RMI2 Depletion Recovers cGAS-STING Signaling to Enhance Immunotherapy in Colorectal Cancer

RMI2 Depletion Recovers cGAS-STING Signaling to Enhance Immunotherapy in Colorectal Cancer | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article RMI2 Depletion Recovers cGAS-STING Signaling to Enhance Immunotherapy in Colorectal Cancer Weixiang Zhan, Runkai Cai, Enmin Huang, Yina Liu, Xiaoshuang Lyu, and 7 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8706982/v1 This work is licensed under a CC BY 4.0 License Status: Under Revision Version 1 posted 9 You are reading this latest preprint version Abstract Microsatellite-stable (MSS) colorectal cancers (CRCs) are largely refractory to immune checkpoint inhibitors (ICIs) due to immunologically "cold" tumor microenvironments (TMEs) characterized by limited T-cell infiltration. While RecQ-mediated genome instability 2 (RMI2) is known for regulating DNA repair, its role in anti-tumor immunity remains unclear. Here, we demonstrate that RMI2 acts as an adaptive oncoprotein in CRC by suppressing innate immune activation through enhanced homologous recombination repair (HRR). Mechanistically, RMI2 stabilizes BRCA1-RAD51 complexes, accelerates DNA double-strand break repair, and limits cytosolic DNA release. Conversely, RMI2 deficiency impairs HRR, causing cytosolic DNA accumulation, cGAS-STING pathway activation, and type I interferon signaling that boosts anti-tumor immunity. Notably, RMI2 knockout in mice synergizes with PD-1 blockade and fluorouracil to induce robust tumor regression and prolonged survival. These findings uncover a previously unrecognized role for RMI2 in maintaining immune evasion through the coordinated regulation of DNA repair and cGAS-STING-dependent innate immune signaling, positioning RMI2 as a promising therapeutic target to convert MSS CRCs from "cold" to "hot" tumors and overcome ICI resistance. Biological sciences/Cancer/Cancer microenvironment Biological sciences/Cell biology/Cell death RMI2 cGAS-STING colorectal cancer DNA damage tumor microenvironment Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Highlights λ RMI2 maintains immune evasion in MSS CRC by suppressing cytosolic DNA sensing. λ RMI2 loss triggers innate immune activation via the cGAS–STING pathway. λ Targeting RMI2 enhances immunotherapy efficacy in MSS CRC. Introduction Colorectal cancer (CRC) remains a leading cause of cancer-related morbidity and mortality worldwide. Although immune checkpoint blockade (ICB) therapy has demonstrated remarkable efficacy in a subset of patients with mismatch repair-deficient (dMMR) or microsatellite instability-high (MSI-H) CRC 1 , 2 , 3 , the majority of CRC cases, particularly those classified as microsatellite stable (MSS), exhibit primary resistance to ICB. This resistance is largely attributed to an immunologically “cold” tumor microenvironment (TME) characterized by low infiltrating CD8⁺ T cell density, limited antigen presentation, and suppressed type I interferon signaling 4 , 5 , 6 , 7 . Consequently, there is an urgent unmet need to identify molecular targets capable of reprogramming the TME to enhance ICB responsiveness in MSS CRC. Genomic instability is a hallmark of cancer and a potential source of tumor neoantigens that elicit antitumor immune responses. Previous studies from our laboratory and others have indicated that RecQ-mediated genome instability 2 (RMI2) is an evolutionarily conserved component of the RMI complex, which, together with BLM and TOP3A, resolves toxic homologous recombination intermediates and maintains genomic stability. Emerging evidence from our group and others indicates that RMI2 functions as a bona fide oncoprotein across multiple malignancies, where it supports tumor cell survival by preventing the accumulation of DNA damage and inhibiting premature senescence or apoptosis 8 , 9 , 10 , 11 , 12 . RMI2 depletion triggers genomic instability, leading to the accumulation of DNA damage 12 . Defects in DNA repair mechanisms or oncogene activation drive genomic instability, resulting in increased somatic nonsynonymous mutations that generate abundant tumor-specific neoantigens through aberrant protein synthesis pathways 13 , 14 , 15 . Genomic instability can enhance the efficacy of ICB by revealing immune vulnerabilities 16 , 17 . The immune system identifies these neoantigens as non-self-antigens, thereby initiating potent T-cell priming and cytotoxic responses 18 , 19 , 20 . However, the precise role of RMI2 in modulating the interplay between genomic instability, DNA sensing, and antitumor immunity remains poorly defined. The stimulator of interferon genes (STING) pathway, activated upon recognition of cytosolic DNA by cyclic GMP-AMP synthase (cGAS), exerts dual immunoregulatory functions in tumorigenesis 21 , 22 . While STING-induced type I interferon signaling can enhance antigen processing and T-cell infiltration within immunologically hostile tumor microenvironments (TMEs), cancer cells frequently circumvent STING-mediated immune surveillance through epigenetic mechanisms that silence STING expression 23 , 24 . STING dimers execute ER-to-Golgi translocation to assemble into oligomeric complexes in the Golgi membrane. Subsequent TBK1 binding to STING oligomers triggers autophosphorylation-mediated activation, facilitating TBK1-mediated phosphorylation of STING and downstream IRF3. Phosphorylated IRF3 dimers undergo nuclear translocation to induce type I interferon gene transcription 25 , 26 , 27 . Interestingly, certain genetic or pharmacological perturbations can restore the STING pathway activity in tumors. For instance, in MSI-H CRC, constitutive activation of the cGAS-STING pathway correlates with enhanced antigen presentation and immune cell infiltration 28 , 29 , 30 . Similarly, SHP2-mediated dephosphorylation of PARP-1 following DNA damage leads to cytosolic dsDNA accumulation and STING activation 31 . The depletion of serine/threonine phosphatases or arginine methyltransferases, such as PRMT6, has also been reported to enhance STING signaling and antitumor immunity 32 , 33 . These findings suggest that targeting pathways that regulate cytosolic DNA processing and STING activation may represent a viable strategy for converting immunologically cold tumors into immunologically responsive ones. In this study, we demonstrated that RMI2 is adaptively upregulated in CRC and suppresses the STING pathway by maintaining genomic stability through enhanced homologous recombination repair. RMI2 deficiency disrupts this protective mechanism, leading to the accumulation of cytosolic DNA, activation of the cGAS-STING pathway, and the subsequent enhancement of antitumor immunity. Our findings uncover a previously unrecognized dual role of RMI2 in simultaneously preserving genomic integrity and suppressing immune surveillance, thereby facilitating immune evasion. Materials and Methods Patients and samples In this study, formalin-fixed paraffin-embedded (FFPE) tumor specimens were collected from 226 consecutive patients with CRC who underwent primary tumor resection between January 2015 and December 2016 at our institution, with comprehensive clinicopathological data obtained during a median follow-up duration of 7 years. Overall survival (OS) was defined as the interval from the date of curative surgery to the date of death from any cause, with surviving patients censored at the date of the last follow-up. To analyze the refractory metastatic MSS CRC immunotherapy cohort, we conducted a rigorous retrospective analysis of prospectively maintained medical records, including patients who met all of the following criteria: histological and imaging studies confirmed metastatic CRC with MSS status, documented disease progression following standard chemotherapy regimens, and subsequent treatment with PD-1 checkpoint inhibitors at our institution. The observational period spanned July 2018 to April 2024. Tumor responses were assessed using serial cross-sectional computed tomography (CT) imaging and evaluated according to RECIST 1.1 criteria. Following standardized protocols, two independent radiologists performed blinded assessments of the CT images to measure and record the longest diameters of all target lesions. Tumor responses were categorized into four distinct groups based on the RECIST 1.1 criteria: (1) Complete Response (CR), defined as the complete disappearance of all target lesions; (2) Partial Response (PR), defined as a ≥ 30% reduction in the sum of the longest diameters of target lesions compared to baseline; (3) Progressive Disease (PD), defined as a ≥ 20% increase in the sum of the longest diameters of target lesions compared to the smallest sum observed since treatment initiation, or the appearance of new lesions; and (4) Stable Disease (SD), defined as neither sufficient shrinkage for PR nor sufficient increase for PD. Tumor responses were further dichotomized into response (R), including CR and PR, and Non-Response (NR), including PD and SD. To ensure interobserver reliability, a single-blinded assessment framework was utilized, wherein radiologists were blinded to the clinical outcomes but had access to baseline and follow-up imaging data. Any discrepancies between the evaluators were resolved through consensus-based discussions, with the final classification determined by the majority consensus. All clinicopathological data were extracted from the institutional CRC registry database according to protocols approved by the Institutional Review Board of the Sixth Affiliated Hospital, Sun Yat-sen University (2022ZSLYEC-002). Written informed consent was waived due to the retrospective nature of this study, and strict patient confidentiality was maintained in compliance with the principles of Declaration of Helsinki. Cell lines All cell lines, including human CRC cell lines (DLD-1, HCT116, SW480, and RKO), murine CRC cell lines (CT26), and HEK293T cells, were obtained from the American Type Culture Collection (ATCC). Cells were cultured at 37°C in a humidified incubator containing 5% CO₂, using DMEM or RPMI-1640 medium (Bio-channel) supplemented with 10% fetal bovine serum (FBS; ExCell Bio) and 1% penicillin-streptomycin. Cell lines were authenticated by short tandem repeat (STR) profiling and confirmed to be free of mycoplasma contamination using standard PCR-based detection methods. All experiments were conducted using cells at low passage numbers (≤ 20 passages). Plasmids Through rigorous molecular cloning methods, we constructed stable expression vectors to investigate the functional roles of RMI2, using the pSin-EF2-puro-oligo lentiviral backbone. This plasmid vector is specifically designed for efficient, long-term transgene expression in mammalian cells, facilitating selection with puromycin. To achieve stable knockdown of human RMI2, short hairpin RNA (shRNA) constructs were generated and cloned into the pLKO.1-puro lentiviral expression vector, containing a puromycin resistance cassette for stable selection. For genome editing applications using the CRISPR-Cas9 system, we designed single-guide RNA (sgRNA) constructs targeting mouse RMI2 and STING, which were cloned into the pLentiCRISPR-V2-puro backbone. The specific sequences of the shRNAs/sgRNAs are listed in Table S4. To ensure the integrity and accuracy of all the genetic constructs, each construct was rigorously validated by DNA sequencing and subjected to strict quality control assessments. Antibodies and reagents Primary antibodies used for Western blotting included: RMI2 rabbit antibody (Abcam, ab122685; 1:1,000 dilution), γH2AX rabbit antibody (Cell Signaling Technology, 9718; 1:500 dilution), STING rabbit antibody (Cell Signaling Technology, 13647; 1:1,000 dilution), Phospho-STING rabbit antibody (Cell Signaling Technology, 50907; 1:500 dilution), TBK1 rabbit antibody (Cell Signaling Technology, 3504; 1:1,000 dilution), Phospho-TBK1 rabbit antibody (Cell Signaling Technology, 5483; 1:500 dilution), IRF-3 rabbit antibody (Cell Signaling Technology, 11904; 1:1,000 dilution), Phospho-IRF-3 rabbit antibody (Cell Signaling Technology, 29047; 1:500 dilution), RAD51 mouse antibody (Abcam, ab88572, 1:1,000 dilution), BRCA1 mouse antibody (Santa Cruz, sc-6954, 1:500 dilution) Flag rabbit antibody (Cell Signaling Technology, 14793; 1:2,000 dilution), GAPDH rabbit antibody (ProteinTech, 10494-1-AP; 1:2,000 dilution), and GAPDH mouse antibody (ProteinTech, 60004-1-Ig; 1:10,000 dilution). For flow cytometry, the following conjugated antibodies were used: PE/Cyanine7-conjugated anti-mouse CD3 antibody (clone 17A2, BioLegend), Brilliant Violet 650-conjugated anti-mouse CD8a antibody (clone 53 − 6.7, BioLegend) or PE-Cyanine5.5-conjugated anti-mouse CD8a antibody (clone 53 − 6.7, Invitrogen), Alexa Fluor 700-conjugated anti-mouse CD4 antibody (clone GK1.5, BioLegend), eFluor 