Intranasal vaccination with engineered BCG expressing CCL2 induces a stronger immune barrier against Mycobacterium tuberculosis than BCG

Abstract

Intradermal Mycobacterium bovis Bacillus Calmette-Guérin (BCG) vaccination is currently the only licensed strategy for preventing tuberculosis (TB). It provides limited protection against pulmonary TB. To enhance the efficacy of BCG, we developed a recombinant BCG expressing exogenous monocyte chemoattractant CC chemokine ligand 2 (CCL2) called rBCG-CCL2. Co-culturing macrophages with rBCG-CCL2 enhances their abilities in migration, phagocytosis, and effector molecule expression. In the mouse model, intranasal vaccination with rBCG-CCL2 induced greater immune cell infiltration and a more extensive innate immune response in lung compared to vaccination with parental BCG, as determined by multiparameter flow cytometry, transcriptomic analysis, and pathological assessments. Moreover, rBCG-CCL2 induced a high frequency of activated macrophages and antigen-specific T helper 1 (Th1) and Th17 T cells in lungs. The enhanced immune microenvironment responded more effectively to intravenous challenge with Mycobacterium tuberculosis (Mtb) H37Ra, leading to significant reductions in H37Ra burden and pathological damage to the lungs and spleen. Intranasal rBCG-CCL2-vaccinated mice rapidly initiated pro-inflammatory Th1 cytokine release and reduced pathological damage to the lungs and spleen during the early stage of H37Ra challenge. The finding that co-expression of CCL2 synergistically enhances the immune barrier induced by BCG provides a model for defining immune correlates and mechanisms of vaccine-elicited protection against TB.

Keywords: BCG; CCL2; Mycobacterium tuberculosis; T cell immunity; Tuberculosis; intranasal vaccination; macrophage.

Graphical abstract

Figure 1

rBCG-CCL2 presents stronger ability to recruit and activate macrophages than parental BCG in vitro (A) The expression of rCCL2 in rBCG-CCL2. Top, CCL2 in rBCG-CCL2 was examined via western blotting with anti-CCL2 monoclonal antibody. Marker, protein ladder (kDa); BCG, supernatant of BCG lysate that was transformed with vector plasmid pMV261, as negative control; rBCG-CCL2, supernatant of rBCG-CCL2 lysate. Bottom, concentration of CCL2 in 10⁶ CFUs rBCG-CCL2 was detected using a mouse MCP-1 (CCL2) ELISA Detection Kit. Data present the absorbance of BCG and rBCG-CCL2 lysate supernatant under 450 nm wavelength (n = 4 biological repeats). (B) Chemotactic activity of rBCG-CCL2 performed by transwell assay. Data were number of cells that migrated through the membrane of insert (n = 5 repeats). (C) Phagocytic activity in macrophages administrated with BCG or rBCG-CCL2 (MOI 10:1) was assessed by detecting the uptake of IgG-opsonized FITC-latex beads. Data were collected as MFI and percentage of FITC⁺ cells that was quantitatively assessed using flow cytometry (n = 4 independent experiments). (D) Concentration of NO released by BCG- or rBCG-CCL2 (MOI 10:1)-treated macrophages. The NO levels in cell pellets and cell culture medium were examined separately using Griess reagent (n = 3 independent experiments). (E) mRNA levels for cytokines, chemokines, Ccr2, and iNos in macrophages relative to β-actin (n = 3 independent experiments). The levels were measured by relative RT-qPCR after Raw264.7 cells were incubated with BCG or rBCG-CCL2 at MOI 10:1 for 24 h. Commercialized CCL2 was used as control. Data are presented as mean ± SD and analyzed using ordinary one-way/two-way ANOVA multiple comparisons (ns, non-significant differences; ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001).

Figure 2

Intranasal vaccination with rBCG-CCL2 leads to increased recruitment of leukocytes into the lungs (A) Schematic of the experiment. Mice were intranasally (i.n.) inoculated with 2 × 10⁶ CFUs BCG or rBCG-CCL2. Lung samples were collected at days 1, 7, and 35 post-inoculation. (B) BCG and rBCG-CCL2 CFUs in lungs at days 1, 7, and 35 post-inoculation (n = 4 mice per group). (C) Number of cells per lung collection for leukocyte populations in each group at days 1, 7, and 35 post-inoculation using flow cytometry (n = 5–6 mice per group). Pooled data from three independent experiments. Data are presented as mean ± SD and analyzed using one-way/two-way ANOVA multiple comparisons (∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001). Antibody panel and gating strategy are shown in Figure S3.

