fisher Scientific) was added and incubated for 1 h at 37 °C. In the case of challenged mice serum, goat anti-mouse IgG1-HRP (Thermo Fisher Scientific) and goat anti-mouse IgG2c-HRP (Abcam) were additionally used. The substrate, tetramethylbenzidine (TMB; Thermo Fisher Scientific), was added to each well, and the plate was incubated at 37 °C for 15–30 min. Thereafter, a stop solution for TMB was added, and the plates were read using a spectrophotometer (Spectramax i3x, Molecular Devices, San Joes, CA, United States) at 450 nm.

To estimate the CFUs of M. tuberculosis in the lungs of infected mice, the lungs were homogenized in 3 mL of PBS at 12 weeks post-challenge. Subsequently, 10-fold dilutions of the tissue homogenates were plated on 7H11 Middlebrook Agar (Difco Laboratories). The plates were incubated at 37°C for 3–4 weeks, and the number of colonies was determined to assess the total CFUs in the lungs. For histopathological analysis, the lungs of infected mice were fixed in formalin (Sigma Aldrich) and embedded in paraffin. Paraffin blocks were cut and stained with hematoxylin and eosin (H&E). A Motic Easy Scan scanner was used to scan the histological sections, and the images were analyzed to quantify the granulomatous inflammation. The percentage of granulomatous tissue in the whole lung was calculated using the ImageJ software version 1.53j. The experiment was conducted twice using the identical mouse immunization and challenge schedule, CFU estimation, and granulomatous inflammation quantification as described above.

Total RNA was isolated using TRIzol reagent (Invitrogen) and RNA quality was evaluated using an Agilent 2100 bioanalyzer using an RNA 6000 Nano Chip (Agilent Technologies, Amstelveen, The Netherlands). RNA was quantified using an ND-2000 Spectrophotometer (Thermo Fisher Scientific), according to the manufacturer’s instructions. An RNA library was constructed using a QuantSeq 3′ mRNA-Seq Library Prep Kit (Lexogen, Inc., Austria) according to the manufacturer’s protocol. High-throughput single-end 75-bp sequencing was performed using NextSeq 500 (Illumina, Inc., San Diego, CA, United States). QuantSeq 3′ mRNA-Seq reads were aligned using Bowtie2 (Langmead and Salzberg, 2012). The alignment file was used to assemble transcripts, estimate their abundance, and detect differentially expressed genes (DEGs) based on counts from unique and multiple alignments using coverage in BEDTools (Quinlan and Hall, 2010). The read count data were processed based on the quantile normalization method using EdgeR within R with the help of Bioconductor (Gentleman et al., 2004). Statistically significant DEGs were arranged into hierarchical clusters using correlation distance and presented in a heatmap. All correlation analyses employed Pearson correlation coefficients, and the Benjamini-Hochberg method was applied for multivariate analysis adjustments, setting the false discovery rate (FDR) threshold at < 0.25. Principal component analysis (PCA) was employed to demonstrate the similarities and differences between samples in the dataset and to detect any outliers. After obtaining a ranked list of DEGs, Gene Ontology (GO) enrichment analysis was performed to categorize and annotate the genes into groups such as a biological process using the Database for Annotation, Visualization, and Integrated Discovery (DAVID)² (Ashburner et al., 2000; Lovero et al., 2022; Sherman et al., 2022).

All graphical visualization and statistical tests were done using GraphPad Prism v10 (GraphPad Software, La Jolla, CA, United States). The statistical significance between respective antigen restimulation was determined with an unpaired Student’s t-test. One-way analysis of variance using Dunnett’s multiple comparison test was used to evaluate significance differences between more than two vaccine groups. The differences with a p-value < 0.05 were considered significant. Data expressed in graphs are presented as the mean ± standard deviation.

The candidate rAd-TB4 multi-antigenic recombinant vaccine was developed by inserting Ag85B, ESAT6, MPT64, and Rv2660c antigens into the pAd/CMV/V5-DEST vector (Supplementary Figure S1A). Its protein expression was confirmed using a poly-antibody specific for Ag85B, ESAT-6, MPT64, and PPD in infected HEK293A cells cultured supernatant and lysates (Supplementary Figure S1B).

Next, we evaluated the immunogenicity of the rAd-TB4 vaccine candidate by analyzing the T-cell response induced by it. First, C57BL/6 mice were subcutaneously vaccinated with BCG, followed by immunization with two doses of rAd-TB4 vaccine (Figure 1A). Compared with the BCG alone immunization, heterologous immunization with BCG priming and rAd-TB4 boosting highly induced IFN-γ secretion in response to stimulation with all antigens (Ag85B, Rv2660c, and ESAT-6) and PPD in lung lymphocytes (Figure 1B). rAd-TB4-immunization increased IFN-γ secretion in splenocytes upon stimulation with the antigens and PPD compared to BCG immunization, and the difference was significant in cells stimulated with Ag85B and ESAT-6 (Figure 1C). Furthermore, in mice vaccinated with BCG/rAd-TB4, the TB antigen-specific IgG titer was higher than that in those vaccinated with BCG alone in Ag85B- and Rv2660c-specific response, and there was no difference in PPD and ESAT-6 specific IgG titer (Figure 1D).