450-conjugated anti-mouse Granzyme B antibody (clone NGZB, Invitrogen), PE-conjugated anti-mouse IFN-γ antibody (clone XMG1.2, BioLegend), and PE conjugated anti-mouse Ki67 antibody (clone 16A8, BioLegend). HRP-conjugated secondary antibodies for mouse and rabbit IgG were obtained from Cell Signaling Technology. Puromycin, PicoGreen, and DAPI staining reagents were obtained from Sigma-Aldrich. Lentivirus and stable expression cell lines construction Here, we describe the construction of lentivirus and stably expressing cell lines using HEK-293T cells. Cells were seeded in six-well plates at approximately 60% confluence. The following day, a transfection cocktail was prepared, consisting of 3 µg of pLKO.1-shRNA or pLentiCRISPR-sgRNA/pSinEF2-cDNA, 2 µg of psPAX2 and 1 µg of pMD2G, dissolved in 24 µL of polyethylenimine (PEI) at a concentration of 2 mg/mL. The cells were then transfected with the cocktail. After incubation for 48 h, the supernatant was collected, filtered through 0.45 µm PVDF filters (Millipore), and used to infect tumor cells in the presence of polybrene (10 µg/mL; Sigma). The infected cultures were centrifuged at 2000 rpm (800×g) for 45 min at 37°C, followed by replacement of the medium with puromycin. Stable cell lines were selected for 1 week using puromycin concentrations optimized for each cell line (0.5 µg/mL for DLD1 and HCT116, and 2 µg/mL for CT26 and MC38). The cells were then analyzed by western blotting to confirm protein expression. All stable cell lines were maintained for no more than 1 month prior to use. Animal experiments All animal experiments were conducted in strict compliance with the "Guide for the Care and Use of Laboratory Animals" and the "Principles for the Utilization and Care of Vertebrate Animals," and were approved by the Animal Research Committee of the Sixth Hospital of Sun Yat-sen University (SYSU-IACUC-2020112701). Male C57BL/6J, BALB/c, and BALB/c nude mice (all sourced from GemPharmatech) were used in this study. These mice were maintained under stringent specific pathogen-free (SPF) conditions with the following environmental controls: relative humidity of 50%, a 12-hour light cycle alternating with 12 hours of darkness, and a temperature range of 25°C to mimic natural diurnal variations. Before the start of the experiments, mice were allowed a minimum 5-day acclimation period to ensure they were comfortable and to minimize the impact of stress-related variables on experimental outcomes. In our experimental design, both subcutaneous and orthotopic tumor models were used to investigate the effects of RMI2 on tumor growth and immune response in vivo. For the subcutaneous model, 5×10 5 CT26 or MC38 tumor cells were subcutaneously injected into the right dorsal region of the randomized mice. Tumor volumes were measured using Vernier calipers according to the following formula: Volume (mm 3 ) = 1/2×(width 2 ×length) Mice were sacrificed when tumors reached a maximum diameter of 15 mm or total volume of 2000 mm³. An orthotopic tumor model was used as described in previous studies. Briefly, after opening the BALB/c mice's skin and muscle layer with sterile surgical tools, the cecum was exposed. Using a 0.3 mL disposable insulin syringe, 5×10 5 CT26 tumor cells in 50 µL (Matrigel: phosphate-buffered saline = 1:4) were slowly injected into the subserosal layer of the cecum, followed by careful closure of the muscle and skin layers. The success of the orthotopic tumor model was confirmed using luciferase assay. To evaluate the effect of CD8 and NK immune cells on the immune response in our mouse model, we employed a depletion strategy using specific antibodies against these cell populations. CD8 + T and NK cells were deleted by using 100 µg of anti-mouse CD8α antibody (clone 2.43, BioXcell) and 100 µg of anti-mouse NK1.1 antibody (clone PK136, BioXcell) intraperitoneally injected on days − 3, 0, 3, 6 and 9. Single-cell RNA sequencing CT26 tumors were enzymatically dissociated into single-cell suspensions, and the cell concentration was adjusted to 600–1200 cells/µL. Single-cell RNA sequencing (scRNA-seq) libraries were prepared using a DNBelab C4 scRNA-seq kit (MGI) with droplet-based mRNA capture. The workflow consisted of cDNA synthesis, PCR amplification, fragment treatment, end repair, A-tailing, and adapter ligation. The resulting libraries were loaded onto DNB nanoarrays and sequenced on the DNBSEQ-T7 platform using Combinatorial Probe-Anchor Synthesis (cPAS). For quality control, Fastp and Seurat were applied using the following criteria: ≥3 cells per gene and ≥ 200 genes per cell. Data integration was performed using Canonical Correlation Analysis (CCA) for batch correction, followed by dimensionality reduction via t-SNE and UMAP. Differential gene expression analysis was conducted using the edgeR package, and tissue-specific markers were identified. The comet assay The comet assays were conducted using the Beyotime Comet Assay Kit (C2041S) following the manufacturer's instructions. Low-melting agarose was heated in water at 70–80°C for 10 min, and subsequently cooled slowly in a 37°C water bath for ≥ 20 min to ensure complete dissolution. Cell suspensions (1×10⁶ cells/mL) were mixed with molten agarose at a 1:7.5 (v/v) ratio, and 70 µL of the mixture was pipetted onto comet slides. Slides were refrigerated at 4°C for 10 min to solidify the agarose matrix, followed by incubation in lysis buffer at 4°C for 2 h or overnight to lyse cells. The lysis buffer was pre-chilled at 4°C for ≥ 20 min prior to use. After neutralization with the neutralization buffer for 30 min, slides were subjected to horizontal electrophoresis at 25 V for 30 min. Subsequently, a DNA precipitation solution was applied to remove proteins and salts. Samples were stained with propidium iodide (PI) for 20 min in the dark and visualized under an epifluorescence microscope (Olympus). Tail moment analysis was quantified using the Comet Assay Software Project (CASP), with appropriate negative/positive controls included. Immunohistochemistry We constructed a tissue microarray (TMA) from CRC tissue blocks and performed immunohistochemistry (IHC) to assess the expression and localization of specific proteins. IHC staining was conducted using a standardized protocol. Initially, tissue sections were deparaffinized and rehydrated under gentle conditions. Antigen retrieval was achieved by boiling the sections in 1×EDTA buffer pH 9.0 for 2.5 min to unmask intracellular antigens. To minimize non-specific binding, sections were blocked with 5% bovine serum albumin (BSA) for 1 h. Subsequently, primary antibodies were applied, including RMI2 rabbit antibody (ab122685, Abcam; 1:300 dilution), γH2AX rabbit antibody (9718, Cell Signaling; 1:200 dilution), CD8A antibody (ZM-0508, ZSGB-bio), IFN-γ antibody (AF5183, Affinity; 1:100 dilution), GZMB antibody (ab255598, Abcam; 1:300 dilution) and Ki-67 antibody (34330, Cell Signaling; 1:200 dilution). After an overnight incubation at 4°C, the secondary antibodies were applied for 1 hour at room temperature. Finally, the staining was visualized using 3,3'-diaminobenzidine (DAB) for 30 s to develop the signal. Multiplex immunofluorescence analysis was performed using 4% paraformaldehyde-fixed, paraffin-embedded tissue sections mounted on poly-L-lysine-coated slides. The protocol incorporated antigen retrieval via EDTA buffer (pH 8.0) microwave treatment (100°C, 2.5 min), followed by BSA blocking (5%, 1 h, 4°C). Primary antibodies against CD8A (ZM-0508, ZSGB-bio) and RMI2 were incubated overnight at 4°C in 1% BSA-PBS, detected with Alexa Fluor-conjugated secondary antibodies, and the signal was amplified through TSA amplification (PANOVUE, RM-2759). Imaging was performed using the TissueFAXS Cytometry System (TissueGnostics), generating quantitative metrics including cell density (cells/mm²), and nuclear morphology indices (area, circularity). RNA sequencing For RNA-seq analysis, we first performed read filtering and alignment using STAR (version 2.0.2), followed by transcript assembly with StringTie2 (version 1.3.5) to construct the transcriptome. For differential gene expression (DGE) analysis, we utilized the edgeR package to identify genes with a log2 fold change > 0.5 and a false discovery rate (FDR) 0.05 in this test were excluded from our final DEG list. To uncover broader biological implications, we performed functional annotation using Gene Set Enrichment Analysis (GSEA), focusing on pathways with an adjusted FDR indicating statistical significance (Table S5). Western Blotting After two washes with cold PBS, cells were lysed using cell lysis buffer (P0013, Beyotime) on ice for 20 min. Lysates were centrifuged at 14,000 ×g for 15 min at 4°C to clarify. The lysate was then heated in gel loading buffer for 10 min and resolved by 11% SDS-PAGE. Proteins were transferred onto a 0.45 µm Immobilon-P PVDF membrane (Millipore) using standard Western blotting techniques. After blocking with PBS containing 5% non-fat milk and 0.1% Tween-20, primary antibodies specific to the target proteins were applied and incubated overnight at 4°C. HRP-conjugated secondary antibodies (W4021 for mouse and W4011 for rabbit, Promega) were then used to detect the target proteins. A high-sensitivity ECL substrate was applied to visualize the bands, which were detected using a MiniChemi imaging system (SageCreation, Beijing). Flow Cytometry In our in vivo analysis of tumor-infiltrating lymphocytes (TILs), tumor tissues were harvested and mechanically dissociated. The tissue fragments were then enzymatically digested using a solution containing 0.5 mg/mL collagenase type IV and 200 U/mL DNase I, followed by incubation at 37°C for 1 hour. This enzymatic treatment effectively disrupted the extracellular matrix, enabling the release of individual cells. The resulting cell suspension was filtered through a 70-µm mesh to achieve a homogeneous single-cell suspension. Red blood cells were subsequently lysed using ACK lysis buffer (Solarbio) for 5 minutes at 4°C. Prior to analysis, the cells were incubated with a panel of surface-specific antibodies for 30 minutes on ice, shielded from light to minimize non-specific binding. To differentiate viable from non-viable cells, we employed the Live/Dead Fixable Aqua dye (Thermo Fisher Scientific). For intracellular staining, cells were first stimulated with the Leukocyte Activation Cocktail (BD Biosciences) for 4 to 6 hours. This was followed by a dual staining protocol targeting both surface and intracellular markers. The Intracellular Fixation & Permeabilization Buffer Set (88-8824-00, eBioscience) was used according to the manufacturer’s guidelines to ensure optimal fixation and permeabilization of the cells. Immunofluorescence Stable cell lines were plated into 15 mm glass-bottom cell culture dishes (801002, NEST). The following day, the cells were fixed with 4% paraformaldehyde for 15 minutes, permeabilized with 0.5% Triton X-100 for 15 minutes, and blocked with goat serum for 30 minutes at room temperature (RT). Between each step, the cells were washed twice with phosphate-buffered saline (PBS). Subsequently, the cells were incubated with primary antibodies overnight at 4°C. After washing, the cells were incubated with secondary antibodies for 1 hour at RT. Nuclei were counterstained with Hoechst 33342 for 2 minutes, followed by three washes with PBS. Finally, the cells were mounted using an antifade mounting medium (Invitrogen). Imaging was performed using a confocal laser scanning microscope (ZEISS, LSM880, ZEN2.6, 63× oil immersion lens). ELISA Human IFN-β (Thermo Fisher Scientific, 414101) and 2’3’-cGAMP (Cayman Chemical, 501700) ELISAs were performed according to the manufacturer’s instructions. Conditioned media collected from cells cultured for 72 hours after seeding (for IFN-β), and cell lysates prepared for cGAMP analysis were processed accordingly. The results represent the average of three replicates from at least two independent experiments, ensuring statistical robustness. Quantitative RT-PCR Total RNA was isolated from cell lysates using the RaPure Total RNA Kit (Magen) according to the manufacturer's instructions. Equal amounts of RNA were reverse-transcribed into cDNA using the HiScript II Q RT SuperMix for qPCR (Vazyme). The synthesized cDNA served as a template for subsequent quantitative PCR analysis, which was performed using the ChamQ Universal SYBR qPCR Master Mix (Vazyme) and gene-specific primers. Expression levels were normalized against GAPDH, a commonly used housekeeping gene for normalization. Details of the specific primers are provided in Table S3 . Proliferation assay Briefly, DLD1, HCT116 and SW480 were seeded at a density of 3,000 cells per well in a 96-well microplate. The culture plate was subsequently transferred to the