The innate immune defence of multicellular organisms against microbial pathogens requires cellular collaboration. Information exchange allowing immune cells to collaborate is generally attributed to soluble protein factors secreted by pathogen-sensing cells. Cytokines, such as type I interferons (IFNs), serve to alert non-infected cells to the possibility of pathogen challenge1. Moreover, in conjunction with chemokines they can instruct specialized immune cells to contain and eradicate microbial infection. Several receptors and signalling pathways exist that couple pathogen sensing to the induction of cytokines, whereas cytosolic recognition of nucleic acids seems to be exquisitely important for the activation of type I IFNs, master regulators of antiviral immunity2. Cytosolic DNA is sensed by the receptor cyclic GMP-AMP (cGAMP) synthase (cGAS), which catalyses the synthesis of the second messenger cGAMP(2′-5′)3,4,5,6,7,8. This molecule in turn activates the endoplasmic reticulum (ER)-resident receptor STING9,10,11, thereby inducing an antiviral state and the secretion of type I IFNs. Here we find in murine and human cells that cGAS-synthesized cGAMP(2′-5′) is transferred from producing cells to neighbouring cells through gap junctions, where it promotes STING activation and thus antiviral immunity independently of type I IFN signalling. In line with the limited cargo specificity of connexins, the proteins that assemble gap junction channels, most connexins tested were able to confer this bystander immunity, thus indicating a broad physiological relevance of this local immune collaboration. Collectively, these observations identify cGAS-triggered cGAMP(2′-5′) transfer as a novel host strategy that serves to rapidly convey antiviral immunity in a transcription-independent, horizontal manner.
Subscribe to Journal
Get full journal access for 1 year
only $3.90 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Tax calculation will be finalised during checkout.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
Sadler, A. J. & Williams, B. R. Interferon-inducible antiviral effectors. Nature Rev. Immunol. 8, 559–568 (2008)
Goubau, D., Deddouche, S. & Reis, E. S. C. Cytosolic sensing of viruses. Immunity 38, 855–869 (2013)
Wu, J. et al. Cyclic GMP-AMP is an endogenous second messenger in innate immune signaling by cytosolic DNA. Science 339, 826–830 (2013)
Sun, L., Wu, J., Du, F., Chen, X. & Chen, Z. J. Cyclic GMP-AMP synthase is a cytosolic DNA sensor that activates the type I interferon pathway. Science 339, 786–791 (2013)
Gao, P. et al. Cyclic [G(2',5′)pA(3′,5′)p] is the metazoan second messenger produced by DNA-activated cyclic GMP-AMP synthase. Cell 153, 1094–1107 (2013)
Diner, E. J. et al. The innate immune DNA sensor cGAS produces a noncanonical cyclic dinucleotide that activates human STING. Cell Rep. 3, 1355–1361 (2013)
Ablasser, A. et al. cGAS produces a 2′-5′-linked cyclic dinucleotide second messenger that activates STING. Nature 498, 380–384 (2013)
Zhang, X. et al. Cyclic GMP-AMP containing mixed phosphodiester linkages is an endogenous high-affinity ligand for STING. Mol. Cell 51, 226–235 (2013)
Ishikawa, H. & Barber, G. N. STING is an endoplasmic reticulum adaptor that facilitates innate immune signalling. Nature 455, 674–678 (2008)
Zhong, B. et al. The adaptor protein MITA links virus-sensing receptors to IRF3 transcription factor activation. Immunity 29, 538–550 (2008)
Sun, W. et al. ERIS, an endoplasmic reticulum IFN stimulator, activates innate immune signaling through dimerization. Proc. Natl Acad. Sci. USA 106, 8653–8658 (2009)
Kasper, C. A. et al. Cell-cell propagation of NF-κB transcription factor and MAP kinase activation amplifies innate immunity against bacterial infection. Immunity 33, 804–816 (2010)
Hamada, N., Matsumoto, H., Hara, T. & Kobayashi, Y. Intercellular and intracellular signaling pathways mediating ionizing radiation-induced bystander effects. J. Radiat. Res. 48, 87–95 (2007)
Patel, S. J., King, K. R., Casali, M. & Yarmush, M. L. DNA-triggered innate immune responses are propagated by gap junction communication. Proc. Natl Acad. Sci. USA 106, 12867–12872 (2009)
