Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

R-loop-derived cytoplasmic RNA–DNA hybrids activate an immune response

Abstract

R-loops are RNA–DNA-hybrid-containing nucleic acids with important cellular roles. Deregulation of R-loop dynamics can lead to DNA damage and genome instability1, which has been linked to the action of endonucleases such as XPG2,3,4. However, the mechanisms and cellular consequences of such processing have remained unclear. Here we identify a new population of RNA–DNA hybrids in the cytoplasm that are R-loop-processing products. When nuclear R-loops were perturbed by depleting the RNA–DNA helicase senataxin (SETX) or the breast cancer gene BRCA1 (refs. 5,6,7), we observed XPG- and XPF-dependent cytoplasmic hybrid formation. We identify their source as a subset of stable, overlapping nuclear hybrids with a specific nucleotide signature. Cytoplasmic hybrids bind to the pattern recognition receptors cGAS and TLR3 (ref. 8), activating IRF3 and inducing apoptosis. Excised hybrids and an R-loop-induced innate immune response were also observed in SETX-mutated cells from patients with ataxia oculomotor apraxia type 2 (ref. 9) and in BRCA1-mutated cancer cells10. These findings establish RNA–DNA hybrids as immunogenic species that aberrantly accumulate in the cytoplasm after R-loop processing, linking R-loop accumulation to cell death through the innate immune response. Aberrant R-loop processing and subsequent innate immune activation may contribute to many diseases, such as neurodegeneration and cancer.

This is a preview of subscription content, access via your institution

Access options

Buy article

Get time limited or full article access on ReadCube.

$32.00

All prices are NET prices.

Fig. 1: The loss of R-loop resolution factors leads to XPG-dependent cytoplasmic RNA–DNA hybrid accumulation.
Fig. 2: cytoDRIP–seq shows that cytoplasmic RNA–DNA hybrids are derived from a subset of nuclear R-loops.
Fig. 3: Cytoplasmic RNA–DNA hybrids derived from R-loop processing activate IRF3 signalling and induce apoptosis.
Fig. 4: R-loop-derived cytoplasmic RNA–DNA hybrids trigger IRF3 signalling through the cGAS and TLR3 receptors.

Data availability

All sequencing data generated in this work have been deposited at the Gene Expression Omnibus (GEO) under accession number GSE178841. For nuclear DRIP–seq, datasets under accession number GSE134084 were used from https://doi.org/10.1093/nar/gkaa500Source data are provided with this paper.

Code availability

Further code information is available on request from the corresponding author.

References

  1. Crossley, M. P., Bocek, M. & Cimprich, K. A. R-loops as cellular regulators and genomic threats. Mol. Cell 73, 398–411 (2019).

    Article  CAS  Google Scholar 

  2. Sollier, J. et al. Transcription-coupled nucleotide excision repair factors promote R-loop-induced genome instability. Mol. Cell 56, 777–785 (2014).

    Article  CAS  Google Scholar 

  3. Makharashvili, N. et al. Sae2/CtIP prevents R-loop accumulation in eukaryotic cells. eLife 7, e42733 (2018).

    Article  Google Scholar 

  4. Cristini, A. et al. Dual processing of R-loops and topoisomerase I induces transcription-dependent DNA double-strand breaks. Cell Rep. 28, 3167–3181 (2019).

    Article  CAS  Google Scholar 

  5. Santos-Pereira, J. M. & Aguilera, A. R loops: new modulators of genome dynamics and function. Nat. Rev. Genet. 16, 583–597 (2015).

    Article  CAS  Google Scholar 

  6. Hatchi, E. et al. BRCA1 recruitment to transcriptional pause sites is required for R-loop-driven DNA damage repair. Mol. Cell 57, 636–647 (2015).

    Article  CAS  Google Scholar 

  7. Skourti-Stathaki, K., Proudfoot, N. J. & Gromak, N. Human senataxin resolves RNA/DNA hybrids formed at transcriptional pause sites to promote Xrn2-dependent termination. Mol. Cell 42, 794–805 (2011).

    Article  CAS  Google Scholar 

  8. Schlee, M. & Hartmann, G. Discriminating self from non-self in nucleic acid sensing. Nat. Rev. Immunol. 16, 566–580 (2016).

    Article  CAS  Google Scholar 

  9. Suraweera, A. et al. Senataxin, defective in ataxia oculomotor apraxia type 2, is involved in the defense against oxidative DNA damage. J. Cell Biol. 177, 969–979 (2007).

    Article  CAS  Google Scholar 

  10. Harding, S. M. et al. Mitotic progression following DNA damage enables pattern recognition within micronuclei. Nature 548, 466–470 (2017).

    Article  ADS  CAS  Google Scholar 

  11. He, Y. et al. NF-κB-induced R-loop accumulation and DNA damage select for nucleotide excision repair deficiencies in adult T cell leukemia. Proc. Natl Acad. Sci. USA 118, e2005568118 (2021).

    Article  CAS  Google Scholar 

  12. Crossley, M. P. et al. Catalytically-inactive, purified RNaseH1: a specific and sensitive probe for RNA-DNA hybrid imaging. J. Cell Biol. 220, e202101092 (2021).

