Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2017 Sep 15;36(18):2742-2757.
doi: 10.15252/embj.201696273. Epub 2017 Aug 29.

The innate immune receptor MDA5 limits rotavirus infection but promotes cell death and pancreatic inflammation

Affiliations

The innate immune receptor MDA5 limits rotavirus infection but promotes cell death and pancreatic inflammation

Yu Dou et al. EMBO J. .

Abstract

Melanoma differentiation-associated protein 5 (MDA5) mediates the innate immune response to viral infection. Polymorphisms in IFIH1, the gene coding for MDA5, correlate with the risk of developing type 1 diabetes (T1D). Here, we demonstrate that MDA5 is crucial for the immune response to enteric rotavirus infection, a proposed etiological agent for T1D. MDA5 variants encoded by minor IFIH1 alleles associated with lower T1D risk exhibit reduced activity against rotavirus infection. We find that MDA5 activity limits rotavirus infection not only through the induction of antiviral interferons and pro-inflammatory cytokines, but also by promoting cell death. Importantly, this MDA5-dependent antiviral response is specific to the pancreas of rotavirus-infected mice, similar to the autoimmunity associated with T1D. These findings imply that MDA5-induced cell death and inflammation in the pancreas facilitate progression to autoimmune destruction of pancreatic β-cells.

Keywords: MDA5; inflammation; innate immunity; interferons; type 1 diabetes.

PubMed Disclaimer

Figures

Figure 1
Figure 1. Mda5 is crucial to limit RV infection in the pancreas
  1. A

    Micrographs of WT, Ifnar1 −/−, and Ifih1 −/− MEFs infected with RV for 24 h, then assessed by immunofluorescence. The viral capsid protein is detected with an anti‐SA11 antibody (red), actin filaments are visualized with CF647 phalloidin (green), and nuclei are visualized with Hoechst stains (blue).

  2. B, C

    RV replication in the indicated MEFs was quantitated by calculating the number and intensities of viral particles per cell (B), or per micrograph field (C).

  3. D

    The relative survival of WT, Ifnar1 −/−, and Ifih1 −/− MEFs after RV infection for 24 h was assessed by direct counting of the cells.

  4. E

    A titration of virus produced in WT, Ifnar1 −/−, and Ifih1 −/− MEFs by immunofluorescence.

  5. F

    The relative induction of an IFNB1‐promoter firefly‐luciferase reporter 24 h after RV infection in HEK293 cells co‐transfected with control empty vector (C), MDA5‐, RIG‐I‐, or LGP2‐expressing constructs. The levels of firefly luciferase were normalized to a constitutively active Renilla luciferase and expressed relative to control cells.

  6. G

    Quantitation of RV from the indicated tissues of WT and Ifih1 −/− mice orally infected with RV for 5 days. Viral infection was quantified by titration against MA104 cells and immunofluorescence.

Data information: All graphs show data as mean ± SEM from repeated experiments. The Student's t‐test or Mann–Whitney test was used to calculate the P‐values from the data produced in vitro (n = 3) or in vivo (n = 6), respectively (NS = P > 0.05, *≤ 0.05, ** 0.01, *** 0.001, ****P ≤ 0.0001).
Figure 2
Figure 2. Mda5 induces IFN‐dependent and IFN‐independent anti‐RV responses
  1. A

    Graphs showing the levels of Ifnβ protein (on left) and the Ifnb1 transcript (on right) induced in WT and Ifih1 −/− MEFs infected with RV assessed after 24 h by ELISA or after 6 h by qRT–PCR.

  2. B

    Immunoblots assessing the levels of Mda5 and P56 relative to β‐actin in lysates from control (C) or infected (RV) WT and Ifih1 −/− MEFs pre‐treated with Ifn1α for 16 h or, alternatively, an Ifnar1‐neutralizing (MAR1) or control (IgG) antibody for 1 h prior to infection with RV for 24 h.

  3. C, D

    Graphs showing the quantitation of the levels of Mda5 and P56 under the identified conditions as detected by immunoblot (see also Fig EV1).

