mTORC2-driven chromatin cGAS mediates chemoresistance through epigenetic reprogramming in colorectal cancer

Cyclic GMP–AMP synthase (cGAS), a cytosolic DNA sensor that initiates a STING-dependent innate immune response, binds tightly to chromatin, where its catalytic activity is inhibited; however, mechanisms underlying cGAS recruitment to chromatin and functions of chromatin-bound cGAS (ccGAS) remain unclear. Here we show that mTORC2-mediated phosphorylation of human cGAS serine 37 promotes its chromatin localization in colorectal cancer cells, regulating cell growth and drug resistance independently of STING. We discovered that ccGAS recruits the SWI/SNF complex at specific chromatin regions, modifying expression of genes linked to glutaminolysis and DNA replication. Although ccGAS depletion inhibited cell growth, it induced chemoresistance to fluorouracil treatment in vitro and in vivo. Moreover, blocking kidney-type glutaminase, a downstream ccGAS target, overcame chemoresistance caused by ccGAS loss. Thus, ccGAS coordinates colorectal cancer plasticity and acquired chemoresistance through epigenetic patterning. Targeting both mTORC2–ccGAS and glutaminase provides a promising strategy to eliminate quiescent resistant cancer cells.


mTORC2-induced phosphorylation at serine 37 promotes cGAS-chromatin localization
We investigated whether cGAS is a direct target of mTORC2 phosphorylation, given mTORC2's ability to phosphorylate substrates.Co-immunoprecipitation (Co-IP) (Extended Data Fig. 2e,f) and BRET assay (Extended Data Fig. 2g) detected interactions between cGAS and SIN1, an essential mTORC2 component for substrate recognition 22 .Using an in vitro cell-free phosphorylation system, we found that activated mTORC2 directly phosphorylates cGAS at serine 37 as demonstrated by mass spectrometry (MS) (Fig. 2e, Extended Data Fig. 2h and Supplementary Table 1).In contrast, inactive mTORC2 or mTORC1 did not mediate this phosphorylation (Extended Data Figs.2i and 3a), validating serine 37 as the specific mTORC2 target site.Consistently, serine 37 phosphorylation of cGAS (p-Ser 37 -cGAS) were reduced in RICTOR-deficient HCT116 cells and rescued by RICTOR reintroduction (Fig. 2f,g and Extended Data Fig. 3b-d).Moreover, p-Ser 37 -cGAS levels, not total cGAS, consistently decreased over time with GDC-0941 treatment (Extended Data Fig. 3e).
Serine 37 is located within the N-terminal nuclear localization signal (NLS) of human cGAS 7 , which contains a common consensus motif for mTOR phosphorylation 19 (Extended Data Fig. 3f).A second NLS in the NTase core region 7 contains potential phosphorylation sites for both AKT1 and mTOR.In vitro phosphorylation assays showed that AKT1 phosphorylated serine 305 (ref.32), whereas mTORC2 did not (Extended Data Fig. 3f,g).BRET assays found that cGAS mutants S305D and S305A did not bind chromatin (Extended Data Fig. 4a).Additional mutant experiments demonstrated that cGAS nuclear localization is independent of serine 305 phosphorylation, but serine 37 phosphorylation promotes nuclear localization of mutants (Extended Data Fig. 4b).These results indicate that mTORC2 specifically phosphorylates the N-terminal NLS at serine 37, which may facilitate human cGAS-chromatin localization.
Immunofluorescence analysis revealed that p-Ser 37 -cGAS co-localizes with histone H2A in the nucleus of HCT116 cells (Extended Data Fig. 4c).MS analysis showed that chromatin-bound but not extranuclear cGAS was phosphorylated at serine 37 (Extended Data Fig. 4d).Immunofluorescence further showed that the S37D phosphomimetic mutant was exclusively localized to chromatin, whereas the non-phosphorylatable S37A mutant was cytoplasmic (Fig. 2h and Extended Data Fig. 4e), despite both retaining chromatin binding capacity during mitosis (Extended Data Fig. 4f).Chromatin fractionation confirmed that the S37D mutation promoted chromatin association, whereas S37A disrupted it (Fig. 2i).In addition, inhibiting mTORC2 by JR-AB2-011 gradually reduced Ser37 phosphorylation and cGASchromatin levels over time (Fig. 2j and Extended Data Fig. 4g,h).These results indicate that mTORC2-mediated serine 37 phosphorylation is essential for cGAS-chromatin localization.

ccGAS recruits the SWI/SNF complex at specific chromatin regions to regulate gene expression
To explore ccGAS functions, we characterized its interacting proteins (Extended Data Fig. 5a,b and Supplementary Table 2).In extranuclear fractions, cGAS interacts with ribosomal subunits (Extended Data Fig. 5c).ccGAS recruits several components of the SWI/SNF chromatin remodelling complex, including SMARCC2, SMARCA4, PBRM1, ARID1A, SMARCA5, SMARCD1 and UCHL5 (Fig. 3a and Extended Data Fig. 5d-f).SWI/SNF alters nucleosome positioning and chromatin accessibility to control gene expression 33 .The nuclease benzonase was included in all buffers to minimize post-lysis DNA binding by cGAS, then Co-IP validated interactions between cGAS and SWI/SNF components such as ARID1A, SMARCA4 and SMARCC2 in colorectal cancer cells (Fig. 3b and Extended Data Fig. 5g).These results suggest that ccGAS recruits the SWI/SNF complex to regulate gene expression.
The mechanistic target of rapamycin (mTOR) pathway integrates extracellular signals to regulate cellular responses to environmental changes 17 .mTOR forms two complexes, mTORC1 and mTORC2.While mTORC1 is well studied, the functions of mTORC2 are less clear, although aberrant activity contributes to several cancer types [18][19][20][21] , implicating oncogenic roles for mTORC2 (ref.22).Recent evidence links mTOR to drug resistance [23][24][25] .Inhibiting mTOR induces a dormant state in mammalian embryos and a reversible diapause-like state in cancer cells 26 , conferring tolerance against chemotherapy 24,27 .Such mTOR blockade protects against chemotherapeutic agents in colorectal cancer and leukaemia [28][29][30] ; however, mechanisms linking mTOR inhibition to acquired chemoresistance remain unclear.Here, through high-throughput compound screening, we discover that mTOR inhibition disrupts cGAS-chromatin tethering.mTORC2 directly phosphorylates cGAS at serine 37, promoting chromatin anchoring.Of note, ccGAS knockdown inhibits colorectal cancer cell growth under normal conditions but induces acquired resistance to fluorouracil, a front-line therapeutic.Mechanistically, ccGAS recruits SWI/SNF complexes at specific gene loci to regulate DNA replication and glutaminolysis genes, and inhibiting the downstream kidney-type glutaminase (KGA) overcomes resistance.Our study provides mechanistic insight into mTOR inhibition-mediated chemoresistance and implies targeting the mTORC2-ccGAS-KGA axis could improve cancer therapy.