Ablasser, A. & Hornung, V. DNA sensing unchained. Cell Res. 23, 585–587 (2013)
Ishikawa, H., Ma, Z. & Barber, G. N. STING regulates intracellular DNA-mediated, type I interferon-dependent innate immunity. Nature 461, 788–792 (2009)
Cavlar, T., Deimling, T., Ablasser, A., Hopfner, K. P. & Hornung, V. Species-specific detection of the antiviral small-molecule compound CMA by STING. EMBO J. 32, 1440–1450 (2013)
Ablasser, A. et al. RIG-I-dependent sensing of poly(dA:dT) through the induction of an RNA polymerase III-transcribed RNA intermediate. Nature Immunol. 10, 1065–1072 (2009)
Laird, D. W. Life cycle of connexins in health and disease. Biochem. J. 394, 527–543 (2006)
Juul, M. H., Rivedal, E., Stokke, T. & Sanner, T. Quantitative determination of gap junction intercellular communication using flow cytometric measurement of fluorescent dye transfer. Cell Adhes. Commun. 7, 501–512 (2000)
Butterweck, A., Gergs, U., Elfgang, C., Willecke, K. & Traub, O. Immunochemical characterization of the gap junction protein connexin45 in mouse kidney and transfected human HeLa cells. J. Membr. Biol. 141, 247–256 (1994)
Langlois, S., Cowan, K. N., Shao, Q., Cowan, B. J. & Laird, D. W. Caveolin-1 and -2 interact with connexin43 and regulate gap junctional intercellular communication in keratinocytes. Mol. Biol. Cell 19, 912–928 (2008)
Mali, P. et al. RNA-guided human genome engineering via Cas9. Science 339, 823–826 (2013)
Cong, L. et al. Multiplex genome engineering using CRISPR/Cas systems. Science 339, 819–823 (2013)
Gall, A. et al. Autoimmunity initiates in nonhematopoietic cells and progresses via lymphocytes in an interferon-dependent autoimmune disease. Immunity 36, 120–131 (2012)
Stetson, D. B. & Medzhitov, R. Recognition of cytosolic DNA activates an IRF3-dependent innate immune response. Immunity 24, 93–103 (2006)
Civril, F. et al. Structural mechanism of cytosolic DNA sensing by cGAS. Nature 498, 332–337 (2013)
Uzé, G. et al. Domains of interaction between alpha interferon and its receptor components. J. Mol. Biol. 243, 245–257 (1994)
Kastenmüller, W. et al. Peripheral prepositioning and local CXCL9 chemokine-mediated guidance orchestrate rapid memory CD8+ T cell responses in the lymph node. Immunity 38, 502–513 (2013)
Schmid-Burgk, J. L., Schmidt, T., Kaiser, V., Honing, K. & Hornung, V. A ligation-independent cloning technique for high-throughput assembly of transcription activator-like effector genes. Nature Biotechnol. 31, 76–81 (2013)
We thank M. Pellegrin for providing us with LL171 cells; W. Kastenmüller for MVA NP-S-GFP; J. Bennink for vaccinia virus; K.-P. Hopfner for recombinant cGAS; and K. Willecke for connexin expression constructs and anti-CX43 antibody. J.L.S.-B. is supported by the Studienstiftung des Deutschen Volkes. I.H. is supported by a BONFOR-funded thesis project. A.A., E.L. and V.H. are supported by the excellence cluster ImmunoSensation. V.H. is supported by grants from the German Research Foundation (SFB670 and SFB704) and the European Research Council (ERC 243046).
The authors declare no competing financial interests.
Extended data figures and tables
Extended Data Figure 1 Stable overexpression of cGAS in HEK cells induces activation of HEK STING cells in trans.
a, HEK cells and HEK STING cells were co-cultured with increasing amounts of HEK cGAS* cells (ratios ranging from 1:0.25 to 1:0.0156 HEK/HEK STING:HEK cGAS*). Co-cultures were transfected with pIFN-β-GLuc and after 20 h transactivation of the reporter construct was assessed. Mean and s.e.m. (biological duplicates) of one representative experiment out of two independent experiments are depicted. b, HEK STING cells were co-cultured with HEK cells or HEK cGAS* cells (ratios HEK STING:HEK/HEK cGAS* = 1:0.25) for 4 h and phosphorylation of IRF3 was determined in the cellular lysates by immunoblotting. c, Kinetics of IRF3 phosphorylation of HEK STING and HEK cGAS* co-cultures (ratio HEK STING:HEK cGAS* = 1:0.25) are depicted. CMA served as a control stimulus. Representative experiments of two independent experiments are shown (b, c).