  4. E

    A quantitation of progeny RV from WT and Ifih1 −/− MEFs treated with Ifn1α or MAR1.

Data information: The data are shown as mean ± SEM. Student's t‐test was used to calculate the P‐values (n = 3). NS = P > 0.05, * 0.05, **≤ 0.01, ***≤ 0.001.
Figure EV1
Figure EV1. Mda5 (Ifih1) and P56 induction in response to RV (related to Fig 2B–D)
  1. Detection (on left) and quantitation (on right) of Mda5 and P56 induction relative to β‐actin in WT, Ifnar1 −/−, and Ifih1 −/− MEFs infected with RV for the indicated times by immunoblot.

  2. Quantitation of the induction of the Ifih1 mRNA in WT, Ifnar1 −/−, and Ifih1 −/− MEFs 6 h post‐infection with RV by qRT–PCR.

Data information: The data are shown as mean ± SEM. Student's t‐test was used to calculate the P‐values (= 3). *≤ 0.05, **≤ 0.01.
Figure 3
Figure 3. Mda5‐dependent activation of NFκB induces cell death
  1. A

    FACS plots showing annexin V and 7‐AAD staining of WT and Ifih1 −/− MEFs mock‐ or RV‐infected for 2 and 8 h.

  2. B–D

    A quantitation of the live, dead, and early apoptotic cells by their relative 7‐AAD and annexin V staining.

  3. E

    A quantitation of the relative survival of WT, Ifnar1 −/−, and Ifih1 −/− MEFs RV‐infected for 24 h, then assessed by staining with crystal violet and spectrophotometry.

  4. F

    A quantitation of apoptosis in HEK293 cells co‐transfected with an empty control (C), MDA5‐, or MAVS‐expressing vectors and either control (C), IκBα, or dominant‐negative IRF3 (IRF3ΔN) constructs to repress the activity of NFκB or IRF3, respectively. These cells were either left untreated (−) or transfected (+) with FuGENE:pIC for 10 h, and then, apoptosis was assessed by measuring the nuclear accumulation of YO‐PRO by co‐localization with the nuclear Hoechst stain (see also Fig EV3).

Data information: The data are a representative result from repeated experiments. In panels (B–D and F), the data are shown as mean ± SD of experimental replicates. In panel (E), the data are shown as mean ± SEM. Student's t‐test was used to calculate the P‐values (n = 3). NS = P > 0.05, *≤ 0.05, **≤ 0.01, ***≤ 0.001.
Figure EV2
Figure EV2. MDA5‐induced cell death in HEK293 cells is NFκB dependent (related to Fig 3F)
  1. Relative cell survival assessed by the pH‐dependent color change of culture medium supporting HEK293 cells transfected with GFP, MDA5 or MAVS expression constructs, and a control backbone vector (C) or dominant‐negative constructs of TANK‐binding kinase 1 (TBK1‐DN), IKK‐related kinase ε (IKKε‐DN), a constitutively active IRF3 construct (IRF3‐5D), or the NFκB inhibitor, IκBα, after 90 h of culture.

  2. A schematic of MDA5‐dependent cell signaling in the cell.

Figure 4
Figure 4. RV infection induces Mda5‐dependent inflammatory signaling
  1. A

    Detection (on left) and quantitation (on right) of NFκB activation by assessing the levels of its cytosolic inhibitor IκBα relative to the β‐actin in WT and Ifih1 −/− MEFs infected with RV for the indicated times.

  2. B, C

    Graphs showing the relative induction of the Il‐6 and Tnfα cytokines and Il6 and Tnf transcripts in WT and Ifih1 −/− primary peritoneal macrophages infected with RV measured by ELISA after 24 h or qRT–PCR after 6 h, respectively.

  3. D, E

    Detection (on left) and quantitation of both immature and processed Il‐1β in lysates from WT and Ifih1 −/− primary peritoneal macrophages that were infected with RV for the indicated times, or treated with lipopolysaccharide (LPS) plus the potassium ionophore nigericin (N) by ELISA (on left) and immunoblot (on right) after 20 h. The levels of Il‐1β are quantitated relative to β‐actin in the immunoblot.