    Article  CAS  Google Scholar 

  13. Nguyen, H. D. et al. Functions of replication protein A as a sensor of R loops and a regulator of RNaseH1. Mol. Cell 65, 832–847 (2017).

    Article  CAS  Google Scholar 

  14. Bernardini, J. P. et al. Parkin inhibits BAK and BAX apoptotic function by distinct mechanisms during mitophagy. EMBO J. 38, e99916 (2019).

    Article  Google Scholar 

  15. Tigano, M., Vargas, D. C., Tremblay-Belzile, S., Fu, Y. & Sfeir, A. Nuclear sensing of breaks in mitochondrial DNA enhances immune surveillance. Nature 591, 477–481 (2021).

    Article  ADS  CAS  Google Scholar 

  16. Köhler, A. & Hurt, E. Exporting RNA from the nucleus to the cytoplasm. Nat. Rev. Mol. Cell Biol. 8, 761–773 (2007).

    Article  Google Scholar 

  17. Chatzidoukaki, O. et al. R-loops trigger the release of cytoplasmic ssDNAs leading to chronic inflammation upon DNA damage. Sci. Adv. 7, eabj5769 (2021).

    Article  ADS  CAS  Google Scholar 

  18. Koo, C. X. et al. RNA polymerase III regulates cytosolic RNA:DNA hybrids and intracellular microRNA expression. J. Biol. Chem. 290, 7463–7473 (2015).

    Article  CAS  Google Scholar 

  19. Smolka, J. A., Sanz, L. A., Hartono, S. R. & Chédin, F. Recognition of RNA by the S9.6 antibody creates pervasive artifacts when imaging RNA:DNA hybrids. J. Cell Biol. 220, e202004079 (2021).

    Article  CAS  Google Scholar 

  20. Petermann, E., Lan, L. & Zou, L. Sources, resolution and physiological relevance of R-loops and RNA-DNA hybrids. Nat. Rev. Mol. Cell Biol. 23, 521–540 (2022).

    Article  CAS  Google Scholar 

  21. Crossley, M. P., Bocek, M. J., Hamperl, S., Swigut, T. & Cimprich, K. A. qDRIP: a method to quantitatively assess RNA-DNA hybrid formation genome-wide. Nucleic Acids Res. 48, e84 (2020).

    Article  CAS  Google Scholar 

  22. Sanz, L. A. et al. Prevalent, dynamic, and conserved R-loop structures associate with specific epigenomic signatures in mammals. Mol. Cell 63, 167–178 (2016).

    Article  CAS  Google Scholar 

  23. Lim, G. & Hohng, S. Single-molecule fluorescence studies on cotranscriptional G-quadruplex formation coupled with R-loop formation. Nucleic Acids Res. 48, 9195–9203 (2020).

    Article  CAS  Google Scholar 

  24. Ginno, P. A., Lott, P. L., Christensen, H. C., Korf, I. & Chédin, F. R-loop formation is a distinctive characteristic of unmethylated human CpG island promoters. Mol. Cell 45, 814–825 (2012).

    Article  CAS  Google Scholar 

  25. Ginno, P. A., Lim, Y. W., Lott, P. L., Korf, I. & Chédin, F. GC skew at the 5′ and 3′ ends of human genes links R-loop formation to epigenetic regulation and transcription termination. Genome Res. 23, 1590–1600 (2013).

    Article  CAS  Google Scholar 

  26. Xu, W. et al. The R-loop is a common chromatin feature of the Arabidopsis genome. Nat. Plants 3, 704–714 (2017).

    Article  CAS  Google Scholar 

  27. Shen, Y. J. et al. Genome-derived cytosolic DNA mediates type I interferon-dependent rejection of B cell lymphoma cells. Cell Rep. 11, 460–473 (2015).

    Article  CAS  Google Scholar 

  28. Coquel, F. et al. SAMHD1 acts at stalled replication forks to prevent interferon induction. Nature 557, 57–61 (2018).

    Article  ADS  CAS  Google Scholar 

  29. Mackenzie, K. J. et al. cGAS surveillance of micronuclei links genome instability to innate immunity. Nature 548, 461–465 (2017).

    Article  ADS  CAS  Google Scholar 

  30. Weinreb, J. T. et al. Excessive R-loops trigger an inflammatory cascade leading to increased HSPC production. Dev. Cell 56, 627–640 (2021).

    Article  CAS  Google Scholar 

  31. Giordano, A. M. S. et al. DNA damage contributes to neurotoxic inflammation in Aicardi–Goutières syndrome astrocytes. J. Exp. Med. 219, e20211121 (2022).

    Article  CAS  Google Scholar 

  32. Suter, M. A. et al. cGAS-STING cytosolic DNA sensing pathway is suppressed by JAK2-STAT3 in tumor cells. Sci. Rep. 11, 7243 (2021).

    Article  CAS  Google Scholar 

  33. Cristini, A. et al. RNase H2, mutated in Aicardi–Goutières syndrome, resolves co-transcriptional R-loops to prevent DNA breaks and inflammation. Nat. Commun. 13, 2961 (2022).

    Article  ADS  CAS  Google Scholar 

  34. Chawla-Sarkar, M. et al. Apoptosis and interferons: role of interferon-stimulated genes as mediators of apoptosis. Apoptosis 8, 237–249 (2003).