Extended Data Figure 2 DNA-triggered cGAS activation induces IFN-β expression in adjacent cells via STING.
a, b, HEK cells and HEK STING cells were co-cultured with increasing amounts of HEK cGASlow cells (a) or primary MEFs (b) as depicted in Fig. 2c, d (ratio of HEK/HEK cGASlow/MEFs was titrated ranging from 1:0.125 to 1:0.0156). Co-cultures were transfected with pIFN-β-GLuc and after 20 h transactivation of the reporter construct was assessed. Mean and s.e.m. of six experiments (a) or eight experiments (b) are depicted (*P < 0.05, **P < 0.01). c, Schematic view of the experimental set-up is shown: primary MEFs were silenced for cGAS expression using two independent siRNAs targeting cGAS or a control siRNA. Forty-eight hours later MEFs were co-cultured with HEK STING cells and then transfected with pIFN-β-GLuc and after an additional period of twenty hours transactivation of the reporter construct was assessed. d, cGAS expression in MEFs treated as in c was analysed by qPCR (data normalized to control siRNA condition). Mean values and s.e.m. of two independent experiments are depicted. e, Mean values and s.e.m. of duplicate measurements of one representative experiment, in which the ratio of HEK STING cells over MEFs was titrated ranging from 1:0.5 to 1:0.0625 is depicted. f, Mean values and s.e.m. (data normalized to control siRNA condition) of three independent experiments are depicted (HEK STING/MEF ratio is 1:0.5) (**P < 0.01).
Extended Data Figure 3 Overexpression of a RIG-I-stimulatory RNA molecule cannot confer activation of bystander cells.
a, HEK cells were transfected with empty vector (Cont.), cGAS–GFP (cGAS) or a construct encoding a RIG-I-stimulatory shRNA molecule (shRNA). Twenty hours after transfection cells were collected, washed and added onto HEK cells or HEK STING cells expressing pIFN-β-GLuc. After 20 h of co-culture luciferase activity was measured. b, HEK STING cells were transfected as in a together with pIFN-β-GLuc and luciferase activity was measured 20 h after transfection. Mean and s.e.m. (biological duplicates) of one representative experiment out of two independent experiments are shown.
Extended Data Figure 4 cGAS-dependent bystander cell activation requires direct cell-to-cell contact.
a, d, Schematic view of the experimental set-up is depicted. HEK STING cells or LL171 cells were left untreated (i) or co-cultured with HEK cells (ii) or HEK cGAS* cells in the presence (iii) or absence (iv) of a trans-well system. b, e, After 4 h of co-culture, phosphorylation of IRF3 was determined in cellular lysates via immunoblotting. c, After 14 h, relative induction of IFNB and CXCL10 in HEK STING cells was analysed via qPCR. In addition, HEK STING cells were transfected with pIFN-β-GLuc 20 h before donor cells were added. After 18 h luciferase activity in HEK STING cells was assessed. f, Relative induction of Ifnb in LL171 cells was determined via qPCR after 4 h. Furthermore, transactivation of an endogenous ISRE-reporter construct was assessed in LL171 cells after 14 h. g, LL171 cells were transfected with siRNA targeting STING or a control siRNA. Forty-eight hours after siRNA transfection relative expression of STING was determined by qPCR. Mean and s.e.m. of duplicate measurements of two independent experiments is shown. h, LL171 cells from g were co-cultured with HEK cGAS* cells and after 6 h phosphorylation of IRF3 was determined by immunoblotting. Mean and s.e.m. (biological duplicates) of one representative experiment out of two independent experiments are shown (c, f) or one representative experiment out of two independent experiments is shown (b, e, h).
a, b, LL171 cells were pre-treated with CBX (100 μM, 150 μM and 200 μM) 3 h before addition of HEK cells or HEK cGAS* cells. In addition, LL171 cells were stimulated with CMA (a) or recombinant IFN-α (250 U ml−1). Phosphorylation of IRF3 (a) and luciferase activity of an endogenous ISRE-reporter construct (b) was determined in the cellular lysate 4 h and 14 h after stimulation, respectively. Mean and s.e.m. (biological duplicates) of one representative experiment out of two independent experiments are shown.
Extended Data Figure 6 Scrape loading assays reveal a direct transfer of cGAMP(2′-5′) through gap junctions.
a, b, HEK STING cells (STING in red) were either incubated with cGAMP(2′-5′), CMA or scratched in the presence of cGAMP(2′-5′). The latter condition was also performed in the presence of 150 μM CBX. STING activation was visualized 8 h later, whereas dashed lines follow the scratch margins and arrows highlight areas of STING complex assembly. Representative images of four independent experiments are shown (a) and STING-activated cells were quantified and depicted in a scatter plot (b). ***P < 0.001.