Figure 3

Transcriptomics from BALF at day 7 post inoculation (A) Heatmap shows the general differentially expressed genes in BALF of rBCG-CCL2-vaccinated mice compared with those of BCG-vaccinated mice. Gene expression level is measured by FPKM value. Comparing FPKM value between samples from rBCG-CCL2-vaccinated mice and those from BCG-vaccinated mice, different genes were defined as∣log₂ FC∣≥1 & Q < 0.05 (FC, fold change). (B) Scatterplot shows the top 20 (p < 0.0001) significantly upregulated Kyoto Encyclopedia of Genes and Genomes enrichment pathways in BALF of rBCG-CCL2-vaccinated mice compared to those of BCG. (C) Expression level of selected genes in BALF from PBS-, BCG-, and rBCG-CCL2-vaccinated mice. Data represent the FPKM value. Data represent one independent experiment (n = 18 mice per group; in the case of PBS group, BALF from nine mice were pooled into one sample due to lower cell number in BAL; in the case of BCG group or rBCG-CCL2 group, BALF from six mice were pooled into one sample). In (C), statistical analyses done using unpaired Student t test (∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001).

Figure 4

Mucosal rBCG-CCL2 vaccination promotes macrophage activation (A) Representative histograms (top) and fluorescence intensity (bottom) of CD86, iNOS, MHCII, and CD206 expression gated on macrophages at day 35 post-inoculation. (B) Tissue sections were stained with HRP-conjugate CD86 and CD206 antibody (left). Data present percentage of area occupied by CD86⁺ or CD206⁺ cells relative to total lung area (right). Data represent two independent experiments (n = 5–6 mice per group). Data are shown as mean ± SEM and analyzed using one-way ANOVA multiple comparisons (ns, non-significant differences; ∗p < 0.05; ∗∗p < 0.01; ∗∗∗∗p < 0.0001).

Figure 5

Intranasal rBCG-CCL2 vaccination stimulates an antigen-specific Th1 and Th17 cell response Flow cytometric analysis of intracellular staining in lung cells of vaccinated mice. The lung cells were collected at day 35 post-inoculation; stimulated with H37Rv WCL in the presence of CD49d, CD28, and brefeldin A; and analyzed for cytokine production using intracellular staining. The single- (A and B) and multi-functional (C and D) antigen-specific T cell responses in lung are shown. IFN-γ-, IL-2-, IL-17A-, or TNF-α-producing CD44⁺ CD4 and CD8 T cells were analyzed. (C and D) Pie charts represent the proportions of total cytokine production comprising each cytokine combination. Data represent two independent experiments (n = 5–6 mice per group). Data are shown as mean ± SD and analyzed by two-way ANOVA multiple comparisons (ns, non-significant differences; ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001). The antibody panel and gating strategy are shown in Figure S5.

Figure 6

Intranasal rBCG-CCL2 vaccination confers superior protection against Mtb infection (A) Schematic of the experiment. The vaccinated mice were challenged intravenously with 5 × 10⁶ CFUs of the Mtb H37Ra strain at day 35 post-inoculation, and the lungs and spleen were collected at days 7 and 60 post-challenge. (B) Mtb burden (CFUs) in right lung and half of a spleen were assessed and expressed as means ± SDs. (C) Representative H&E-stained lung sections from PBS-, BCG-, and rBCG-CCL2-vaccinated mice. Green arrows represent lymphocyte infiltration. Red arrows represent thickening of alveolar septa and fibrous connective tissue hyperplasia. (D) Histological scores of lung tissue sections measured at days 0, 7, and 60 post-challenge. Data represent two independent experiments at days 0, 7, and 60 (n = 5 mice per group) and were analyzed using two-way ANOVA multiple comparisons (ns, non-significant differences; ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001).

Figure 7

Intranasal rBCG-CCL2 vaccination enhances innate cytokine production and antigen-responsive T cell response after Mtb challenge (A) Concentration of cytokines in serum measured by ELISA at days 0, 7, and 60 post-intravenous Mtb challenge. Data are combined from three independent experiments (n = 16 mice per group). (B and C) Frequency of cytokine-producing CD44⁺ CD4 and CD8 T cells in lung at day 60 post-challenge were shown (d.p.c., days post-challenge). Mononuclear cells were isolated and stimulated with H37Rv WCL. Cytokine expression was assessed with intracellular fluorescence-conjugated antibody staining and examined with flow cytometry. (D and E) Frequency of polyfunctional CD44⁺ CD4 and CD8 T cells in lung that express multi-cytokines. Data represent two independent experiments at day 60 (n = 6 mice per group). The antibody panel and gating strategy are shown in Figure S5. Data are shown as means ± SDs and analyzed using two-way ANOVA multiple comparisons (ns, non-significant differences; ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001).