High-throughput screening identifies PI3K-mTOR pathway regulation of cGAS-chromatin localization
To explore how cGAS localization is regulated, we analysed its subnuclear distribution in cancer cells.We found that cGAS localized constitutively to the nucleus independent of cell cycle timing or nucleotidyltransferase activity (Extended Data Fig. 1a-c).When fractionating HCT116 cell lysates, we found cGAS to be predominantly chromatin-bound (Fig. 1a).To identify potential regulatory molecules, we performed high-throughput compound screening in HCT116 cells to monitor cGAS-chromatin interaction using bioluminescence resonance energy transfer (BRET) technology 31 (Fig. 1b).In a library of 8,326 compounds, BEZ-235, GDC-0941 and KU-0063794 significantly inhibited this interaction (Fig. 1c), indicating involvement of the PI3K-mTOR pathway.Stable isotope labelling identified cGAS and lymphoid-specific helicase HELLS most significantly reduced on chromatin by the PI3Ki GDC-0941 treatment (Extended Data Fig. 1d).Unlike HELLS fluctuated on chromatin after GDC-0941 exposure, cGAS levels on chromatin consistently decreased with increasing treatment time (Fig. 1d and Extended Data Fig. 1e), suggesting that cGAS is a specific PI3K-mTOR chromatin target.
To determine the specific components driving cGAS-chromatin localization, we treated HCT116 cells with PI3K-mTOR pathway inhibitors (Extended Data Fig. 2a).Immunofluorescence analysis showed that inhibition of PI3K and mTOR, but not AKT, mTORC1 or S6K, reduced cGAS-chromatin association (Fig. 1e and Extended Data Fig. 2b).This was confirmed using the extended bioluminescence resonance energy transfer (eBRET2) assay (Fig. 1f).We then used short hairpin RNA (shRNA) to knockdown RAPTOR and RICTOR, essential subunits of mTORC1 and mTORC2, respectively (Extended Data Fig. 2c).RICTOR knockdown, but not RAPTOR, inhibited cGAS-chromatin localization (Fig. 2a and Extended Data Fig. 2d).BRET (Fig. 2b) and chromatin fractionation (Fig. 2c,d) assays also showed that mTOR and RICTOR knockdown, but not RAPTOR, disrupted the cGAS-chromatin interaction.Moreover, the interaction between the enzyme-inactive S213D cGAS mutant and chromatin was regulated by mTORC2 (Fig. 2b,d).Collectively, these data demonstrate that cGAS-chromatin association is dependent on mTORC2, but not its nucleotidyltransferase activity.
Consistent with CUT&Tag data showing ccGAS interaction with the KGA promoter (Extended Data Fig. 6i), chromatin immunoprecipitation (ChIP) assays found that ccGAS localized to the KGA promoter region in colorectal cancer cells (Fig. 3g).ccGAS localization decreased with mTOR or RICTOR knockdown but not RAPTOR knockdown (Fig. 3g).STING knockout or cGAS deactivation did not impact ccGAS localization (Extended Data Fig. 8g).Compared to wild-type cGAS, the S37D mutant, but not S37A, maintained KGA promoter localization (Fig. 3h and Extended Data Fig. 8h).These results validate the cGAS genomic binding data and suggest that KGA transcription is strictly regulated by ccGAS.
Inhibiting PI3K by GDC-0941 increased both KGA protein and mRNA levels in a time-dependent manner correlated with reduced ccGAS protein levels (Fig. 3i).KGA mRNA and protein levels also significantly elevated in cGAS-knockout cells compared with controls, but re-expressing wild-type or S213D mutant cGAS rescued this (Extended Data Fig. 8i), indicating cGAS enzymatic activity is dispensable for regulating KGA.Overexpressing cGAS mutant S37D, but not S37A, reduced elevated KGA mRNA and protein levels in cGAS-knockout cells (Fig. 3j).Additionally, Seahorse analysis showed that cGAS knockout increased glutamine-dependent oxygen consumption, an effect rescued by S37D but not S37A mutant expression (Extended Data Fig. 8j).Together, these results demonstrate that ccGAS depletion induces KGA expression and promotes glutaminolysis.

ccGAS depletion induces a diapause-like state in colorectal cancer cells
Given cancer cells' addiction to glutaminolysis and DNA replication 35 , we infer that ccGAS was hypothesized to be essential for cancer progression.cGAS knockdown induced G1/S-phase arrest in HCT116 cells after nocodazole synchronization (Extended Data Fig. 9a,b).Overexpressing shRNA-resistant S37D or S213D cGAS rescued this arrest, whereas S37A mutant or STING activator (ADU-S100) supplementation did not (Fig. 4a and Extended Data Fig. 9c).Plate cloning and cell counting assays confirmed that cGAS knockdown inhibited HCT116 cell growth, rescued by S37D but not S37A overexpression (Fig. 4b-d).Therefore, these results demonstrated that ccGAS depletion induces colorectal cancer cell cycle arrest.

