Mitochondrial DNA stress primes the antiviral innate immune response

Abstract

Mitochondrial DNA (mtDNA) is normally present at thousands of copies per cell and is packaged into several hundred higher-order structures termed nucleoids. The abundant mtDNA-binding protein TFAM (transcription factor A, mitochondrial) regulates nucleoid architecture, abundance and segregation. Complete mtDNA depletion profoundly impairs oxidative phosphorylation, triggering calcium-dependent stress signalling and adaptive metabolic responses. However, the cellular responses to mtDNA instability, a physiologically relevant stress observed in many human diseases and ageing, remain poorly defined. Here we show that moderate mtDNA stress elicited by TFAM deficiency engages cytosolic antiviral signalling to enhance the expression of a subset of interferon-stimulated genes. Mechanistically, we find that aberrant mtDNA packaging promotes escape of mtDNA into the cytosol, where it engages the DNA sensor cGAS (also known as MB21D1) and promotes STING (also known as TMEM173)-IRF3-dependent signalling to elevate interferon-stimulated gene expression, potentiate type I interferon responses and confer broad viral resistance. Furthermore, we demonstrate that herpesviruses induce mtDNA stress, which enhances antiviral signalling and type I interferon responses during infection. Our results further demonstrate that mitochondria are central participants in innate immunity, identify mtDNA stress as a cell-intrinsic trigger of antiviral signalling and suggest that cellular monitoring of mtDNA homeostasis cooperates with canonical virus sensing mechanisms to fully engage antiviral innate immunity.

Extended Data Figure 1. TFAM deficiency induces mtDNA depletion, nucleoid stress, elevated ISG expression, and augmented type I interferon responses, but does not drastically alter oxygen consumption and mt-transcription rates

a, qPCR analysis of relative mtDNA copy number from WT and Tfam+/− MEFs. b, Basal oxygen consumption rate of WT and Tfam+/− MEFs as determined by Seahorse Bioscience XF96 Extracellular Flux assay. c, qRT-PCR of mtDNA-encoded rRNA (16s) and mRNA (ND6, Cytb, Cox1) transcripts in WT and Tfam+/− MEFs. d-f, Untransfected Tfam+/− (d) or WT MEFs transfected with control [si-Ctrl] or Tfam [si-Tfam] siRNAs (d-f) were stained with anti-Hsp60 [Mito] and anti-DNA [DNA] antibodies. Nucleoid area from multiple independent images was calculated, stratified into groups, and graphed as percentage of the total number of nucleoids counted for each sample (d). Inset panels are magnified 3X to enhance viewing of mitochondrial network and nucleoid architecture (e). TFAM and ISG mRNA expression were measured by qRT-PCR (f). g-i, Tfam^flox/flox ERCre− or Tfam^flox/flox ERCre+ BMDM were incubated in 4OHT for 96hrs to induce TFAM depletion. Immunofluorescence staining was performed as described above (g). ISG mRNA and protein expression was monitored by qRT-PCR and western blotting (h). qRT-PCR analysis of type I interferon and Il6 expression in 4OHT-treated Tfam flox ERCre−/+ BMDM two hours post cytosolic delivery of interferon-stimulatory DNA [ISD] or poly(I:C) [PIC] (i). Error bars indicate ± s.e.m. of triplicates and data are representative of three independent experiments.. *=p<0.05, **=p<0.01, ***=p<0.001, ns=not significant. Unless noted p<0.05.

Extended Data Figure 2. TFAM deficiency promotes accumulation of cytosolic mtDNA

a, WT or Tfam+/− MEFs were subjected to digitonin fractionation as described in the Methods and whole cell extracts [WCE], pellets [Pel], or cytosolic extracts [Cyt] were blotted using the indicated antibodies. b, DNA was extracted from digitonin extracts of WT and Tfam+/− MEFs or Tfam^flox/flox ERCre− or Tfam^flox/flox ERCre+ BMDM incubated in 4OHT for 72hrs. Cytosolic mtDNA was quantitated via qPCR using the mt-Dloop3 primer set. Normalization was performed as described in the Methods. c, Samples were prepared as described in b, and cytosolic mtDNA was quantitated via qPCR using the indicated primer sets. Error bars indicate ± s.e.m. of triplicates and data are representative of three independent experiments.**=p<0.01, ***=p<0.001.

Extended Data Figure 3. Mitochondrial hyperfusion regulates the accumulation of mtDNA nucleoid stress in TF^D MEFs

a-b, WT MEFs were transfected with control or TFAM siRNAs for 96hrs. Cells were fixed and processed for EM analysis (a). Mitochondrial perimeter measurements were obtained from multiple independent images, stratified into groups, and graphed as percentage of the total number of mitochondria counted for each sample (b). c-e, WT MEFs were transfected with control, Mfn1, and/or TFAM siRNAs for 96hrs. Cells were fixed and stained with anti-Hsp60 antibody [Mito] and anti-DNA antibody [DNA] for confocal microscopy (c). Nucleoid area from multiple independent images was calculated as previously described (d). RNA was extracted for ISG expression analysis by qRT-PCR (e). f, WT and Tfam+/− MEFs were transfected with the indicated siRNAs for 96 hours and ISG expression analyzed by qRT-PCR. Error bars indicate ± s.e.m. of triplicates and data are representative of two independent experiments. **=p<0.01, ***=p<0.001.