    Article  CAS  Google Scholar 

  35. Boulares, A. H. et al. Role of poly(ADP-ribose) polymerase (PARP) cleavage in apoptosis. Caspase 3-resistant PARP mutant increases rates of apoptosis in transfected cells. J. Biol. Chem. 274, 22932–22940 (1999).

    Article  CAS  Google Scholar 

  36. Porter, A. G. & Jänicke, R. U. Emerging roles of caspase-3 in apoptosis. Cell Death Differ. 6, 99–104 (1999).

    Article  CAS  Google Scholar 

  37. Wajant, H., Pfizenmaier, K. & Scheurich, P. Tumor necrosis factor signaling. Cell Death Differ. 10, 45–65 (2003).

    Article  CAS  Google Scholar 

  38. Zhou, Y., He, C., Wang, L. & Ge, B. Post-translational regulation of antiviral innate signaling. Eur. J. Immunol. 47, 1414–1426 (2017).

    Article  CAS  Google Scholar 

  39. Mankan, A. K. et al. Cytosolic RNA:DNA hybrids activate the cGAS-STING axis. EMBO J. 33, 2937–2946 (2014).

    Article  CAS  Google Scholar 

  40. Blasius, A. L. & Beutler, B. Intracellular toll-like receptors. Immunity 32, 305–315 (2010).

    Article  CAS  Google Scholar 

  41. Abu-Remaileh, M. et al. Lysosomal metabolomics reveals V-ATPase- and mTOR-dependent regulation of amino acid efflux from lysosomes. Science 358, 807–813 (2017).

    Article  ADS  CAS  Google Scholar 

  42. Yeo, A. J. et al. R-loops in proliferating cells but not in the brain: implications for AOA2 and other autosomal recessive ataxias. PLoS ONE 9, e90219 (2014).

    Article  ADS  Google Scholar 

  43. Kanagaraj, R. et al. Integrated genome and transcriptome analyses reveal the mechanism of genome instability in ataxia with oculomotor apraxia 2. Proc. Natl Acad. Sci. USA 119, e2114314119 (2022).

    Article  Google Scholar 

  44. Semmler, L., Reiter-Brennan, C. & Klein, A. BRCA1 and breast cancer: a review of the underlying mechanisms resulting in the tissue-specific tumorigenesis in mutation carriers. J. Breast Cancer 22, 1–14 (2019).

    Article  Google Scholar 

  45. Bai, G. et al. HLTF promotes fork reversal, limiting replication stress resistance and preventing multiple mechanisms of unrestrained DNA synthesis. Mol. Cell 78, 1237–1251 (2020).

    Article  CAS  Google Scholar 

  46. Choi, J.-H., Kim, S.-Y., Kim, S.-K., Kemp, M. G. & Sancar, A. An integrated approach for analysis of the DNA damage response in mammalian cells: nucleotide excision repair, DNA damage checkpoint, and apoptosis. J. Biol. Chem. 290, 28812–28821 (2015).

    Article  CAS  Google Scholar 

  47. Long, H. K. et al. Loss of extreme long-range enhancers in human neural crest drives a craniofacial disorder. Cell Stem Cell 27, 765–783 (2020).

    Article  CAS  Google Scholar 

  48. Guerra, J. et al. Lysyl-tRNA synthetase produces diadenosine tetraphosphate to curb STING-dependent inflammation. Sci. Adv. 6, eaax3333 (2020).

    Article  ADS  CAS  Google Scholar 

  49. Cristini, A., Groh, M., Kristiansen, M. S. & Gromak, N. RNA/DNA hybrid interactome identifies DXH9 as a molecular player in transcriptional termination and R-loop-associated DNA damage. Cell Rep. 23, 1891–1905 (2018).

    Article  CAS  Google Scholar 

  50. Hamperl, S., Bocek, M. J., Saldivar, J. C., Swigut, T. & Cimprich, K. A. Transcription-replication conflict orientation modulates R-loop levels and activates distinct DNA damage responses. Cell 170, 774–786 (2017).

    Article  CAS  Google Scholar 

  51. Natsume, T., Kiyomitsu, T., Saga, Y. & Kanemaki, M. T. Rapid protein depletion in human cells by auxin-inducible degron tagging with short homology donors. Cell Rep. 15, 210–218 (2016).

    Article  CAS  Google Scholar 

  52. Sanjana, N. E., Shalem, O. & Zhang, F. Improved vectors and genome-wide libraries for CRISPR screening. Nat. Methods 11, 783–784 (2014).

    Article  CAS  Google Scholar 

Download references

Acknowledgements

We thank J. Wysocka, S. Hamperl and J. Sollier for discussions and comments; and M.-S. Tsai (for GFP–dRH) and the members of the Straight laboratory for help in designing AID–XPG in HeLa cells. This work was supported by the Leukemia and Lymphoma Society (5455-17 to M.P.C.); the National Institutes of Health (GM119334 to K.A.C., S10OD018220 to the Stanford Functional Genomics Facility, T32-CA09302 to M.J.B., T32-HG000044 to C.L., DP2-CA271386 to M.A.-R.); Stanford Cancer Institute, an NCI-designated Comprehensive Cancer Center, to M.A.-R. and M.P.C.; the Korea Research Institute of Standards and Science (KRISS-GP2021-0003-10 to J.-H.C.), the National Research Foundation of Korea (MSIT) (NRF-2020R1A2C1101575 to J.-H.C.); the National Science Foundation (GRFP to C.L.), Jane Coffin Childs Memorial Fund for Medical Research (61-1755 to J.R.B.), the Gravitation Program CancerGenomiCs.nl from the Netherlands Organisation for Scientific Research (NOW) and the Oncode Institute, which is partly financed by the Dutch Cancer Society to W.V., and the V Foundation (D2018-017 to K.A.C.). M.A.-R. is a Terman Fellow and Pew-Stewart Scholar. K.A.C. is an ACS research professor.