ccGAS depletion restricts tumour growth and induces chemoresistance
While diapause-like cancer cells exhibit reduced proliferation, they often acquire drug resistance properties [36][37][38][39][40][41][42] .We investigated whether ccGAS is required for chemosensitivity in colorectal cancer.Fluorouracil (5-FU) is a widely used chemotherapeutic for colorectal cancer 43,44 .cGAS knockdown inhibited HCT116 cell proliferation under normal growth conditions but desensitized cells to 5-FU-induced death (Fig. 4f).Re-expressing wild-type cGAS resensitized cells to 5-FU, an effect blocked by inhibiting mTORC2 with JR-AB2-011 (Fig. 4g).Expressing S37D or S213D mutants also resensitized cells to 5-FU, whereas Fig. 4 | ccGAS depletion induces a diapause-like state and chemoresistance in colorectal cancer cells.a, Quantification data of cell cycle distribution for HCT116 cells are presented.shRNA-resistant cGAS mutants were transfected into HCT116 cells stably expressing cGAS shRNA.ADU-S100, an activator of STING, 10 µM.b, Representative images of colony formation assays showed the proliferative capacity of HCT116 cells stably expressing indicated cGAS mutants.c, Colony formation assays were quantified in HCT116 cells stably expressing indicated cGAS mutants.d, HCT116 cells stably expressing indicated cGAS mutants were measured for cell proliferation.e, The relative size of cGASknockout HCT116 cells with cGAS mutants restoration was calculated.n = 2 × 10 7 cells per group.f, HCT116 cells stably expressing cGAS shRNA were treated with the indicated concentrations of 5-FU and cell viability was assessed after 24 h.g, cGAS-knockdown HCT116 cells stably transfected cGAS or pretreated with 10 µM JR-AB2-011 for 6 h were treated with the indicated concentrations of 5-FU, and cell viability was assessed after 24 h.JR, JR-AB2-011, a selective mTORC2 inhibitor.h, HCT116 cells stably expressing indicated cGAS mutants were treated with the indicated concentrations of 5-FU and cell viability was assessed after 24 h.i, Colony formation assays were performed in HCT116 cells in the presence of 2 µM 5-FU in DMSO.j, HCT116 cells stably expressing indicated shRNAs were treated with the indicated concentrations of 5-FU and cell viability was assessed after 24 h.k, Cell viability assays measuring idarubicin response of THP-1 cells stably expressing indicated cGAS mutants.Data are shown as mean ± s.e.m. from three independent experiments.Unpaired two-tailed t-test (a, c, e, i) and Kruskal-Wallis one-way analysis of variance (ANOVA) followed by Dunn's multiple comparison tests (d, f-h, j, k).Experiments were repeated four times (b) with similar results.Numerical source data are available.Fig. 3 | ccGAS recruits SWI/SNF to regulate genes involved in DNA replication and glutaminolysis.a, Functional enrichment (cell component) analysis of 33 unique cGAS-interacting intranuclear proteins.b, IB analysis of immunoprecipitation (IP) and WCLs derived from HCT116 cells.c, The cGAS CUT&Tag peaks were annotated by ChIPseeker, and 16.64% were found to be located on gene promoters.d, KEGG pathway analysis of promoterlocalized cGAS peaks showed that these genes were closely related to cell cycle and amino acid metabolism.e, Functional enrichment analysis of 53 downregulated proteins under cGAS knockdown in HCT116 cells.f, Eighty-one proteins significantly altered in abundance under cGAS knockdown, of which 61 proteins,18 upregulated (orange) and 43 downregulated (blue), were annotated to 142 reliable protein-protein interactions (grey line) based on databases such as BioGrid and StringDB.Node size indicates the degree of altered abundance.g, ChIP-qPCR analysis of cGAS occupancy at the KGA promoter in HCT116 cells lentivirally infected with the indicated shRNAs.h, ChIP-qPCR analysis of ARID1A occupancy at the KGA promoter in cGAS-knockout HCT116 cells stably expressing control or cGAS mutants.i, HCT116 cells were treated with 2 µM GDC-0941 for the indicated times, mRNA levels of indicated genes were quantified by qPCR analysis and normalized to ACTB control and WCLs were analysed by IB. j, cGAS mutants were transfected into cGAS-knockout HCT116 cells and KGA mRNA levels and WCLs were analysed.Data are shown as mean ± s.e.m. from three independent experiments (g-j).Unpaired two-tailed t-test.Experiments were repeated three times (b) with similar results.Numerical data and unprocessed blots are available as source data.
While murine cGAS is predominantly nuclear, mTORC2 contributes minimally to its chromatin localization in mice (Extended Data Fig. 10h).This differs from humans likely due to lack of N-terminal NLS conservation between species.Notably, the fundamental ccGAS-KGA pathway regulating glutaminolysis and chemotherapy response remained functionally conserved in murine colorectal cancer cells (Fig. 7a and Extended Data Fig. 10i-k).An azoxymethane (AOM)/ dextran sodium sulfate (DSS) model of colitis-associated cancer in cGAS-KO mice was used (Fig. 7b), mimicking human disease pathologically 46 .Colorectal tumours developed in both cGAS +/+ and cGAS −/− mice after AOM/DSS treatment, predominantly in the distal colon (Fig. 7c).Of note, cGAS deficiency significantly increased tumour burden, with greater numbers and larger sizes of colorectal tumours in cGAS-KO mice compared with their wild-type littermates (Fig. 7c,d).This contrasted with ccGAS knockdown inhibiting cancer cell growth in vitro, suggesting cGAS deficiency may promote in vivo tumour growth by relieving anti-tumourigenic effects through the cGAS-STING pathway.
Notably, as observed in cell lines, cGAS deficiency also conferred resistance to 5-FU in the AOM/DSS tumour model.5-FU treatment significantly reduced tumour numbers and sizes in wild-type littermate control mice but not cGAS-KO mice (Fig. 7c,d).However, combining 5-FU and BPTES markedly attenuated tumour burden in KOs (Fig. 7c,d).Together, these results indicate that KGA inhibition can overcome cGAS deficiency-induced chemoresistance and that targeting both ccGAS and KGA may represent a promising therapeutic strategy for colorectal cancer.

Discussion
In this study, we uncovered a mechanism by which mTORC2-induced cGAS phosphorylation at serine 37 promotes its chromatin recruitment and functions.This post-translational modification represents a key regulatory node influencing cGAS-dependent processes in cancer cells.By modulating cGAS-chromatin localization, PI3K-mTORC2 signalling supports proliferation under normal conditions but its disruption provokes acquired resistance to chemotherapy (Fig. 7e).This elucidates the importance of the PI3K-mTORC2-ccGAS pathway in tumour growth and treatment response.
We found that inhibiting the mTORC2-ccGAS axis drives colorectal cancer cells into a diapause-like state of chemoresistance.This provides insights into how mTOR inhibition clinically elicits drug tolerance 47,48 .Emerging evidence associates such chemoresistance with cell plasticity [49][50][51][52] , though mechanisms were unclear.Tracing resistance back to the mTORC2-ccGAS node, we discovered diapause-like plasticity underlies resistance upon ccGAS depletion.Adding ccGAS as a biomarker may help optimize strategies combining mTOR inhibitors with KGA blockade to eliminate persistent quiescent chemoresistance tumours.
Our study also revealed an epigenetic mechanism, whereby mTORC2 modifies chromatin through ccGAS.ccGAS selectively recruits the SWI/SNF complex to regions regulating DNA replication and glutaminolysis.Depleting ccGAS strongly induced KGA expression, validating links between SWI/SNF defects and glutaminase inhibition sensitivity 53 .Characterizing ccGAS cistromes and targets may elucidate plasticity governance across contexts.While targeting SWI/SNF is challenging, selectively inhibiting downstream nodes such as KGA may improve precision oncology.
While our findings demonstrate mTORC2-mediated S37 phosphorylation promotes cGAS-chromatin localization, we cannot exclude contributions of additional factors downstream of PI3K-mTORC2 signalling.Our findings require validation across diverse models and clinical settings, as additional ccGAS regulators remain unknown.Validating findings using multi-omic patient data and longitudinal analyses strengthens translational relevance.In summary, we provide provisional evidence that the mTORC2-ccGAS-KGA axis mediates cell plasticity and acquired chemoresistance in cancer.Further exploring modulatory factors and pathways may optimize precision strategies against adaptive survival, pending validation.Continued investigation holds promise to refine the clinical impact of cGAS biology.

Online content
Any methods, additional references, Nature Portfolio reporting summaries, source data, extended data, supplementary information, acknowledgements, peer review information; details of author contributions and competing interests; and statements of data and code availability are available at https://doi.org/10.1038/s41556-024-01473-0.https://doi.org/10.1038/s41556-024-01473-0pipetted in triplicates into a 96-well plate (COSTAR Lumiplates Flat White; Corning) with 20 µl per well.The luciferase substrate Coelenterazine 400a (CLZ400a; 0.5 mg ml −1 in 100% ethanol, NanoLight) was diluted 1:100 in eBRET2 assay buffer (PBS supplemented with 1 g l −1 d-glucose monohydrate (Roth), 0.1 g l −1 calcium chloridedihydrate (Merck) and 0.1 g l −1 magnesium chloride-hexahydrate (Merck)) and incubated for 20 min at room temperature under light protection to avoid light emission caused by substrate oxidation.Then, 100 µl of the substrate solution was added per well with the injector of the Tecan Infinite M1000Pro plate reader (Tecan).For eBRET2 measurements the dual colour luminescence mode of the Tecan plate reader was used with two filters in 395 nm (blue filter) and 510 nm (green filter).The measurements were carried out with an integration time of 1 s.

In vitro phosphorylation assays
HCT293T cells transfected with HA-RICTOR were cultured under serum starvation (for 48 h) or insulin stimulation (100 nM insulin for 30 min) conditions.HA immunoprecipitation was then performed in CHAPS buffer (50 mM Tris-HCl (pH 7.5), 120 mM NaCl and 0.3% CHAPS).The immunoprecipitate (mTORC2) was washed four times in CHAPS buffer and supplied as the kinase sources for in vitro phosphorylation assay.The immunoprecipitate (from 4 mg total proteins) was incubated with 4 µg Flag-cGAS and 200 µM ATP in the kinase assay buffer (10 mM Tris-HCl (pH 7.5), 10 mM MgCl 2 , 0.1 mM EDTA and 2 mM DTT) at 30 °C for 1 h.The reactions were gently tapped every 15 min to mix the reaction well, then subjected to IB or MS analysis.