Incucyte® Live-Cell Analysis System (Sartorius, Germany) for continuous monitoring. The system was configured to acquire phase-contrast images at 2-hour intervals under standard culture conditions (37°C, 5% CO2). Real-time cell proliferation kinetics were quantitatively assessed through automated image analysis, measuring temporal changes in cellular confluence (%) using integrated image processing algorithms. Post-experimental data analysis included generation of proliferation curves and calculation of growth parameters, specifically population doubling time (PDT), through proprietary Incucyte software analysis modules. Colony formation assay Briefly, DLD1, HCT116, and SW480 cells were seeded into 6-well plates at a density of 500 cells per well, in triplicate. The cells were cultured under standard conditions for no fewer than 12 days to allow sufficient time for proliferation and colony formation. After the culture period, cells were gently washed twice with PBS to remove excess medium and debris. To preserve cell morphology and their attachment to the plate, cells were fixed using ethanol for approximately 30 minutes. Following fixation, cells were stained with a 1% solution of methyl violet in PBS for 60 minutes to facilitate colony visualization. After staining, plates were washed with PBS to remove excess dye. Finally, colonies formed by each cell line were counted. Statistical analysis All statistical analyses were performed using GraphPad Prism 8 (La Jolla, CA). Data are presented as mean ± standard error of the mean (SEM). Inter-group correlations in tissue microarray gene expression profiles were quantified using Spearman's rank correlation coefficient. For longitudinal tumor growth dynamics, two-way ANOVA with Tukey-Kramer post hoc comparisons was applied to assess time-dependent differences across multiple treatment cohorts. Kaplan-Meier survival curves were compared using the log-rank test. Between-group comparisons for normally distributed data met parametric assumptions and were evaluated using unpaired Student's t-tests (two-tailed). Comparative analysis of three or more independent groups was conducted using one-way ANOVA for multiple comparisons. Statistical significance was defined as p < 0.05. Results High RMI2 expression is associated with poor clinical outcomes in CRC To investigate whether RMI2 is associated with the clinical and pathological features of CRC progression, we assessed RMI2 protein expression by immunohistochemistry (IHC) using a human CRC tissue microarray (TMA) (Table S1 ) . Consistent with public transcriptomic data from The Cancer Genome Atlas (TCGA), CRC tissues exhibited significantly higher RMI2 staining intensity compared to matched normal tissues ( Fig. 1 A, 1 B and S1A ) . Notably, elevated RMI2 expression was observed in tumors with deeper invasion, advanced tumor stage, and lymph node metastasis ( Fig. 1 C-E ) . Furthermore, high RMI2 expression was correlated with poor prognosis in the overall cohort and patient subgroups, as well as across other cancer types, based on TCGA database analysis ( Fig. 1 F-H, and S1B ) . RMI2 expression was also significantly higher in metastatic tissues than in matched primary CRC tissues (Fig. 1 I and 1 J). These findings suggest that RMI2 may serve as a potential biomarker of CRC progression. Next, we investigated whether RMI2 expression is associated with the tumor immune microenvironment. We focused on a cohort of patients with refractory MSS CRC who received anti-PD-1 therapy or combination immunotherapy (Fig. S1 C and table S2 ) . Computed tomography (CT) imaging revealed a higher response rate to anti-PD-1 therapy in liver metastases of patients with low RMI2 expression than in those with high RMI2 expression. Notably, the objective response rate (ORR) was significantly higher in the RMI2-low group than in the RMI2-high group (28.56% vs. 11.54%, P = 0.0039) ( Fig. 1 K and 1 L ) . The RMI2-low group also exhibited a significantly longer median progression-free survival (PFS) compared with the RMI2-high group (10 months versus 2.7 months, P = 0.0012) ( Fig. 1 M ) . Given that tumor-infiltrating T lymphocytes (T-TILs) are the primary immune cell population responsible for tumor cell recognition and elimination, we further analyzed the correlation between intratumoral T-TIL infiltration and RMI2 expression using data from TCGA database. We found that RMI2 expression was negatively correlated with T-TIL abundance across 32 cancer types, including colon adenocarcinoma (COAD) ( Fig. 1 N and 1 O ) . Additionally, we evaluated the prognostic impact of RMI2 expression in a clinical immunotherapy cohort using Kaplan-Meier Plotter analysis. Among pan-cancer patients treated with anti-PD-1 therapy, those in the RMI2-high group showed significantly shorter overall survival (OS) and PFS than those in the RMI2-low group ( Fig. 1 P ) . Collectively, these findings indicate that high RMI2 expression is associated with poor clinical outcomes in CRC and may serve as a predictive biomarker for the efficacy of ICIs. RMI2 loss suppresses tumor growth and modulates the immune tumor microenvironment Next, we investigated the functional role of RMI2 in CRC. RMI2 expression was generally higher in CRC cells than in the human intestinal epithelial cell line CCD18-Co (Fig. S2 A) . To assess the biological effect of RMI2 on CRC cells, we knocked down RMI2 in HCT116, DLD1 and SW480 cells (Fig. S2 B) . RMI2 knockdown significantly inhibited the proliferation and migration of these cell lines ( Fig. S2 B-D ). Conversely, lentivirus-mediated RMI2 overexpression (OE) promoted cell proliferation ( Fig. S2 E and 2F ), indicating that RMI2 contributes to CRC progression. In vivo, control (NC) or RMI2 knockout (RMI2 sg1/2) MC38 cells were subcutaneously implanted into immunocompetent C57BL/6J and immunodeficient BALB/c nude mice. RMI2 knockout markedly suppressed tumor growth in both immunocompetent and immunodeficient mice (Fig. S3 A-D) . Notably, RMI2 knockout resulted in an approximately 78% reduction in tumor volume when MC38 cells were implanted in C57BL/6J mice, whereas the inhibition rate was approximately 60% in immunodeficient BALB/c nude mice (Fig. S3 E) . Furthermore, overexpression of RMI2 in C57BL/6J mice promoted tumor growth by approximately 60%, whereas only about 40% tumor growth promotion was observed in immunodeficient mice (Fig. S3 F-J) . To investigate the impact of RMI2 depletion on the immune microenvironment of MSS CRC, we performed single-cell RNA sequencing (scRNA-seq) to profile the transcriptomic changes in sgRMI2 versus NC CT26 syngeneic tumors ( Fig. 2 A and 2 B ) . After rigorous quality control, 42,999 high-quality single-cell transcriptomes were obtained from both tumor groups. Notably, inferCNV analysis confirmed that cancer cells exhibited copy number instability, whereas immune populations displayed genomic stability (Fig. S4A) . Unsupervised clustering combined with uniform manifold approximation and projection (UMAP) visualization revealed distinct cellular populations ( Fig. 2 C and S4B ) . Gene Ontology (GO) enrichment analysis using the Kyoto Encyclopedia of Genes and Genomes (KEGG) database revealed significant upregulation of type I/II interferon response pathways and inflammatory response-related signaling pathways in sgRMI2-derived cancer cells ( Fig. 2 D and 2 E ) . Furthermore, marked expansion of CD8⁺ T cell and natural killer (NK) cell clusters was observed in sgRMI2 tumors ( Fig. 2 C and S4C ) . To further dissect the functional differences between the two groups, we performed separate GO enrichment analyses of CD8⁺ T cells and NK cells. In sgRMI2-derived CD8⁺ T cells, multiple pathways associated with T cell activation, proliferation, cytotoxicity, antigen receptor-mediated signaling, and cytokine production were significantly enriched ( Fig. 2 F ) . Similarly, NK cells from sgRMI2 tumors showed prominent enrichment of pathways related to the inflammatory response, cytotoxic activity, and bacterial response ( Fig. 2 G ) . These data demonstrate that high RMI2 expression promotes tumor cell growth and contributes to an immunosuppressive microenvironment. Loss of RMI2 enhances infiltration and activation of cytotoxic CD8⁺ T cells Consistent with scRNA-seq, RMI2 knockout in MC38 cells significantly increased the infiltration of cytotoxic CD8⁺ T cells and natural killer (NK) cells into the tumor microenvironment, while reducing the proportion of immunosuppressive Ly6G⁺ myeloid-derived suppressor cells (MDSCs). However, the overall abundance of CD45⁺ leukocytes, dendritic cells, and CD4⁺ T cells within the tumors remained unchanged upon RMI2 deletion ( Fig. 3 A, 3 B and S5A ) . To identify the key immune effector cells mediating tumor growth inhibition under RMI2 knockout models, we selectively depleted CD8⁺ T cells or NK cells in vivo using anti-CD8 or anti-NK1.1 neutralizing antibodies, respectively. Tumor growth was significantly restored upon CD8⁺ T cell depletion, whereas NK cell depletion had no significant effect on tumor progression ( Fig. 3 C and 3 D ) . Moreover, RMI2 knockout led to a significant increase in the proportion of effector CD8⁺ T cells, including those expressing granzyme B (GZMB), interferon-gamma (IFN-γ), and proliferation marker Ki67 ( Fig. 3 E and S5B ) . IHC staining confirmed that RMI2 knockout tumors exhibited increased infiltration of CD8⁺ T cells, along with elevated expression of GZMB and IFN-γ ( Fig. 3 F and 3 G ) . Analysis of refractory MSS CRC patient tissues (cohort 3) and MSS CRC patient (cohort 1) further revealed a significant negative correlation between RMI2 expression and CD8⁺ T cell abundance ( Fig. 3 H and S5C ) . Similarly, RMI2 mRNA expression was negatively correlated with the levels of CD3, and CD8A T cell markers in CRC samples from TCGA database (Fig. S5D) . In addition, in vitro co-culture experiments demonstrated that CD8⁺ T cells exhibited enhanced cytotoxic activity and functional activation when co-cultured with RMI2 knockdown CT26 cells ( Fig. 3 I-K ) . Collectively, these findings indicate that RMI2 suppresses the recruitment and activation of cytotoxic CD8⁺ T cells. Loss of RMI2 leads to elevate DNA damage and impair homologous recombination As previously reported, RMI2 deficiency results in increased DNA damage, leading to genomic instability 12 . A higher tumor mutational burden, driven by an increased frequency of somatic nonsynonymous coding mutations, generates a greater abundance of tumor-specific mutant peptides, which may enhance tumor immunogenicity and render these cancers more responsive to immune checkpoint therapies 34 , 35 . Therefore, we sought to determine whether RMI2 plays a role in regulating DNA damage in CRC cells. Western blot and immunofluorescence staining showed increased expression and formation of γH2AX foci, a marker of unrepaired DNA lesions, in RMI2 knockdown cells ( Fig. 4 A-C ) . Consistently, the comet assay demonstrated significantly elevated levels of DNA damage following RMI2 knockdown ( Fig. 4 D and 4 E ) . In addition, RMI2 knockdown increased apoptosis ( Fig. 4 F and S6A ) . IHC staining of CRC tissues further revealed a significant negative correlation between RMI2 expression and the abundance of γH2AX-positive cells ( Fig. 4 G ) . Given that most CRCs harbor loss-of-function mutations in TP53, a key regulator of genomic stability 36 , 37 , RMI2 has emerged as a promising therapeutic target that selectively exacerbate DNA damage in these tumors. Next, we assessed the impact of RMI2 on the DNA damage repair pathways. RMI2 knockdown significantly impaired homologous recombination (HR) repair efficiency, but it had no significant effect on non-homologous end joining (NHEJ) ( Fig. 4 H and 4 I ) . Furthermore, the formation of BRCA1 and RAD51 foci, two critical mediators of the HR repair process 38 , 39 , was markedly reduced in RMI2-deficient cells compared to controls ( Fig. 4 J, 4 K, S6C and S6D ) . Co-immunoprecipitation assays further demonstrated that RMI2 knockdown significantly decreased the endogenous interaction between BRCA1 and RAD51 ( Fig. 4 L and S6E ) . Collectively, these findings indicate that RMI2 is essential for maintaining genomic stability by supporting homologous recombination-mediated DNA repair, and that its loss leads to the accumulation of unrepaired DNA damage and impaired HR efficiency. RMI2 deficiency increases cytosolic DNA and activates the cGAS-STING pathway Next, we stained cytosolic double-stranded DNA (dsDNA) using PicoGreen and observed a significantly higher proportion of cytosolic DNA-positive cells in RMI2 knockdown cells ( Fig. 5 A and 5 B ) . The percentage of micronuclei also increased in the RMI2 knockdown cells (Fig. S7A and S7B) . These findings indicated that RMI2 deficiency promotes the release of nuclear DNA