Author information

Authors and Affiliations

Authors

Contributions

M.P.C., C.S., M.J.B. and K.A.C. designed the study. M.P.C., C.S., J.-H.C., M.J.B., J.R.B., G.B. and C.L. performed the experiments and data analyses. M.P.C. and M.J.B. performed the bioinformatic analyses. K.A.C. and M.A.-R. supervised the experiments and data analyses. J.K. and A.S. provided technical support. H.L. and W.V. designed and prepared HCT116 XPG-AID cells. M.P.C., C.S. and K.A.C. prepared the manuscript with contributions from the other authors.

Corresponding author

Correspondence to Karlene A. Cimprich.

Ethics declarations

Competing interests

K.A.C. is a scientific advisory board member of RADD Pharmaceuticals and IDEAYA Biosciences. M.A.-R. is a scientific advisory board member of Lycia Therapeutics. The other authors declare no competing interests.

Peer review

Peer review information

Nature thanks the anonymous reviewers for their contribution to the peer review of this work.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data figures and tables

Extended Data Fig. 1 Cytoplasmic RNA–DNA hybrids are induced upon multiple cellular perturbations.

(a) Western blot showing knockdown efficiency of siRNAs targeting SETX and BRCA1 in HeLa cells. (b) Images showing segmentation of nuclear and cytoplasmic compartments in HeLa cells, using DAPI and whole cell stain as masks, respectively. Scale bar is 20 μm. (c) Images showing the lack of GFP protein binding on fixed HeLa cells. Scale bar is 10 μm. (d) Western blot showing fractionation of siCtrl and siSETX-treated HeLa cells into soluble nuclear and cytoplasmic compartments with Lamin B1 and GAPDH as markers, respectively. (e) cytoDRIP blot showing cytoplasmic hybrid accumulation in HeLa cells following SETX knockdown using a second siRNA. In vitro RH treatment was performed prior to pull-down. (f) cytoDRIP blot showing cytoplasmic hybrid accumulation in siSETX-treated HCT116 cells. In vitro RH treatment was performed prior to pull-down. (g) cytoDRIP blot showing cytoplasmic hybrid accumulation in siBRCA1-treated HCT116 cells. In vitro RH treatment was performed prior to pull-down. (h) cytoDRIP blot showing cytoplasmic hybrid accumulation in PlaB-treated (500nM, 3h) HeLa cells. In vitro RH treatment was performed prior to pull-down. (i) Western blot showing knockdown efficiency of siRNAs targeting XPG and XPF in HeLa cells. (j) cytoDRIP blot showing the role of XPG and XPF in cytoplasmic hybrid production after SETX or BRCA1 knockdown in HeLa cells. (k) Left, images of HeLa cells after SETX and XPG or XPF knockdown probed with GFP–dRH protein after fixation, following mock or RH pre-treatment. Scale bar is 10 μm. Right, quantification of cytoplasmic GFP–dRH intensities; p-values are shown; two-sided Mann Whitney U test: n-values from left to right: 611, 573, 659, 686. Centre line, median; box limits, 75 and 25 percentiles, whiskers, min and max values. (l) As in (k) but after BRCA1 knockdown in HeLa cells. Two-sided Mann Whitney U test: n-values from left to right: 526, 502, 633, 653. Centre line, median; box limits, 75 and 25 percentiles; whiskers, min and max values. (m) Schematic of the XPG auxin-inducible degron (AID) system. (n) Western blots showing XPG degradation after knockdown of SETX (left) and BRCA1 (right) in HCT116 cells. (o) cytoDRIP blot showing cytoplasmic hybrid accumulation after knockdown of SETX or BRCA1 and impact of auxin-induced XPG degradation in HCT116 cells.

Extended Data Fig. 2 Dynamics of cytoplasmic hybrid production.