MS analysis for cGAS phosphorylation
In vitro cGAS phosphorylation reactions were stopped by the addition of 8 M urea and 5 mM DTT, incubated at 37 °C for 60 min and alkylated with 15 mM iodoacetamide for 30 min in the dark at room temperature.Then, 50 mM Tris-HCl (pH 7.8) and 1 mM CaCl 2 were added to make the final urea concentration 1 M, and then digested with trypsin (trypsin:protein, 1:50) at 37 °C overnight.After overnight digestion, the sample was acidified by adding formic acid to a final concentration of 2% to a pH of 2-3 to stop the enzymatic activity.Desalt peptides by using Pierce peptide desalting spin columns (Thermo, 89852).Phosphorylated peptides were enriched by High-Select Fe-NTA phosphopeptide enrichment kit (Thermo, A32992).The enriched products were analysed by high-precision MS and data retrieval was completed by professional proteome discoverer software.
For in vivo cGAS phosphorylation detection, proteins in different cell components were obtained through a Chromatin Extraction Kit (Abcam, ab117152) and ProteoExtract kit (Millipore, 539790).Anti-cGAS antibody-mediated IP was performed with whole-cell lysates (WCLs) derived from three 10-cm dishes of HCT116 cells.The IP proteins were resolved by SDS-PAGE and identified by Coomassie staining.The bands containing cGAS were reduced with 10 mM DTT for 30 min, alkylated with 55 mM iodoacetamide for 45 min and in-gel-digested with trypsin enzymes.The resulting peptides were extracted from the gel and analysed by microcapillary reversed-phase (C18) liquid chromatography (LC)-MS/MS using a high-resolution Orbitrap Elite (Thermo Fisher Scientific) in positive ion DDA mode (Top 6) via higher energy collisional dissociation coupled to a Proxeon EASY-nLc II nano-HPLC.MS/ MS data were searched against the reviewed UniProt Human protein database (v.20230119 containing 20,308 entries) using Mascot v.2.5.1 (Matrix Science) and data analysis was performed using Scaffold v.4.4.8 software (Proteome Software).Peptides and modified peptides were accepted if they passed a 1% false discovery rate threshold.

SILAC medium preparation and cell culture conditions
All standard stable isotope labelling with amino acids in cell culture (SILAC) medium preparation and labelling steps were followed as previously described.In brief, the base medium for DMEM (Macgene) was divided into two parts and to each added l-arginine (Arg 0 ) and l-lysine (Lys 0 ) (light) or 13 C 6 15 N 4 -l-arginine (Arg 10 ) and 13 C 6 15 N 2 -l-Lysine (Lys 8 ) (heavy) to generate the two SILAC-labelling mediums.Each medium with the full complement of amino acids at the standard concentration, was sterile filtered through a 0.22-µm filter (Milipore).Cells were grown in the corresponding labelling medium, prepared as described above, supplemented with 2 mM l-glutamine (Gibco) and 10% dialysed foetal bovine serum (Sigma) plus antibiotics (Gibco), in a humidified atmosphere with 5% CO 2 at 37 °C.Cells were cultured in labelling medium for at least six cell divisions.

TMT labelling
Protein at 100 µg from each sample was reduced with 100 mM DTT at 56 °C for 30 min, followed by the addition of 20 mM iodoacetamide (Sigma) at room temperature while in the dark with alkylating solution for 45 min.Then, the protein sample was diluted by adding 100 mM triethylammonium bicarbonate (TEAB; Sigma) to a urea concentration below 2 M. Finally, each protein sample was digested with trypsin at a mass ratio of 1:50 (trypsin:protein) for the first digestion overnight for 16-18 h and at 1:100 (trypsin:protein) for the second 4 h digestion.After trypsin digestion, peptides were desalted with a Strata X C18 SPE column (Phenomenex) and vacuum dried.Subsequently, peptides were reconstituted in 0.5 M TEAB and processed according to the manufacturer's protocol for the six-plex TMT kit (Thermo Fisher Scientific).In brief, one unit of TMT reagent (defined as the amount of reagent required to label 100 µg of proteins) was equilibrated at room temperature; 100 µg of each sample was resuspended in 24 µl anhydrous acetonitrile and TMT reagent was added to the peptides dissolved in https://doi.org/10.1038/s41556-024-01473-00.5 M TEAB.After 2 h at room temperature, 8 µl 5% hydroxylamine (w/v) was added and incubated for 15 min.The samples were then combined, desalted and dried via vacuum centrifugation.

LC-MS/MS analysis
The dried peptides were then fractionated using a high-pH reverse-phase HPLC system fitted with an Agilent 300Extend C18 column (5-µm particles, 4.6 mm internal diameter, 250 mm length).In brief, the peptides were first separated with a gradient of 2% to 60% acetonitrile in 10 mM ammonium bicarbonate (pH 10) over 80 min into 51 fractions.Then, the peptides were combined into 18 fractions and dried with vacuum centrifugation.The peptides were dissolved in 0.1% solvent A (formic acid), directly loaded onto a reversed-phase pre-column (Thermo Scientific, Acclaim PepMap 100).Peptide separation was performed using a reversed-phase analytical column (Thermo Scientific, Acclaim PepMap RSLC).The gradient was increased with solvent B (0.1% formic acid and 90% anhydrous acetonitrile) from 8% to 26% in 22 min, from 26% to 40% in 12 min and increased to 80% in 3 min to remain at 80% for the last 3 min.This was carried out at a constant flow rate of 400 nl min −1 using an EASY-nLC 1000 UPLC system.The peptides were then analysed on a Q Exactive plus hybrid quadrupole-orbitrap mass spectrometer (Thermo Fisher Scientific).The peptides were analysed via the Q Exactive plus (Thermo Scientific) with a positive ion model and data-dependent acquisition.The resolution of the MS scan was 70,000, and the ion fragment was 17,500.Based on the MS scan, the top 20 precursor ions were selected to fragment with 30 s of dynamic exclusion.The electrospray voltage applied was 2.0 kV.The MS/MS spectra were generated using automatic gain control to prevent overfilling of the Orbitrap and accumulation of 5E4 ions.For the MS scans, the m/z scan range was set from 350 to 1,800.The first fixed mass was set at 100 m/z.

CUT&Tag
The Epicypher protocol for Cleavage under targets and tagmentation was used with slight modifications 34 .Concanavalin A (BioMag Plus, 86057) beads were activated with bead activation buffer and stored on ice until further use.Cells (100,000) were collected and washed with cold PBS.Cells were spun at 800g for 5 min at 4 °C and PBS supernatant was removed from the cell pellet.Nuclear extraction buffer was added to the tube and the pellet was gently resuspended by pipetting to lyse cells and extract nuclei.Activated Concanavalin A beads and nuclei were incubated were mixed and incubated at room temperature for 10 min.The nuclei-conjugated bead complexes were resuspended in antibody binding buffer and adding primary antibody rotating on a nutator overnight at 4 °C (2, 1 and 1 µg of IgG, cGAS and HA were added, respectively).The primary antibody mixture was removed and nuclei-bead complexes were incubated with 0.5 µg secondary antibody in digitonin 150 buffer for 1 h at room temperature on the nutator.After secondary antibody incubation, samples were washed with digitonin 150 buffer and resuspended in digitonin 300 buffer supplemented with 2 µl of CUTANA pAG-Tn5 (Epicycpher, 15-1117) added per sample.Samples were incubated with Tn5 for 1 h at room temperature on the nutator.Digitonin 300 buffer was added two times to remove excess enzyme from samples.Targeted chromatin tagmentation was completed following the Epicypher protocol.Libraries were amplified with 14 PCR cycles and purified by single sided 1.3× AMPure bead purification.The NextSeq500 and 35 base-pair paired-end sequencing parameters were used for library sequencing.

