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
. 2022 Mar;603(7899):145-151.
doi: 10.1038/s41586-022-04421-w. Epub 2022 Jan 19.

The cGAS-STING pathway drives type I IFN immunopathology in COVID-19

Affiliations

The cGAS-STING pathway drives type I IFN immunopathology in COVID-19

Jeremy Di Domizio et al. Nature. 2022 Mar.

Abstract

COVID-19, which is caused by infection with SARS-CoV-2, is characterized by lung pathology and extrapulmonary complications1,2. Type I interferons (IFNs) have an essential role in the pathogenesis of COVID-19 (refs 3-5). Although rapid induction of type I IFNs limits virus propagation, a sustained increase in the levels of type I IFNs in the late phase of the infection is associated with aberrant inflammation and poor clinical outcome5-17. Here we show that the cyclic GMP-AMP synthase (cGAS)-stimulator of interferon genes (STING) pathway, which controls immunity to cytosolic DNA, is a critical driver of aberrant type I IFN responses in COVID-19 (ref. 18). Profiling COVID-19 skin manifestations, we uncover a STING-dependent type I IFN signature that is primarily mediated by macrophages adjacent to areas of endothelial cell damage. Moreover, cGAS-STING activity was detected in lung samples from patients with COVID-19 with prominent tissue destruction, and was associated with type I IFN responses. A lung-on-chip model revealed that, in addition to macrophages, infection with SARS-CoV-2 activates cGAS-STING signalling in endothelial cells through mitochondrial DNA release, which leads to cell death and type I IFN production. In mice, pharmacological inhibition of STING reduces severe lung inflammation induced by SARS-CoV-2 and improves disease outcome. Collectively, our study establishes a mechanistic basis of pathological type I IFN responses in COVID-19 and reveals a principle for the development of host-directed therapeutics.

PubMed Disclaimer

Conflict of interest statement

A.A. is a scientific co-founder of IFM Due. The other authors declare no competing interests.