(a) cytoDRIP blot showing cytoplasmic hybrids in mock or PlaB treated (500 nM, 3 h) BAX−/−BAK−/− HeLa cells with or without in vitro RH treatment. (b) Flow cytometry analysis of asynchronous or serum-starved MCF10A cells following incubation with BrdU. Cells were segmented based on DNA content (propidium iodide staining) and BrdU intensity. The percentage of cells in G1, S and G2/M are indicated. At least 50,000 cells were quantified per condition. (c) Western blot showing fractionation of asynchronous (asynch) and serum-starved (starved) MCF10A cells into soluble nuclear and cytoplasmic compartments with Lamin B1 and GAPDH as markers, respectively. (d) RT-qPCR from asynchronous or serum-starved MCF10A cells showing increased unspliced mRNA following PlaB treatment (500 nM, 3 h). Shown is the mean ± s.d. from three independent biological replicates (n = 3), p-values are indicated in the figure; unpaired, two-tailed t-test. (e) As in (c) but for foreskin fibroblasts. (f) Cell cycle quantification from high-content imaging of foreskin fibroblasts after EdU incorporation, using DAPI staining for DNA content. Shown is the mean ± s.d. from three independent biological replicates (n = 3). (g) As in (d) but for foreskin fibroblasts. (h) cytoDRIP blot showing cytoplasmic hybrids extracted from equal numbers of asynchronous or serum-starved foreskin fibroblasts following DMSO or PlaB treatment (500 nM, 3 h), with mock and RH treatment prior to pull-down. (i) As in (b) but for BAX−/−BAK−/− MCF10A cells. (j) Western blot showing fractionation of asynchronous and serum-starved BAX−/−BAK−/− MCF10A cells into soluble nuclear and cytoplasmic compartments with Lamin B1 and GAPDH as markers, respectively. (k) cytoDRIP blot showing cytoplasmic hybrids in asynchronous and serum-starved BAX−/−BAK−/− MCF10A cells with DMSO or PlaB treatment (500 nM, 3 h). Each sample was treated with RH in vitro to confirm the specificity of the hybrid IP. (l) Left, experimental workflow. Right, blots showing hybrids isolated from the cytoplasm or nucleoplasm of siCtrl or siSETX-treated HeLa cells, with mock or LMB treatment (3 h, 5 nM) prior to harvest. (m) Left, experimental workflow. Middle, western blot as in (c) from HeLa cells treated with vehicle control (DMSO) or PlaB (500 nM, 3 h). Right, blot showing hybrids as in (i) but in HeLa cells treated with LMB (2 h, 5 nM) followed by PlaB + LMB for a further 3 h. (n) Representative images showing cyclin B1 localization in fixed HeLa cells treated with LMB (5 h, 5 nM) or vehicle control (EtOH). Scale bar is 10 μm. (o) RT-qPCR from HeLa cells showing increased unspliced mRNA following treatment with PlaB (500 nM) for the times indicated. Shown is the mean ± s.d. from three independent biological replicates (n = 3). (p) As in (o) but cells were treated with PlaB (500 nM, 3 h) and then fresh media was added following PlaB withdrawal for the times indicated. Shown is the mean ± s.d. from four independent biological replicates (n = 4).

Extended Data Fig. 3 Characteristics of cytoDRIP peaks.

(a) Scatter plots showing high reproducibility between cytoDRIP–seq siCtrl (left) and siSETX (right) replicates; Pearson’s correlation: R = 0.96 and 0.95 respectively; p < 1e-16 (machine precision limit). (b) Bar blot showing proportion of deduplicated sequencing reads mapping to the nuclear and mitochondrial (mito) genomes in cytoDRIP–seq samples. Data from two biological replicates are shown. (c) Table showing peak characteristics in cytoDRIP–seq, nuclear DRIP–seq, and nuclear DRIP–seq following RH treatment (RHR DRIP). Numbers of peaks (peak count), genomic space covered by peaks (peak area), size of peaks (mean and median), percent of genome covered by peaks (coverage) are shown. IQR is interquartile range. (d) Venn diagram of genome areas (in megabases) occupied by peaks identified in siCtrl and siSETX cytoDRIP–seq samples. (e) Bar plot showing enrichment of cytoplasmic hybrid sites by qPCR after S9.6 pull-down, relative to IgG. OPN3 was only found in the nucleus; the other sites were found in the nucleus and cytoplasm. Shown is the mean ± s.d. from three independent biological replicates (n = 3). (f) cytoDRIP-qPCR in HeLa cells after depletion of SETX at cytoDRIP–seq sites and nuclear R-loop forming sites. RH treatment was performed in vitro, prior to hybrid pull-down. ‘Nuc+ Cyto+’ sites were found in the nucleus and cytoplasm, while ‘Nuc+ Cyto-’ sites were only found in the nucleus. Gene names are shown; IG1IG5 are intergenic sites. Shown is the mean ± s.d. from four independent biological replicates (n = 4); p-values are shown in the figure, unpaired two-tailed t-test. (g) Scatter plots showing increased cytoDRIP–seq signal upon depletion of SETX in genic sites (upper) and intergenic sites (lower). Dashed line represents x = y. (h) Genome browser views of genic (top) and intergenic (bottom) cytoDRIP–seq sites. From top to bottom normalized tracks are: IgG, siCtrl (2 replicates), siSETX (2 replicates), nuclear DRIP–seq, nuclear DRIP–seq + RH. Red indicates negative strand signal, blue indicates positive strand signal.

Extended Data Fig. 4 cytoDRIP sites map to genic and intergenic regions.

