Innate Immunity to the Pathogenic Fungus Coccidioides posadasii Is Dependent on Toll-Like Receptor 2 and Dectin-1

Reagents.

Salmonella enterica serovar Minnesota Re 595 LPS (LPS) was prepared as previously described (29). The monoclonal antibody (MAb) 2A11 is an anti-mouse Dectin-1 antibody that was described by us previously (5). Control mouse immunoglobulin G2b (IgG2b) was obtained from Caltag. All reagents were tested for LPS contamination with the QCL-1000 chromogenic LAL assay (BioWhittaker).

Mice.

Female C57BL/6J, C57BL/10J, C3H/HeJ, and C3H/OuJ mice were purchased from Jackson Laboratories (Bar Harbor, Maine). TLR2^−/−, MyD88^−/−, and TLR4^−/− mice on a C57BL/6J background created by Shizuo Akira (Osaka University) were provided by Peter Tobias (Research Institute of Scripps Clinic, La Jolla, Calif.) and Eyal Raz (University of California, San Diego, Calif.). C57BL/10/ScCr mice, which have a deletion of the TLR4 gene, were purchased from the National Cancer Institute, National Institutes of Health. All mice were 6 to 8 weeks of age, housed in a pathogen-free facility, and handled according to the recommended guidelines.

C. posadasii.

C. posadasii (isolate C735) was grown as mycelial phase on a slant of 3.6% Mycosel agar, 0.5% yeast extract, and 50 μg of gentamicin/ml at room temperature. After the mycelia covered most of the surface of the slant, arthroconidia were isolated in water and washed twice in water by centrifugation at 2,200 × g at 4°C for 15 min. The pellet was used to inoculate 125 ml of modified Converse medium (15.96 mM ammonium acetate, 3.7 mM KH₂PO₄, 3.0 mM K₂HPO₄, 1.6 mM MgSO₄, 0.0125 mM ZnSO₄, 0.24 mM NaCl, 0.0204 mM CaCl₂, 0.143 mM NaHCO₃, 0.5 g of Tamol SN/liter, 4.0 g of glucose/liter, 0.05 g of N-Z amine/liter). The culture was incubated in a 180-rpm shaking incubator at 39.5°C in the presence of 10% CO₂. After 6 days, when the spherules had matured and some began to rupture, the culture was harvested by centrifugation at 2,200 × g for 10 min. The formalin-killed spherules (FKS) were prepared by resuspending the pellet in 5% formaldehyde in 0.9% NaCl overnight. The formaldehyde was removed by three washes in 0.9% NaCl, and the FKS were resuspended in 0.9% saline or tissue culture medium.

All preparations were made using pyrogen-free reagents and solutions. All lots of FKS contained less than 4.3 pg of endotoxin/10⁶ spherules.

Other stimuli.

Candida albicans was a clinical isolate made at the Veterans Administration hospital. The cells were grown as yeast cells in yeast extract-peptone-dextrose broth (Difco) at 37°C overnight, harvested by centrifugation, washed twice in phosphate-buffered saline (PBS), and fixed in formaldehyde as described above. Staphylococcus aureus in heat-killed form was obtained from Calbiochem. Zymosan was purchased from Sigma.

Activation of mouse macrophages.

To obtain elicited mouse peritoneal macrophages, mice were injected intraperitoneally with 1 ml of 5 mM sodium periodate in PBS 4 days prior to harvesting of macrophages. The peritoneal macrophages were harvested in PBS, washed with Dulbecco's modified Eagle high-glucose medium supplemented with 10% fetal calf serum (FCS), 10 mM HEPES, and 2 mM glutamine, and seeded at 2 × 10⁵ cells per well in a 96-well plates. The plate was incubated for 1 to 2 h at 37°C to allow the cells to adhere. The wells were washed once with medium. Adherent cells were stimulated with various reagents. In the experiment where we tested the effect of MAb anti-Dectin-1, 2 μg of the MAb or control IgG2b/ml was added to the wells and incubated for 30 min at 37°C prior to adding other stimuli. After 16 h of incubation, the supernatant was harvested and assayed for mouse cytokines using ELISA kits for TNF-α, interleukin 6 (IL-6) (OptEIA Set; Pharmingen), IL-10 (BD Biosciences), and MIP-2 (Duo Set; R&D Systems), according to the manufacturers' protocol.

Mouse alveolar macrophages were obtained from bronchoalveolar lavage, using the method previously described (22). Briefly, mice were euthanized by asphyxia in a high-CO₂ environment. The trachea was intubated using a 1.7-mm-outside-diameter polyethylene catheter. One-milliliter aliquots of Dulbecco's PBS (Invitrogen) containing 5 mM EDTA were instilled and removed to a total of 15 ml per mouse. Approximately 10 ml of lavage fluid was retrieved per mouse. The cells were washed with Dulbecco's modified Eagle high-glucose medium supplemented with 10% FCS, 10 mM HEPES, and 2 mM glutamine, seeded at 3 × 10⁵ cells per well in 96-well plates, and activated with FKS. The supernatant was harvested and assayed for cytokines as described above.

