Increased pulmonary tumor necrosis factor alpha, interleukin-6 (IL-6), and IL-17A responses compensate for decreased gamma interferon production in anti-IL-12 autovaccine-treated, Mycobacterium bovis BCG-vaccinated mice

Abstract

Interleukin-12 (IL-12) and IL-23 (which share a p40 subunit) are pivotal cytokines in the generation of protective Th1/Th17-type immune responses upon infection with the intracellular pathogen Mycobacterium tuberculosis. The role of IL-12 and IL-23 in protection conferred by the tuberculosis vaccine Mycobacterium bovis bacillus Calmette-Guérin (BCG) is, however, less well documented. By using an autovaccine approach, i.e., IL-12p70 cross-linked with ovalbumin and PADRE peptide formulated with the GSK proprietary adjuvant system AS02(V), we could specifically neutralize IL-12 while leaving the IL-23 axis intact. Neutralization of IL-12 before M. tuberculosis challenge rendered C57BL/6 mice highly susceptible, resulting in 30-fold-higher CFU in spleen and lungs and accelerated mortality. In contrast, neutralization of IL-12 in BCG-vaccinated mice prior to M. tuberculosis challenge only marginally affected vaccine-mediated protection. Analysis of cytokine production in spleen and lungs 3 weeks post-TB challenge by enzyme-linked immunosorbent assay and functional and flow cytometric assays showed significantly reduced mycobacterium-specific gamma interferon (IFN-γ) responses in M. tuberculosis-infected and BCG-vaccinated mice that had been treated with the autovaccine. Purified protein derivative-induced tumor necrosis factor alpha (TNF-α), IL-6, and IL-17A levels, however, were highest in lungs from BCG-vaccinated/IL-12-neutralized animals, and even unstimulated lung cells from these mice produced significant levels of the three cytokines. Mycobacterium-specific IL-4 and IL-5 production levels were overall very low, but IL-12 neutralization resulted in increased concanavalin A-triggered polyclonal secretion of these Th2-type cytokines. These results suggest that TNF-α, IL-6, and IL-17A may be more important pulmonary effector molecules of BCG-mediated protection than IFN-γ in a context of IL-12 deficiency.

FIG. 1.

(A) Anti-IL-12 and anti-IL-23 inhibitory titers in auto-IL-12 vaccine-treated mice. C57BL/6 mice, immunized or not with M. bovis BCG, were vaccinated with IL-12-(OVA)-PADRE complexes emulsified in AS02_V adjuvant. Control mice received only the adjuvant. Anti-IL-12 (closed symbols) and anti-IL-23 (open symbols) inhibitory titers were determined as described in Materials and Methods, with bioassays using Ba/F3 cells transfected with murine IL-12Rα/IL-12Rβ1 chains and murine IL-12Rβ1/IL-23R chains, respectively. Inhibitory titers were calculated as the serum dilution that inhibited 50% of the biological activity of 1 ng/ml IL-12 or 1 ng/ml IL-23 (as measured in a proliferation assay using tritiated thymidine and a hexosaminidase colorimetric assay, respectively). (B and C) Effects of in vivo IL-12 neutralization on susceptibility to intravenous M. tuberculosis. Bacterial replication is shown for spleen (B) and lungs (C) of untreated (open squares) or IL-12-neutralized (filled triangles) C57BL/6 mice at day 1 (n = 2), day 30 (n = 4), and day 50 (n = 4 for untreated and n = 2 for IL-12-neutralized mice) post-TB challenge. Data are expressed as means ± the SD of the log₁₀ mRLU/organ. NS, not statistically significant; ***, P < 0.001. (Results are from experiment 1.)

FIG. 2.

Effect of in vivo IL-12 neutralization on M. bovis BCG-mediated protection against M. tuberculosis. Bacterial replication in the spleen and lungs of untreated (white dots) or IL-12-neutralized (black dots) B6 mice (unvaccinated or vaccinated with BCG) at week 3 (upper panel) or week 6 (lower panel) post-TB challenge is illustrated. Data are expressed as the log₁₀ milli-RLU/organ. Each dot represents one animal, and the bar represents the mean value for each group (n = 4 to 6). (Results are from experiment 4.) ns, not statistically significant; **, P < 0.01; ***, P < 0.001.

FIG. 3.

Effect of in vivo IL-12 neutralization on M. bovis BCG-mediated protection against M. tuberculosis. Bacterial replication in the spleen (upper panel) and lungs (lower panel) of naïve or IL-12-neutralized B6 mice (unvaccinated or vaccinated with BCG) at days 21, 42, and 63 post-TB challenge. Data are expressed as the mean ± SD log₁₀ mRLU/organ of four mice tested/group. ns, not statistically significant; **, P < 0.01; ***, P < 0.001; #, all mice succumbed to TB infection at this time point. (Results are from experiment 3.)

FIG. 4.

Effect of anti-IL-12 autovaccine on IFN-γ production in M. tuberculosis-infected and in M. bovis BCG-vaccinated mice. Spleen cells from four individual mice and lung cells (pooled per group) were collected at 3 weeks post-TB challenge (experiment 4) and cultured without stimulation (white bars) or in the presence of PPD (gray bars) or ConA (black bars). The 72-h culture supernatants were assessed for IFN-γ activity by ELISA. Data are reported as mean titers in pg/ml (± SD, for spleen only).

FIG. 5.

Effects of anti-IL-12 autovaccine on IL-5 production in M. tuberculosis-infected and in M. bovis BCG-vaccinated mice. Spleen cells from four individual mice and lung cells (pooled per group) were collected at 6 weeks post-TB challenge and cultured without stimulation (white bars) or in the presence of PPD (gray bars) or ConA (black bars). The 72-h culture supernatants were assessed for IL-5 activity by ELISA. Data are reported as mean titers in pg/ml (± SD, for spleen only).

FIG. 6.

Effects of anti-IL-12 autovaccine on TNF-α production in M. tuberculosis-infected and in M. bovis BCG-vaccinated mice. Spleen cells from four individual mice and lung cells (pooled per group) were collected at 3 weeks post-TB challenge and cultured without stimulation (white bars) or in the presence of PPD (gray bars) or ConA (black bars). The 72-h culture supernatants were assessed for TNF-α activity in a bioassay on WEHI 164 cells. Data are reported as mean titers in pg/ml (± SD, for spleen only).

FIG. 7.

Effects of anti-IL-12 autovaccine on IL-17A production in M. tuberculosis-infected and in M. bovis BCG-vaccinated mice. Spleen cells from four individual mice and lung cells (pooled per group) were collected at 3 weeks post-TB challenge and cultured without stimulation (white bars) or in the presence of PPD (gray bars) or ConA (black bars). The 72-h culture supernatants were assessed for IL-17A activity by ELISA. Data are reported as mean titers in pg/ml (± SD, for spleen only).

FIG. 8.

Effects of anti-IL-12 autovaccine on IL-6 production in M. tuberculosis-infected and in M. bovis BCG-vaccinated mice. Spleen cells from four individual mice and lung cells (pooled per group) were collected at 3 weeks post-TB challenge and cultured without stimulation (white bars) or in the presence of PPD (gray bars) or ConA (black bars). The 72-h culture supernatants were assessed for IL-6 activity in a bioassay using 7TD1 cells. Data are reported as mean titers in pg/ml (± SD, for spleen only).