(a) Histogram showing distribution of cytoDRIP peaks over genes. Positions of transcription start site (TSS) and transcription end site (TES) are indicated. (b) Blue histograms show expected overlaps between repeat elements and randomly sampled peak sets (matched in size and number from cytoDRIP peaks) from within all nuclear R-loop peaks. Red dashed line indicates the observed overlap for cytoDRIP peaks. (c) Z-scores for the overlaps calculated in (b); individual p-values are shown on the right. (d) Proportion of nuclear DRIP–seq and cytoDRIP–seq reads aligning to consensus regions for rDNA, alpha satellite for centromeres and telomeric repeats. Data are from two independent biological replicates (n = 2) per condition, black lines show the mean. (e) Bar plot showing proportion of cytoDRIP peaks overlapping nuclear DRIP and/or RNase H resistant hybrid (RHR) sites. (f) Histogram of peak lengths comparing cytoDRIP–seq (siSETX condition) (Cyto) and nuclear DRIP–seq (Nuc) peaks. (g) Scatter plot correlating nuclear DRIP–seq signal at cytoDRIP regions with cytoDRIP–seq signal, Pearson’s correlation: R = 0.24. (h) Scatter plot correlating nascent transcription by global run-on sequencing at cytoDRIP regions with cytoDRIP–seq signal, Pearson’s correlation: R = 0.14. (i,j) Genome browser views showing lack of cytoDRIP signal at sites with robust nuclear R-loop formation (i) ACTB, (j) RPL13A. From top to bottom normalized tracks are: IgG, siCtrl (2 replicates), siSETX (2 replicates), nuclear DRIP–seq, nuclear DRIP–seq + RH. Red indicates negative strand signal, blue indicates positive strand signal. (k) Western blot showing HeLa cells stably expressing GFP-tagged XPG. GAPDH is the loading control. (l) GFP ChIP-qPCR in HeLa cells following knockdown of SETX and/or XPG, showing GFP-XPG binding at hybrid sites. ‘Nuc+ Cyto+’ sites were found in the nucleus and cytoplasm, while ‘Nuc+ Cyto-’ sites were only found in the nucleus. Gene names are shown; IG1 and IG2 are intergenic sites. Shown is the mean ± s.d. from three independent biological replicates (n = 3); unpaired two-tailed t-test; p-values are shown.

Extended Data Fig. 5 Cytoplasmic hybrids are derived from long-lived and partially RNase H-resistant nuclear R-loops.

(a) Nuclear DRIP-qPCR after actinomycin D treatment. Nuclear R-loop sites with short, average and long half-lives21 are indicated, as well as cytoDRIP sites. R-loopneg indicates a nuclear site with low R-loop abundance. Gene names are indicated. IG1 is an intergenic site. Shown is the mean ± s.d. from three independent biological replicates (n = 3). P = 9.81e-12 between nuclear sites with short or medium lifetimes and cytoDRIP sites (two-tailed Mann Whitney U test). (b) Nuclear DRIP-qPCR after actinomycin D treatment showing example fits of exponential decay to derive RNA–DNA hybrid half-lives. Shown is the mean from three independent biological replicates (n = 3). (c) Aggregate plots around cytoDRIP regions showing nuclear DRIP–seq signal following low (red) or high (purple) RH treatment in vitro. Input signal is grey. Each line is the mean of 1762 genic peaks (n = 1762). Error bands represent 95% CI of the mean. (d) Genome browser views of previously identified21 long-lived nuclear R-loop sites. From top to bottom normalized tracks are: IgG, siCtrl (2 replicates), siSETX (2 replicates), nuclear DRIP–seq, nuclear DRIP–seq + in vitro RH. Red indicates negative strand signal, blue indicates positive strand signal.

Extended Data Fig. 6 Cytoplasmic hybrids are characterized by switches in nucleotide skew.

(a) Model of converging transcription at cytoDRIP peaks. RNA–DNA hybrids form as a result of sense and antisense transcription in regions of high GC-skew on the non-template strand. Example nucleotide sequences that fit the observed skew pattern are shown. (b) Violin plots of AT content (left) and GC content (right) for cytoDRIP (n = 2911) and nuclear DRIP (n = 65,541) regions, two-tailed Mann Whitney U between cytoDRIP and nuclear DRIP regions: p = 3.6e-8 (left), p = 3.5e-16 (right). Centre line, median; box limits, 75 and 25 percentiles, whiskers, min and max values. (c) Aggregate plots around genic nuclear DRIP regions (n = 56,433) showing GC skew (left) and AT skew (right). Error bands represent 95% CI of the mean. (d) Aggregate plots around genic siCtrl cytoDRIP regions showing GC skew (left) and AT skew (right). Means of 282 peaks (n = 282) are shown; error bands represent 95% CI of the mean. (e) Aggregate plots around genic siCtrl cytoDRIP regions showing cytoDRIP–seq signal (left), nuclear DRIP–seq signal (middle) and nuclear RHR signal (right). Means of 282 peaks (n = 282) are shown; error bands represent 95% CI of the mean. (f) Violin plots of AT content (left) and GC content (right) for siSETX (n = 2629) and siCtrl (n = 282) cytoDRIP regions, two-tailed Mann Whitney U between cytoDRIP and nuclear DRIP regions: p = 0.003 (left), p = 0.003 (right). Centre line, median; box limits, 75 and 25 percentiles, whiskers, min and max values.

Extended Data Fig. 7 Different perturbations inducing cellular R-loops trigger IRF3 signalling and apoptosis.