Figures

Fig. 1
Fig. 1. Type I IFN-producing macrophages surround damaged endothelial cells in COVID-19 skin lesions.
a, Immune gene expression profiles of skin lesions from individuals with COVID-19 (n = 10) and individuals with CLE (n = 11), and skin from healthy donors (HD; n = 5). Unbiased clustering was performed. b, Immunohistochemistry quantification of macrophages, neutrophils, plasmacytoid dendritic cells (pDCs) and T cells (stained for CD163, MPO, CD123 and CD3) in CLE (n = 5) and COVID-19 (n = 10) skin lesions. c, Confocal microscopy images of representative COVID-19 skin lesion stained for CD163 (green) and IFNβ (red). Scale bars, 20 μm. d, Contribution of CD163+ macrophages and CD31+ endothelial cells to IFNβ expression in CLE (n = 5) and COVID-19 (n = 10). e, Confocal microscopy images of representative COVID-19 skin lesion stained for CD31 (green) and IFNβ (red). Scale bars, 20 μm. f, Proportions of CD163+ macrophages, CD31+ endothelial cells and other cells among IFNβ-producing cells for each CLE and COVID-19 sample. g, Confocal microscopy images of a representative COVID-19 skin sample stained for CD163 (green) and CD31 (red) to depict macrophages and endothelial cells. Scale bar, 50 μm. h, Transmission electron microscopy of dermal vessels in purpuro-necrotic (left) and maculopapular (middle) COVID-19 skin lesions, and in healthy skin (right). Arrows show disrupted endothelial cells (COVID-19 skin lesions) and intact endothelial cells (healthy skin). Scale bars, 2 μm (left); 5 μm (middle); 1 μm (right). i, Immunohistochemistry for cleaved caspase-3 (cl. caspase-3) in COVID-19 skin lesions (nuclear staining indicated by arrow). Scale bar, 20 μm. j, Percentage of CD31+ endothelial cells with cleaved caspase-3 staining in healthy skin (n = 8) and in CLE (n = 6) and COVID-19 (n = 10) skin lesions. k, Correlation between cleaved-caspase-3-positive nuclei and overall staining intensity of IFNβ measured in COVID-19 skin samples (n = 8). Spearman correlation and two-tailed statistical significance were performed. Data are mean ± s.d. (b, d, j). P values obtained with two-tailed Student’s t-test and one-way ANOVA followed by Tukey’s multiple comparisons test (b, d, j). Source data
Fig. 2
Fig. 2. cGAS–STING-dependent type I IFN signature in COVID-19 skin and lung pathology.
a, Confocal microscopy images of a representative COVID-19 skin sample stained for CD163 (green), IFNB1 mRNA (red) and DNA (blue). Scale bars, 10 μm. Arrows indicate cytosolic DNA particles. b, Quantification of CD163+ macrophages containing cytosolic DNA particles in COVID-19 skin lesions (n = 10) and in healthy skin (n = 9). c, Quantification of cGAMP in lysates of COVID-19 skin lesions (n = 10) and healthy skin (n = 3). d, Confocal microscopy images of a representative COVID-19 skin sample stained for CD163 (green) and p-STING (red). Blood vessels, dashed line. Arrows show p-STING+ endothelial cells. Scale bars, 20 μm (left); 5 μm (right two images). e, Quantification of p-STING+ macrophages in COVID-19 skin lesions (n = 10) and in healthy skin (n = 10). f, Confocal microscopy images of a representative COVID-19 skin sample stained for CD31 (green) and p-STING (red). Blood vessels, dashed line. Scale bars, 20 μm. g, Proportions of CD163+ macrophages and CD31+ endothelial cells among p-STING+ cells in COVID-19 skin lesions (n = 7). h, Expression of ISGs (IFI35, IRF7 and MX1) in cultured healthy skin (n = 3) and COVID-19 skin explants (n = 3), treated or not with H-151. i, Confocal microscopy images of representative post-mortem lungs with early (fewer than 10 days; left) or late (more than 14 days, right) DAD, stained for p-STING (red) and CD163 (green). Scale bars (left to right): 20 μm, 10 μm, 50 μm, 10 μm. j, Quantification of p-STING+CD163+ macrophages in post-mortem lungs with early and late DAD (n = 4). k, Immunohistochemistry of representative post-mortem lungs with early (left) or late (right) DAD, stained for MxA. Scale bar, 50 μm. l, Percentage of tissue area with MxA positivity in early and late DAD samples (n = 4). Data are mean ± s.d. (b, c, e, h, j, l). P values obtained with two-tailed Student’s t-test (c, e, h, j, l) and with Mann–Whitney test (b). Source data
Fig. 3
Fig. 3. STING-dependent type I IFN production and cell death after SARS-CoV-2 infection in endothelial cells.