FACS cell cycle analysis
HCT116 cells were synchronized in thymidine 2.5 mM for 24 h and released in nocodazole 100 nM for 16 h.Mitotic arrested cells were collected by shake off and plated in nocodazole for 56 h, then released and seeded for subsequent analysis.
Cells were incubated with 33 µM bromodeoxyuridine for 20 min, collected and centrifuged at 250g for 10 min.The pellet was resuspended in 750 µl PBS 1× and fixed by adding 2,250 µl of ice-cold (−20 °C) pure ethanol dropwise while vortexing.Samples were washed once in 1% BSA/PBS and resuspended in 1 ml 2 N HCl and incubated for 25 min at room temperature allowing DNA denaturation.Then 3 ml 0.1 M sodium borate (pH 8.5) was added to neutralize the acidic pH of the HCl solution and samples were incubated at room temperature for 2 min, centrifuged and washed twice in 1% BSA/PBS.Samples were then transferred to an Eppendorf tube and centrifuged at 800g for 5 min.Pellets were resuspended in 100 µl pure anti-bromodeoxyuridine antibody (Life Technologies) diluted 1:5 in 1%BSA/PBS and incubated for 1 h at room temperature in the dark.Samples were washed with 1% BSA/PBS and resuspended in 100 µl anti-mouse FITC (Life Technologies) diluted 1:50 in 1% BSA/PBS for 1 h at room temperature in the dark.After washing once with 1% BSA/PBS pellets were resuspended in 1 ml propidium iodide (2.5 µg ml −1 ) and RNase (250 µg ml −1 ) (ribonuclease A from bovine pancreas, Sigma) and incubated overnight at 4 °C.Acquisition was made with FACScalibur and data analysis was performed with FlowJo software.

Cell viability measurements
Cell viability was typically assessed in 96-well format by Cell Counting Kit-8 (CCK-8; Dojindo) and alamarBlue Cell Viability Reagent Blue.In brief, cells were seeded onto 96-well plates at a density of 2 × 10 4 per well.Subsequently, cells exposed to 10 µl CCK-8 reagent (100 µl medium per well) for 1 h at 37 °C, 5% CO 2 in an incubator.The absorbance at a wavelength of 450 nm was determined using a FLUOstar Omega microplate reader (BMG Labtech).The alamarBlue (Invitrogen) fluorescence (ex/em 530/590) was measured on a Victor3 plate reader (PerkinElmer).In some experiments, Trypan blue dye exclusion counting was performed by using an automated cell counter (ViCell, Beckman-Coulter).Cell viability under test conditions is reported as a percentage relative to the negative control treatment.

Fluorescence microscopy
For detection of subcellular localization by immunofluorescence, after being fixed with 4% paraformaldehyde and permeabilized in 0.2% Triton X-100 (PBS), cells were incubated with the indicated antibodies (dilution 1:50) for 8 h at 4 °C, followed by incubation with TRITC-conjugated or FITC-conjugated secondary antibody (dilution 1:200; Cwbio) for 1 h at 25 °C.F-actin was detected with Acti-stain 555 phalloidin (Cytoskeleton).The nuclei were stained with DAPI (Sigma) and images were visualized with a Zeiss LSM 510 Meta inverted confocal microscope.

ChIP-qPCR
The EZ-Magna ChIP G kit (17-409; Sigma-Aldrich, Merck KGaA) was used to analyse the binding between cGAS and GLS1 promoter.In brief, 1 × 10 7 HCT116 cells were digested and collected for protein-DNA crosslinking in formaldehyde.The crosslinking was terminated by glycine.Then, the cells were lysed and ultrasonicated for DNA truncation.The lysates were probed with the cGAS antibody (1:100 dilution, 31659; CST) or rabbit IgG (1:100 dilution, ab171870; Abcam).Protein and DNA were de-crosslinked by proteinase K treatment.DNA was collected and purified, in which the enrichment of GLS1 promoter fragments was analysed by qPCR analysis.Primer sequences used for ChIPs are listed in Supplementary Table 6.

RNA preparation and quantitative real-time PCR analysis
HCT116 cells were lysed in Trizol (Invitrogen).Total RNA was recovered from cells following the manufacturer's protocol.Two micrograms of purified RNA from each sample was reverse transcribed to single-stranded complementary DNA with an All-In-One RT MasterMix (ABM, G486).The newly synthesized cDNA was mixed with TransStart Top Green qPCRSuperMix (AQ131, Transgen Biotech) in a volume of 20 µl.For quantitative PCR, a Real Time PCR Detection system (ABI 7500) was using to detect each gene in triplicate.Fold changes were https://doi.org/10.1038/s41556-024-01473-0analysed (quantified) relative to the internal control gene ACTB on the basis of the 2 −ΔΔCT method.The qPCR primer pairs are listed in Supplementary Table 7.

Dot immunoblot assays
Peptides were spotted onto nitrocellulose membranes, allowing the solution to penetrate (usually 3-4 mm diameter) by applying it slowly at a volume of 1 µl.The membrane was dried and blocked in non-specific sites by soaking in TBST buffer with 5% non-fat milk for IB analysis as described previously.

Glutaminase activity measurement assay
Glutaminase activity was determined using a GLS Assay kit (Biomedical Research Service).In brief, 2 million cells were washed with ice-cold PBS and lysated by 100 µl 1× Cell Lysis buffer on ice for 5 min with gentle agitation.The supernatant was collected after centrifugation at maximum for 3 min.Followed by measuring the protein concentration, samples were diluted to 0.2-2 mg ml −1 and 10 µl was used for GLS assay.Samples were combined with 40 µl fresh glutamine solution and incubated in a humidified 37 °C non-CO 2 incubator for at least 2 h.Followed by adding 50 µl TA Assay solution and incubating for another 2 h, the reaction was stopped by adding 50 µl 3% acetic acid.GLS activity was measured by absorbance at OD 492 using a plate reader (Versamax).

Colony formation assays
Cells were seeded into 12-well plates (1,000 or 2,000 cells per well) and left for 8-12 days (37 °C and 5% CO 2 ) until the formation of visible colonies.Colonies were washed with PBS, fixed with 10% acetic acid/10% methanol for 20 min and then stained with 0.4% crystal violet in 20% ethanol for 20 min.After staining, the plates were washed and air-dried, and colony numbers were counted.Three independent experiments were performed to generate the standard error of the difference (SED).