into the cytoplasm. To further explore the underlying mechanisms, we performed RNA sequencing (RNA-seq) of CRC samples from TCGA. Consistent with our scRNA-seq data, gene set enrichment analysis (GSEA) revealed significant upregulation of immune-associated gene sets such as IFN-α response, IFN-γ response, and inflammatory response in RMI2-deficient cells ( Fig. 5 C and 5 D ) . It is well established that cytoplasmic DNA resulting from DNA damage can activate innate immune responses, including antimicrobial and antitumor immunity 40 . Next, we investigated whether RMI2 modulates the activity of the cGAS-STING signaling pathway. The baseline phosphorylation levels of endogenous STING, TBK1 and interferon regulatory factor 3 (IRF3) were higher in the RMI2 knockdown cells ( Fig. 5 E ) . Moreover, the proportion of cells with detectable cytosolic DNA accumulation was significantly higher in RMI2 knockdown cells compared to the control and RMI2-rescued cells (Fig. S7C) . Since some CRC cell lines such as SW480 and DLD1 express STING, whereas RKO robustly silence STING, we first determined the consequences of RMI2 knockdown across these different cell lines ( Fig. 5 F ) . RMI2 knockdown increased IFN-β secretion in both SW480 and DLD1 cells, but not in RKO cells, which lack functional STING ( Fig. 5 G ) . Consistently, the expression of canonical STING target genes, including IFN-β, IFN-γ, CXCL10, ISG15, and IRF7, were increased in RMI2 knockdown cells, whereas their expression in RMI2-rescued cells remained comparable to that in control cells ( Fig. 5 H, 5 I and S7D ) . To directly assess whether the STING pathway mediates enhanced IFN signaling induced by RMI2 depletion, we generated double-knockout cells lacking both RMI2 and STING in SW480 and CT26 models. As expected, STING knockout completely abolished RMI2 knockdown-induced IFN-β secretion, as well as the upregulation of STING target genes, including IFN-β, IFN-γ, CXCL10, ISG15, and IRF7, in both cell lines ( Fig. 5 J, 5 K and S7E ) . Furthermore, STING deficiency abrogated the inhibitory effect of RMI2 knockout on tumor growth in CT26 cells ( Fig. 5 L-N ) . The proportions of effector T cells, including GZMB⁺ CD8⁺ T cells and IFN-γ⁺ CD8⁺ T cells, differed significantly between the sgRMI2 and double-knockout groups, suggesting that the tumor microenvironment in the double-knockout group exhibited a more immunosuppressive phenotype (Fig. S7F and S7G) . In addition, overexpression of RAD51 attenuated RMI2 knockdown-induced γH2AX foci formation and reduced the expression of STING target genes, including IFN-β, IFN-γ, CXCL10, ISG15, and IRF7 (Fig. S8A-C) . Collectively, these findings demonstrate that RMI2 suppresses cytosolic DNA sensing and IFN signaling in cancer cells by inhibiting the cGAS-STING pathway, and that STING is required for the full immunostimulatory effects of RMI2 loss. RMI2 deficiency sensitizes the tumor response to anti-PD-1 therapy We further investigated whether RMI2 deficiency enhanced the therapeutic efficacy of anti-PD-1 blockade. In the CT26 orthotopic tumor model, RMI2-deficient tumors exhibited significantly reduced tumor volumes and prolonged survival following treatment with anti-PD-1 antibodies ( Fig. 6 A and 6 B ) . Similarly, in subcutaneous tumor models established with either MC38 or CT26 cells, RMI2-deficient tumors treated with anti-PD-1 antibodies showed markedly suppressed tumor growth and reduced tumor weight ( Fig. 6 D, 6 E, S9A and S9B ) . Flow cytometry and histological analysis revealed that the proportions of key immune cell populations, including NK cells, CD4⁺ T cells, and CD8⁺ T cells, were significantly increased in RMI2-deficient tumors upon anti-PD-1 treatment ( Fig. 6 F and S9C-E ) . Furthermore, the frequency of effector CD8⁺ T cells expressing GZMB, IFN-γ, and Ki67 was significantly increased elevated in RMI2-deficient tumors following anti-PD-1 therapy ( Fig. 6 G and S9F-H ) . IHC staining confirmed that combined RMI2 deficiency and anti-PD-1 treatment led to increased expression of GZMB, and IFN-γ in tumor tissues ( Fig. 6 H ) . Consistent with these findings, quantitative PCR analysis revealed significant upregulation of TNF-α, GZMB, and IFN-γ mRNA levels in RMI2-deficient tumors treated with anti-PD-1 antibodies (Fig. S9I) . Together, these results demonstrate that RMI2 deficiency enhances the responsiveness of colorectal cancer to anti-PD-1 immunotherapy, suggesting that targeting RMI2 may represent a promising strategy to improve the efficacy of checkpoint blockade. RMI2 deficiency sensitizes the tumor response to chemotherapy Given that both 5-fluorouracil (5-FU) and ionizing radiation (IR) induce DNA damage 41 , 42 , we investigated whether RMI2 deficiency could enhance the efficacy of these treatments. Notably, western blot and immunofluorescence analyses revealed that RMI2-deficient cells exhibited increased expression and formation of γH2AX foci, and the cGAS-STING signaling pathway upon treatment with either IR or 5-FU ( Fig. 7 A, S10A and S10B ). Consistent with this, the formation of γH2AX foci and the proportion of cytosolic DNA positivity were higher in RMI2 deficiency cells treated with 5-FU ( Fig. 7 B, 7 C and S10C ) . In addition, 5-FU treatment induced a marked increase in IFN-β secretion in RMI2-deficient cells ( Fig. 7 D and S10D ) . The mRNA levels of key interferon-stimulated genes (ISGs), including IFN-β, IFN-γ, CXCL10, ISG15, and IRF7, were also significantly upregulated following 5-FU treatment, with effects further potentiated by RMI2 loss ( Fig. 7 E and S10E ) . To determine whether RMI2 deficiency enhances the sensitivity of CRC tumors to 5-FU chemotherapy, we evaluated tumor growth in subcutaneous models using CT26 cells. RMI2-deficient tumors exhibited significantly reduced growth and tumor weight following 5-FU treatment ( Fig. 7 F and 7 G ) . Importantly, the abundance of infiltrating immune cells, including NK cells, CD4⁺ T cells, and CD8⁺ T cells, was significantly higher in RMI2-deficient tumors treated with 5-FU ( Fig. 7 H and 7 I ) . Furthermore, the proportions of effector CD8⁺ T cells expressing GZMB, IFN-γ, and Ki67 were also significantly increased ( Fig. 7 J ) . IHC validated that RMI2-deficient tumors treated with 5-FU displayed elevated expression of GZMB, and IFN-γ ( Fig. 7 K and 7 L ) . Consistent with these findings, the mRNA levels of TNF-α, GZMB, and IFN-γ were also increased in these tumors (Fig. S10F) . In a clinical cohort receiving neoadjuvant 5-FU-based chemotherapy, low RMI2 expression correlated with higher treatment response rates, an inverse association between RMI2 negativity and CD8 + T-cell infiltration, and significantly prolonged overall survival ( Fig. 7 M-O, and S10G ) . Collectively, these findings demonstrate that RMI2 deficiency sensitizes CRC tumors to 5-FU chemotherapy and enhances the associated antitumor immune response, suggesting its potential as a therapeutic target to improve chemotherapeutic efficacy and immune activation. Discussion Tumor cells employ multiple strategies to evade antitumor immunity, including the suppression of cytosolic nucleic acid sensing pathways 43 , 44 . In this study, we demonstrate that RMI2 deficiency disrupts homologous recombination-mediated DNA repair, leads to the accumulation of cytosolic double-stranded DNA (dsDNA), and activates intrinsic antitumor and immune signaling pathways (Fig. S11) . Collectively, these alterations compromise the genomic stability and enhance the immunogenicity of tumor cells. Notably, these tumor-intrinsic changes significantly improve the efficacy of ICB therapy, particularly in immunologically cold tumors such as MSS CRC. Through comprehensive analysis of clinical data, we identified RMI2 expression as an independent prognostic biomarker in patients with MSS CRC patients receiving immune checkpoint inhibitors. High RMI2 expression was correlated with poorer therapeutic responses and significantly shorter progression-free survival (PFS; HR = 3.045, 95% CI: 1.569–5.909; P = 0.0012). Specifically, patients with low RMI2 expression exhibited a median PFS nearly threefold longer than that of patients with high RMI2 expression (10 months vs. 2.7 months). These findings highlight RMI2 as a potential predictive biomarker for ICI response and underscore its role in shaping the tumor immune microenvironment. A key factor contributing to the limited efficacy of immunotherapy in cold tumors is the lack of an inflamed, immunogenic tumor microenvironment, which is often characterized by reduced infiltration of cytotoxic CD8⁺ T cells and defective type I interferon signaling 43 , 45 . Our data from immunocompetent mouse models reveal that RMI2 deficiency not only inhibits tumor growth but also enhances the recruitment and activation of multiple immune cell populations, most notably CD8⁺ T cells. This shift significantly sensitizes CRC tumors to anti-PD-1 therapy. Mechanistically, RMI2 loss activates the cGAS-STING pathway, leading to increased phosphorylation of STING, TBK1, and IRF3; elevated secretion of IFN-β, and upregulation of downstream interferon-stimulated genes (ISGs), including ISG15, CXCL10, and IRF7. Furthermore, RMI2 depletion selectively enhanced the infiltration of CD8⁺ T cells into the tumor microenvironment, although no significant increase was observed in macrophages. These findings indicate that RMI2 deficiency transforms the tumor immune landscape from immunosuppressive to immunostimulatory. In addition to its effects in immunotherapy, we found that RMI2 deficiency potentiates the therapeutic efficacy of conventional DNA-damaging agents. In models treated with IR or 5-FU, RMI2-deficient cells exhibited elevated γH2AX foci formation, which is indicative of increased DNA damage and impaired repair. Consistent with these observations, RMI2 loss enhanced activation of the cGAS-STING pathway following IR or 5-FU treatment, as evidenced by increased STING pathway phosphorylation and IFN-β production. In the context of 5-FU-based chemotherapy, RMI2-deficient tumors demonstrated significantly reduced growth and tumor weight, accompanied by increased infiltration of cytotoxic immune cells, including NK cells, CD4⁺ T cells, and CD8⁺ T cells. Among CD8⁺ T cells, the proportion of effector cells expressing GZMB, and IFN-γ was also significantly elevated. Immunohistochemical analysis further confirmed the increased expression of GZMB and IFN-γ in RMI2-deficient, 5-FU-treated tumors, along with upregulated tumor mRNA levels of TNF-α, GZMB, and IFN-γ. Clinically, in a cohort of patients receiving neoadjuvant 5-FU-based chemotherapy, low RMI2 expression is associated with improved response rates and significantly prolonged overall survival. These data suggest that RMI2 deficiency not only enhances tumor cell intrinsic susceptibility to DNA-damaging agents but also augments antitumor immunity, thereby synergizing with conventional chemotherapy. Given the essential role of RMI2 in maintaining genome stability across various cellular compartments 46 , 47 , and its overexpression in multiple cancer types, including CRC, where high RMI2 expression correlates with advanced tumor stage and poor prognosis, we emphasize the need to carefully evaluate the systemic effects and safety profile of RMI2-targeted strategies 8 , 9 , 10 , 11 , 12 . Notably, RMI2 exhibits relatively low expression in normal tissues, suggesting therapeutically exploitable vulnerability. This differential expression pattern provides a tumor-selective therapeutic window in which RMI2 inhibition preferentially targets rapidly proliferating cancer cells while sparing normal tissues. Mechanistically, RMI2 depletion exerts a dual therapeutic effect: (1) it induces DNA damage and impairs homologous recombination, leading to oncogene addiction-dependent cytotoxicity; and (2) it activates the cGAS-STING pathway, enhances type I interferon signaling and promotes the recruitment of cytotoxic immune cells such as NK cells and CD8⁺ T cells. This dual mechanism not only drives direct tumor cell killing but also synergizes with immune checkpoint blockade to enhance antitumor immunity. Recent research has demonstrated that genomic instability can effectively enhance the efficacy of immunotherapy in MSS colon cancer 48 . This dual-action paradigm represents a promising strategy for improving response rates in immunologically cold tumors and warrants further exploration in combinatorial therapeutic regimens. However, several important questions remain. The precise dose-dependent effects of RMI2 modulation on immune activation, identification of predictive biomarkers for patient stratification, and assessment of potential combinatorial toxicities in broader preclinical models require further systematic investigation. Future studies should also explore the