Mycobacterium tuberculosis triggers apoptosis in peripheral neutrophils involving toll-like receptor 2 and p38 mitogen protein kinase in tuberculosis patients

Abstract

Polymorphonuclear neutrophils (PMN) exposed to Mycobacterium tuberculosis display bactericidal responses and produce inflammatory proteins. This PMN-mediated inflammatory response is regulated by an activation of the apoptotic program, which collaborates to avoid tissue injury. In vitro, circulating PMN from patients with tuberculosis (TB) show an increased spontaneous apoptosis, and M. tuberculosis-induced activation accelerates the PMN apoptosis. In this study, we evaluated the mechanisms involved in spontaneous and M. tuberculosis-induced apoptosis. We demonstrate that apoptosis of PMN is not induced by lipoarabinomannan or by a whole-cell lysate of M. tuberculosis and that neither tumor necrosis factor alpha nor CD11b, CD14, and Fcgamma receptors are involved. Apoptosis of PMN from patients with active TB (TB-PMN) is induced by the interaction with the whole M. tuberculosis via Toll-like receptor 2 (TLR2), and, in contrast to spontaneous apoptosis, it involves the p38 mitogen-activated protein kinase (MAPK) pathway. These results correlate with a high expression of phosphorylated p38 (p-p38) in circulating TB-PMN and with the ability of M. tuberculosis to induce in vitro the expression of p-p38 in PMN. Therefore, when the bacterial burden is low, TB-PMN could be detecting nonopsonized M. tuberculosis via TLR2, leading to the activation of the p38 MAPK pathway, which in turn would induce PMN activation and apoptosis. This mechanism needs further confirmation at the site of infection.

FIG. 1.

M. tuberculosis-induced apoptosis in PMN depends on M. tuberculosis/PMN ratio. N-PMN (▴) and TB-PMN (▪) (3 × 10⁶/ml) were cultured in medium alone (control) or stimulated with M. tuberculosis (Mtb) at different M. tuberculosis/PMN ratios (1:3, 1:1, 10:1, and 20:1) for 18 h. Thereafter, apoptosis was evaluated on the basis of AV-FITC binding. Results are expressed as means (± standard errors of the means) of the percentage of AV⁺ cells (n = 10). Statistical differences for control versus M. tuberculosis: *, P < 0.0001; ϕ, P < 0.0005; #, P < 0.01. Statistical differences for N-PMN versus TB-PMN: Ψ, P < 0.02, Φ, P < 0.005.

FIG. 2.

Endogenous TNF-α is not involved in the M. tuberculosis-induced apoptosis of TB-PMN. PMN (3 × 10⁶/ml) cells were treated with anti-TNF-α (aTNFα) or an irrelevant isotype antibody following incubation with or without M. tuberculosis (Mtb) (10⁶ /ml) for 18 h. The percentage of apoptotic cells was evaluated and expressed as means (± standard errors of the means) of the percentage of AV⁺ cells (n = 10). *, P < 0.001 (TB-PMN control versus M. tuberculosis).

FIG. 3.

CD11b, CD14, and Fcγ receptors are not involved in M. tuberculosis-induced apoptosis. N-PMN and TB-PMN (3 × 10⁶/ml) were treated with specific antibodies against CD11b (aCD11b), CD14, CD16, CD32, and CD64 (solid bars) or their corresponding isotype controls (open bars) following incubation without (control) or with M. tuberculosis (Mtb) (10⁶/ml) for 18 h. Apoptotic cells were evaluated by AV-FITC binding assay as described in Materials and Methods. Results are expressed as means (± standard errors of the means) of the percentage of AV⁺ cells (n = 8). IgG, immunoglobulin G.

FIG. 4.

TLR2 triggers apoptosis in TB-PMN. (A) Blocking of TLR2 abolishes M. tuberculosis-induced apoptosis in TB-PMN. PMN (3 × 10⁶/ml) from eight controls and eight TB patients were treated with specific antibody against TLR2 (aTLR2) or its corresponding isotype following incubation with or without M. tuberculosis (Mtb) (10⁶ /ml) for 18 h. Apoptotic cells were evaluated by AV-FITC binding assay as described in Materials and Methods). *, P < 0.001 (M. tuberculosis versus control); #, P < 0.04 (M. tuberculosis versus M. tuberculosis plus anti-TLR2. (B) TLR2 expression in freshly isolated PMN. Expression of TLR2 in TB-PMN () and N-PMN (—) assessed by using an FITC-anti-TLR2 antibody and FACScan. Isotype-control antibody-stained cells are shown as a dotted line histogram. An example from 13 experiments done in each group are shown. (C) TLR expression in N-PMN (▴) and TB-PMN (▪) cultured with () or without (—) M. tuberculosis at 0, 3, and 18 h. Expression of TLR2 was evaluated and expressed as relative MFI as detailed in Materials and Methods. *, P < 0.04; **, P < 0.002 (control versus M. tuberculosis N-PMN) (n = 8). ψ, P < 0.02; ψψ, P < 0.0004 (TB-PMN vs. N-PMN) (n = 8).

FIG. 5.

p38 MAPK is required for M. tuberculosis-induced apoptosis in TB-PMN. PMN (3 × 10⁶/ml) were pretreated with 25 μM SB203580 or 50 μM PD98059 and incubated with medium alone (control) or M. tuberculosis (Mtb) (10⁶ /ml) for 18 h. Apoptotic cells were evaluated by AV-FITC binding assay as described in Materials Methods. Results are expressed as means (± standard errors of the means) of the percentage of AV⁺ cells (n = 15). *, P < 0.05 (M. tuberculosis versus control); †, P < 0.03; ‡, P < 0.05 (inhibitor versus untreated PMN).

FIG. 6.

p38 MAPK activation in circulating PMN and in M. tuberculosis-stimulated PMN. (A) Expression of p-p38 in recently isolated or cultured PMN from N- and TB-PMN was evaluated. As detailed in Materials and Methods, 3 × 10⁶ cells/ml were pretreated with 25 μM SB203580 and incubated with medium alone (control) or M. tuberculosis (Mtb) (10⁶/ml) for 2 h, and the p-p38 expression was evaluated by flow cytometry. Results are expressed as the means (± standard errors of the means) of the MFI (n = 10). Statistical differences: *, P < 0.03 (M. tuberculosis versus control); †, P < 0.03 (2 h versus 0 h); ‡, P < 0.001 (TB-PMN versus N-PMN); §, P < 0.02; ‖, P < 0.05 (M. tuberculosis versus M. tuberculosis-SB). (B) Representative histograms of p-p38 expression in control (—), M. tuberculosis-stimulated (—), and M. tuberculosis-SB-treated (- - -) PMN in a 2-h culture. (C) Equal amounts of cellular proteins were separated by SDS-PAGE and blotted with anti-active p38 MAPK (upper panel). To ascertain that each lane received similar amounts of p38 MAPK, the blots were reprobed with anti-total-p38 MAPK antibody (lower panel). Results are for freshly isolated PMN from one control (N) and two patients (TB1 and TB2) (0 h) or from a control and one TB patient and cultured with or without M. tuberculosis (2 h). The intensities of the autoradiographic bands were estimated by densitometric scanning, and the relative densities (ROD) of the p-p38 bands were referred to N-PMN.