Data information: The data are shown as mean ± SEM. Student's t‐test was used to calculate the P‐values (n = 3). NS = > 0.05, *≤ 0.05, **≤ 0.01, ***≤ 0.001, ****≤ 0.0001.
Figure EV3
Figure EV3. Mda5‐induced nuclear translocation of IRF3 in MEFs infected with RV (related to Fig 4)
Micrographs showing the cellular distribution of IRF3 (green) in WT and Ifih1 −/− MEFs detected with an anti‐IRF3 antibody 4 h after infection with RV. Arrows indicate the more intense cytosolic, perinuclear enrichment of IRF3 in the WT and Ifih1 −/− cells before infection. This fluorescence pattern changes to a relatively more intense nuclear (blue) staining of IRF3 in RV‐infected WT cells but not infected Ifih1 −/− cells. The nucleus is visualized with the Hoechst DNA stain.
Figure 5
Figure 5. RV infection induces Mda5‐dependent IFN signaling in the pancreas
  1. A–D

    Measures of the induction of the Mda5 and P56 proteins and the Ifih1 and Ifnb1 transcripts in the indicated tissues of 5‐week‐old WT and Ifih1 −/− mice infected with RV for 5 days, detected by ELISA or qRT–PCR, respectively. The levels of the transcripts were quantified and normalized to the levels of the 18S ribosomal RNA (Rn18s). The data are shown as mean ± SEM. The Mann–Whitney test was used to calculate the P‐values (n = 6). NS = P > 0.05, *≤ 0.05, **≤ 0.01.

Figure 6
Figure 6. RV infection induces Mda5‐dependent inflammation in the pancreas
  1. A–D

    Measures of the induction of the Il‐6 and Il‐1β proteins and the Tnf and Il1b transcripts in the indicated tissues of 5‐week‐old WT and Ifih1 −/− mice infected with RV for 5 days as detected by ELISA or qRT–PCR, respectively. The levels of the transcripts were quantified and normalized to the levels of the 18S ribosomal RNA (Rn18s). The data are shown as mean ± SEM. The Mann–Whitney test was used to calculate the P‐values (n = 6). NS = > 0.05, *≤ 0.05, **≤ 0.01.

Figure 7
Figure 7. Allelic variation alters MDA5 activity
  1. A

    Micrographs showing Venus fluorescence (green) produced by homotypic interaction between split‐Venus fusion constructs of MAVS at 56 h (far right) or MDA5 after 48 and 56 h (first and second columns) and heterologous interactions between MDA5 and MAVS at 56 h (third column) in HEK293 cells. Mitochondria are visualized with MitoTracker (magenta), and cell nuclei are stained with Hoechst (blue). Arrows indicate the co‐localization of protein complexes (Venus) and mitochondria.

  2. B, C

    Graphs quantitating the association between the major (MDA5) and minor IFIH1 alleles (T946, V923, and X627) and the complementary split‐Venus tag alone (C), MDA5 or MAVS tagged with the separate halves of the split‐Venus fluorophore and assessed by the fluorescence intensity in the cell after 48 h (n = 3) (see also Fig EV4).

  3. D–F

    The relative induction of IFNB1‐, IFIT1‐, and IL‐8‐promoter firefly‐luciferase reporters 24 h after mock or RV infection in HEK293 cells co‐transfected with the control LV‐GFP (C) or the indicated MDA5‐GFP constructs (= 3). The levels of firefly luciferase were normalized to a constitutively active Renilla luciferase and expressed relative to control (GFP only) cells that did not express MDA5 (see also Fig EV5).

  4. G

    The relative survival of HEK293 cells expressing the indicated MDA5‐GFP or control GFP (C) constructs labeled with annexin V and 7‐AAD and assessed by FACS (n = 3) (see also Fig EV5).

  5. H

    The relative production of recombinant control LV‐GFP (C) or the indicated MDA5‐GFP lentivirus (LV) measured by quantitating GFP fluorescence in HEK293 cells (n = 3).

  6. I

    Titration of RV produced from HEK293 cells transfected with the control LV‐GFP (C) or the indicated MDA5‐GFP constructs (n = 3) (see also Fig EV6).