therapeutic potential of combining RMI2-targeted approaches with additional immunotherapies, cytotoxic agents, or STING agonists to maximize the clinical benefit. In summary, our study reveals that RMI2 acts as a critical negative regulator of cytosolic DNA sensing and repair. Genetic perturbation of RMI2 disrupts genomic integrity while simultaneously activating innate immune responses, thereby converting immunologically cold tumors into immunologically active (“hot”) tumors. These tumor-intrinsic changes significantly enhance the responsiveness to immune checkpoint blockade and chemotherapy. Collectively, our findings suggest that RMI2 is a novel and promising therapeutic target for improving the efficacy of immunotherapy in MSS CRC and other immunologically cold malignancies. Declarations Fundings This work was supported by the National Science Fund for Distinguished Young Scholars (82425045), National Nature Science Foundation in China (NSFC) (82272800, 82102952, 82403266 and 82103080), Noncommunicable Chronic Diseases of National Science and Technology Major Project (2024ZD0520304), and the CSCO-Tongshu Gene Tumor Research Foundation Youth Project (No. Y-tongshu2021/qn-0303). Conflict of interest The authors declare no competing financial interests. Ethics approval All human tissue studies were approved by the Institutional Review Board of The Sixth Affiliated Hospital, Sun Yat-sen University (2022ZSLYEC-002). All animal experiments were performed in accordance with the Guidelines for the Care and Use of Laboratory Animals and approved by the Animal Research Committee of The Sixth Hospital of Sun Yat-sen University (SYSU-IACUC-2020112701). Data availability The RMI2 gene expression dataset and corresponding clinical survival information were retrieved from the Kaplan-Meier Plotter database (https://kmplot.com). Patients were stratified into high- and low-expression cohorts based on the median RMI2 expression levels, and overall survival differences were analyzed using Kaplan-Meier survival curves with log-rank testing. Raw and processed sc-RNA data have been deposited with the Gene Expression Omnibus (GEO accession GSE 308382). References Wang S, Zheng R, Li J, Zeng H, Li L, Chen R, et al. Global, regional, and national lifetime risks of developing and dying from gastrointestinal cancers in 185 countries: a population-based systematic analysis of GLOBOCAN. The lancet Gastroenterology & hepatology 2024, 9(3): 229–237. André T, Shiu KK, Kim TW, Jensen BV, Jensen LH, Punt C, et al. Pembrolizumab in Microsatellite-Instability-High Advanced Colorectal Cancer. The New England journal of medicine 2020, 383(23): 2207–2218. Hu H, Kang L, Zhang J, Wu Z, Wang H, Huang M, et al. Neoadjuvant PD-1 blockade with toripalimab, with or without celecoxib, in mismatch repair-deficient or microsatellite instability-high, locally advanced, colorectal cancer (PICC): a single-centre, parallel-group, non-comparative, randomised, phase 2 trial. The lancet Gastroenterology & hepatology 2022, 7(1): 38–48. Pelka K, Hofree M, Chen JH, Sarkizova S, Pirl JD, Jorgji V, et al. Spatially organized multicellular immune hubs in human colorectal cancer. Cell 2021, 184(18): 4734–4752.e4720. Roerden M, Spranger S. Cancer immune evasion, immunoediting and intratumour heterogeneity. Nature reviews Immunology 2025. Westcott PMK, Sacks NJ, Schenkel JM, Ely ZA, Smith O, Hauck H, et al. Low neoantigen expression and poor T-cell priming underlie early immune escape in colorectal cancer. Nature cancer 2021, 2(10): 1071–1085. Ochoa de Olza M, Navarro Rodrigo B, Zimmermann S, Coukos G. Turning up the heat on non-immunoreactive tumours: opportunities for clinical development. The Lancet Oncology 2020, 21(9): e419-e430. Xiao D, Dong S, Yang S, Liu Z. CKS2 and RMI2 are two prognostic biomarkers of lung adenocarcinoma. PeerJ 2020, 8: e10126. Zheng Ms B, Wang Ms H, Wang Ms JX, Liu Ms ZH, Zhang Md P, Zhang Md D. The Clinical Significance of RMI2 in Hepatocellular Carcinoma. Technology in cancer research & treatment 2021, 20: 15330338211045496. Wei W, Ying X, Chen L, Sun Q, Lu X, Xia Y, et al. RecQ mediated genome instability 2 (RMI2): a potential prognostic and immunological biomarker for pan-cancers. Aging 2022, 14(9): 4107–4136. Zhang L, Hao C, Han B, Zeng G, Han L, Cao C, et al. RMI2 is a novel prognostic and predictive biomarker for breast cancer. Cancer medicine 2023, 12(7): 8331–8350. Zhan W, Liu Y, Gao Y, Gong R, Wang W, Zhang R, et al. RMI2 plays crucial roles in growth and metastasis of lung cancer. Signal transduction and targeted therapy 2020, 5(1): 188. Concannon K, Morris BB, Gay CM, Byers LA. Combining targeted DNA repair inhibition and immune-oncology approaches for enhanced tumor control. Molecular cell 2023, 83(5): 660–680. Wang D, Elenbaas B, Murugesan K, Shah K, Montesion M, Gounaris I, et al. Relationship among DDR gene mutations, TMB and PD-L1 in solid tumour genomes identified using clinically actionable biomarker assays. NPJ precision oncology 2023, 7(1): 103. Germano G, Lamba S, Rospo G, Barault L, Magrì A, Maione F, et al. Inactivation of DNA repair triggers neoantigen generation and impairs tumour growth. Nature 2017, 552(7683): 116–120. Ben-David U, Amon A. Context is everything: aneuploidy in cancer. Nature reviews Genetics 2020, 21(1): 44–62. Kuang X, Li J. Chromosome instability and aneuploidy as context-dependent activators or inhibitors of antitumor immunity. Frontiers in immunology 2022, 13: 895961. Tubbs A, Nussenzweig A. Endogenous DNA Damage as a Source of Genomic Instability in Cancer. Cell 2017, 168(4): 644–656. Wang Z, Zhao J, Wang G, Zhang F, Zhang Z, Zhang F, et al. Comutations in DNA Damage Response Pathways Serve as Potential Biomarkers for Immune Checkpoint Blockade. Cancer research 2018, 78(22): 6486–6496. Keenan TE, Burke KP, Van Allen EM. Genomic correlates of response to immune checkpoint blockade. Nature medicine 2019, 25(3): 389–402. Zhang B, Xu P, Ablasser A. Regulation of the cGAS-STING Pathway. Annual review of immunology 2025. Chen Q, Sun L, Chen ZJ. Regulation and function of the cGAS-STING pathway of cytosolic DNA sensing. Nature immunology 2016, 17(10): 1142–1149. Falahat R, Berglund A, Perez-Villarroel P, Putney RM, Hamaidi I, Kim S, et al. Epigenetic state determines the in vivo efficacy of STING agonist therapy. Nature communications 2023, 14(1): 1573. Konno H, Yamauchi S, Berglund A, Putney RM, Mulé JJ, Barber GN. Suppression of STING signaling through epigenetic silencing and missense mutation impedes DNA damage mediated cytokine production. Oncogene 2018, 37(15): 2037–2051. Shang G, Zhang C, Chen ZJ, Bai XC, Zhang X. Cryo-EM structures of STING reveal its mechanism of activation by cyclic GMP-AMP. Nature 2019, 567(7748): 389–393. Ergun SL, Fernandez D, Weiss TM, Li L. STING Polymer Structure Reveals Mechanisms for Activation, Hyperactivation, and Inhibition. Cell 2019, 178(2): 290–301.e210. Zhang C, Shang G, Gui X, Zhang X, Bai XC, Chen ZJ. Structural basis of STING binding with and phosphorylation by TBK1. Nature 2019, 567(7748): 394–398. Lu C, Guan J, Lu S, Jin Q, Rousseau B, Lu T, et al. DNA Sensing in Mismatch Repair-Deficient Tumor Cells Is Essential for Anti-tumor Immunity. Cancer cell 2021, 39(1): 96–108.e106. Guan J, Lu C, Jin Q, Lu H, Chen X, Tian L, et al. MLH1 Deficiency-Triggered DNA Hyperexcision by Exonuclease 1 Activates the cGAS-STING Pathway. Cancer cell 2021, 39(1): 109–121.e105. Kaneta A, Nakajima S, Okayama H, Matsumoto T, Saito K, Kikuchi T, et al. Role of the cGAS-STING pathway in regulating the tumor-immune microenvironment in dMMR/MSI colorectal cancer. Cancer immunology, immunotherapy: CII 2022, 71(11): 2765–2776. Wei B, Xu L, Guo W, Wang Y, Wu J, Li X, et al. SHP2-Mediated Inhibition of DNA Repair Contributes to cGAS-STING Activation and Chemotherapeutic Sensitivity in Colon Cancer. Cancer research 2021, 81(12): 3215–3228. Duan J, Chen T, Li Q, Zhang Y, Lu T, Xue J, et al. Protein arginine methyltransferase 6 enhances immune checkpoint blockade efficacy via the STING pathway in MMR-proficient colorectal cancer. Journal for immunotherapy of cancer 2025, 13(3). Saha S, Ghosh M, Li J, Wen A, Galluzzi L, Martinez LA, et al. Serine Depletion Promotes Antitumor Immunity by Activating Mitochondrial DNA-Mediated cGAS-STING Signaling. Cancer research 2024, 84(16): 2645–2659. Rizvi NA, Hellmann MD, Snyder A, Kvistborg P, Makarov V, Havel JJ, et al. Cancer immunology. Mutational landscape determines sensitivity to PD-1 blockade in non-small cell lung cancer. Science (New York, NY) 2015, 348(6230): 124–128. Schumacher TN, Schreiber RD. Neoantigens in cancer immunotherapy. Science (New York, NY) 2015, 348(6230): 69–74. Chin KV, Ueda K, Pastan I, Gottesman MM. Modulation of activity of the promoter of the human MDR1 gene by Ras and p53. Science (New York, NY) 1992, 255(5043): 459–462. Brannon AR, Vakiani E, Sylvester BE, Scott SN, McDermott G, Shah RH, et al. Comparative sequencing analysis reveals high genomic concordance between matched primary and metastatic colorectal cancer lesions. Genome biology 2014, 15(8): 454. Daboussi F, Dumay A, Delacôte F, Lopez BS. DNA double-strand break repair signalling: the case of RAD51 post-translational regulation. Cellular signalling 2002, 14(12): 969–975. Lord CJ, Ashworth A. The DNA damage response and cancer therapy. Nature 2012, 481(7381): 287–294. Härtlova A, Erttmann SF, Raffi FA, Schmalz AM, Resch U, Anugula S, et al. DNA damage primes the type I interferon system via the cytosolic DNA sensor STING to promote anti-microbial innate immunity. Immunity 2015, 42(2): 332–343. Shah MA, Schwartz GK. Cell cycle-mediated drug resistance: an emerging concept in cancer therapy. Clinical cancer research: an official journal of the American Association for Cancer Research 2001, 7(8): 2168–2181. Sharma RA, Plummer R, Stock JK, Greenhalgh TA, Ataman O, Kelly S, et al. Clinical development of new drug-radiotherapy combinations. Nature reviews Clinical oncology 2016, 13(10): 627–642. Zou SS, Qiao Y, Zhu S, Gao B, Yang N, Liu YJ, et al. Intrinsic strategies for the evasion of cGAS-STING signaling-mediated immune surveillance in human cancer: How therapy can overcome them. Pharmacological research 2021, 166: 105514. Roulois D, Loo Yau H, Singhania R, Wang Y, Danesh A, Shen SY, et al. DNA-Demethylating Agents Target Colorectal Cancer Cells by Inducing Viral Mimicry by Endogenous Transcripts. Cell 2015, 162(5): 961–973. Jhunjhunwala S, Hammer C, Delamarre L. Antigen presentation in cancer: insights into tumour immunogenicity and immune evasion. Nature reviews Cancer 2021, 21(5): 298–312. Singh TR, Ali AM, Busygina V, Raynard S, Fan Q, Du CH, et al. BLAP18/RMI2, a novel OB-fold-containing protein, is an essential component of the Bloom helicase-double Holliday junction dissolvasome. Genes & development 2008, 22(20): 2856–2868. Xu D, Guo R, Sobeck A, Bachrati CZ, Yang J, Enomoto T, et al. RMI, a new OB-fold complex essential for Bloom syndrome protein to maintain genome stability. Genes & development 2008, 22(20): 2843–2855. Tan PB, Verschoor YL, van den Berg JG, Balduzzi S, Kok NFM, Ijsselsteijn ME, et al. Neoadjuvant immunotherapy in mismatch-repair-proficient colon cancers. Nature 2025. Additional Declarations (Not answered) Supplementary Files Uncroppedoriginalwesternblots.docx Uncropped original western blots SupplementarymaterialTableS5.xlsx RMI2 Depletion Recovers cGAS-STING Signaling to Enhance Immunotherapy in Colorectal Cancer-SM Supplementarymaterial.docx RMI2 Depletion Recovers cGAS-STING Signaling to Enhance Immunotherapy in Colorectal Cancer-SM Cite Share Download PDF Status: Under Revision Version 1 posted Editorial decision: revise 10 Mar, 2026 Review # 2 received at journal 08 Mar, 2026 Review # 1 received at journal 01 Mar, 2026 Reviewer # 2 agreed at journal 16 Feb, 2026 Reviewer # 1 agreed at journal 16 Feb, 2026 Reviewers invited by journal 13 Feb, 2026 Submission checks completed at journal 28 Jan, 2026 Editor assigned by journal 27 Jan, 2026 First submitted to journal 27 Jan, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8706982","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":581780128,"identity":"f0c0aae6-489a-402c-bb1f-33ff70fd5221","order_by":0,"name":"Weixiang Zhan","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABE0lEQVRIiWNgGAWjYDACCTBpI8fAwNh48ANMlIewljRjoJaGwxIkaDmc2AAkD8BV4tMiP7v52WPeNub0te2HGw5IMNQlzp+RwPjgbRuDvDkOLYxzjpkb85xhy912JrHhQAHQug03EpgN57YxGO5swK6FWSLBTDqngid324FEkC0HEjdIJLBJ87YxJBgcwK6FTSL9m3SOgUS62fmHDUC/gB3G/hufFh6JHJAtBglmNxJBWpgTG24ksDHj0yIhkVMm/edMguG2Gw+BgWxw2HjDmYfNknPOSRhuwKFFfkb6NsmZbf/lzc6nP3z4oaJOdn578sEPb8ps5HHZggYMGBwbgHHKAIsvooA98UpHwSgYBaNgpAAAirNd4W5rZNIAAAAASUVORK5CYII=","orcid":"","institution":"the Sixth Affiliated Hospital, Sun Yat-sen