(a) Western blot of pIRF3 upon PlaB treatment (500 nM) in HeLa cells. GAPDH is the loading control. (b) Western blot showing knockdown efficiency of a second siRNA to BRCA1. GAPDH serves as the loading control. (c) Western blot showing pIRF3 levels upon knockdown of SETX or BRCA1 using a second siRNA in HeLa cells. (d) Effect of PlaB treatment (500 nM, 3 h) on pIRF3 in HCT116 cells. (e) Top: Schematic of nuclear localization signal (NLS)-tagged wild-type (WT) or catalytically-inactive (D210N) RNase H1. HBD = hybrid binding domain, CD = connection domain. Bottom: cellular localization of GFP-tagged NLS-RNaseH1 WT/D210N. Scale bar, 20 μm. (f) RT-qPCR measurements of IRF3 effectors upon knockdown of SETX or BRCA1 in MCF10A cells. (g) Western blot of pIRF3 after auxin-induced XPG degradation and SETX knockdown in HCT116 cells. GAPDH is the loading control. (h) Western blot showing C-PARP levels upon knockdown of SETX or BRCA1 in MCF10A cells. (i) Western blot showing the impact of auxin-induced XPG degradation on C-PARP in siSETX- or siBRCA1-treated HCT116 cells. The same GAPDH blot, which is the loading control, is used in Extended Data Fig. 1n. (j) Left, RT-qPCR showing the knockdown efficiency of TNFα in HeLa cells. Right, western blots showing levels of C-PARP upon knockdown of TNFα in siSETX-treated HeLa cells. (k) Western blot showing levels of pIRF3 upon knockdown of SETX or BRCA1 in BAX−/−BAK−/− HeLa cells. (l) Top: Schematic of nuclear export signal (NES)-tagged RNase HI. Bottom: cellular localization of GFP-tagged RNase HI-NES in HeLa cells. Scale bar, 20 μm. (m) cytoDRIP blot showing cytoplasmic hybrids upon knockdown of SETX or BRCA1 in mock-treated HeLa cells and HeLa cells stably expressing NES-tagged RNase HI. Bar graphs are mean ± s.d. from 3 independent biological replicates (n = 3) (unpaired two-tailed t-test with CI = 95%). P values are shown at the top of the graphs.

Extended Data Fig. 8 cGAS and TLR3 cooperate to activate IRF3 signalling.

(a) Western blot showing pIRF3 levels upon siRNA-mediated knockdown of SETX and either RIG1 or MDA5 in HeLa cells. GAPDH is the loading control. (b) RT-qPCR showing the knockdown efficiency of RIGI and MDA5 in HeLa cells. (c) Western blot showing the knockdown efficiency of two different TLR3 siRNAs in HeLa cells. (d) RT-qPCR measurements of IRF3 effectors upon TLR3 knockdown with two different siRNAs in siSETX-treated BAX−/−BAK−/− HeLa cells. (e) and (f) Western blot showing levels of pIRF3 in two negative control (neg) clones and either cGAS knockout clones (e) or TLR3 knockout clones (f) generated using the CRISPR–Cas9 system in HeLa cells. c1 = clone 1, c2 = clone 2. GAPDH serves as the loading control. (g) RT-qPCR measurements of IRF3 effectors upon single or combined inhibition/knockdown of cGAS and TLR3 in control and siSETX-treated HeLa cells. (h) As in (g) but in BAX−/−BAK−/− HeLa cells. (i) Caspase 3 activity assay after knockdown of SETX and either XPG knockdown or the combination of cGAS inhibition and TLR3 knockdown. (j) cGAS and TLR3 protein levels upon siRNA-mediated knockdown of TLR3 or cGAS in HeLa cells. (k) Agarose gel showing DNA (60 nt) and RNA (60 nt) oligonucleotides can anneal to form a DNA-RNA hybrid. (l) Gel shift assays of cGAS binding to double-stranded DNA (dsDNA) (left) and TLR3 binding to double-stranded RNA (dsRNA) (right). (m) Gel shift assays show binding of human RNaseH1 D210N catalytically-inactive mutant and GFP protein to RNA–DNA hybrids which are used as positive and negative controls, respectively. NP stands for no protein. Bar graphs are mean ± s.d.from three independent biological replicates (n = 3) (unpaired two-tailed t-test with CI = 95%). P values are shown at the top of the graphs.

Source data

Extended Data Fig. 9 cGAS and TLR3 bind directly to cytoplasmic RNA–DNA hybrids.

