Critical role of MDA5 in the interferon response induced by human metapneumovirus infection in dendritic cells and in vivo

Abstract

Human metapneumovirus (hMPV) is a respiratory paramyxovirus of global clinical relevance. Despite the substantial knowledge generated during the last 10 years about hMPV infection, information regarding the activation of the immune response against this virus remains largely unknown. In this study, we demonstrated that the helicase melanoma differentiation-associated gene 5 (MDA5) is essential to induce the interferon response after hMPV infection in human and mouse dendritic cells as well as in an experimental mouse model of infection. Our findings in vitro and in vivo showed that MDA5 is required for the expression and activation of interferon (IFN) regulatory factors (IRFs). hMPV infection induces activation of IRF-3, and it regulates the expression of IRF-7. However, both IRF-3 and IRF-7 are critical for the production of type I and type III IFNs. In addition, our in vivo studies in hMPV-infected mice indicated that MDA5 alters viral clearance, enhances disease severity and pulmonary inflammation, and regulates the production of cytokines and chemokines in response to hMPV. These findings are relevant for a better understanding of the pathogenesis of hMPV infection.

Fig 1

Expression of MDA5 regulates IFN production in moDC infected with hMPV. Human moDC were infected with hMPV at an MOI of 5 or left uninfected in growth medium only (U). (A and B) After 3, 6, 9, 12, and 24 h of infection, MDA5 and RIG-I mRNA expression was determined by real-time qRT-PCR (A) and Western blotting (n = 3) (B). (C) Prior to hMPV infection, MDA5 and RIG-I expression was knocked down with specific siRNAs. The efficiency of gene silencing was determined by real-time qRT-PCR (n = 12). The concentrations of IFN-α, -β, -ω, -γ, and -λ were determined after 24 h of hMPV infection by VeriPlex ELISA (n = 12). (D) The same samples were analyzed by real-time RT-PCR for the expression of ISG56, ISG54, ISG60, and MX1 (n = 12). Bar graphs represent means ± standard errors of the means. *, P < 0.05; **, P < 0.01.

Fig 2

Characterization and activation of BMDC after hMPV infection. BMDC were differentiated from marrow cells cultured in the presence of murine GM-CSF and IL-4 for 6 days as described in Materials and Methods. (A) Dot plots show recovered BMDC stained with anti-CD11c or isotype-matched antibodies. (B) Morphology of BMDC recovered. Bar, 30 μm. (C) BMDC from MDA5^−/− or WT mice were infected with hMPV at an MOI of 3 for 24 h. Two-color flow cytometry analysis was applied using anti-CD11c in combination with anti-anti-I-A/I-E (MHC-II), anti-CD40, anti-CD80, or anti-CD86 antibodies. Gated CD11c⁺ cells were analyzed for the expression of each additional surface molecule. Isotype control (shaded histograms), uninfected BMDC (dotted histograms), and BMDC infected with hMPV (bold lines) are shown. A representative experiment from two similar experiments is shown.

Fig 3

Lack of MDA5 alters the production of IFN in response to hMPV infection in mouse DC. (A) BMDC were differentiated from marrow cells cultured in the presence of murine GM-CSF and IL-4 for 6 days. BMDC from MDA5^−/− or WT mice were infected with hMPV at an MOI of 3 for 24 h, and production of IFN-α and IFN-β was determined by ELISA (n = 3). (B) Lung conventional DC were isolated by collagenase digestion, and IFN-β production was determined by ELISA. Bar graphs represent means ± standard errors of the means. *, P < 0.05.

Fig 4

Expression and activation of IRF-3 and IRF-7 in moDC infected with hMPV. (A) Human moDC were infected with hMPV at an MOI of 5. After 3, 6, 9, 12, 24, and 48 h of hMPV infection, the expression of IRF-3, IRF-5, and IRF-7 was determined by real-time qRT-PCR (n = 3). (B) MDA5 expression was knocked down in moDC with specific siRNA prior to hMPV infection. Scrambled siRNA was used as a control. Expression of IRF-7 was determined by real-time qRT-PCR (n = 8). (C) moDC were treated with siRNA IRF-7 or scrambled siRNA for 24 h and infected with hMPV at an MOI of 5 for additional 24 h. Production of IFN-α and IFN-β was measured by ELISA (n = 3). (D) IRF-3 activation was measured by DN binding activity in hMPV-infected moDC (n = 3). Bar graphs represent means ± standard errors of the means. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Fig 5

IRF-3 and IRF-7 regulate the IFN response in hMPV infection. (A) BMDC from WT, IRF3^−/−, and IRF-7^−/− mice were infected with hMPV at an MOI of 3 for 24 h. IFN-α and IFN-β production were measured by ELISA (n = 4). (B) BMDC from WT and MDA5^−/− mice were infected with hMPV for 24 h. The expression levels of IRF-3, IRF-5, and IRF-7 were measured by real-time qRT-PCR (n = 3). (C) After hMPV infection, activation of IRF-3 was measured in a DNA binding assay in BMDC from WT and MDA5^−/− mice (n = 3). (D) BMDC from WT, IRF3^−/−, and IRF-7^−/− mice were infected at an MOI of 3, and the expression levels of IRF-7 and IRF-3 were assessed by real-time qRT-PCR (n = 4). Bar graphs represent means ± standard errors of the means. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Fig 6

MDA5 expression is critical for the IFN response and viral clearance in vivo. C57BL/6J mice were inoculated with PBS (mock) or infected i.n. with 1 × 10⁷ PFU of hMPV. Mice were sacrificed at the indicated day after infection, and lung tissue was collected. (A and B) MDA5 expression (A) or IFN-β expression (B) levels were determined by real-time qRT-PCR (n = 3 to 4 mice/group). (C to E) In a separate set of experiments, WT and MDA5^−/− mice were treated as described above, and BAL samples were collected at 24 h after hMPV infection. ELISAs were performed to determine the concentration of (C) IFN-α; (D) IFN-β; and (E) IFN-λ (IL-28A/IL-28B). Data are representative of three independent experiments. n = 4–6 mice/group. (F) Lung tissue samples from WT and MDA5^−/− mice were collected at days 5, 7, and 10 after hMPV infection. Expression of the hMPV N gene was determined by real-time qRT-PCR. Data are representative of two independent experiments with similar results (n = 3 to 6 mice/group). The bar graphs represent means ± standard errors of the means. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Fig 7

MDA5 regulates disease severity and pulmonary inflammation in hMPV-infected mice. WT and MDA5^−/− mice were infected i.n. with 1 × 10⁷ PFU of hMPV. (A) Mice were monitored daily, and body weight change was calculated based on the original weight before the infection. (B) Lungs were harvested at day 7 after hMPV infection, fixed for slide preparation, and H&E stained. Representative stained lung tissue sections from the indicated treatment are shown. Arrows indicate cells infiltrating the perivascular and peribronchial spaces. Bar, 200 μm. (C) BAL samples were collected from each group of mice at day 7 after infection (same as lung tissue) and assessed for cytokine/chemokine production by using a multiplex cytokine detection system. Data are representative of three independent experiments with similar results (n = 6 to 9 mice/group). Bar graphs represent means ± standard errors of the means. *, P < 0.05; **, P < 0.01.