University","correspondingAuthor":true,"prefix":"","firstName":"Weixiang","middleName":"","lastName":"Zhan","suffix":""},{"id":581780129,"identity":"adce0bb8-eeac-4b33-84fd-96fb60cd0d2e","order_by":1,"name":"Runkai Cai","email":"","orcid":"","institution":"the Sixth Affiliated Hospital, Sun Yat-sen University","correspondingAuthor":false,"prefix":"","firstName":"Runkai","middleName":"","lastName":"Cai","suffix":""},{"id":581780130,"identity":"bbbab7f2-37e9-4e19-8b77-1c393d3d61ad","order_by":2,"name":"Enmin Huang","email":"","orcid":"","institution":"the Sixth Affiliated Hospital, Sun Yat-sen University","correspondingAuthor":false,"prefix":"","firstName":"Enmin","middleName":"","lastName":"Huang","suffix":""},{"id":581780131,"identity":"4c9d1615-957c-498b-af85-c4f1af9c8081","order_by":3,"name":"Yina Liu","email":"","orcid":"","institution":"Sun Yat-Sen University Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Yina","middleName":"","lastName":"Liu","suffix":""},{"id":581780132,"identity":"c917d591-ab9e-417a-822a-2c0c68158a98","order_by":4,"name":"Xiaoshuang Lyu","email":"","orcid":"","institution":"the Sixth Affiliated Hospital, Sun Yat-sen University","correspondingAuthor":false,"prefix":"","firstName":"Xiaoshuang","middleName":"","lastName":"Lyu","suffix":""},{"id":581780133,"identity":"c8675cb1-f585-46e2-9d8a-eb723a8ed47e","order_by":5,"name":"Yang Fu","email":"","orcid":"https://orcid.org/0000-0003-0202-4893","institution":"The Sixth Affiliated Hospital of Sun Yat-sen University","correspondingAuthor":false,"prefix":"","firstName":"Yang","middleName":"","lastName":"Fu","suffix":""},{"id":581780134,"identity":"b39dd96c-4c4c-487e-9983-15aeb87d2650","order_by":6,"name":"Fan Bai","email":"","orcid":"","institution":"the Sixth Affiliated Hospital, Sun Yat-sen University","correspondingAuthor":false,"prefix":"","firstName":"Fan","middleName":"","lastName":"Bai","suffix":""},{"id":581780135,"identity":"f23408cd-b077-4ec0-9124-7282481c8c10","order_by":7,"name":"Chenxu Guo","email":"","orcid":"","institution":"the Sixth Affiliated Hospital, Sun Yat-sen University","correspondingAuthor":false,"prefix":"","firstName":"Chenxu","middleName":"","lastName":"Guo","suffix":""},{"id":581780136,"identity":"04c68e10-22b8-4650-88a4-7671b167644a","order_by":8,"name":"Ge Qin","email":"","orcid":"","institution":"The Sixth Affiliated Hospital of Sun Yat-sen University","correspondingAuthor":false,"prefix":"","firstName":"Ge","middleName":"","lastName":"Qin","suffix":""},{"id":581780137,"identity":"26deb8db-9aa5-45ed-9894-4107865a388f","order_by":9,"name":"Yuqian Xie","email":"","orcid":"","institution":"the Sixth Affiliated Hospital, Sun Yat-sen University","correspondingAuthor":false,"prefix":"","firstName":"Yuqian","middleName":"","lastName":"Xie","suffix":""},{"id":581780138,"identity":"92ac5590-6ebd-4bdc-bbe6-3061ae384a5b","order_by":10,"name":"Jianwei Zhang","email":"","orcid":"","institution":"The Sixth Affiliated Hospital, Sun Yat-sen University","correspondingAuthor":false,"prefix":"","firstName":"Jianwei","middleName":"","lastName":"Zhang","suffix":""},{"id":581780139,"identity":"0f7d5ae5-a266-40a1-a4ed-1dcf6542797c","order_by":11,"name":"Yanhong Deng","email":"","orcid":"","institution":"the Sixth Affiliated Hospital, Sun Yat-sen University","correspondingAuthor":false,"prefix":"","firstName":"Yanhong","middleName":"","lastName":"Deng","suffix":""}],"badges":[],"createdAt":"2026-01-27 07:16:59","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8706982/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8706982/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":103049817,"identity":"46747197-7575-4e4c-8228-e32d01c9e799","added_by":"auto","created_at":"2026-02-20 07:46:34","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":570708,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eElevated RMI2 levels are associated with adverse clinical outcomes in CRC.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Representative images of immunohistochemical (IHC) staining for RMI2 in CRC tissue microarrays (TMA) (n=226). Scale bars, 100 µm. (B) Quantification of RMI2 expression in human normal colorectal tissues and CRC tissues, based on RMI2 IHC scores. (C) Quantification of RMI2 expression in T1-2 and T3-4 CRC tissues, based on RMI2 IHC scores. (D) Quantification of RMI2 expression in stage I-II and III-IV CRC tissues, based on RMI2 IHC scores. (E) Quantification of RMI2 expression in with and without lymph node metastasis CRC tissues, based on RMI2 IHC scores. (F) Overall survival (OS) analysis comparing patients with low and high RMI2 expression in the CRC cohort. (G) OS analysis of patients without lymph node metastasis in CRC cohort 1 based on RMI2 expression. (H) OS analysis of patients with lymph node metastasis in CRC cohort 1 based on RMI2 expression. (I) Representative images of IHC staining for RMI2 in CRC cohort 2 (n=20). (J) Quantification of RMI2 in tissues from patients in in CRC cohort2. Scale bars, 100 µm. (K) Representative computed tomography (CT) images of patients with liver metastasis of CRC before and after anti-PD-1 mAb (PD-1) treatment in our refractory microsatellite-stable (MSS) CRC cohort 3. Yellow circles indicate the location of liver metastases. (L) Histogram showing the overall response rate in patients with low and high RMI2 expression levels in the refractory MSS CRC cohort 3 (n=40). (M) Progression-free survival (PFS) analysis of patients in the refractory MSS CRC cohort 3, based on RMI2 expression. (N) Correlation between RMI2 expression and tumor-infiltrating lymphocytes (TILs) in TCGA-COAD dataset. (O) Pearson correlation between TILs proportion and RMI2 expression across 33 cancer types from TCGA. (P) Kaplan-Meier survival curves for overall survival and progression-free survival based on RMI2 expression levels in the Kaplan-Meier Plotter Immunotherapy cohort, which includes pan-cancer samples treated with anti-PD-1 therapy. Data are presented as means ± SEM. Statistical analyses were performed using unpaired Student’s t-tests (B-E), Fisher’s exact test (J and L), and log-rank tests (F-H, M and P). *, P \u0026lt; 0.05; **, P \u0026lt; 0.01; ***, P \u0026lt; 0.001.\u003c/p\u003e","description":"","filename":"image1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8706982/v1/216407a01b7411f7891cf2f7.jpeg"},{"id":102994793,"identity":"3c9e9e8f-feec-4895-aaf7-8e6e3a201ff6","added_by":"auto","created_at":"2026-02-19 11:59:15","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":642882,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eRMI2 alters the immune landscape and impairs T-cell and NK cell response.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) The impact of RMI2 knockout on CT26 tumor growth was assessed in BALB/c mice (n\u003cem\u003e \u003c/em\u003e= 6/group). (B) Representative images of tumors (left) and tumor weight (right) for the indicated groups, as shown in (A). For tumor volume, p values were calculated by two-way ANOVA. For tumor weight, p values were calculated by two-tailed student’s t-test versus the NC. Data are means ± s.e.m. of tumor volume and weight. (C) scRNA-seq was performed on CT26 tumors (NC=2, sgRMI2=2). The UMAP plot shows the distribution of clusters, colored by annotated cell type combined from both groups, and the relative fraction of each cluster was shown on the right. (D) Pathways enrichment analysis in cancer cells from sgRMI2 versus NC tumors. (E) Upregulation of type I/II interferon and inflammatory response-related pathways in sgRMI2 tumor cells. (F and G) Gene ontology biological process enrichment analysis (GOBP) in reclustered CD8\u003csup\u003e+\u003c/sup\u003e T cells. (F) GOBP in NK cells. *, P \u0026lt; 0.05; **, P \u0026lt; 0.01; ***, P \u0026lt; 0.001.\u003c/p\u003e","description":"","filename":"image2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8706982/v1/18f8b4a290929e29ee8601e0.jpeg"},{"id":102994790,"identity":"b45c1058-5cbb-435d-b4f7-7ef26348b365","added_by":"auto","created_at":"2026-02-19 11:59:15","extension":"jpeg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":473873,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eRMI2 deficiency promotes infiltration and activation of cytotoxic CD8\u003c/strong\u003e\u003csup\u003e\u003cstrong\u003e+\u003c/strong\u003e\u003c/sup\u003e\u003cstrong\u003e T cells.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Flow cytometry analysis of subcutaneously implanted MC38 tumors with non-targeting control (NC) or RMI2 knockout. The percentages of CD45\u003csup\u003e+\u003c/sup\u003e, NK, DC, and Ly6G\u003csup\u003e+\u003c/sup\u003e MDSC cells within the tumors were analyzed (n\u003cem\u003e \u003c/em\u003e= 5/group). (B) Flow cytometric analysis of the proportions of CD4\u003csup\u003e+\u003c/sup\u003e T cells and CD8\u003csup\u003e+\u003c/sup\u003e T cells in MC38 subcutaneous tumors from C57BL/6J mice. (C and D) Tumor volume, representative tumor images, and tumor weight from RMI2 knockout MC38 tumor cells in C57BL/6J mice depleted of CD8\u003csup\u003e+\u003c/sup\u003e T or NK cells (n\u003cem\u003e \u003c/em\u003e= 5/group). s.c represents subcutaneous injection. (E) Flow cytometric analysis of the proportions of GZMB\u003csup\u003e+\u003c/sup\u003e CD8\u003csup\u003e+\u003c/sup\u003e T cells, IFN-γ\u003csup\u003e+\u003c/sup\u003e CD8\u003csup\u003e+\u003c/sup\u003e T cells, and Ki67\u003csup\u003e+\u003c/sup\u003e CD8\u003csup\u003e+\u003c/sup\u003e T cells in MC38 subcutaneous tumors from C57BL/6J mice. (F) Infiltration of CD8\u003csup\u003e+\u003c/sup\u003e T cells in CT26 xenografts by IHC analysis (n\u003cem\u003e \u003c/em\u003e= 5/group). Tumor tissues were stained using anti-CD8. Blue: hematoxylin\u003csup\u003e+\u003c/sup\u003e cells, brown: CD8\u003csup\u003e+\u003c/sup\u003e cells. (G) IHC staining of GZMB, IFN-γ and Ki67 in CT26 xenografts (n\u003cem\u003e \u003c/em\u003e= 5/group). Scale bars, 20 μm. (H) Histogram showing the CD8 positive rate in patients with low and high RMI2 expression levels in the refractory MSS CRC cohort 3. (I) A schematic representation of the in vitro coculture assay involving tumor and immune cells. (J) In vitro cytotoxicity assays for activated CD8\u003csup\u003e+\u003c/sup\u003e T cells against RMI2 knockout or control CT26 cells. (K) Post-coculture of tumor cells with splenocytes, the percentages of GZMB\u003csup\u003e+\u003c/sup\u003e, IFN-γ\u003csup\u003e+\u003c/sup\u003e, and Ki67 cells within the CD8\u003csup\u003e+\u003c/sup\u003e T cell population were determined by flow cytometry. Three independent repeated experiments. Data are presented as means ± SEM. Statistical analyses were performed using unpaired Student’s t-tests (A, B, D, E-G and K), two-way ANOVA (C), and Fisher’s exact test (H). *, P \u0026lt; 0.05; **, P \u0026lt; 0.01; ***, P \u0026lt; 0.001. ns, not significant. NC, negative control; GZMB, granzyme B; IFN-γ, Interferon-γ.\u003c/p\u003e","description":"","filename":"image3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8706982/v1/1cfa841c7fb875bd0a128a0b.jpeg"},{"id":102994799,"identity":"305ecfff-defa-42b3-86df-0fc7e6ffa744","added_by":"auto","created_at":"2026-02-19 11:59:16","extension":"jpeg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":455888,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eRMI2 depletion increases DNA damage and impairs homologous recombination.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Expression levels of the indicated proteins were assessed by Western blot in stably transfected HCT116 and DLD1 cells. (B and C) Representative images (B) and quantification (C) of γH2AX immunofluorescence staining in the indicated cells. Scale bars, 20 µm. (D and E) Comet assays (D) and analysis (E) of tail moment in the indicated cells. Scale bars, 20 µm. (F) Quantification of apoptotic (annexin V\u003csup\u003e+\u003c/sup\u003e) cells in NC and shRMI2 cells in the indicated CRC cell lines. (G) Representative images of IHC staining for RMI2 and γH2AX (Left). Scale bars, 100 µm. Statistical analysis of γH2AX levels and RMI2 expression in the CRC cohort 1 (Right). (H) Schematic representation of the HR reporter (Upper). HR efficiency was measured in CRC cells by determining GFP expression via flow cytometry (Bottom). (I) Schematic representation of the NHEJ reporter (Upper). NHEJ efficiency was measured in CRC cells by determining GFP expression via flow cytometry (Bottom). (J and K) Immunostaining with anti-RAD51 (J) and anti-BRCA1 (K) was performed in indicated cells. Scale bars, 15 μm. (L) RAD51 and BRCA1 were immunoprecipitated from cell lysates, followed by immunoblotting with the indicated antibodies. Data are presented as means ± SEM. Data were analyzed using unpaired Student’s t test (C, E, F, H and I), and Fisher’s exact test (G). **, P \u0026lt; 0.01; ***, P \u0026lt; 0.001.