PPARγ controls Dectin-1 expression required for host antifungal defense against Candida albicans

Abstract

We recently showed that IL-13 or peroxisome proliferator activated receptor gamma (PPARgamma) ligands attenuate Candida albicans colonization of the gastrointestinal tract. Here, using a macrophage-specific Dectin-1 deficient mice model, we demonstrate that Dectin-1 is essential to control fungal gastrointestinal infection by PPARgamma ligands. We also show that the phagocytosis of yeast and the release of reactive oxygen intermediates in response to Candida albicans challenge are impaired in macrophages from Dectin-1 deficient mice treated with PPARgamma ligands or IL-13. Although the Mannose Receptor is not sufficient to trigger antifungal functions during the alternative activation of macrophages, our data establish the involvement of the Mannose Receptor in the initial recognition of non-opsonized Candida albicans by macrophages. We also demonstrate for the first time that the modulation of Dectin-1 expression by IL-13 involves the PPARgamma signaling pathway. These findings are consistent with a crucial role for PPARgamma in the alternative activation of macrophages by Th2 cytokines. Altogether these data suggest that PPARgamma ligands may be of therapeutic value in esophageal and gastrointestinal candidiasis in patients severely immunocompromised or with metabolic diseases in whom the prevalence of candidiasis is considerable.

Figure 1. Dectin-1 and the Mannose Receptor are implicated in antifungal functions of macrophages treated with IL-13 or PPARγ ligand.

Peritoneal macrophages were cultured with IL-13 (50 ng/mL) (A and B) or rosiglitazone (5 µM) (C and D). Mannan (mann) and/or soluble β-glucan (laminarin, lam) solutions were incubated at 4°C for 20 min until the phagocytosis and respiratory burst experiments. (A and C) The phagocytosis of non-opsonized C.albicans (ratio 1∶6) by macrophages was measured at 37°C after exposure to FITC-labeled C.albicans for 60 min. The amount of fluorescence was determined using a FACS based approach. The distinction between internalized yeast cells and those attached to macrophage surface was done via quenching the FITC-fluorescence with trypan blue. Data are expressed as percentage relative to untreated control macrophages and are means±SE of three separate experiments. (B and D) Non-opsonized C.albicans-induced respiratory burst of macrophages (ratio 1∶3) was measured by chimiluminescence. Total chemiluminescence emission (area under the curve expressed in counts x seconds) was observed continuously for 60 min in the presence or absence of non-opsonized C. albicans. The data are the means±SE of three separate experiments. ** (p<0.01) and * (p<0.05) indicates a significant difference compared with the untreated macrophages. ## (p<0.01) and # (p<0.05) indicates a significant difference compared with the treated control macrophages.

Figure 2. Characterization of Dectin-1 conditional knock-out macrophages.

(A) The Dectin-1 mRNA level of Dectin-1 control (Cre 0) and Dectin-1 knockout (Cre Tg) peritoneal macrophages was quantified by quantitative real-time RT-PCR. Values are representative of data obtained from three mice. (B) The surface protein level of Dectin-1 was measured by flow cytometry on the Dectin-1 control (Cre 0) and Dectin-1 knockout (Cre Tg) peritoneal macrophages. Plots are representative of data obtained from six mice. (C) The protein level of several receptors (CD11b, Mannose Receptor MR, TLR4, TLR2, SIGNR1, CD36) and markers (F4/80 and CD14) on the macrophage surface was determined by flow cytometry on the Dectin-1 control (Cre 0) and Dectin-1 knockout (Cre Tg) peritoneal macrophages. (D and E) Released [3H]arachidonic acid is expressed as the percentage of [3H]arachidonic acid in the culture medium divided by the total incorporated [3H]arachidonic acid in murine peritoneal macrophages. The release of [3H]arachidonic acid was determined after incubation for 120 min of peritoneal Dectin-1-control and Dectin-1-knockout macrophages with non-opsonized C.albicans (ratio 1∶3) or PMA (100 nM). The data represent the means±SE of three separate experiments. (F) TNFα production by Dectin-1 control (Cre 0) and Dectin-1 knockout (Cre Tg) macrophages after 24 h of stimulation with heat-killed C.albicans (ratio 1∶3), ZNO (2 µg/mL) or LPS (100 ng/mL). Data are the means±SE of three separate experiments. ** (p<0.01) and * (p<0.05) indicates a significant difference compared with the respective Cre 0 or Cre Tg control. § (p<0.05) indicates a significant difference between Cre 0 and Cre Tg.