Extended Data Figure 7 Deep sequencing results of CX43/CX45-targeted HEK STING cells generated by CRISPR/Cas9-mediated genome editing and western blot analysis of HEK STING CX43/45DKO cells.
a, c, For generating HEK STING CX43/45DKO cells, a targeting strategy was devised based on hybrid gRNA sequences targeting Cas9 to the first coding exons of the respective genes. The open reading frame of CX43 (a) and CX45 (c) are delineated in red. PAM, protospacer adjacent motif. b, d, Deep-sequencing-based allele calls of targeted HEK STING cell lines as well as control cell lines are presented. Mutations are indicated in red letters, whereas the numbers in brackets indicate the net frame shifts. e, HEK STING CX43/45WT cells and HEK STING CX43/45DKO cells were analysed for CX43 and CX45 expression via immunoblotting. Of note, HEK STING CX43/45DKO cell line 2 harbours an in-frame deletion for CX43 (−12 bp) and for CX45 (−18 bp), which probably accounts for the faint signal observed in the immunoblot (asterisk). Data are representative of three independent experiments.
Extended Data Figure 8 Scrape loading of cGAMP(2′-5′) into HEK STING CX43/45WT and CX43/45DKO cells and overexpression of distinct connexin members in HEK STING CX43/45DKO cells.
a, Fluorescence images of HEK STING CX43/45WT and CX43/45DKO cells (STING in red) wounded and overlaid with cGAMP(2′-5′). Wounded cells without addition of cGAMP(2′-5′) served as controls. Dashed line outlines the scratch margins. Representative images of n = 2 experiments are shown. b, Fluorescence images of HEK STING CX43/45DKO co-cultured with HEK cGAS* and transfected with empty vector (pCI) and distinct members of human or murine connexins as indicated are depicted (pCI, CMA and mmCx45 as depicted in Fig. 4 are shown). Multimerization of STING was visualized 20 h after transfection. CMA stimulation for 8 h served as positive control. Representative images of n = 2 experiments are depicted.
Extended Data Figure 9 MVA-infected MEFs activate HEK STING cells in trans in a gap-junction dependent fashion.
a, Schematic view of the experimental set-up for b, c: MEFs were infected with MVA–GFP for 3 h, washed three times and then loaded onto HEK cells or HEK STING cells that were then incubated overnight. Subsequently, human IFN-β expression was analysed by qPCR. b, c, A representative experiment with a titration of MVA–GFP (1.6 × 106, 0.8 × 106 and 0.16 × 106 virus particles per ml) is depicted (b) and mean values and s.e.m. of three independent experiments at a concentration of 1.6 × 106 virus particles/ml are shown (c). d, Experiments were conducted as in b, now using HEK STING CX43/CX45WT and HEK STING CX43/CX45DKO cell lines as recipient cells. One representative experiment out of two independent experiments using 3.2 × 106 and 1.6 × 106 virus particles per ml is depicted in d.
Extended Data Figure 10 Schematic model of the mechanism of gap-junction-mediated local immune collaboration.
On infection with a DNA virus, a cell senses the presence of cytosolic viral DNA through the receptor cGAS, thus activating the synthesis of the second messenger cGAMP(2′-5′). cGAMP(2′-5′) can pass through gap junctions into the cytosol of neighbouring cells, where it is detected by STING. The subsequent induction of an antiviral transcriptional program thus protects bystander cells from viral infection after the virus has successfully replicated in the cell initially infected.
HEK cGAS* cells loaded with calcein (in green) were added onto HEK STING cells (STING in red) and dye transfer and STING multimerization were continuously analyzed by confocal fluorescence microscopy for up to 210 min. Suppl. Video 1 shows the complete visual field. (AVI 605 kb)
HEK cGAS* cells loaded with calcein (in green) were added onto HEK STING cells (STING in red) and dye transfer and STING multimerization were continuously analyzed by confocal fluorescence microscopy for up to 210 min. Suppl. Video 2 highlights one particular region in higher resolution. (AVI 8129 kb)
About this article
Cite this article
Ablasser, A., Schmid-Burgk, J., Hemmerling, I. et al. Cell intrinsic immunity spreads to bystander cells via the intercellular transfer of cGAMP. Nature 503, 530–534 (2013). https://doi.org/10.1038/nature12640
Virus Research (2021)
Fundamental Research (2021)
Signal Transduction and Targeted Therapy (2021)
European Journal of Immunology (2021)
Trends in Biochemical Sciences (2021)