\u003c/p\u003e","description":"","filename":"image4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8706982/v1/53f3d7694925e60ab6403a69.jpeg"},{"id":102994795,"identity":"1e567382-62d6-43d7-b525-e31e88904759","added_by":"auto","created_at":"2026-02-19 11:59:15","extension":"jpeg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":671453,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eRMI2 deficiency leads to elevated levels of cytosolic DNA, thereby activating the cGAS-STING pathway.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A and B) Representative images (left) and quantification (right) of cytoplasmic DNA immunofluorescence staining in stably transfected DLD1 and SW480 cells. Scale bars, 20 µm. (C) Gene set enrichment analysis (GSEA) highlighting significantly differentially expressed pathways. (D) GSEA of immune-related signaling pathways comparing the NC and RMI2 knockdown groups. (E) Immunoblot of the indicated proteins in NC, shRMI2, and RMI2-rescued cells. (F) Immunoblot of the indicated proteins in CRC cells. (G) ELISA analysis of human IFN-β in conditioned media derived from CRC cells transduced with the indicated vectors. (H) RT-qPCR analysis of mRNA expression levels of IFN-β, IFN-γ, ISG15, CXCL10, and IRF7 in CRC cells. (I) RT-qPCR analysis of mRNA expression levels of IFN-β, IFN-γ, ISG15, CXCL10, and IRF7 in NC, shRMI2, and RMI2-rescued cells. (J) ELISA analysis of human IFN-β in conditioned media derived from CRC cells transduced with the indicated vectors. (K) RT-qPCR analysis of mRNA expression levels of IFN-β, IFN-γ, ISG15, CXCL10, and IRF7 in CRC cells. (L) Immunoblot of the indicated proteins in mouse CRC CT26 cells. (M) The effect of RMI2 and STING knockdown on CT26 tumor growth in syngeneic BALB/c mice. (N) Representative images of tumors (left) and tumor weight (right) for the indicated groups, as shown in (M). Data are presented as means ± SEM. Statistical analyses were performed using unpaired Student’s t-tests (A, B, G–I, and K), one-way ANOVA with Tukey’s multiple comparison correction (J and N) and two-way ANOVA (M). **, P \u0026lt; 0.01; ***, P \u0026lt; 0.001. ns, not significant.\u003c/p\u003e","description":"","filename":"image5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8706982/v1/d6aeea33819d116b9603180e.jpeg"},{"id":102994792,"identity":"3354d4a6-7256-49da-bab6-1d4fc4b98ff1","added_by":"auto","created_at":"2026-02-19 11:59:15","extension":"jpeg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":497067,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eRMI2 deficiency enhances the tumor's sensitivity to anti-PD-1 therapy.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Schematic diagram describing the construction and treatment of the tumor orthotopic model in BALB/c mice. Representative images of bioluminescent imaging of the indicated groups in the CT26 orthotopic model on day 5 and day 25 (Right). (B) Relative fluorescence intensity fold change by bioluminescent imaging in the CT26 orthotopic model (n\u003cem\u003e \u003c/em\u003e= 5/group). (C) Overall survival of CT26 orthotopic tumor-bearing mice from the indicated groups. (D)Tumor growth curves of subcutaneous tumors at the indicated time points. (E) Representative images of tumors (left) and tumor weight (right) for the indicated groups (n\u003cem\u003e \u003c/em\u003e= 6/group), as shown in (D). (F) Flow cytometric analysis of the proportions of NK cells, CD4\u003csup\u003e+\u003c/sup\u003e T cells, and CD8\u003csup\u003e+\u003c/sup\u003e T cells in MC38 subcutaneous tumors. (G) Flow cytometric analysis of the proportions of GZMB\u003csup\u003e+\u003c/sup\u003e CD8\u003csup\u003e+\u003c/sup\u003e T cells, IFN-γ\u003csup\u003e+\u003c/sup\u003e CD8\u003csup\u003e+\u003c/sup\u003e T cells, and Ki67\u003csup\u003e+\u003c/sup\u003e CD8\u003csup\u003e+\u003c/sup\u003e T cells in MC38 subcutaneous tumors. (H) IHC staining of GZMB, IFN-γ and Ki67 in CT26 xenografts (n\u003cem\u003e \u003c/em\u003e= 5/group). Scale bars, 20 μm. Data are presented as means ± SEM. Statistical analyses were performed using one-way ANOVA with Tukey’s multiple comparison correction (B and E-H), two-way ANOVA (D), and log-rank test (C). *, P \u0026lt; 0.05; **, P \u0026lt; 0.01; ***, P \u0026lt; 0.001.\u003c/p\u003e","description":"","filename":"image6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8706982/v1/fe89056efaa5a26a2e1e2f50.jpeg"},{"id":103049914,"identity":"25ccb34b-157c-425c-9eb2-3764d0746550","added_by":"auto","created_at":"2026-02-20 07:47:12","extension":"jpeg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":750785,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eRMI2 deficiency sensitizes tumors to 5-fluorouracil (5-FU).\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Immunoblot of the indicated proteins in SW480 CRC cells after 5-FU treatment. (B) Representative images of γH2AX immunofluorescence staining in the indicated cells. Scale bars, 20 µm. (C) Representative images of cytoplasmic DNA immunofluorescence staining in the indicated cells. Scale bars, 20 µm. (D) ELISA analysis of human IFN-β in conditioned media derived from CRC cells transduced with the indicated treatment. (E) RT-qPCR analysis of mRNA expression levels of IFN-β, IFN-γ, ISG15, CXCL10, and IRF7 in SW480 cells. (F) Tumor growth curves of subcutaneous tumors at the indicated time points. (G) Representative images of tumors (left) and tumor weight (right) for the indicated groups, as shown in (F). (H and I) Flow cytometric analysis of the proportions of CD45 (H), NK cells, CD4\u003csup\u003e+\u003c/sup\u003e T cells, and CD8\u003csup\u003e+\u003c/sup\u003e T cells (I) in CT26 subcutaneous tumors. (J) Flow cytometric analysis of the proportions of GZMB\u003csup\u003e+\u003c/sup\u003e CD8\u003csup\u003e+\u003c/sup\u003e T cells, IFN-γ\u003csup\u003e+\u003c/sup\u003e CD8\u003csup\u003e+\u003c/sup\u003e T cells, and Ki67\u003csup\u003e+\u003c/sup\u003e CD8\u003csup\u003e+\u003c/sup\u003e T cells in CT26 subcutaneous tumors. (K and L) IHC staining of CD8 (K), and GZMB, IFN-γ and Ki67 (L) in CT26 xenografts (n\u003cem\u003e \u003c/em\u003e= 6/group). Scale bars, 20 μm. (M) Representative IHC staining images of RMI2 in a CRC TMA from cohort 4 (n=154) of patients who received neoadjuvant chemotherapy. Scale bars, 1 mm. (N) The correlation between RMI2 expression and CD8-positive cells is shown. (O) Overall survival analysis comparing patients with low and high RMI2 expression in the CRC cohort 4. Data are presented as means ± SEM. Statistical analyses were performed using one-way ANOVA with Tukey’s multiple comparison correction (D, E, and G-L), two-way ANOVA (F), Pearson’s correlation test (N), Fisher’s exact test (M), and log-rank test (O). *, P \u0026lt; 0.05; **, P \u0026lt; 0.01; ***, P \u0026lt; 0.001. ns, not significant.\u003c/p\u003e","description":"","filename":"image7.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8706982/v1/a90d4d252cd7e2da809009b2.jpeg"},{"id":103050928,"identity":"2d8b9f3f-716a-4616-bb82-8485a1af70b8","added_by":"auto","created_at":"2026-02-20 07:57:08","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":5487191,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8706982/v1/8c300cda-8a09-40f7-85ae-dbfab038b640.pdf"},{"id":102994794,"identity":"6c91728f-d757-44af-9058-4f952c5e7559","added_by":"auto","created_at":"2026-02-19 11:59:15","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":421892,"visible":true,"origin":"","legend":"Uncropped original western blots","description":"","filename":"Uncroppedoriginalwesternblots.docx","url":"https://assets-eu.researchsquare.com/files/rs-8706982/v1/3c0a67eefdfd515a99744d77.docx"},{"id":102994789,"identity":"3de7ea6b-72ad-406f-a18e-fba053b69be1","added_by":"auto","created_at":"2026-02-19 11:59:15","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":955813,"visible":true,"origin":"","legend":"RMI2 Depletion Recovers cGAS-STING Signaling to Enhance Immunotherapy in Colorectal Cancer-SM","description":"","filename":"SupplementarymaterialTableS5.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-8706982/v1/2aad5a44757a01b659d25d90.xlsx"},{"id":103049366,"identity":"c95654f8-0f75-4c42-8acf-5e795cd07cf9","added_by":"auto","created_at":"2026-02-20 07:40:21","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":4357212,"visible":true,"origin":"","legend":"RMI2 Depletion Recovers cGAS-STING Signaling to Enhance Immunotherapy in Colorectal Cancer-SM","description":"","filename":"Supplementarymaterial.docx","url":"https://assets-eu.researchsquare.com/files/rs-8706982/v1/16751a22a03358064fe1b986.docx"}],"financialInterests":"(Not answered)","formattedTitle":"RMI2 Depletion Recovers cGAS-STING Signaling to Enhance Immunotherapy in Colorectal Cancer","fulltext":[{"header":"Highlights","content":"\u003cp\u003eλ RMI2 maintains immune evasion in MSS CRC by suppressing cytosolic DNA sensing.\u003c/p\u003e\u003cp\u003eλ RMI2 loss triggers innate immune activation via the cGAS\u0026ndash;STING pathway.\u003c/p\u003e\u003cp\u003eλ Targeting RMI2 enhances immunotherapy efficacy in MSS CRC.\u003c/p\u003e"},{"header":"Introduction","content":"\u003cp\u003eColorectal cancer (CRC) remains a leading cause of cancer-related morbidity and mortality worldwide. Although immune checkpoint blockade (ICB) therapy has demonstrated remarkable efficacy in a subset of patients with mismatch repair-deficient (dMMR) or microsatellite instability-high (MSI-H) CRC\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e, the majority of CRC cases, particularly those classified as microsatellite stable (MSS), exhibit primary resistance to ICB. This resistance is largely attributed to an immunologically \u0026ldquo;cold\u0026rdquo; tumor microenvironment (TME) characterized by low infiltrating CD8⁺ T cell density, limited antigen presentation, and suppressed type I interferon signaling\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e. Consequently, there is an urgent unmet need to identify molecular targets capable of reprogramming the TME to enhance ICB responsiveness in MSS CRC.\u003c/p\u003e \u003cp\u003eGenomic instability is a hallmark of cancer and a potential source of tumor neoantigens that elicit antitumor immune responses. Previous studies from our laboratory and others have indicated that RecQ-mediated genome instability 2 (RMI2) is an evolutionarily conserved component of the RMI complex, which, together with BLM and TOP3A, resolves toxic homologous recombination intermediates and maintains genomic stability. Emerging evidence from our group and others indicates that RMI2 functions as a bona fide oncoprotein across multiple malignancies, where it supports tumor cell survival by preventing the accumulation of DNA damage and inhibiting premature senescence or apoptosis\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. RMI2 depletion triggers genomic instability, leading to the accumulation of DNA damage \u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. Defects in DNA repair mechanisms or oncogene activation drive genomic instability, resulting in increased somatic nonsynonymous mutations that generate abundant tumor-specific neoantigens through aberrant protein synthesis pathways\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e. Genomic instability can enhance the efficacy of ICB by revealing immune vulnerabilities\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e. The immune system identifies these neoantigens as non-self-antigens, thereby initiating potent T-cell priming and cytotoxic responses\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e. However, the precise role of RMI2 in modulating the interplay between genomic instability, DNA sensing, and antitumor immunity remains poorly defined.\u003c/p\u003e \u003cp\u003eThe stimulator of interferon genes (STING) pathway, activated upon recognition of cytosolic DNA by cyclic GMP-AMP synthase (cGAS), exerts dual immunoregulatory functions in tumorigenesis\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e. While STING-induced type I interferon signaling can enhance antigen processing and T-cell infiltration within immunologically hostile tumor microenvironments (TMEs), cancer cells frequently circumvent STING-mediated immune surveillance through epigenetic mechanisms that silence STING expression\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e. STING dimers execute ER-to-Golgi translocation to assemble into oligomeric complexes in the Golgi membrane. Subsequent TBK1 binding to STING oligomers triggers autophosphorylation-mediated activation, facilitating TBK1-mediated phosphorylation of STING and downstream IRF3. Phosphorylated IRF3 dimers undergo nuclear translocation to induce type I interferon gene transcription\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e. Interestingly, certain genetic or pharmacological perturbations can restore the STING pathway activity in tumors. For instance, in MSI-H CRC, constitutive activation of the cGAS-STING pathway correlates with enhanced antigen presentation and immune cell infiltration\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e. Similarly, SHP2-mediated dephosphorylation of PARP-1 following DNA damage leads to cytosolic dsDNA accumulation and STING activation\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e. The depletion of serine/threonine phosphatases or arginine methyltransferases, such as PRMT6, has also been reported to enhance STING signaling and antitumor immunity\u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e. These findings suggest that targeting pathways that regulate cytosolic DNA processing and STING activation may represent a viable strategy for converting immunologically cold tumors into immunologically responsive ones.