Figure 3. Dectin-1 and the Mannose Receptor are required in different stages of C.albicans clearance.

Dectin-1 control (Cre 0) and Dectin-1 knockout (Cre Tg) peritoneal macrophages were cultured with IL-13 (50 ng/mL) (A–C) or with rosiglitazone (RZ) (5 µM) (D–F). Mannan (mann) solution was incubated at 4°C for 20 min until the binding, phagocytosis and respiratory burst experiments. (A and D) The binding experiment of non-opsonized C.albicans by macrophages was measured at 4°C after challenge with FITC-labeled C.albicans for 20 min (ratio 1∶6). The amount of fluorescence was determined using a FACS based approach. Data are expressed as the percentage relative to the untreated Dectin-1 control (Cre 0) macrophages and are the means±SE of three separate experiments. (B and E) The phagocytosis of non-opsonized C.albicans by macrophages was measured at 37°C after challenge with FITC-labeled C.albicans for 60 min (ratio 1∶6). The amount of fluorescence was determined using a FACS based approach. The distinction between internalized yeast cells and those attached to macrophage surface was done via quenching the FITC-fluorescence with trypan blue. Data are expressed as the percentage relative to the untreated Dectin-1 control (Cre 0) macrophages and are the means±SE of three separate experiments. (C and F) The respiratory burst of the Dectin-1 control (Cre 0) and Dectin-1 knockout (Cre Tg) macrophages induced by non-opsonized C.albicans was measured by chimiluminescence (ratio 1∶3). Total chemiluminescence emission (area under the curve expressed in counts x seconds) was observed continuously for 60 min. Data are the means±SE of three separate experiments. ** (p<0.01) and * (p<0.05) indicates a significant difference compared with the Cre 0 untreated macrophages. ## (p<0.01) indicates a significant difference compared with the Cre Tg untreated macrophages. § (p<0.05) indicates a significant difference between Cre 0 and Cre Tg.

Figure 4. Dectin-1 expression depends on PPARγ activation by IL-13 or PPARγ-specific ligands.

(A) The surface protein level of Dectin-1 on peritoneal macrophages was measured by flow cytometry after treatment with IL-13 (50 ng/mL), rosiglitazone (RZ) (5 µM), 15d-PGJ2 (1 µM), MCC555 (5 µM) or GW1929 (1 µM). The changes in Dectin-1 induction were normalized to the untreated control value. Data are the means±SE of three separate experiments. (B) Representative FACS profiles of Dectin-1 (filled histograms) and isotype control labeling (unfilled histograms) in treated macrophages. Representative Dectin-1 FACS profiles of untreated macrophages (unfilled histograms) and treated (filled histograms) macrophages. (C) The mRNA level of Dectin-1 on peritoneal macrophages was quantified by quantitative real-time RT-PCR after treatment with IL-13 (50 ng/mL), rosiglitazone (RZ) (5 µM) or 15d-PGJ2 (1 µM). Data are the means±SE of three separate experiments. ** (p<0.01) and * (p<0.05) indicates a significant difference compared with the untreated macrophages. (D) The protein level of Dectin-1 on the murine cell line RAW264.7 transiently transfected with pCMV-luciferase (CMV-luc) or with pCMV-mPPARγ (CMV-PPARγ) and after treatment with IL-13 or rosiglitazone (RZ). Representative Dectin-1 FACS profiles of untreated (unfilled histograms) and treated (filled histograms) macrophages were obtained by flow cytometry. The changes in Dectin-1 induction were normalized to the untreated RAW 264.7 cells transfected with pCMV-luc. Data are the means±SE of three separate experiments. ** (p<0.01) and * (p<0.05) indicates a significant difference compared with the untreated RAW 264.7 cells transfected with pCMV-luc.

Figure 5. PPARγ inhibition in M2 polarized macrophages abolishes the increase of Dectin-1.