Soft-agar colony formation assay
In brief, the assays were performed using six-well plates where the solid medium consisted of two layers.A total of 1 × 10 4 or 3 × 10 4 cells were resuspended in DMEM containing 0.35% low-melting agarose (Sigma) and 10% FBS and seeded onto a coating of 0.7% low-melting agarose in DMEM containing 10% FBS.Plates were incubated at 37 °C and 5% CO 2.Then, 500 µl of complete DMEM was added every 7 days to keep the top layer moist and 4 weeks later, the cells were stained with iodonitrotetrazolium chloride (1 mg ml −1 ) (Sigma, I10406) for colony visualization and counting.Colonies larger than 0.1 mm in diameter were scored as positive.Three independent experiments were performed to generate the SED.

Tumour growth assay
BALB/c nude mice (6 weeks old, 18.0 ± 2.0 g) were randomly divided into the indicated groups; the mice in the groups were subcutaneously injected with the indicated cells stably expressing the indicated shRNAs or constructs (2 × 10 6 cells in a volume of 100 µl PBS).Tumour growth was measured twice weekly with a caliper for the length (L, largest diameter) and perpendicular width (W), including the skin fold.The volume was calculated using the formula V = W 2 × L/2.The tumour latency/lag phase was recorded and defined as the duration between tumour implantation and the first palpable tumour detection.

PDX establishment
Tumour specimens were collected in serum-free DMEM and used within 24 h.Fragments (4-8 mm) or cell pellets were mixed with 10% Matrigel (Corning, 354234) at 4 °C and implanted subcutaneously into the flanks of 6-8-week-old male NOD-SCID mice (five mice per patient sample).Tumour growth was measured twice weekly with a caliper for the length (L, largest diameter) and perpendicular width (W), including the skin fold.The volume was calculated using the formula V = W 2 × L/2.
The tumour latency/lag phase was recorded and defined as the duration between tumour implantation and the first palpable tumour detection.The initial implant of the patient tumour fragment into the mouse host was defined as passage 0 (P0), followed by serial propagation of tumour fragments in subsequent new hosts.Tumours were collected once they reached a humane end point size of 1.5 cm in largest diameter and were divided for further studies.Engraftment was successful if tumours were established in the initial tumour implant (P0) and as stable engrafters in at least another two successive passages.

Mouse models of colitis-associated colorectal cancer
cGAS-KO mice were provided by F. You (Peking University Health Science Centre, Beijing) and used and genotyped according to the protocols provided by The Jackson Laboratory.To induce colitis-associated colorectal cancer, mice were intraperitoneally injected with 10 mg kg −1 BW AOM (Sigma-Aldrich, A5486) and subsequently treated with three cycles of DSS (36,000-50,000 MW; MP Biomedicals, 0216011010) starting on day 7. Animals received 2.5% DSS in the drinking water for 7 days (first two cycles) or 5 days (third cycle) ad libitum followed by a recovery phase of 14 or 16 days, respectively.

Statistics and reproducibility
Experimental data were analysed and processed in Excel (2016) and plotted using GraphPad Prism v.9.4.0 (GraphPad Software) using unpaired two-tailed t-tests and one-way ANOVA with Dunnett correction for multiple comparisons.All results are shown as the mean ± s.e.m. of multiple independent experiments, not technical replicates.The number of experiments, sample size and statistic tests are reported in the respective figure legends.No statistical method was used to predetermine sample size.Animals were randomly assigned to the experimental groups to prevent bias in group allocation.The order of experimental conditions or stimulus presentation was not randomized.No data were excluded from the analyses.The investigators were not blinded to allocation during experiments and outcome assessment.

Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.Extended Data Fig. 2 | mTORC2 induces cGAS chromatin localization.a. Immunoblot analysis of whole-cell lysates (WCLs) from HCT116 cells treated with PI3K-mTOR pathway inhibitors (2 µM, 12 hours).b.Automated quantification of cGAS intensities in individual cell nuclei and cytosol from (B) using CellProfiler.Nuclear/cytosolic ratios are plotted as violin plots with mean for each condition.n = 50 from 3 independent experiments.Unpaired two-tailed t-test.c.Immunoblot analysis of whole cell lysates from HCT116 cells lentivirally infected with scrambled (sh-Src) or shRNAs targeting mTOR components.d.Immunofluorescence analysis and quantification cGAS subcellular distribution in HCT116 cells lentivirally infected with the scramble control or gene-targeted shRNAs.Nuclear/cytosolic ratios are plotted as violin plots with means for each condition.n = 50 from 3 independent experiments.Unpaired two-tailed t-test.e. Co-immunoprecipitation and immunoblot analysis of cGAS and mTORC2 component interaction in HCT116 cells.f.Co-immunoprecipitation of Myc-tagged cGAS mutants from HCT116 cells transfected with the indicated constructs.g. eBRET2 signal reporting the interaction between cGAS and SIN1 (mTORC2 component) in HCT116 cells expressing a GFP2-cGAS and Rluc8-SIN1 biosensor.Bars represent the means ± s.e.m. from three independent experiments.Unpaired two-tailed t-test.h.Schematic of the workflow establishing an in vitro cGAS phosphorylation system using purified mTORC2 complexes and AKT1 as a control substrate.i. Mass spectrometry analysis of in vitro cGAS phosphorylation by purified mTORC1 or mTORC2 complexes from starved (48 h) or insulin-stimulated (30 min with 100 nM insulin) HCT293T cells.Experiments were repeated three times (a, c, e, f, h) with similar results.Numerical source data and unprocessed blots are available as source data.
Extended Data Fig. 3 | mTORC2 induces cGAS phosphorylation at serine 37. a.In vitro cGAS phosphorylation assays by means of LC-MS/MS showed that purified mTORC1 could not directly phosphorylate recombinant His-cGAS.b.Generation of a pSer37-cGAS polyclonal antibody by immunizing rabbits with a phosphorylated peptide antigen and affinity purifying the antibodies.The antigen sequence used for immunization was cGAS aa 29-47 (GAPMDPTES*PAAPEAALPK).S* stands for phosphorylated serine residue in these synthetic peptides.The antibodies were affinity purified using the antigen peptide column, but they were not counter selected on unmodified antigen.c.Dot blot analysis of antibody specificity using serial dilutions of phosphorylated and non-phosphorylated cGAS peptides.This experiment was performed three times with similar results.d.Correlation between cell lysate protein concentration and absorbance signal in a pSer37-cGAS ELISA using HCT116 and MC38 cell extracts, demonstrating cell type-specific expression.This experiment was performed three times with similar results.e. HCT116 cells were treated with GDC-0941 for times indicated, then WCLs were analysed by immunoblot (IB) with the antibodies shown.This experiment was performed three times with similar results.f.Purified mTORC2 or AKT1 was incubated with recombinant His-cGAS to establish in vitro cGAS phosphorylation system, and the phosphorylation sites were analysed by means of LC-MS/MS.g.In vitro cGAS phosphorylation assays by means of LC-MS/MS showed that purified AKT1, but not mTORC2, directly phosphorylated cGAS at serine 305 in vitro.Unprocessed blots are available as source data.
Extended Data Fig. 4 | Serine 37 phosphorylation promotes cGAS chromatin localization.a. eBRET2 signal reporting on chromatin (H2A) binding of cGAS mutants following tansfection into cGAS knockout HCT116 cells.b.Immunofluorescence analysis of subcellular localization of cGAS S37A and S37D mutants re-expressed by lentiviral transduction in cGAS-knockout HCT116 cells.Scale bar, 20 µm.c.Immunofluorescence analysis of p-Ser 37 -cGAS (serine 37 phosphorylation) and H2A (chromatin marker) subcellular distribution in HCT116 cells.d. cGAS proteins were immunoprecipitation from extranuclear and intranuclear (chromatin-bound) fractions of HCT116 cells, and the phosphorylation sites were analysed by means of LC-MS/MS analysis.e. Immunofluorescence analysis of extranuclear versus intranuclear localization of cGAS S37D and S37A mutants re-expressed in cGAS-knockout HCT116 cells.f.Immunofluorescence analysis of cGAS S37D and S37A mutant localization in mitotic HCT116 cells, showing chromatin binding capacity.g.ELISA quantification of cGAS protein levels in chromatin fractions isolated using a Chromatin Extraction Kit from STING-/-HCT116 cells treated with 10 µM JR-AB2-011 for indicated times.h.eBRET2 signal measured the interaction between H2A-Rluc8 and cGAS-GFP in STING-/-HCT116 cells treated with 10 µM JR-AB2-011 for indicated times.Data are shown as means ± s.e.m. from three independent experiments (a, g, h).Unpaired two-tailed t-test.Experiments were repeated three times (b, c, e, f) with similar results.Numerical source data are available as source data.
Extended Data Fig. 5 | Interactome analysis identifies SWI/SNF complex recruitment by ccGAS.a. Schematic of experimental workflow for identification of cGAS interacting proteins.ProteoExtract®Kit and Chromatin Extraction Kit were used to obtain proteins of intracellular and extracellular cell fractions, and then the interaction proteins of cGAS in different cell fractions were obtained by immunoprecipitation. cGAS interactome was analysed by means of LC-MS/ MS. b.LC-MS/MS analysis detected the binding proteins of cGAS in different cell fractions, and the frequency distribution histogram described the signal distribution.By comparing with the control group, proteins that significantly interact with cGAS in cytoplasm and nucleus could be screened out.c.Gene Ontology enrichment of cGAS interacting proteins in the cytoplasm.d.PPI network of 102 nuclear cGAS interactors including 89 nodes and 451 interactions annotated from public databases.among which 6 are cGAS and their Direct interactors are highlighted in orange.e. Gene Ontology enrichment of cGAS interacting proteins in the nucleus.f.PPI network of 33 unique intranuclear cGAS interactors including 30 nodes and 95 interactions, showing enrichment for SWI/SNF complex components.g.Co-immunoprecipitation and immunoblot validation of ARID1A binding to indicated cGAS mutants.This experiment was performed three times with similar results.Statistics source data and unprocessed blots are available as source data.f.Genomic distribution of cGAS peaks in relation to essential genes with necessary for cell survival.g.Motif enrichment analysis within top 500 cGASbound regions sorted by signal intensity.h.KEGG pathway analysis of all cGAS CUT&Tag peaks.i. Visualization of cGAS CUT&Tag signals at the promoter of KGA  (chr2: 191,792,416-191,830,270) and PTER (chr10: 16,401,685-16,456,017). j.Gene Ontology enrichment of genes with cGAS CUT&Tag peaks in promoter regions.Statistics source data are available as source data.