a, Schematic of the three-cell component LoC model. b, c, Representative 3D images of the vascular face of uninfected or SARS-CoV-2-infected LoCs with or without vascular H-151 perfusion. ‘3C’, three-cell component (epithelial cells, endothelial cells and macrophages) in b; ‘2C’, two-cell component (epithelial cells and endothelial cells) in c. CD45+ macrophage (green), IFNβ (bright pink), cleaved caspase-3 (amber), actin (azure) and nuclear (purple) stainings are shown. Scale bars, 20 μm. d, Representative 3D images of p-STING+ endothelial cells (yellow). Scale bars, 20 μm. e, Expression levels of the indicated genes in uninfected (n = 4), infected (n = 5) and H-151-treated (n = 5) LoCs. CD31 is also known as PECAM1. RE, relative expression. f, Representative volumetric electron microscopy images, 3D reconstructions and quantification of the surface-area-to-volume ratio of endothelial cell mitochondria from uninfected (n = 45) and infected (n = 43) LoCs. Solid line, mean; dashed lines, quartiles. Scale bars, 1 μm (left images); grey hexagons (middle images) represent 1 μm3. g, Representative 3D images of the vascular face of infected LoCs with or without vascular VBIT-4 perfusion. Scale bars, 50 μm. Statistics for quantification: b, IFNβ: uninfected (n = 6 fields of view (FOV)), infected (n = 7 FOV) and H-151-treated (n = 4 FOV) LoCs, n = 2 LoCs each; c, IFNβ/cleaved caspase-3: infected (n = 4 FOV in each case) and H-151-treated (n = 4/n = 5 FOV) LoCs, n = 2 LoCs each for both markers; f, data from n = 4 endothelial cells each from n = 1 uninfected and n = 2 infected LoCs; g, IFNβ: infected (n = 7 FOV) and VBIT-4-treated (n = 6 FOV) LoCs, n = 2 LoCs each. Data acquired at 3 dpi; mean ± s.e.m.; P values calculated by one-way ANOVA followed by Tukey’s multiple comparisons tests (b, c, e) or two-tailed Mann–Whitney test (f, g). Source data
Fig. 4
Fig. 4. STING inhibition reduces SARS-CoV-2-induced inflammation in mice.
a, Schematic of SARS-CoV-2 infection (intranasal; 1 × 104 plaque-forming units (PFU) per mouse) and intraperitoneal administration of vehicle or H-151 (starting at 1 day before infection), related to data from bd. b, Left, representative haematoxylin and eosin (H&E) images of lungs from vehicle- and H-151-treated mice. Scale bars, 500 μm. Right, average inflamed area in SARS-CoV-2 infected mice. c, mRNA expression levels of the indicated genes in uninfected and infected lungs at 6 dpi, analysed by RT–qPCR. d, Relative weight loss in mice after SARS-CoV-2 infection. e, Schematic of SARS-CoV-2 infection (intranasal; 1 × 104 PFU per mouse) and intraperitoneal administration of vehicle or H-151 (starting at 2 dpi), related to data from fh. f, Left, representative H&E images of lungs from vehicle- and H-151-treated mice. Scale bars, 500 μm, Right, average inflamed area in SARS-CoV-2 infected mice. g, h, Relative weight loss (g) and survival (h) in mice after SARS-CoV-2 infection with post-infection regimen. Numbers are: ac, uninfected (n = 4), vehicle and H-151 (n = 7); d, uninfected (n = 8), vehicle and H-151 (n = 12); eg, uninfected, vehicle and H-151 (n = 5); h, vehicle and H-151 (n = 15). Throughout the figure, data are mean ± s.e.m.; P values calculated by one-way ANOVA followed by Tukey multiple comparison tests (b, c, d, f, g), or by Mantel–Cox survival analysis (h). Mice infected with SARS-CoV-2 were age-matched (12–16 weeks) female K18-hACE2 mice. Source data
Extended Data Fig. 1
Extended Data Fig. 1. Clinical characteristics of patients with COVID-19 with associated skin manifestations.
a, Photographs of the skin lesions. b, Clinical parameters and demographics of the 10 patients with COVID-19 selected for this study. HRD, Heart Rhythm Disorder; Overweight (BMI > 25 kg/m2, but < 30 kg/m2); COPD, chronic obstructive pulmonary disease; CKD, chronic kidney disease; DM, diabetes mellitus; AIH, autoimmune hepatitis.
Extended Data Fig. 2
Extended Data Fig. 2. Immune gene expression profiling of COVID-19 skin lesions and common inflammatory skin diseases.
a, Immune gene expression profiles of patients with lichen planus (n = 5), cutaneous lupus erythematosus (CLE, n = 10), COVID-19 associated skin lesions (n = 10), plaque-type psoriasis (n = 21), and atopic dermatitis (AD, n = 16) compared to healthy skin (HD, n = 4) assessed by NanoString assay. Differentially expressed genes between different pairwise comparisons (i.e. each disease group vs other skin inflammatory diseases) were used to generate disease-related gene signatures. P value  <  0.05, and fold change  > 2 were used as cutoffs to choose specific classifiers. b, Volcano plot of upregulated genes in COVID-19 compared with CLE skin lesions. Source data
Extended Data Fig. 3
Extended Data Fig. 3. Macrophages accumulate around vessels exhibiting prominent endotheliopathy in COVID-19 skin lesions.
a, Confocal microscopy images of CD3+ T cells, CD123+ plasmacytoid dendritic cells, MPO+ neutrophils, and CD163+ macrophages in skin lesions from CLE (top row) and COVID-19 (bottom row). Images are representative of 10 patients with COVID-19 and 5 patients with CLE. b, Percentages of CD163+ macrophages present at different distances from blood vessel in COVID-19 skin lesions (n = 9). c, Representative histopathology image of a dermal blood vessel in COVID-19 skin lesions (H&E stain). Blood vessel, dashed line. d, Endothelial cell swelling index, a measure of endotheliopathy, quantified in COVID-19 (n = 10) and CLE skin lesions (n = 6). P values were obtained with one-way ANOVA followed by Tukey’s multiple comparison test (b) and two-tailed Student’s t-test (d). Source data
