Volume 18, Issue 2, Pages 198-209 (August 2015) Pathogenic NLRP3 Inflammasome Activity during Candida Infection Is Negatively Regulated by IL-22 via Activation of NLRC4 and IL-1Ra  Monica Borghi, Antonella De Luca, Matteo Puccetti, Martin Jaeger, Antonella Mencacci, Vasilis Oikonomou, Marilena Pariano, Cecilia Garlanda, Silvia Moretti, Andrea Bartoli, Jack Sobel, Frank L. van de Veerdonk, Charles A. Dinarello, Mihai G. Netea, Luigina Romani  Cell Host & Microbe  Volume 18, Issue 2, Pages 198-209 (August 2015) DOI: 10.1016/j.chom.2015.07.004 Copyright © 2015 Elsevier Inc. Terms and Conditions

Cell Host & Microbe 2015 18, 198-209DOI: (10.1016/j.chom.2015.07.004) Copyright © 2015 Elsevier Inc. Terms and Conditions

Figure 1 Inflammasome Activity in Vaginal Candidiasis C57BL/6 mice (n = 6) were intravaginally inoculated with 5 × 106 C. albicans blastoconidia. (A) Caspase-1 activation by immunoblotting of vaginal lysates with specific antibodies. Normalization was performed by probing the membrane with mouse-monoclonal anti-β-actin antibody. (B) Pyropoptosis was quantified as LDH activity (% cytotoxicity). (C and D) (C) Cytokine levels (pg/mg, cytokine/total proteins for each sample) and (D) cytokine gene expression (real-time RT-PCR) at different days postinfection (dpi). (E) Periodic acid-Schiff-staining, immunohistochemistry (scale bars, 100 μm), and immunofluorescence staining (scale bars, 25 μm) of vaginal sections with antibodies to NLRP3, NLRC4, and p (phospho) NLRC4. Sections were stained with the relevant primary antibody overnight at 4°C followed by secondary biotinylated or fluorescent antibodies. Images were acquired with a 40× objective and the analySIS image processing software. Insets, 100× objective. 4′-6-Diamino-2-phenylindole was used to counterstain tissues and to detect nuclei. (F and G) (F) Gene and (G) protein analysis (RT-PCR and western blotting, respectively) on ex vivo vaginas. Data (mean ± SD) are pooled or representative (histology/immunostaining/western blotting) from four independent experiments. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, infected versus uninfected (dpi 0) mice. See also Figure S1. Cell Host & Microbe 2015 18, 198-209DOI: (10.1016/j.chom.2015.07.004) Copyright © 2015 Elsevier Inc. Terms and Conditions

Figure 2 NLRP3 and NLRC4 Have Distinct Roles in VVC (A) Nlrp3–/– or Nlrc4–/– mice (n = 6) were intravaginally inoculated with 5 × 106 C. albicans blastoconidia and assessed for vaginal fungal burden (Log10 CFU/100μl VF ± SD). (B) PMN recruitment in VF (identified by nuclear morphology and enumerated per field at 100× magnification). (C) Vaginal pathology (periodic acid-Schiff-staining of sections; scale bars, 100 μm) and immunofluorescence staining with anti-NLRP3 or anti-pNLRC4 antibodies (scale bars, 25 μm). Sections were stained with the relevant primary antibody overnight at 4°C followed by secondary FITC or TRITC antibody. Images were acquired using a fluorescence microscope with a 40× objective and the analySIS image processing software. 4′-6-Diamino-2-phenylindole was used to counterstain tissues and to detect nuclei. (D) Caspase-1 activation, by immunoblotting of vaginal lysates with specific antibodies. (E) LDH activity (% cytotoxicity). (F and G) (F) IL-1β, (G) IL-17A, and IL-17F (pg/mg, cytokine/total proteins for each sample) in VF. (H) NLRP3 and pNLRC4 protein expression (western blotting) on ex vivo vaginas from Nlrp3–/– or Nlrc4–/– mice. Note the negative expression of NLRP3 and pNLRC4 in Nlrp3–/– or Nlrc4–/– mice, respectively). Results represent mean cytokine levels from samples pooled from three experiments (n = 4–6 total samples per group). (I–K) (I) PMNs quantification (by nuclear morphology) in VF, (J) vaginal pathology (periodic acid-Schiff-staining of sections; scale bars, 100 μm; red arrows indicate fungi, and yellow arrows indicate neutrophils), and (K) fungal burden in bone marrow chimera mice after vaginal infection with C. albicans. Data (mean ± SD) are pooled or representative (histology/immunostaining/western blotting) from four independent experiments. ∗p < 0.05, chimeric Nlrp3–/– or Nlrc4–/– mice versus C57BL/6→C57BL/6 mice. See also Figures S2 and S3. Cell Host & Microbe 2015 18, 198-209DOI: (10.1016/j.chom.2015.07.004) Copyright © 2015 Elsevier Inc. Terms and Conditions