Extended
Extended Data Fig. | TMT proteomics identifies DNA replication proteins and kidney-type glutaminase regulated downstream of ccGAS.a. Schematic of experrimental workflow for identification of ccGAS downstream proteins.Proteins from HCT116 cells lentivirally infected with the indicated shRNAs or cGAS-S37A mutants were labelled with TMT reagents and analysed by LC-MS/ MS. b.Protein signal distribution after cGAS knockdown determined by tandem mass tag (TMT).c.After cGAS knockdown, 81 proteins with significant changes in abundance were quantitatively screened from 6523 proteins, of which 28 were upregulated (orange) and 53 were downregulated (blue).d.The interaction frequency of 81 proteins with significant changes in abundance was significantly higher than that of randomized controls.e. Gene Ontology enrichment of 53 downregulated proteins, associated with DNA replication.Statistics source data are available as source data.
Extended Data Fig. 8 | ccGAS depletion regulates expression of the CDC45-MCM-GINS (CMG) complex and KGA.a. Immunoblot analysis of whole cell lysates from HCT116 cells infected with cGAS shRNA or control lentiviruses.b. qPCR analysis of indicated gene expression normalized to ACTB in cGAS knockdown HCT116 cells.c.TMT proteomics quantification of KGA and GAC protein levels in cGAS knockdown HCT116 cells.d.Immunoblot analysis of WCL derived from cGAS-knockdown HCT116 cells infected with lentiviruses encoding shRNA-resistant cGAS.e. qPCR analysis of indicated gene expression in cGAS rescued knockdown HCT116 cells.f.Glutaminase activity assay of parental and cGAS knockout cells expressing cGAS mutants.g.ChIP-qPCR analysis of cGAS occupancy at KGA promoter in indicated HCT116 cells.h.ChIP-qPCR analysis of cGAS occupancy at the KGA promoter in cGAS knockout colorectal cancer cells stably expressing indicated cGAS mutants.i.Wild-type and S213D mutant cGAS were transfected into cGAS-knockout HCT116 cells, and KGA mRNA levels and chromatin-bound proteins were analysed.j.Contribution of glutamine to oxygen consumption in indicated cell lines analysed by Seahorse.Data are shown as means ± s.e.m. from three independent experiments (b, e-j).Unpaired twotailed t-test.Experiments were repeated three times (a, d) with similar results.Numerical source data and unprocessed blots are available as source data.
Extended Data Fig. 9 | ccGAS depletion induces diapause-like state and chemoresistance in colorectal cancer cells.(A) Flow cytometric analysis of cell cycle profiles in HCT116 cells lentivirally infected with the indicated shRNAs.(B) Quantification of cell cycle distributions in HCT116 cells lentivirally infected with the indicated shRNAs.(C) cGAS or STING was transfected into HCT116 cells stably expressing cGAS shRNA.Quantification of cell cycle distributions in cGAS knockdown HCT116 cells stably expressing cGAS, STING, or treated with STING activator ADU-S100 (10 µM).(D) cGAS-S37D/A mutants were transfected into cGAS-knockout HCT116 cells, the subcellular distribution of cGAS mutants was analysed by immunofluorescence, and the morphology of the corresponding cells was observed.(E) Immunofluorescence analysis of cell morphology in cGAS-knockout HCT116 cells with cGAS-S37D mutants restoration.F-actin stain used to label cytoskeleton.(F) Immunofluorescence analysis of cell morphology in cGAS-knockout HCT116 cells with cGAS-S37A mutants restoration.(G) Cell viability assays measuring 5-FU response in cGAS knockdown HCT116 cells stably expressing shRNA-resistant cGAS-S213D mutants or pretreated with 10 µM ADU-S100 for 6 hours.Data are shown as means ± s.e.m. from three independent experiments (b, c, g).Unpaired two-tailed t-test (b, c) and Kruskal-Wallis oneway ANOVA followed by Dunn's multiple comparison tests (g).Experiments were repeated four times (d-f) with similar results.Numerical source data are available as source data.
Extended Data Fig. 10 | The ccGAS-KGA signalling axis is conserved in murine colorectal cancer cells.a. CDKN1A mRNA levels were quantified by qPCR analysis and normalized to ACTB control in HCT116 cells stably expressing indicated shRNAs.b.Cell viability assays measuring 5-FU response in HCT116 cells stably transfected MLH1 or pretreated with 10 µM JR-AB2-011 for 6 hours.JR ( JR-AB2-011), a selective mTORC2 inhibitor.c. qPCR analysis of indicated gene expression in HCT116 cells stably transfected MLH1.d.Colony formation assays measuring drug response of HT29 cells.e. Colony formation assays measuring drug response of SW480 cells.f.HCT116 cells stably expressing indicated shRNAs were treated with the indicated concentrations of 5-FU and cell viability was assessed after 24 hours.g.Glutaminase activity assay of PDX models (RICTOR+/+ and RICTOR−/−).h.ELISA quantification of cGAS protein levels in chromatin fractions isolated using a Chromatin Extraction Kit from MC38 cells stably expressing indicated shRNAs.i. Immunofluorescence analysis of human cGAS S37A or S37D mutant localization in MC38 cells.Scale bar, 10 µm.j. qPCR analysis of gene expression in indicated MC38 cells normalized to ACTB.k.Contribution of glutamine to oxygen consumption in the indicated MC38 cells analysed by Seahorse.Data are shown as means ± s.e.m. from three independent experiments (a-h, j, k).Unpaired two-tailed t-test (a, c-e, g, h, j, k) and Kruskal-Wallis oneway ANOVA followed by Dunn's multiple comparison tests (b, f).Experiments were repeated four times (i) with similar results.Numerical source data and unprocessed blots are available as source data.