Extended Data Fig. 4
Extended Data Fig. 4. Perivascular macrophages engulf dying cells.
a, b, Confocal microscopy images of representative COVID-19 skin lesion stained for CD163 (green), cleaved caspase-3 (red) and DNA (DAPI). Images are representative of 10 patients with COVID-19. Blood vessel, dashed line.
Extended Data Fig. 5
Extended Data Fig. 5. Patient characteristics and histopathological analyses of a post-mortem COVID-19-affected lung.
a, Representative histopathology image of a COVID-19 lung in the early phase of DAD with extensive hyaline membranes (left) or in the late phase of DAD with fibrosis obliterating the alveolar lumina (right) (H&E stain). Arrows indicate hyaline membranes. b, Clinical parameters of the 8 patients with COVID-19 selected for the study. HC, hydroxychloroquine; Phase of the diffuse alveolar damage defined based on pure presence of hyaline membranes (exudative) or fibrotic changes (proliferative); * limit of detection is 20.8 copies per reaction (c/r) for RdRp gene, and 5.4 c/r for E gene; ** spike and nucleocapsid antibody; MxA-staining defined as high (>50% cells with intermediate to strong positive staining), or low (< 50%). c, Confocal microscopy images of representative COVID-19 lung section stained for CD31 (green) and p-STING (red). Arrow indicates an endothelial cell with activated STING. d, Proportions of CD163+ macrophages and CD31+ endothelial cells among p-STING+ cells in COVID-19 lungs (n = 4). Source data
Extended Data Fig. 6
Extended Data Fig. 6. Responses from distinct cell types after SARS-CoV-2 infection in the LoC system.
a, mRNA expression levels of SARS-CoV-2 N gene in epithelial and endothelial cells at 3 dpi in 2-cell component LoCs. b, Representative 3D views of the airway surface of a LoC infected with SARS-CoV-2 at 3 dpi. Areas with high levels of IFNβ (bright pink) are shown as surfaces. Macrophage surfaces are depicted in green, and nuclear labelling in purple. c, mRNA expression levels of indicated genes in the epithelial cells at 3 dpi in uninfected (n = 3), infected (n = 5), and H-151-treated (n = 5) 2-cell component LoCs. d, Representative 3D views of the vascular face from uninfected and infected LoCs; PMA-activated WT and cGAS−/− THP-1 cells were added to vascular layer 2 days after infection. LoCs were analysed at 3 dpi by quantifying the total volume with high IFNβ expression/volume with high IFNβ expression within macrophages from uninfected (n = 6/ n = 4 fields of view), infected chips with WT THP-1 cells added (n = 9/ n = 6 fields of view) and infected chips with cGAS−/− THP-1 cells added (n = 6/ n = 5 fields of view) across n = 2 LoCs in each case. e, Representative 3D views of the vascular face from uninfected and infected 3-component LoCs; volumes with high levels of cleaved caspase-3 (amber) are shown as surfaces. Quantification of the total volume with high cleaved caspase-3 expression from uninfected (n = 3 fields of view) and infected (n = 5 fields of view) from n = 1 LoC in each case. ‘3C’ refers to 3-cell component (epithelial cells, endothelial cells, and macrophages) LoCs. Bars represent mean ± s.e.m.; P values were calculated by a two-tailed Mann–Whitney test (a) or one-way ANOVA followed by Tukey’s multiple comparison tests (c, d). Source data
Extended Data Fig. 7
Extended Data Fig. 7. Effect of innate immune sensors on endothelial cell response after SARS-CoV-2 infection in the LoC model.
a, Representative images of the epithelial and endothelial cells on LoC infected with SARS-CoV-2 at 3 dpi (above). Western blot characterization of STING expression in epithelial and endothelial cells treated with control (ctrl) and STING shRNAs (below, left). Modal value of CD31 expression in the endothelial layer of LoCs reconstituted with control or STING shRNA treated epithelial or endothelial cells (below, right). Each data point was calculated from a maximum intensity projection of CD31 expression from n = 7 (sh Ctrl - epithelial/sh STING - endothelial), n = 8 (sh Ctrl – epithelial/sh Ctrl – endothelial and sh STING – epithelial/sh Ctrl – endothelial) and n = 9 (sh STING – epithelial/sh STING = endothelial) fields of view from n = 1 LoC in each case. b, Representative 3D views at 3 dpi of the vascular surface of a LoCs reconstituted with endothelial cells treated with ctrl or MAVS shRNA and infected with SARS-CoV-2. Volumes with high levels of IFNβ (bright pink) are shown as surfaces. Quantification of the volume with high IFNβ expression from ctrl (n = 5 fields of view) and MAVS shRNA (n = 4 fields of view) from n = 1 LoC in each case. Reduction of MAVS mRNA in endothelial cells after shRNA transduction (right). ‘2C’ refers to 2-cell component (epithelial cells, endothelial cells) LoCs. Bars represent mean ± s.e.m.; P values were calculated by a one-way ANOVA test followed by Tukey multiple comparison tests (a) or a two-tailed Mann–Whitney test (b). For gel source data, see Supplementary Fig. 1. Source data