Extended Data Figure 4. mtDNA stress in TF^D MEFs and BMDM potentiates type I interferon responses to viral infection and enhances viral clearance

a-b, WT and Tfam+/− MEFs were infected with VSV-GFP (a) or MHV68-GFP (b) and after the indicated times, cytokine and ISG mRNA expression determined by qRT-PCR, or cytokine secretion determined by ELISA. c-f, Tfam^flox/flox ERCre− or Tfam^flox/flox ERCre+ BMDM were incubated in 4OHT for 96hrs to induce TFAM depletion. Cells were infected with HSV1-GFP (c, e-f) or VSV-GFP (d-e), incubated for the indicated times, and viral gene expression was determined by qRT-PCR (c-d) and western blotting (e), or cytokine and ISG mRNA expression determined by qRT-PCR (f). Error bars indicate ± s.e.m. of triplicates and data are representative of two independent experiments. **=p<0.01, ***=p<0.001, ns=not significant.

Extended Data Figure 5. Tissues fromTfam+/− mice display elevated ISG expression, and ddC abrogates mtDNA stress, ISG expression and viral resistance phenotypes of TF^D cells

a, RNA was extracted from the liver and kidneys of 8 week old WT and Tfam+/− mice (n=2 each) and subjected to qRT-PCR analysis for basal ISG expression. b-d, Relative mtDNA copy number (b), mtDNA nucleoid area (c) and ISG expression (d) of WT and Tfam+/− MEFs exposed to ddC for 96hrs. e-f, mtDNA nucleoid area (e) and ISG expression (f) of WT MEFs transfected with control or TFAM siRNAs for 96 hrs in the presence or absence of ddC. g-i, Tfam^flox/flox ERCre− or Tfam^flox/flox ERCre+ BMDM were incubated in 4OHT for 96hrs to induce TFAM depletion in the presence of ddC. ddC was washed out and cells allowed to recover overnight before infection. Cells were infected with VSV-GFP (g) or HSV1-GFP (h) at MOI 1, or WT BMDM transfected with poly(I:C) or ISD (i), and incubated for the indicated times. Ifnb expression or viral gene expression was determined by qRT-PCR. Error bars indicate ± s.e.m. of triplicates and data are representative of two independent experiments. *=p<0.05, **=p<0.01, ***=p<0.001, ns=not significant. Unless noted p<0.05.

Extended Data Figure 6. Alpha- and gamma-herpesviruses induce mtDNA stress, but RNA viruses Influenza and LCMV do not

a, Relative mtDNA copy number of WT MEFs 24 hours post infection with VSV-GFP, HSV1-GFP, or mock at the indicated MOIs. b, WT MEFs were infected with MHV68-GFP at a MOI 0.5 and after the indicated times, cells were stained and subjected to confocal microscopy or relative mtDNA copy number determined. c, WT MEFs were infected with HSV2, Flu-GFP, or LCMV-GFP, at a MOI 10, and after 6 hours, cells were stained and subjected to confocal microscopy. d, WT MEFs were infected with Vaccinia virus at a MOI 10 or 1, and after the indicated times, cells were stained and subjected to confocal microscopy or relative mtDNA copy number determined. Error bars indicate ± s.e.m. of triplicates and data are representative of two independent experiments. *=p<0.05, **=p<0.01, ***=p<0.001, ns=not significant.

Extended Data Figure 7. HSV1 UL12 M185 expression is sufficient to trigger mtDNA stress, TFAM depletion, and antiviral priming in BMDM; infection with UL12-deficient HSV-1 fails to induce mtDNA stress, elicits lower vaginal type I interferon responses, and spreads more readily to DRG

a, WT BMDM were transduced with HSV1 UL12 M185 expressing or empty retroviruses (RV) and relative mtDNA abundance, protein expression, and ISG mRNA expression determined. b, WT MEFs were infected HSV1 (UL12-FLAG) or UL12-deficient HSV1 (UL12Δ + UL98-FLAG) at MOI 10 for 3 hours and analyzed by confocal microscopy. c, WT MEFs were infected HSV1 (UL12-FLAG) or UL12-deficient HSV1 (UL12Δ + UL98-FLAG) at MOI 2 for 24 hours and mtDNA abundance was determined by qPCR. d, The vaginas of WT mice (n=3 per condition) were inoculated with 10⁶ p.f.u. of HSV1 (UL12-FLAG) or UL12-deficient HSV1 (UL12Δ + UL98-FLAG), and 24 hours post infection, vaginal RNA was extracted and gene expression analyzed by qRT-PCR. e, Mice (n=3 per condition) were infected as previously described, and 10 days post infection, DNA from DRG was isolated for mtDNA and HSV-1 genome abundance measurements by qPCR. Error bars indicate ± s.e.m. of triplicates and data are representative of two independent experiments. *=p<0.05, **=p<0.01, ***=p<0.001; unless noted p<0.05, ns=not significant.