5 Fig. 1 |Fig. 2 |
Fig. 1 | High-throughput screening identifies PI3K-mTOR pathway regulation of cGAS-chromatin localization.a, Cell fractionation and ELISA were used to quantify cGAS protein levels in subcellular fractions of HCT116 cells.n = 5 independent experiments per group.b, A chromatin cGAS biosensor composed of pcDNA3.1(+)-GFP2-cGASand pcDNA3.1(+)-Rluc8-H2Awas generated.The eBRET2 signal ratios indicate cGAS-H2A interactions.c, The biosensor underwent high-throughput drug screening in HCT116 cells.Ratios of eBRET2 signals with drugs (Bioactive Compound Library) versus dimethylsulfoxide (DMSO) indicate effects on cGAS-chromatin interactions.d, HCT116 cells were treated with GDC-0941 for the times indicated, then soluble nuclear and chromatin fractions were analysed by immunoblot (IB) with the antibodies shown.e, Immunofluorescence analysis of cGAS subcellular distribution in HCT116 cells treated with PI3K-mTOR inhibitors (2 µM, 12 h).MK-2206, a highly selective AKT inhibitor; rapamycin, an allosteric mTORC1 inhibitor; PF-4708671, a p70 ribosomal S6 kinase inhibitor; Scale bars, 20 µm.f, HCT116 cells transfected with the biosensor were treated with 2 µM of the indicated compounds for 12 h, then eBRET2 signals were measured.Ratios of compounds versus control indicate effects on cGAS-chromatin interactions.n = 3 independent experiments per group.Data are shown as mean ± s.e.m.Unpaired two-tailed t-test.Experiments were repeated three times (d, e) or twice (c) with similar results.Numerical data and unprocessed blots are available as source data.

cFig. 5 |
Fig.5| mTORC2-driven ccGAS directs colorectal cancer plasticity and acquired chemoresistance in vivo.a, mTOR-knockdown HCT116 cells stably transfected mTOR or pretreated with 10 µM JR-AB2-011 for 6 h were treated with the indicated concentrations of 5-FU, and cell viability was assessed after 24 h.b, mTOR-knockdown HCT116 cells stably expressing indicated cGAS mutants were treated with the indicated concentrations of 5-FU and cell viability was assessed after 24 h.c, RICTOR-knockdown HCT116 cells stably expressing indicated cGAS mutants were treated with the indicated concentrations of 5-FU and cell viability was assessed after 24 h.d, HCT116 cells lentivirally infected with indicated shRNAs or cGAS mutants were tested for tumour formation in nude mice (2 × 10 7 cells per mouse).n = 8 mice per group.e, Mice were killed

Fig. 6 |
Fig. 6 | KGA inhibition overcomes chemoresistance induced by disruption of mTORC2-ccGAS axis.a, Mouse xenograft experiments were performed with HCT116 cells stably expressing cGAS shRNA.When the tumour diameter reached 5 mm, 5-FU (23 mg kg −1 , twice a week) was intraperitoneally injected for five consecutive weeks.Tumour growth curves were calculated.n = 5 mice per group.b, HCT116 cells stably expressing cGAS shRNA were treated with 10 µM BPTES and the indicated concentrations of 5-FU and cell viability was assessed after 24 h.n = 3 independent experiments per group.c, HCT116 cells stably expressing the indicated shRNA or cGAS mutants were treated with 10 µM BPTES and the indicated concentrations of 5-FU and cell viability was assessed after 24 h.n = 3 independent experiments per group.d, Immunoblot analysis of WCLs derived from PDX models (RICTOR +/+ and RICTOR −/− ).e, Immunohistochemical

Fig. 7 |
Fig. 7 | Targeting KGA re-establishes chemotherapy sensitivity in tumours of cGAS-deficient mice.a, MC38 cells stably expressing indicated hcGAS mutants were treated with the indicated concentrations of 5-FU and cell viability was assessed after 24 h.n = 3 independent experiments per group.b, The colitisassociated colorectal cancer model was established with AOM/DSS in cGAS-WT and knockout (KO) littermate mice.After 21 days of treatment, intraperitoneal injection (i.p.) was performed using 5-FU (23 mg kg −1 , twice per week) and BPTES (30 mg kg −1 , twice per week) for 7 weeks.c, Representative images of AOM/DSS-

Extended Data Fig. 1 |
PI3K-mTOR pathway inhibition reduces cGAS chromatin localization.a. Immunofluorescence analysis of cGAS localization (green) in HCT116 cells throughout the cell cycle as defined by DAPI staining (blue).Scale bar, 5 µm.b.Immunofluorescence analysis of subcellular localization of cGAS mutants versus wild-type in cGAS knockout HCT116 cells.Scale bar, 10 µm.c. eBRET2 signal measured the chromatin binding affinity of cGAS mutants versus wild-type, as indicated by the interaction between H2A-Rluc8 and cGAS-GFP following transfection into cGAS knockout HCT116 cells.d.SILAC proteomics identified cGAS and HELLS as proteins whose chromatin association is most sensitive to treatment with the PI3K inhibitor GDC-0941.The and Heavy (treated) and light (control) peptide signals are shown inyellow and blue, respectively.e. ELISA quantification of cGAS protein levels in different cell fractions from HCT116 cells treated with GDC-0941 for the indicated times.Fractionation was performed using ProteoExtract® Kit and Chromatin Extraction Kit.Data are means ± SEM of three independent experiments.Data are shown as means ± s.e.m. from three independent experiments (c, e).Unpaired two-tailed t-test.Experiments were repeated three times (a, b) with similar results.Numerical source data are available as source data.

Data Fig. 6 |
CUT&Tag profiling reveals genes regulated by ccGAS associated with cell cycle and amino acid metabolism.a. Schematic of CUT&Tag workflow to map genomic cGAS binding sites.b.Spearman correlation heatmap comparing biological replicates.c.Scatter plot and correlation of peak calls between technical replicates, demonstrating reproducibility.d.Visualization of cGAS CUT&Tag signals on Chromosome 1. e. Metagene analysis showing cGAS CUT&Tag enrichment at transcription start sites (TSS).