Figure 3 NLRP3 Contributes to Pathogenic Inflammation in VVC (A) Tir8–/– or Il1r1–/– mice (n = 6) were intravaginally inoculated with 5 × 106 C. albicans blastoconidia and assessed for vaginal fungal burden (Log10 CFU/100μl VF ± SD). (B) PMNs quantification (by nuclear morphology) in VF. (C) Vaginal pathology (periodic acid-Schiff-staining of sections; scale bars, 100 μm) and immunofluorescence staining with NLRP3 and pNLRC4 antibodies (scale bars, 25 μm). Sections were stained with the relevant primary antibody overnight at 4°C followed by secondary FITC or TRITC antibody. Images were acquired using a fluorescence microscope with a 40× objective and the analySIS image processing software. 4′-6-Diamino-2-phenylindole was used to counterstain and to detect nuclei. (D) Caspase-1 activation by immunoblotting of vaginal lysates with anti-murine antibodies. (E) IL-1β levels (pg/mg, cytokine/total proteins for each sample, mean ± SD) in the vaginal fluids. (F and G) (F) Gene and (G) protein expression analysis (RT-PCR and western blotting, respectively) on ex vivo vaginas. Data (mean ± SD) are pooled or representative (histology/immunostaining/western blotting) from four independent experiments. ∗p < 0.05, ∗∗p < 0.01, Tir8–/– or Il1r1–/– mice versus WT mice. Cell Host & Microbe 2015 18, 198-209DOI: (10.1016/j.chom.2015.07.004) Copyright © 2015 Elsevier Inc. Terms and Conditions

Figure 4 IL-22 Activates Epithelial NLRC4 in Response to Candida Il22–/– mice (n = 6) were intravaginally inoculated with 5 × 106 C. albicans blastoconidia. (A) IL-1β levels (pg/mg, cytokine/total proteins for each sample) in VF. (B) Caspase-1 activation by immunoblotting of vaginal lysates with anti-murine antibodies. (C) Vaginal pathology (periodic acid-Schiff-staining of sections; scale bars, 100 μm) and immunofluorescence staining with NLRP3 and pNLRC4 antibodies followed by secondary FITC or TRITC antibody (scale bars, 25 μm). (D) Western blotting of NLRP3 and pNLRC4 on ex vivo vaginas. (E and F) PKCδ activation (immunoblotting with anti-PKC delta [phospho T505] antibody) in ex vivo vaginal cultures exposed to flagellin (E) or LPS, live germinated Candida (CA) cells (F) in the presence of IL-22 for 4 hr and/or the PKCδ inhibitor rottlerin. Untreated, flagellin only. (G) Nos2 expression (RT-PCR) in vaginal cells. (H) IL-1Ra levels (pg/mg, cytokine/total proteins, mean ± SD) in vaginal culture supernatants. (I) NLRP3 and pNLRC4 expression (immunofluorescence staining), Lcn2 expression (RT-PCR), and LDH activity (% cytotoxicity) (7 dpi) in the vaginas of infected C57BL/6 mice intravaginally treated with IL-22. Control mice received PBS. ∗p < 0.05, Il22–/– mice versus WT mice, and treated versus untreated. (J–L) (J) Immunoblotting of PKCδ in vaginal lysates, (K) vaginal pathology (periodic acid-Schiff-staining of sections; scale bars, 100 μm), and immunofluorescence staining with NLRP3 and pNLRC4 antibodies (scale bars, 25 μm) and (L) PMNs quantification (by nuclear morphology) in VF of infected mice treated with rottlerin. Control mice received PBS. Data (mean ± SD) are pooled or representative (histology/immunostaining/western blotting) from three independent experiments. ∗p < 0.05, treated versus untreated mice. Cell Host & Microbe 2015 18, 198-209DOI: (10.1016/j.chom.2015.07.004) Copyright © 2015 Elsevier Inc. Terms and Conditions