(A) The protein level of Dectin-1 on peritoneal macrophages was measured by flow cytometry after treatment with IL-13 (50 ng/mL) or rosiglitazone (RZ) (5 µM) in the presence of the PPARγ antagonists (GW9662 (5 µM) and T007 (2 µM)). Data are the means±SE of three separate experiments. (B) Dectin-1 mRNA level of peritoneal macrophages was quantified by quantitative real-time RT-PCR after treatment with IL-13 (50 ng/mL) or rosiglitazone (RZ) (5 µM) in the presence of the PPARγ antagonist (GW9662 (5 µM)). Data are the means±SE of three separate experiments. ** (p<0.01) and * (p<0.05) indicates a significant difference compared with the untreated macrophages. (C) The surface protein level of Dectin-1 on peritoneal macrophages transfected with siRNA targeting PPARγ (PPARγ siRNA) or control siRNA (control siRNA) and stimulated by IL-13. Representative Dectin-1 FACS profiles of untreated (unfilled histograms) and treated (filled histograms) macrophages were obtained by flow cytometry. The changes in Dectin-1 receptor levels were normalized to the untreated macrophages transfected with the siRNA control. Data are the means±SE of three separate experiments. ** (p<0.01) and * (p<0.05) indicates significant difference compared with the untreated macrophages transfected with the siRNA control.

Figure 6. cPLA2 is involved in Dectin-1 induction by IL-13.

(A) The Dectin-1 mRNA level of peritoneal macrophages was measured by quantitative real-time RT-PCR after treatment of peritoneal macrophages by an irreversible cPLA2 antagonist (MAFP) and by IL-13. Data are the means±SE of three separate experiments. (B) The protein level of Dectin-1 was measured by flow cytometry on peritoneal macrophages after treatment with MAFP (10 µM and 20 µM) and with IL-13 (50 ng/mL). Data are the means±SE of three separate experiments. (C) The protein level of Dectin-1 was measured by flow cytometry on peritoneal macrophages after treatment with MAFP and IL-13 and/or 15d-PGJ2 (1 µM). Data are the means±SE of three separate experiments. ** (p<0.01) and * (p<0.05) indicates a significant difference compared with the IL-13-treated macrophages.

Figure 7. Dectin-1-knockout mice are more susceptible than Dectin-1-wildtype mice to C. albicans gastrointestinal infection.

(A and B) Quantification of C. albicans fungal burden in the gastrointestinal tract (stomach and cecum) of Dectin-1-control mice Cre 0 (filled circles) and Dectin-1-knockout mice Cre Tg (open circles) at 5 day after oral infection with 5.10⁶ CFU (A, n = 6) or with 5.10⁷ CFU (B, n = 4) in standard conditions or after treatment with rosiglitazone (RZ) (2.8 µg/g of mouse). Each symbol represents an individual mouse. § (p<0.05) indicates a significant difference between group of mice. (C) Phagocytosis and ROS production were measured on peritoneal macrophages from Dectin-1 knockout (Cre Tg) mice at 5 day after oral infection with 5.10⁶ CFU in standard conditions or after treatment with rosiglitazone (RZ). The phagocytosis of non-opsonized C.albicans by macrophages was measured at 37°C after exposure to FITC-labeled C.albicans for 60 min (ratio 1∶6). The amount of fluorescence was determined using a FACS based approach. The distinction between internalized yeast cells and those attached to macrophage surface was done via quenching the FITC-fluorescence with trypan blue. Data are expressed as the percentage relative to untreated Dectin-1 control (Cre 0) macrophages and are the means±SE (n = 6). The respiratory burst of macrophages induced by non-opsonized zymosan (ZNO) (2 µg/mL) was measured by chimiluminescence. Total chemiluminescence emission (area under the curve expressed in counts x seconds) was observed continuously for 60 min. Data are the means±SE (n = 6). ** (p<0.01) indicates a significant difference compared with the Cre 0 untreated macrophages. § (p<0.05) indicates a significant difference between Cre 0 and Cre Tg. (D) The MR surface protein level was measured by flow cytometry on peritoneal macrophages from Dectin-1 control (Cre 0) or Dectin-1 knockout (Cre Tg) mice at day 5 after oral infection with 5.10⁷ CFU in standard conditions or after treatment with rosiglitazone (RZ). Data are the means±SE (n = 4). ** (p<0.01) indicates a significant difference compared with the Cre 0 control. § (p<0.05) indicates a significant difference between Cre 0 and Cre Tg.