To assess CD4⁺ and CD8⁺ T-cell responses to rAd-TB4 immunization, we used ICS to quantify cytokine frequencies. Ag-specific stimulated T-cells were stained with intracellular Th1-type cytokines (IL-2, TNF-α, and IFN-γ) and separated into triple-, double-, or single-positive populations based on a combination of secreted cytokines. Lung lymphocytes in rAd-TB4-immunized mice contained a higher proportion of triple-positive multifunctional CD4⁺ T-cells than those in BCG-immunized mice (Figure 2A). Similarly, the CD8⁺ T-cell population showed numerous triple- and double-positive T-cells in the rAd-TB4-immunized group (Figure 2B). These results suggest that rAd-TB4 promotes CD4⁺ and CD8⁺ T-cell responses and implicates the induction of M. tuberculosis-specific IgG production.

Establishing an appropriate regimen is a crucial step in investigating vaccine efficacy. Therefore, we determined the optimal route of administration and dose to establish the maximum benefits of rAd-TB4, which showed high immunogenicity in BCG-primed mice. Mice were vaccinated with rAd-TB4 via SC (BCG/rAd-TB4-SC group) or IM (BCG/rAd-TB4-IM group) injections 10 weeks after BCG priming (Table 1). Lung lymphocytes in the SC and IM groups stimulated with the Ag85B peptide produced more IFN-γ than those in the BCG group. Similarly, IFN-γ production in splenocytes of the SC and IM groups significantly increased compared to that in those of the BCG group (Figure 3A).

Based on the multifunctional features of rAd-TB4 (Figure 2), we quantified the frequencies of cytokine (IFN-γ, TNF-α, and IL-2) responses, which comprised three (3+), two (2+), or one (1+) cytokine(s) in both the lungs and spleen. In the case of lung lymphocytes, SC immunization of rAd-TB4 elicited a higher population of 3+ cytokine-secreting CD4⁺ and CD8⁺ T cells compared to IM immunization. IM immunization showed identical levels of BCG groups (Figure 3B). Similarly, in CD4⁺ T cells of splenocytes, the highest proportion of 3+ T cells was observed in SC-immunized mice. In the case of CD8⁺ T cells, however, it was difficult to compare the differences between the three groups (Figure 3C). Moreover, among cytokine-secreting T cells, the proportion of 3+ T cells was highest in the BCG/rAd-TB4-SC group. In contrast, BCG and rAd-TB4-IM groups elicited only marginal T-cell responses in both organs (Figures 3B,C, pie chart).

Next, lung lymphocytes of rAd-TB4-vaccinated mice were stimulated ex vivo to profile Th1-, Th2-, and Th17-related cytokine productions (Figure 4). The production of Th1-related cytokines (IFN-γ, TNF-α, and IL-2) was elevated in both SC and IM groups compared to that in the BCG group. The increases in IFN-γ and IL-2 production were more significant in the SC group (p < 0.0001 for both) than those in the IM group (p < 0.05 for both), whereas the increase in TNF-α was not significant. The level of IL-17A, which has recently attracted attention for its essential role in vaccine-induced immune responses, also increased in the SC group compared to that in the BCG group; however, the increase was not significant. In addition, representative pro-inflammatory cytokines, such as granulocyte-macrophage colony-stimulating factor (p > 0.05) and IL-12p40 (p < 0.01) were induced after SC administration of rAd-TB4 (Figure 4).