Extended Data Fig. 8
Extended Data Fig. 8. Disruption of mitochondrial homeostasis in endothelial cells after SARS-CoV-2 infection.
a, Functional interactions between proteins with significantly altered expression identified by a pairwise analysis depicted via an interaction network generated from StringDB and clustered with the MCL algorithm. Functional annotations (GO Biological Processes/ KEGG pathways) relevant to mitochondrial function are indicated. b, Heat map of data from mitochondrial proteins with significantly altered expression identified via a time-course analysis. c, d, Serial block scanning electron microscope image of endothelial cells in LoC system (c) or transmission electron microscopy image of endothelial cells in COVID-19 skin lesion (d). Arrows indicate damaged mitochondria with loss of cristae morphology with a magnified inset at bottom right. As control, arrowhead indicates an intact mitochondrium with a magnified inset at top right. e, Representative 3D views at 3 dpi of the vascular surface of a LoCs reconstituted with endothelial cells treated or untreated with 20 μM ddC for 7 days in total and infected with SARS-CoV-2. Total mtDNA in endothelial cells was quantified by qPCR (right). Volumes with high levels of IFNβ (bright pink) are shown as surfaces. Quantification of the volume with high IFNβ expression from infected (n = 11 fields of view) and ddC treated infected (n = 8 fields of view) from n = 1 and n = 2 LoCs respectively. ‘2C’ refers to 2-cell component (epithelial cells, endothelial cells) LoCs. Bars represent mean ± s.e.m.; P values were calculated by a two-tailed Mann–Whitney test (e). Source data
Extended Data Fig. 9
Extended Data Fig. 9. Prophylactic STING inhibition reduces pathology and inflammatory gene expression in the late stages of SARS-CoV-2 infection.
ad, Mice were infected with SARS-CoV-2 infection (intranasal; 1x104 PFU/mouse) and intraperitoneal administration of vehicle or H-151 was started at 1 day prior to infection. TUNEL assay performed on the infected lung sections collected at 3 or 6 dpi is shown (a). mRNA levels of indicated genes isolated from uninfected and infected mouse lungs at 6 dpi were analysed by RT–qPCR (b). Tissue lysates from the lungs were subjected to Western blotting (c). Viral burden in the lungs and brains was analysed at 6 dpi by plaque assay for infectious virus (d). Numbers are for a and b uninfected (n = 4), day 3 samples (n = 4), day 6 samples (n = 7); for d, uninfected (n = 4) and infected (n = 5). Throughout the figure, bars represent mean ± s.e.m.; P values were calculated by one-way ANOVA followed by Tukey multiple comparison tests. For gel source data, see Supplementary Fig. 1. Source data
Extended Data Fig. 10
Extended Data Fig. 10. Therapeutic inhibition of STING reduces pathology and inflammatory gene expression after SARS-CoV-2 infection.
a, b, Mice were infected with SARS-CoV-2 infection (intranasal; 1 × 104 PFU per mouse) and intraperitoneal administration of vehicle or H-151 was started at 2 dpi. mRNA was isolated from uninfected and infected mouse lungs and relative expression of indicated genes were analysed by RT–qPCR (a). Viral burden in the lungs and brains was analysed at 6 dpi by plaque assay for infectious virus (b). c, Model of the involvement of the cGAS–STING pathway in severe SARS-CoV-2 infection created with biorender.com. Numbers are uninfected (n = 4), infected (n = 5) (a, b). Throughout the figure, bars represent mean ± s.e.m.; P values were calculated by one-way ANOVA followed by Tukey multiple comparison tests. Source data

Comment in

Similar articles

Cited by

References

    1. Huang C, et al. Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China. Lancet. 2020;395:497–506. - PMC - PubMed
    1. Gupta A, et al. Extrapulmonary manifestations of COVID-19. Nat. Med. 2020;26:1017–1032. - PubMed
    1. Lucas C, et al. Longitudinal analyses reveal immunological misfiring in severe COVID-19. Nature. 2020;584:463–469. - PMC - PubMed
    1. Nienhold R, et al. Two distinct immunopathological profiles in autopsy lungs of COVID-19. Nat. Commun. 2020;11:5086. - PMC - PubMed
    1. Park A, Iwasaki A. Type I and type III interferons—induction, signaling, evasion, and application to combat COVID-19. Cell Host Microbe. 2020;27:870–878. - PMC - PubMed

Publication types

MeSH terms