(a) The purity of the cytoplasmic fraction used for the S9.6 co-IP was assessed by western blot. (b) S9.6 co-IP from the cytoplasmic fraction showing LysRS binds to cytoplasmic hybrids in our methods. LysRS has been reported to interact with cytoplasmic hybrids and serves as the positive control. (c) S9.6 co-IP from the cytoplasmic fraction showing cGAS and TLR3 associate with RNA–DNA hybrids isolated from siBRCA1-treated cells, as well as the impact of 37 °C no enzyme mock control and in vitro RNase H treatment before the IP step. RNase H treatment, 50 U ml−1 for 1 h at 37 °C. (d) S9.6 co-IP from cytoplasmic fraction showing cGAS binding to hybrids induced by siSETX is disrupted by 1 μM hybrid competitor in IP reaction, and TLR3 binding to hybrids is disrupted by 3 μM hybrid competitor in an IP reaction. hyb = hybrid. (e) Western blot validating the TLR3 IP efficiency in experiments to detect TLR3-associated cytoplasmic hybrids by performing TLR3 IP followed by S9.6 IP (Fig. 4h). (f) Western blot assessing the purity of the endolysosomal fraction after isolation following HA immunoprecipitation in control or SETX-depleted HA-TMEM192 HEK293T cells. Flag-TMEM192 HEK293T cells were used as a negative control for the LysoIP. Proteins marking the lysosome (Lyso), Golgi apparatus (Golgi), endoplasmic reticulum (ER) and mitochondria (Mito) are indicated. (g) Western blot showing pIRF3 and C-PARP levels induced by SETX knockdown in HA-TMEM192 HEK293T cells, as was observed in HeLa cells. This result suggests this cell line is suitable for the study of R-loop-induced immune activation. This experiment is a control for the LysoIP (Fig. 4i). (h) cytoDRIP blot showing cytoplasmic hybrids levels are elevated upon knockdown of SETX in HA-TMEM192 HEK293T cells. In vitro RNase H digestion was used to ensure IP specificity. This experiment is also a control for the LysoIP. (i) cytoDRIP blot showing cytoplasmic hybrids upon knockdown of SETX in HA-TMEM192 HEK293T cells with or without knockdown of XPG. (j) co-IP testing the interaction between Flag-tagged cGAS and endogenous TLR3. (k) co-IP testing the interaction between endogenous TLR3 and cGAS. (l) Working model. Left: in wild-type cells, nuclear R-loops are efficiently resolved by RNase H or RNA–DNA helicases, such as SETX. Only a small number of R-loops are processed by XPG and converted to cytoplasmic hybrids, so that cytoplasmic hybrid levels are below the threshold required for activation of IRF3 signalling. Right: under certain perturbations, including depletion of SETX/BRCA1, or under pathological conditions that deregulate R-loops, a subset of nuclear R-loops that may not be efficiently resolved are processed by XPG, leading to RNA–DNA hybrid accumulation in the cytoplasm. These hybrids are then recognized by cGAS and TLR3 in the cytosol and endolysosome, activating IRF3-mediated immune signalling and apoptosis.

Extended Data Fig. 10 R-loop-induced cytoplasmic RNA–DNA hybrid accumulation and innate immune activation in patient-derived disease cell models.

(a) RT-qPCR showing the XPG siRNA knockdown efficiency in AOA2 patient-derived fibroblasts. (b) cytoDRIP blot showing cytoplasmic hybrids in control and AOA2 patient-derived fibroblasts with or without knockdown of XPG. (c) cytoDRIP blot showing cytoplasmic hybrids in control and AOA2 patient-derived fibroblasts with or without in vitro RNase H treatment prior to hybrid IP. (d) RT-qPCR showing the knockdown efficiency of TLR3 in AOA2 patient-derived fibroblasts. (e) RT-qPCR measurements of immune effectors upon single or combined inhibition and knockdown of cGAS and TLR3, respectively, in control and AOA2 patient-derived fibroblasts. (f) RT-qPCR showing the SETX siRNA knockdown efficiency in control fibroblasts. (g) RT-qPCR measurements of IFNβ and ISGs upon knockdown of SETX in control fibroblasts. (h) Western blot showing the fractionation of UWB1.289 and UWB1.289+BRCA1 cells into soluble nuclear and cytoplasmic compartments with Lamin B1 and GAPDH as markers, respectively. (i) cytoDRIP blot showing cytoplasmic hybrids in UWB1.289 and UWB1.289+BRCA1 cells with or without in vitro RNaseH treatment prior to hybrid IP. (j) Cellular localization of GFP-tagged NES-RNaseHI in UWB1.289 and UWB1.289+BRCA1 cells. Scale bar, 20 μm. (k) RT-qPCR measurements of immune effectors in UWB1.289 and UWB1.289+BRCA1 cells stably expressing GFP (mock) or NES-tagged RH (RH-NES). (l) RT-qPCR showing the SAMHD1 siRNA knockdown efficiency in HeLa cells. (m) cytoDRIP blot showing cytoplasmic hybrids in control and SAMHD1-deficient HeLa cells with or without in vitro RNase H treatment. (n) cytoDRIP blot showing cytoplasmic hybrids in control and SAMHD1-deficient HeLa cells with or without XPG knockdown. (o) Left: western blots showing levels of pIRF3 and C-PARP upon knockdown of XPG in siSAMHD1-treated HeLa cells. Right: western blots showing pIRF3 and C-PARP level upon knockdown of SAMHD1 in mock and RH-NES HeLa stable cell lines. siSAM = siSAMHD1. Bar graphs are mean ± s.d. from three independent biological replicates (n = 3) (unpaired, two-tailed t-test with CI = 95%). P values are shown at the top of the graphs.

Source data

Supplementary information

Supplementary Figs. 1 and 2

Uncropped gel images from western blots, cytoDRIP blot and agarose gels (Supplementary Fig. 1) and example gating strategy for flow cytometry (Supplementary Fig. 2).

Reporting Summary

Supplementary Table 1

Primers, siRNAs and antibodies used in this study.

Source data

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Crossley, M.P., Song, C., Bocek, M.J. et al. R-loop-derived cytoplasmic RNA–DNA hybrids activate an immune response. Nature 613, 187–194 (2023). https://doi.org/10.1038/s41586-022-05545-9

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s41586-022-05545-9

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing