Receptor interacting protein kinase 2-mediated mitophagy regulates inflammasome activation during virus infection

Abstract

NOD2 receptor and the cytosolic protein kinase RIPK2 regulate NF-κB and MAP kinase signaling during bacterial infections, but the role of this immune axis during viral infections has not been addressed. We demonstrate that Nod2(-/-) and Ripk2(-/-) mice are hypersusceptible to infection with influenza A virus. Ripk2(-/-) cells exhibited defective autophagy of mitochondria (mitophagy), leading to enhanced mitochondrial production of superoxide and accumulation of damaged mitochondria, which resulted in greater activation of the NLRP3 inflammasome and production of IL-18. RIPK2 regulated mitophagy in a kinase-dependent manner by phosphorylating the mitophagy inducer ULK1. Accordingly, Ulk1(-/-) cells exhibited enhanced mitochondrial production of superoxide and activation of caspase-1. These results demonstrate a role for NOD2-RIPK2 signaling in protection against virally triggered immunopathology by negatively regulating activation of the NLRP3 inflammasome and production of IL-18 via ULK1-dependent mitophagy.

Figure 1. RIPK2 deficiency leads to hyper-inflammation

(a) Mice were infected intra-nasally (i.n.) with 750 PFU of the PR8 virus and examined daily for survival and clinical features of disease. (b) H&E-stained lung sections from mice infected with PR8 virus on day 2. Cellular infiltrates and necrotic debris occluding the airways are indicated with arrows. (c) H&E-stained lung sections were prepared on day 2 post-infection and scored for disease severity and necrosis. (d) The extent of cell death on day 7 was estimated visually on H&E-stained lung sections (necrosis score), and by flow cytometry for annexin V-stained lung cell populations. (e) Immunohistochemistry (IHC) staining of neutrophils occluding the airways of PR8 infected Ripk2^−/− mice. (f-g) Flow cytometric analysis of neutrophil, DC and macrophage cell numbers on days 2 and 7. (h) Pulmonary viral titers on days 2 and 7 after infection in Ripk2^−/− and WT mice. a,f,g,h: Data are cumulative from 3 independent experiments (g-h: n=10–12). b-e: Data are representative of 2-3 independent experiments n=3–7 per group/experiment (mean +/−SEM, *p<0.05, **p<0.01, ***p<0.001).

Figure 2. RIPK2 modulates cytokine and chemokine production

Supernatants of whole lung homogenates taken on d2 and d7 after infection with the PR8 virus were analyzed for IL-6 and TNF (a), IL-18 and IFN-γ (b), and a set of chemokines (c). Data were normalized for total protein from lung homogenates (pg/mg total protein (t.p.)). Data are cumulative from 3 independent experiments, n=10–12 mice. (mean +/−SEM, *p<0.05, **p<0.01).

Figure 3. IL-18 mediates hyper-inflammation in Ripk2^−/− mice

Ripk2^−/− mice were injected i.p. with anti-IL-18 neutralizing antiserum (αIL-18Ab) 3h prior to infection with the PR8 virus. Mice were examined on d2 for IL-18, IFN-γ, and CCL2 levels in whole lung homogenates (a), and for neutrophil numbers by IHC (b) and by flow cytometry (c). (d) Analysis of survival of infected Ripk2^−/− mice that were treated with IL-18 neutralizing antibodies. (e) IL-18 was deleted from Ripk2^−/− mice (Ripk2^−/−Il18^−/−mice) and neutrophilia examined on d2 following PR8 infection. (f) Survival of Ripk2^−/−Il18^−/− mice compared to Ripk2^−/− or WT mice. Data are representative of 2 independent experiments, n=6–8 per group/experiment. (mean +/−SEM, *p<0.05, **p<0.01, ***p<0.001). (t.p.=total protein)

Figure 4. Both hematopoietic and lung epithelial cells contribute to hypercytokinemia

RIPK2 bone marrow chimeras were analyzed at d2 after PR8 infection for contribution to elevated IL-18 production by using (a) IL-18 IHC and (b) IL-18 ELISA of whole lung homogenates. (c) Flow cytometric analysis of NK cell counts in Ripk2^−/− and WT mice. (d) Analysis of WT and Ripk2^−/− NK cells for IFN-γ production on a per cell basis. (e) Accumulation of T and B cells in lungs. (f) Analysis of CD8⁺ T cells for IFN-γ and CD44 expression in WT and Ripk2^−/− mice. Data are representative of 2–3 independent experiments, n=6–8 per group/experiment. (mean +/−SEM, *p<0.05, **p<0.01). For (a), blue arrows denote IL-18⁺ macrophages, and red arrows denote IL-18⁺ bronchial epithelium.