Figure 5 NLRC4 Dampens Pathogenic NLRP3 Activity via IL-1Ra (A) IL-1Ra levels (pg/mg, cytokine/total proteins for each sample) in the VF of mice (n = 6) intravaginally infected with 5 × 106 C. albicans blastoconidia. Results represent mean cytokine levels (±SD) from samples pooled from three experiments (n = 4–6 total samples per group). (B) Vaginal pathology (periodic acid-Schiff-staining of sections; scale bars, 100 μm). (C–G) (C) Vaginal fungal burden (Log10 CFU/100μl VF ± SD), (D) PMN recruitment in VF (identified by nuclear morphology and enumerated per field at 100× magnification), (E) caspase-1 activation by immunoblotting, (F) LDH activity (% cytotoxicity), and (G) immunofluorescence staining of vaginas with NLRP3 and pNLRC4 antibodies (scale bars, 25 μm) in Il1ra–/– mice (n = 6) intravaginally inoculated with 5 × 106 C. albicans blastoconidia and treated with 10 mg/kg anakinra intraperitoneally daily for 7 days, starting the day of the infection. Assays were done at 7 dpi. Data (mean ± SD) are pooled or representative (histology/immunostaining/western blotting) from three independent experiments. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, Nlrp3–/–, Il1r1–/–, Il22–/–, Nlrc4–/–, and Tir8–/– mice versus WT mice and treated versus untreated. Cell Host & Microbe 2015 18, 198-209DOI: (10.1016/j.chom.2015.07.004) Copyright © 2015 Elsevier Inc. Terms and Conditions

Figure 6 IL-22 Activates NLRC4 in Response to Candida in Human Vaginal Cells (A) HeLa cells were exposed to live germinated C. albicans cells (magnified in the insets), flagellin (Flag), or nigericin + LPS (Nig) in the presence or not of human recombinant IL-22 and assessed for immunofluorescence staining with anti-NLRP3 or anti-pNLRC4 antibodies. (B–D) (B and C) PKCδ activation (immunoblotting with anti-PKC delta [phospho-T505] antibody) and (D) pNLRC4 expression (immunoblotting with anti-pNLRC4 antibody) of cells treated as above and silenced for PKCδ. (E) Cytokine (pg/mg, cytokine/total protein) production in the vaginal fluids of healthy women or patients with RVVC. Phalloidine staining highlights the NLRP3 cytoplasmic localization. ∗∗p < 0.01; ∗p < 0.05, controls versus RVVC. See also Figure S4. Cell Host & Microbe 2015 18, 198-209DOI: (10.1016/j.chom.2015.07.004) Copyright © 2015 Elsevier Inc. Terms and Conditions

Figure 7 Proposed Model for the Role of the IL-22/NLRC4/IL-1Ra Axis in Mucosal Candidiasis Activation of NLRP3 in response to the fungus and/or DAMPs contributes to neutrophil recruitment and inflammation, an activity that is restrained by IL-22. Produced via AhR, IL-22 activates NLRC4 on epithelial cells via PKCδ phosphorylation, which results, among other functions, in a sustained production of the IL-1 receptor antagonist, IL-1Ra, that restrains NLRP3 activity. Thus, the IL-22/NLRC4/IL-1Ra axis may restrain pathogenic inflammasome activity to the fungus at mucosal surfaces, thus offering insights into molecular mechanisms regulating commensalism and infectivity by the fungus. AhR, aryl hydrocarbon receptor. Cell Host & Microbe 2015 18, 198-209DOI: (10.1016/j.chom.2015.07.004) Copyright © 2015 Elsevier Inc. Terms and Conditions