The apoptotic response to pneumolysin is Toll-like receptor 4 dependent and protects against pneumococcal disease

Abstract

Pneumolysin, the cholesterol-dependent cytolysin of Streptococcus pneumoniae, induces inflammatory and apoptotic events in mammalian cells. Toll-like receptor 4 (TLR4) confers resistance to pneumococcal infection via its interaction with pneumolysin, but the underlying mechanisms remain to be identified. In the present study, we found that pneumolysin-induced apoptosis is also mediated by TLR4 and confers protection against invasive disease. The interaction between TLR4 and pneumolysin is direct and specific; ligand-binding studies demonstrated that pneumolysin binds to TLR4 but not to TLR2. Involvement of TLR4 in pneumolysin-induced apoptosis was demonstrated in several complementary experiments. First, macrophages from wild-type mice were significantly more prone to pneumolysin-induced apoptosis than cells from TLR4-defective mice. In gain-of-function experiments, we found that epithelial cells expressing TLR4 and stimulated with pneumolysin were more likely to undergo apoptosis than cells expressing TLR2. A specific TLR4 antagonist, B1287, reduced pneumolysin-mediated apoptosis in wild-type cells. This apoptotic response was also partially caspase dependent as preincubation of cells with the pan-caspase inhibitor zVAD-fmk reduced pneumolysin-induced apoptosis. Finally, in a mouse model of pneumococcal infection, pneumolysin-producing pneumococci elicited significantly more upper respiratory tract cell apoptosis in wild-type mice than in TLR4-defective mice, and blocking apoptosis by administration of zVAD-fmk to wild-type mice resulted in a significant increase in mortality following nasopharyngeal pneumococcal exposure. Overall, our results strongly suggest that protection against pneumococcal disease is dependent on the TLR4-mediated enhancement of pneumolysin-induced apoptosis.

FIG. 1.

Pneumolysin interacts with TLR4 but not with TLR2 in a solid-phase binding assay. (A) Ninety-six-well ELISA plates were coated with pneumolysin, blocked, and then overlaid with increasing amounts of lysates from HEK293-TLR2-YFP or HEK293-TLR4-YFP cells. After extensive washing, TLR-YFP bound to the coating protein was detected by an anti-GFP antibody. (Inset) Immunoblot with anti-GFP antibody with equal amount of lysates used in the interaction assay. (B) Pneumolysin coated at a concentration of 0.17 μg/well and maximum HEK293-TLR lysate concentration of ∼11 μg/well. (C) BSA coated at concentration of 0.2 μg/well and maximum HEK293-TLR lysate concentration of 50 μg/well. The results shown are representative of three or more experiments. HRP, horseradish peroxidase; IgG, immunoglobulin G; Ply, pneumolysin.

FIG. 2.

Pneumolysin-induced apoptosis is TLR4 dependent. (A) Selective TLR4 antagonist B1287 blocks pneumolysin-induced apoptosis. TLR4 wild-type (C3H/HeOuJ) macrophages were stimulated with increasing concentrations of pneumolysin in the presence or absence of B1287. Apoptosis was assessed by enumerating cells with hypodiploid DNA by flow cytometry; the assay was repeated twice, and the results of a representative experiment are shown. (B) Macrophages with functional TLR4 (C3H/HeOuJ) exhibit significantly more pneumolysin-induced apoptosis than TLR4-defective (C3H/HeJ) macrophages. Macrophages derived from wild-type and TLR4-defective mice were exposed to increasing concentrations of pneumolysin and assayed for apoptosis by enumerating cells with hypodiploid DNA as described above. An asterisk indicates that the P value is 0.037 as determined by a t test. (Inset) Human epithelial cells transfected with TLR4 (HEK-TLR4) are more susceptible to apoptosis than cells transfected with TLR2 (HEK-TLR2). Stably transfected cells were exposed to increasing concentrations of pneumolysin, and apoptosis was assessed by measurement of histone-associated DNA by flow cytometry. The assay was repeated more than three times, and the results of a representative experiment are shown.

FIG. 3.

Pneumolysin-induced apoptosis is caspase dependent. The apoptotic response to pneumolysin is inhibited by the pan-caspase inhibitor zVAD-fmk. RAW264.7 macrophages were exposed to increasing concentrations of pneumolysin in the presence or absence of the pan-caspase inhibitor (20 μg/ml for 18 h), and apoptosis was assessed by TUNEL staining followed by flow cytometry as described in Materials and Methods. Asterisks indicate that the P value is 0.047. The results shown are representative of three or more experiments. Ply, pneumolysin.

FIG. 4.

TLR4 mediates apoptosis by pneumolysin-producing pneumococci in the upper respiratory tree of mice. Wild-type (C3H/HeOuJ) and TLR4-defective (C3H/HeJ) mice were intranasally inoculated with strain WU2 (five mice per group) or its isogenic, pneumolysin-negative derivative WU2-PLA (four mice per group); 18 h after challenge, the mice were sacrificed, and cells recovered from retrograde tracheal washes were assessed for apoptosis by TUNEL staining. Cells were first examined by fluorescence (detection at 515 to 565 nm [green]) and bright-field microscopy, which was followed by enumeration by flow cytometry. (A and B) Collages of micrographs showing phenotypes of recovered cells. Cells exhibiting bright TUNEL staining (A), indicative of DNA fragmentation associated with apoptosis, were characteristically shrunken and were much smaller than cells that remained unstained and possessed numerous apoptotic bodies in the cytoplasm. Cells that did not show any TUNEL staining due to the absence of DNA fragmentation (B) were much larger and possessed an intact cell membrane. Photographs were taken at the same magnification and are representative of the cell populations observed. (C) Flow cytometry analysis revealed that wild-type mice challenged with WU2 have a significantly larger proportion of apoptotic cells than TLR4-defective mice (P = 0.016); no differences were noted when pneumolysin-deficient strain WU2-PLA was used (P = 0.29).

FIG. 5.

In vivo inhibition of apoptosis in the upper respiratory tree favors progression of invasive pneumococcal disease. Wild-type (C3H/HeOuJ) mice received either the pan-caspase inhibitor zVAD-fmk (n = 12) or vehicle alone (n = 12) intranasally every 12 h prior to and following intranasal inoculation with strain WU2, a pneumolysin-producing type 3 pneumococcus. Survival was monitored twice daily for 8 days. The difference in survival between zVAD-fmk- and vehicle-treated mice was significant (P = 0.043, as determined by a Kaplan-Meier test).