Activation of RAW 264.7 cells.

RAW 264.7 cells, stably transfected with mouse Dectin-1 (RAW-Dectin) and vector control (RAW-pFB), were constructed as described previously (4). The cells are grown in RPMI with 10% FCS, 2 mM glutamine, 10 mM HEPES, 50 μg of gentamicin/ml, and 1 mg of G418/ml. The cells were subcultured 1 day before the assay. The transfected cells were washed, resuspended in medium without antibiotics, and plated at 2 × 10⁵ cells per well in a 48-well plate. After the cells had adhered, the medium was removed and replaced with fresh medium containing LPS, FKS, or other components in a final volume of 200 μl/well. Supernatants were harvested after 18 h of incubation and assayed for cytokines by ELISA. When live spherules were used, cells were incubated with spherules in the BSL-3 laboratory, and the supernatant was filtered through a 0.22-μm syringe filter before being assayed for cytokines by ELISA.

TaqMan RT-PCR analysis for mouse cytokine expression.

To assess the level of cytokine gene expression, C57BL/6 elicited mouse peritoneal macrophages were plated at 6 × 10⁶ cells per well of a six-well tissue culture plate in a total volume of 2 ml. After stimulation with 5 × 10³ C. posadasii FKS/ml for 4 h, total RNA was extracted by using the RNeasy Mini kit (QIAGEN). cDNA was synthesized from the RNA by using the ThermoScript reverse transcription-PCR (RT-PCR) system with random hexamers as primers treated with RNase H (Invitrogen) and subjected to quantitative PCR analysis, using the ABI Prism 7700 and Sequence Detection System software, version 1.6.3 (Applied Biosystems).

Purification of β-glucan.

The C. posadasii spherules were grown up by inoculating arthroconidia in modified Converse medium as described above. The culture containing spherules was autoclaved for 20 min, allowed to cool, and centrifuged at 2,200 × g for 10 min. β-Glucan was extracted by using a previously described method (28). The precipitate was washed twice in saline, resuspended in chloroform-methanol (2:1), and mixed on a rotator at room temperature for 2 h. After centrifugation, the pellet was mixed with 1 M NaOH on a rotator for 1 h and then washed three times with water. The spherule pellet was then extracted with equal volumes of the following series of solvents: ethanol, acetone, and diethyl ether. The final pellet was then washed twice with water, dried under a Speedvac, and weighed. The pellet was reconstituted with water to a solution of 10 mg/ml. The solution was tested for endotoxin contamination by using the QCL-1000 kit (Biowhittaker) and found to be free of endotoxin.

Statistical analyses.

The data were analyzed in a one-way analysis of variance with posthoc Tukey tests for pairwise comparisons to avoid type I error inflation, using the Prism software (GraphPad). Probabilities of less than 0.05 were considered significant.

C. posadasiiFKS induced cytokine production in mouse macrophages.

Elicited C57BL/6 mouse peritoneal macrophages were incubated with FKS, and total RNA was isolated for Taqman RT-PCR analysis after 4 h of stimulation. Figure 1A shows that in the stimulated cells there were more than threefold-higher levels of mRNA of IL-6, IL-12, MIP-2, and TNF-α than in unstimulated cells. In contrast, we observed no significant increase in IL-4, IL-10, and gamma interferon transcripts. When macrophages were activated overnight with various concentrations, from 5 × 10² to 5 × 10⁴ FSK/ml, TNF-α and MIP-2 (Fig. 1B) were produced in a dose-dependent manner. We observed IL-6 production in a similar fashion (data not shown). A small amount of IL-10 could also be detected (data not shown). We also found that like peritoneal macrophages, alveolar macrophages and a murine macrophage cell line, RAW 264.7, secreted TNF-α, MIP-2, and IL-6 in response to FKS in a dose-dependent manner (data not shown).

The role of TLR4 in the stimulation of proinflammatory cytokines by C. posadasii formalin-killed spherules.