Extended Data Figure 8. Model illustrating mtDNA stress-dependent antiviral priming

TFAM depletion, induced genetically or during Herpes virus infection, triggers mtDNA stress, characterized by nucleoid loss and enlargement. This results in the release of fragmented mtDNA that recruits and activates peri-mitochondrial cGAS to generate the second messenger cyclic GMP-AMP (cGAMP) and activate ER-resident STING. STING then activates TBK1, which phosphorylates IRF3 to induce dimerization and nuclear translocation. Active IRF3 elevates basal gene expression of ISGs with antiviral signaling and effector functions. Signaling ISGs, such as IRF7, ISG15, STAT1, and STAT2 cooperate with IRF3 to potentiate RIG-I-like receptor (RLR), interferon-stimulatory DNA (ISD), and type I interferon (IFN-I) responses, and effector ISGs, such as IFI44, IFIT1, IFIT3, and OASL2, augment viral resistance. Both outcomes collectively and robustly boost innate antiviral defenses to dampen viral replication.

Figure 1. Tfam+/− cells exhibit mtDNA stress, elevated ISG expression, and augmented type I interferon responses

a, Confocal microscopy images of MEFs stained with anti-DNA (DNA) and anti-HSP60 (Mito) antibodies. b, Heatmaps of microarray analyses. Genes in Tfam+/− MEFs exhibiting statistically significant (p<0.05), two-fold or greater increases over WT are shown. c-d, qRT-PCR (c) and western blots (d) of basal ISG expression in two littermate WT and Tfam+/− MEF lines. e, qRT-PCR analysis of type I interferon expression in MEFs 9 hrs post cytosolic delivery of poly(I:C). Error bars indicate ± s.e.m. of triplicates and are representative of three independent experiments. ***=p<0.001.

Figure 2. mtDNA stress triggers ISG expression in a cGAS- and STING-dependent fashion

a-b, ISG expression in Tfam+/− MEFs transfected with the indicated siRNAs (top panels), or WT, cGas−/− (a), and Sting−/− (b) MEFs transfected with TFAM siRNAs (bottom panels). c-d, ISG expression in Tfam+/− MEFs transfected with the indicated siRNAs for 96 hrs. e, Western blots of whole cell and nuclear extracts of WT and Tfam+/− MEFs or Tfam flox ERCre−/+ BMDM exposed to 4OHT for 96 hrs. f-gcGAS−/− MEFs reconstituted with cGAS-HA were transfected with the indicated siRNAs for 96 hrs, then stained with anti-DNA, anti-Hsp60 [Mito], and anti-HA antibodies and imaged. cGAS colocalization scoring was performed as described in the Methods. Error bars indicate ± s.e.m. of triplicates and are representative of three independent experiments. *=p<0.05, **=p<0.01, ***=p<0.001.

Figure 3. mtDNA stress potentiates viral resistance

a, Viral GFP expression in MEFs infected with HSV1-GFP or VSV-GFP at MOI 0.5 for 24 hrs. b, MHV68-GFP abundance in MEFs infected at MOI 0.5. c, LCMV Armstrong gene expression 4d after i.v. infection of WT and Tfam+/− mice; n=4. d-f, Nucleoid area (d) or ISG expression (e) of MEFs exposed to ddC for 96 hrs. ddC-exposed MEFs were infected with HSV1-GFP or VSV-GFP at MOI 0.1 and imaged after 24 hrs. Error bars represent ± s.e.m. of triplicates (b,e) or quadruplicates (c) and all panels are representative of three independent experiments. **=p<0.01, ***=p<0.001.

Figure 4. HSV-1 induces mtDNA stress and TFAM depletion sufficient to trigger ISG expression and necessary to fully engage antiviral immunity

a-c, WT MEFs were mock infected or infected with HSV1-GFP or VSV-GFP at MOI 10 for the indicated times, and imaged after staining with anti-DNA, anti-Hsp60 [Mito], and anti-HSV [HSV1] or GFP [VSV] antibodies (a). mtDNA nucleoid area was calculated as described in the Methods (b). Extracts were blotted as indicated (c). d-e, WT MEFs were transduced with HSV1 UL12 M185-FLAG-expressing or empty retroviruses (RV) and cells were stained with anti-DNA, anti-Hsp60 [Mito] and anti-FLAG antibody [UL12 M185] (d), and protein or ISG expression examined after 24 hrs (e). f-g, Protein and RNA expression in BMDM infected with HSV1 (UL12-FLAG) or UL12-deficient HSV1 (UL12Δ + UL98-FLAG) at MOI 2 for the indicated times,. h, HSV1 genome abundance in L929 cells were infected as in f-g. Error bars indicate ± s.e.m. of triplicates and are representative of three independent experiments. ***=p<0.001.