Figure 5. Elevated IL-18 in Ripk2^−/− cells is NLRP3 inflammasome dependent

(a) Increased activation of caspase-1 (casp-1) in Ripk2^−/− and Nod2^−/− BMDCs following PR8 infection was analyzed by probing for active casp-1 p20 in immunoblots compared to GAPDH loading controls. (b) IL-18 release in PR8-infected Ripk2^−/− and Nod2^−/− BMDCs. (c, d) BMDCs infected with 10 MOI Listeria monocytogenes for 4h were analyzed for casp-1 activation by immunoblot and IL-18 production. (e, f) IL-18 and casp-1 activation in Ripk2^−/− BMDCs treated with the NLRP3 inflammasome specific inhibitor glyburide during PR8 infection. (g) Immunoblot for casp-1 p20 during Listeria infection of WT and Aim2^−/− BMDCs. Data are representative of 3–6 independent experiments with n=3 per experiment. (mean +/−SEM; *p<0.01, **p<0.001).

Figure 6. RIPK2 modulates inflammasome activation through autophagy

(a) Ripk2^−/− and Nod2^−/− BMDCs were analyzed by LC3-II imunoblotting as a marker for autophagy after 18h PR8 infection. (b, c) LC3-II⁺ puncta were counted in at least 100 cells from 5 random fields from confocal images of BMDCs following PR8 infection and/or chloroquine treatment. (d, e) Immunoblots of LC3-II and casp-1 and IL-18 production were determined in WT and Ripk2^−/−BMDCs following treatment with rapamycin (Rap). (f) LC3-II levels in the lungs of PR8 infected WT and Ripk2^−/− mice. (g, h) Lung neutrophil infiltration, IL-18 and IFN-γ levels in IAV infected Ripk2^−/− mice treated with rapamycin. (a–e) Data are representative of 2–6 independent experiments with n=2–3 per experiment. (f–h) Data are representative of 2 independent experiments for a total n=10–14 (mean +/−SEM; *p<0.05, **p<0.01, ***p<0.001).

Figure 7. RIPK2 specifically regulates mitophagy and accumulation of damaged mitochondria to modulate inflammasome activation

(a, b) Staining of BMDCs with the mitochondrial superoxide (SOX) specific stain MitoSOX during IAV or Listeria infection. (c) Staining of WT and Ripk2^−/− BMDCs with the general mitochondrial stain MitoTracker Green during IAV infection. (d, e) Analysis of IL-18 production and casp-1 activation in IAV-infected BMDCs that were pre-treated with the ROS inhibitor N-acetyl cysteine (NAC). (f) Analysis of casp-1 p20 in cell lysates of IAV-infected BMDCs pre-treated with the mitochondrial SOX inhibitor mitoTEMPO. (g) Confocal microscopy was used to analyze BMDCs stained for mitochondria (red) and LC3-II (green) for colocalization as an indicator of mitophagy. (a–f) Data are representative of 3–9 independent experiments with n=2–3 per experiment. (g) Data are representative of 2 independent experiments (mean +/−SEM; *p<0.01).

Figure 8. RIPK2 regulates activation of the critical autophagy protein ULK1 in response to IAV infection

(a, b) Casp-1 p20 and LC3-II immunoblots and IL-18 levels in PR8-infected WT BMDCs treated with the p38 and RIPK2 kinase inhibitor SB203580. (c) Immunoblots of phosphorylation of ULK1 at Ser555 in WT and Ripk2^−/− BMDCs following PR8 infection. (d) Casp-1 activation in untreated or NAC treated WT and Ulk1^−/− BMDCs following PR8 infection. (e) MitoSOX staining by flow cytometry in PR8-infected Ulk1^−/− BMDCs. (f) MitoTracker Green staining of PR8-infected Ulk1^−/− BMDCs. (g) Schematic diagram of the NOD2/RIPK2 dependent mechanism for mitophagy and subsequent NLRP3 inflammasome regulation. (a–f) Data are representative of 2–3 independent experiments with total n=4–6. (mean +/−SEM; *p<0.05).