To evaluate the role of TLR4 in macrophage responses to Coccidioides, we incubated LPS, FKS, or formalin-killed C. albicans with elicited peritoneal macrophages of C3H/HeJ mice. We compared those cells, which have a dominant-negative point mutation in the signaling domain of TLR4, to cells from control C3H/OuJ mice. We measured secreted TNF-α, IL-6, and MIP-2 after 18 h of incubation by ELISA. Figure 2A shows that C3H/HeJ peritoneal macrophages responded poorly to concentrations of LPS up to 100 μg/ml, as expected. In contrast, C3H/HeJ cells responded normally to C. albicans. C. posadasii FKS stimulated C3H/HeJ macrophages to secrete TNF-α as efficiently as they stimulated C3H/OuJ macrophages, although small differences between the responses of the two cells can be observed at the highest dose of spherules. The same results were obtained when the supernatants were assayed for MIP-2 and IL-6 (data not shown). To further test the role of TLR4 in response to FKS, we compared the cytokine responses of peritoneal macrophages from TLR4 knockout (TLR4^−/−) and wild-type (WT) mice (C57BL/6). We found that the macrophages from TLR4^−/− mice produced a level of MIP-2 similar to that for the WT in response to 5 × 10⁴ FKS/ml, but they did not respond as well to 5 × 10⁵ FKS/ml (Fig. 2B). Similar results were obtained when comparing the C57BL/10/ScCr mice, which lack TLR4, and the control C57BL/10J mice (data not shown). To rule out endotoxin contamination as the explanation for these results, we tested C. posadasii FKS in an LAL assay and found that it was free of LPS. In addition, we performed an experiment where C. posadasii FKS was preincubated with 50 μg of polymyxin B/ml prior to macrophage stimulation. No significant effect on cytokine production was observed (data not shown). These data suggest that TLR4 is not necessary for the macrophage response to C. posadasii FKS. We cannot explain the change in the dose-response curve for the TLR4-deficient cells.

Response to C. posadasii is dependent on TLR2.

To investigate the role of TLR2 in response to spherules, we compared peritoneal macrophages from mice lacking TLR2 (TLR2^−/−) with cells from normal B6 mice. As a positive control, we included heat-killed Staphylococcus as a TLR2 ligand. LPS, which activates cells independently of TLR2, was also included. We found that the macrophages of TLR2^−/− mice secreted significantly less TNF-α and MIP-2 in response to all concentrations of FKS used than the wild type mice (Fig. 3). As expected, heat-killed S. aureus stimulated TNF-α and MIP-2 release only from the wild-type macrophages, and LPS stimulated similar amounts of TNF-α and MIP-2 from both cell types.

Macrophage cytokine responses to C. posadasii signal through MyD88.

We determined that the cellular signaling pathway required the adapter molecule MyD88 by using peritoneal macrophages from MyD88^−/− mice. We found that C. posadasii FKS, LPS, and heat-killed Staphylococcus stimulated significantly more MIP-2 and TNF-α release from the wild-type macrophages than from the MyD88^−/− macrophages (Fig. 4). TLR2 and TLR4 both signal through the recruitment of the cytoplasmic adaptor protein MyD88 (1). These data demonstrated that the majority of the response to C. posadasii FKS is via a MyD88-dependent signaling pathway.

The role of Dectin-1 in peritoneal macrophage-mediated cytokine production.

Dectin-1 is a receptor for β-glucans, which are major structural components of fungal cell walls (5). Dectin-1 mediates cellular response to β-glucans by collaboration with TLR-2 (11). Since we found that C. posadasii FKS elicited cytokine responses in macrophages in a TLR-2-dependent manner, we investigated whether the response was also dependent on Dectin-1. To study the role of Dectin-1, we transduced RAW 264.7 cells with a plasmid expressing Dectin-1 (RAW-Dectin). As a control, RAW cells were transduced with a control plasmid. As shown in Fig. 5A, RAW-Dectin macrophages produced significantly more TNF-α than RAW-pFB control macrophages after activation with C. posadasii FKS (left panel). While RAW-Dectin enhanced the TNF-α response to zymosan as expected, there was no significant increase in the amount of TNF-α secreted by RAW-Dectin and RAW-pFB after simulation with heat-killed S. aureus (right panel). We observed the same effect when assaying for MIP-2 responses (data not shown). Next we examined whether the formalin fixation of C. posadasii spherules had somehow altered the way spherules interacted with Dectin-1. To exclude this possibility, we stimulated RAW-Dectin and RAW-pFB with intact, live spherules and assayed for TNF-α and MIP-2 levels in the supernatant. Figure 5B shows that live spherules and FKS both activated RAW-pFB and RAW-Dectin to produce TNF-α and MIP-2. Overexpression of Dectin-1 greatly enhanced the cytokine production by RAW cells, whether the stimulus was live spherules or FKS.

Mouse peritoneal macrophages express high levels of Dectin-1 on the cell surface (27). We next investigated whether Dectin-1 is involved in FKS stimulation of mouse peritoneal macrophages. We used anti-Dectin-1 MAb, 2A11, which was previously shown to inhibit binding of zymosan to macrophages (5) and alveolar macrophage killing of Pneumocystis carinii (25). Figure 5C showed that MAb 2A11 at a concentration of 2 μg/ml significantly inhibited MIP-2, TNF-α, and IL-6 production from mouse peritoneal macrophages activated with FKS.

Dectin-1 mediated cellular activation by C. posadasii glucan.

We sought to identify the cellular components of C. posadasii that stimulated mouse macrophages in the Dectin-1-dependent manner. β-Glucan was purified from C. posadasii spherules. By weighing the final product, we determined that β-glucan constituted approximately 60% of C. posadasii dry weight. We activated RAW-Dectin macrophages overexpressing Dectin-1 with the purified β-glucan and found that a significantly higher level of TNF-α was produced from these cells than from RAW-pFB control macrophages (Fig. 6).