Data information: The data are shown as mean ± SEM. Student's t‐test was used to calculate the P‐values. NS = > 0.05, *≤ 0.05, **≤ 0.01, ***≤ 0.001.
Figure EV4
Figure EV4. Visualization of MDA5 expression and associations (related to Fig 7A–C)
  1. A, B

    Micrographs showing GFP fluorescence in HEK293 cells transfected with the indicated LV‐MDA5‐GFP variants after 48 (A) and 56 h (B), showing that the cytosolic expressed fusion protein coalesces to a perinuclear signal.

  2. C, D

    Micrographs showing Venus fluorescence in cells co‐transfected with split‐Venus constructs separately fused to the indicated MDA5 variants and either (C) MDA5 to detect homotropic interactions or (D) MAVS to detect hetero‐oligomerization.

  3. E

    A quantitation of the extent of the association of protein complexes with the mitochondria assessed by measuring the co‐localization of Venus fluorescence with the signal from the MitoTracker stain as regions of interest (ROI). The data are shown as mean ± SD. Student's t‐test was used to calculate the P‐values from six measures for each protein pair at each time point. NS = > 0.05, *≤ 0.05, **≤ 0.01, ***≤ 0.001, ****≤ 0.0001.

Figure EV5
Figure EV5. Transient transfection assays with MDA5 lentiviral constructs (related to Fig 7D–G)
  1. A, B

    Immunoblots detecting the induction of P56 to demonstrate the responsiveness of (A) HEK293 compared to (B) HEK293T (at top) and HeLa (at bottom) cells expressing MDA5 (Ifih1), RIG‐I, or MAVS when treated with FuGENE:pIC to activate the RLRs. β‐Actin is used as a loading control.

  2. C

    FACS plots of HEK293 cells expressing LV‐GFP or the indicated LV‐MDA5‐GFP variants stained with annexin V and 7‐AAD. The data represent cellular viability in both transfected (GFP+) and non‐transfected cells (GFP). The GFP‐, annexin V‐, and 7‐AAD‐positive cells are graphed in Fig 7G.

Figure EV6
Figure EV6. MDA5‐induced cell death (related to Fig 7H and I)
  1. Immunoblot detection of the expression of the indicated LV‐MDA5‐GFP constructs in HEK293 cells using an anti‐GFP antibody. Unmarked lanes are mutant MDA5 constructs that were used as additional controls in the study that are not described in the text.

  2. A graph showing the relative GFP fluorescence produced in HEK293 cells transfected with the indicated LV‐MDA5‐GFP constructs. The data is shown as mean ± the SD (n = 3).

Similar articles

Cited by

References

    1. Angelosanto JM, Blackburn SD, Crawford A, Wherry EJ (2012) Progressive loss of memory T cell potential and commitment to exhaustion during chronic viral infection. J Virol 86: 8161–8170 - PMC - PubMed
    1. Besch R, Poeck H, Hohenauer T, Senft D, Hacker G, Berking C, Hornung V, Endres S, Ruzicka T, Rothenfusser S, Hartmann G (2009) Proapoptotic signaling induced by RIG‐I and MDA‐5 results in type I interferon‐independent apoptosis in human melanoma cells. J Clin Invest 119: 2399–2411 - PMC - PubMed
    1. Broquet AH, Hirata Y, McAllister CS, Kagnoff MF (2011) RIG‐I/MDA5/MAVS are required to signal a protective IFN response in rotavirus‐infected intestinal epithelium. J Immunol 186: 1618–1626 - PubMed
    1. Bruns AM, Leser GP, Lamb RA, Horvath CM (2014) The innate immune sensor LGP2 activates antiviral signaling by regulating MDA5‐RNA interaction and filament assembly. Mol Cell 55: 771–781 - PMC - PubMed
    1. Carrero JA, Calderon B, Towfic F, Artyomov MN, Unanue ER (2013) Defining the transcriptional and cellular landscape of type 1 diabetes in the NOD mouse. PLoS ONE 8: e59701 - PMC - PubMed

Publication types

Substances