Next, we challenged C57BL/6 mice (n = 5) with the M. tuberculosis H37Rv strain via aerosol exposure to evaluate the protective efficacy of rAd-TB4. Eight weeks after infection, we collected mouse serum to measure antigen specific IgG titer and the mouse lungs to determine the bacterial load in each individual animal and stained them with H&E to measure the area occupied by inflammation (Figure 5A). Inflamed area and CFU measurements were conducted twice to verify protective effectiveness. Along with the above T-cell response and cytokine production result followed by SC immunization, BCG/rAd-TB4 immunization produced a significantly higher Ag85B-specific IgG and IgG1 compared with BCG immunization (Figure 5B). Notably, IgG2c production which is engaged in Th1 response was induced in rAd-TB4 immunization. In the case of Experiment 1, BCG/rAd-TB4 immunization elicited a significant protective effect against H37Rv (5.013 log₁₀) compared to that with PBS (5.889 log₁₀) or BCG (5.342 log₁₀) immunization (Figure 5C). The histopathology result showed reduction in the inflamed area of BCG/rAd-TB4 mice groups; however, no significant differences were present between the unvaccinated and BCG-vaccinated groups (Figure 5D). Due to the inconsistent results observed in the inflamed area, additional experiments were conducted to revalidate these findings. C57BL/6 mice (n = 3) were challenged with the H37Rv strain via identical challenge method. In Experiment 2, the rAd-TB4 boosting resulted in a reduction of both the inflamed area and CFU values compared to those of the PBS and BCG groups (PBS: 5.970 log₁₀; BCG: 5.314 log₁₀; BCG/rAd-TB4: 5.118 log₁₀) (Figures 5C,D). Representative images of H&E staining lung images from Experiment 1 also showed reduced inflammatory regions (Figure 5E). To evaluate the protective efficacy of rAd-TB4, a challenge study was conducted using the clinical isolated strain HN878. C57BL/6 mice (n = 6) were challenged with HN878 following an identical regimen and infection schedule of H37Rv challenge experiment (Figure 6A). Eight weeks post-infection, lungs and spleens were harvested to assess bacterial load and to identify T cell phenotype analysis. The BCG/rAd-TB4 group exhibited a significantly lower bacterial burden in the lungs and spleens compared to the BCG-only group (Figure 6B) and showed reduced inflammatory lesions in the rAd-TB4 boosted group (Figure 6C). To further characterize the cellular response to M. tuberculosis infection, a T cell phenotype analysis was performed on CD4⁺ T cells isolated from lung lymphocytes and splenocytes. To identify phenotypic differences, we analyzed the distributions of the canonical CD4⁺ T cell subsets. Lung lymphocytes of rAd-TB4 boosted mice contained significantly higher numbers of Ag85B-specific central memory T (Tcm) cell and effector memory T (Tem) cell compared to those in BCG vaccinated mice (Figure 6D). Notably, splenocytes from rAd-TB4 boosted mice demonstrated a remarkable increase in the effector T (Teff) cell population (Figure 6E).

Next, we determined the transcriptome profile associated with BCG/rAd-TB4 immunization in M. tuberculosis-infected mice 12 weeks post-challenge. Whole blood samples collected at pre-infection, 1 (p.i.1week) and 4 weeks post-infection (p.i.4weeks) from both groups were used for RNA isolation and transcriptome profiling. PCA using RNA-sequencing (RNA-seq) data revealed clustering of all groups, except for the rAd-TB4 p.i.4weeks group (Figure 7A). We identified 23282 DEGs [fold change, 2.0; normalized data (log₂), 2.0; p ≤ 0.05] using 18 RNA-seq libraries (triplicates for each group). Among these DEGs, 143, 104, and 84 genes were significantly upregulated, and 186, 131, and 182 genes were significantly downregulated in the rAd-TB4-immunized group at pre-infection, p.i.1week, and p.i.4weeks, respectively, compared to those of the respective BCG-administered groups (Figure 7B; Table 2). Normalized data from each group were hierarchically clustered to gain further insight. The BCG-administered groups at pre-infection and p.i.1week showed minimal changes, while gene expression was most variable in the rAd-TB4-immunized group at p.i.4weeks compared to those in pre-infection and p.i.1week (Figure 7C). In addition, we hierarchically clustered the data based on the fold-change between the rAd-TB4- and BCG-vaccinated groups at each time point. Gene expression dynamics were more discrete at p.i.4weeks than at the other two time points (Figure 7D). These results demonstrated that the expression pattern of the transcriptome changed over time after the M. tuberculosis challenge in the rAd-TB4-vaccinated group.

GO functional enrichment analysis of the DEGs identified in the BCG- and rAd-TB4-vaccinated mice was performed to screen for specific expression during infection. Among the top 10 enriched biological processes during pre-infection, “innate immune response,” “inflammatory response,” and “positive regulation of cell migration (Matsumiya et al., 2014; Santoro et al., 2018)” were positively regulated in rAd-TB4-vaccinated mice compared with those in the BCG-vaccinated mice (Figures 8A,B). At p.i.4weeks, “positive regulation of the canonical Wnt signaling pathway” was downregulated in the rAd-TB4-vaccinated groups compared to that in the BCG-vaccinated group (Figure 8C). We compared the mRNA expression levels of genes related to the canonical Wnt/β-catenin pathway in rAd-TB4-vaccinated mice and BCG-vaccinated mice to precisely identify the affected signaling moieties. The findings demonstrated that vps35, col1a1, and sulf2 were downregulated in rAd-TB4-vaccinated mice compared to those in BCG-vaccinated mice at p.i.4weeks (Supplementary Figure S3A). Moreover, these genes showed reduced fold changes after the pathogen challenge (Supplementary Figure S3B). These observations indicate that rAd-TB4 immunization engages in down-regulation of canonical Wnt/β-catenin pathway.