\u003c/p\u003e \u003cp\u003eIn this study, we demonstrated that RMI2 is adaptively upregulated in CRC and suppresses the STING pathway by maintaining genomic stability through enhanced homologous recombination repair. RMI2 deficiency disrupts this protective mechanism, leading to the accumulation of cytosolic DNA, activation of the cGAS-STING pathway, and the subsequent enhancement of antitumor immunity. Our findings uncover a previously unrecognized dual role of RMI2 in simultaneously preserving genomic integrity and suppressing immune surveillance, thereby facilitating immune evasion.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003ePatients and samples\u003c/h2\u003e \u003cp\u003eIn this study, formalin-fixed paraffin-embedded (FFPE) tumor specimens were collected from 226 consecutive patients with CRC who underwent primary tumor resection between January 2015 and December 2016 at our institution, with comprehensive clinicopathological data obtained during a median follow-up duration of 7 years. Overall survival (OS) was defined as the interval from the date of curative surgery to the date of death from any cause, with surviving patients censored at the date of the last follow-up.\u003c/p\u003e \u003cp\u003eTo analyze the refractory metastatic MSS CRC immunotherapy cohort, we conducted a rigorous retrospective analysis of prospectively maintained medical records, including patients who met all of the following criteria: histological and imaging studies confirmed metastatic CRC with MSS status, documented disease progression following standard chemotherapy regimens, and subsequent treatment with PD-1 checkpoint inhibitors at our institution. The observational period spanned July 2018 to April 2024. Tumor responses were assessed using serial cross-sectional computed tomography (CT) imaging and evaluated according to RECIST 1.1 criteria. Following standardized protocols, two independent radiologists performed blinded assessments of the CT images to measure and record the longest diameters of all target lesions. Tumor responses were categorized into four distinct groups based on the RECIST 1.1 criteria: (1) Complete Response (CR), defined as the complete disappearance of all target lesions; (2) Partial Response (PR), defined as a\u0026thinsp;\u0026ge;\u0026thinsp;30% reduction in the sum of the longest diameters of target lesions compared to baseline; (3) Progressive Disease (PD), defined as a\u0026thinsp;\u0026ge;\u0026thinsp;20% increase in the sum of the longest diameters of target lesions compared to the smallest sum observed since treatment initiation, or the appearance of new lesions; and (4) Stable Disease (SD), defined as neither sufficient shrinkage for PR nor sufficient increase for PD. Tumor responses were further dichotomized into response (R), including CR and PR, and Non-Response (NR), including PD and SD. To ensure interobserver reliability, a single-blinded assessment framework was utilized, wherein radiologists were blinded to the clinical outcomes but had access to baseline and follow-up imaging data. Any discrepancies between the evaluators were resolved through consensus-based discussions, with the final classification determined by the majority consensus.\u003c/p\u003e \u003cp\u003eAll clinicopathological data were extracted from the institutional CRC registry database according to protocols approved by the Institutional Review Board of the Sixth Affiliated Hospital, Sun Yat-sen University (2022ZSLYEC-002). Written informed consent was waived due to the retrospective nature of this study, and strict patient confidentiality was maintained in compliance with the principles of Declaration of Helsinki.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eCell lines\u003c/h3\u003e\n\u003cp\u003eAll cell lines, including human CRC cell lines (DLD-1, HCT116, SW480, and RKO), murine CRC cell lines (CT26), and HEK293T cells, were obtained from the American Type Culture Collection (ATCC). Cells were cultured at 37\u0026deg;C in a humidified incubator containing 5% CO₂, using DMEM or RPMI-1640 medium (Bio-channel) supplemented with 10% fetal bovine serum (FBS; ExCell Bio) and 1% penicillin-streptomycin. Cell lines were authenticated by short tandem repeat (STR) profiling and confirmed to be free of mycoplasma contamination using standard PCR-based detection methods. All experiments were conducted using cells at low passage numbers (\u0026le;\u0026thinsp;20 passages).\u003c/p\u003e\n\u003ch3\u003ePlasmids\u003c/h3\u003e\n\u003cp\u003eThrough rigorous molecular cloning methods, we constructed stable expression vectors to investigate the functional roles of RMI2, using the pSin-EF2-puro-oligo lentiviral backbone. This plasmid vector is specifically designed for efficient, long-term transgene expression in mammalian cells, facilitating selection with puromycin. To achieve stable knockdown of human RMI2, short hairpin RNA (shRNA) constructs were generated and cloned into the pLKO.1-puro lentiviral expression vector, containing a puromycin resistance cassette for stable selection. For genome editing applications using the CRISPR-Cas9 system, we designed single-guide RNA (sgRNA) constructs targeting mouse RMI2 and STING, which were cloned into the pLentiCRISPR-V2-puro backbone. The specific sequences of the shRNAs/sgRNAs are listed in Table S4. To ensure the integrity and accuracy of all the genetic constructs, each construct was rigorously validated by DNA sequencing and subjected to strict quality control assessments.\u003c/p\u003e\n\u003ch3\u003eAntibodies and reagents\u003c/h3\u003e\n\u003cp\u003ePrimary antibodies used for Western blotting included: RMI2 rabbit antibody (Abcam, ab122685; 1:1,000 dilution), γH2AX rabbit antibody (Cell Signaling Technology, 9718; 1:500 dilution), STING rabbit antibody (Cell Signaling Technology, 13647; 1:1,000 dilution), Phospho-STING rabbit antibody (Cell Signaling Technology, 50907; 1:500 dilution), TBK1 rabbit antibody (Cell Signaling Technology, 3504; 1:1,000 dilution), Phospho-TBK1 rabbit antibody (Cell Signaling Technology, 5483; 1:500 dilution), IRF-3 rabbit antibody (Cell Signaling Technology, 11904; 1:1,000 dilution), Phospho-IRF-3 rabbit antibody (Cell Signaling Technology, 29047; 1:500 dilution), RAD51 mouse antibody (Abcam, ab88572, 1:1,000 dilution), BRCA1 mouse antibody (Santa Cruz, sc-6954, 1:500 dilution) Flag rabbit antibody (Cell Signaling Technology, 14793; 1:2,000 dilution), GAPDH rabbit antibody (ProteinTech, 10494-1-AP; 1:2,000 dilution), and GAPDH mouse antibody (ProteinTech, 60004-1-Ig; 1:10,000 dilution). For flow cytometry, the following conjugated antibodies were used: PE/Cyanine7-conjugated anti-mouse CD3 antibody (clone 17A2, BioLegend), Brilliant Violet 650-conjugated anti-mouse CD8a antibody (clone 53\u0026thinsp;\u0026minus;\u0026thinsp;6.7, BioLegend) or PE-Cyanine5.5-conjugated anti-mouse CD8a antibody (clone 53\u0026thinsp;\u0026minus;\u0026thinsp;6.7, Invitrogen), Alexa Fluor 700-conjugated anti-mouse CD4 antibody (clone GK1.5, BioLegend), eFluor 450-conjugated anti-mouse Granzyme B antibody (clone NGZB, Invitrogen), PE-conjugated anti-mouse IFN-γ antibody (clone XMG1.2, BioLegend), and PE conjugated anti-mouse Ki67 antibody (clone 16A8, BioLegend). HRP-conjugated secondary antibodies for mouse and rabbit IgG were obtained from Cell Signaling Technology. Puromycin, PicoGreen, and DAPI staining reagents were obtained from Sigma-Aldrich.\u003c/p\u003e\n\u003ch3\u003eLentivirus and stable expression cell lines construction\u003c/h3\u003e\n\u003cp\u003eHere, we describe the construction of lentivirus and stably expressing cell lines using HEK-293T cells. Cells were seeded in six-well plates at approximately 60% confluence. The following day, a transfection cocktail was prepared, consisting of 3 \u0026micro;g of pLKO.1-shRNA or pLentiCRISPR-sgRNA/pSinEF2-cDNA, 2 \u0026micro;g of psPAX2 and 1 \u0026micro;g of pMD2G, dissolved in 24 \u0026micro;L of polyethylenimine (PEI) at a concentration of 2 mg/mL. The cells were then transfected with the cocktail.\u003c/p\u003e \u003cp\u003eAfter incubation for 48 h, the supernatant was collected, filtered through 0.45 \u0026micro;m PVDF filters (Millipore), and used to infect tumor cells in the presence of polybrene (10 \u0026micro;g/mL; Sigma). The infected cultures were centrifuged at 2000 rpm (800\u0026times;g) for 45 min at 37\u0026deg;C, followed by replacement of the medium with puromycin. Stable cell lines were selected for 1 week using puromycin concentrations optimized for each cell line (0.5 \u0026micro;g/mL for DLD1 and HCT116, and 2 \u0026micro;g/mL for CT26 and MC38). The cells were then analyzed by western blotting to confirm protein expression. All stable cell lines were maintained for no more than 1 month prior to use.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eAnimal experiments\u003c/h2\u003e \u003cp\u003eAll animal experiments were conducted in strict compliance with the \"Guide for the Care and Use of Laboratory Animals\" and the \"Principles for the Utilization and Care of Vertebrate Animals,\" and were approved by the Animal Research Committee of the Sixth Hospital of Sun Yat-sen University (SYSU-IACUC-2020112701).\u003c/p\u003e \u003cp\u003eMale C57BL/6J, BALB/c, and BALB/c nude mice (all sourced from GemPharmatech) were used in this study. These mice were maintained under stringent specific pathogen-free (SPF) conditions with the following environmental controls: relative humidity of 50%, a 12-hour light cycle alternating with 12 hours of darkness, and a temperature range of 25\u0026deg;C to mimic natural diurnal variations. Before the start of the experiments, mice were allowed a minimum 5-day acclimation period to ensure they were comfortable and to minimize the impact of stress-related variables on experimental outcomes.\u003c/p\u003e \u003cp\u003eIn our experimental design, both subcutaneous and orthotopic tumor models were used to investigate the effects of RMI2 on tumor growth and immune response in vivo. For the subcutaneous model, 5\u0026times;10\u003csup\u003e5\u003c/sup\u003e CT26 or MC38 tumor cells were subcutaneously injected into the right dorsal region of the randomized mice. Tumor volumes were measured using Vernier calipers according to the following formula:\u003c/p\u003e \u003cp\u003eVolume (mm\u003csup\u003e3\u003c/sup\u003e)\u0026thinsp;=\u0026thinsp;1/2\u0026times;(width\u003csup\u003e2\u003c/sup\u003e\u0026times;length)\u003c/p\u003e \u003cp\u003eMice were sacrificed when tumors reached a maximum diameter of 15 mm or total volume of 2000 mm\u0026sup3;.\u003c/p\u003e \u003cp\u003eAn orthotopic tumor model was used as described in previous studies. Briefly, after opening the BALB/c mice's skin and muscle layer with sterile surgical tools, the cecum was exposed. Using a 0.3 mL disposable insulin syringe, 5\u0026times;10\u003csup\u003e5\u003c/sup\u003e CT26 tumor cells in 50 \u0026micro;L (Matrigel: phosphate-buffered saline\u0026thinsp;=\u0026thinsp;1:4) were slowly injected into the subserosal layer of the cecum, followed by careful closure of the muscle and skin layers. The success of the orthotopic tumor model was confirmed using luciferase assay.\u003c/p\u003e \u003cp\u003eTo evaluate the effect of CD8 and NK immune cells on the immune response in our mouse model, we employed a depletion strategy using specific antibodies against these cell populations. CD8\u003csup\u003e+\u003c/sup\u003e T and NK cells were deleted by using 100 \u0026micro;g of anti-mouse CD8α antibody (clone 2.43, BioXcell) and 100 \u0026micro;g of anti-mouse NK1.1 antibody (clone PK136, BioXcell) intraperitoneally injected on days\u0026thinsp;\u0026minus;\u0026thinsp;3, 0, 3, 6 and 9.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eSingle-cell RNA sequencing\u003c/h3\u003e\n\u003cp\u003eCT26 tumors were enzymatically dissociated into single-cell suspensions, and the cell concentration was adjusted to 600\u0026ndash;1200 cells/\u0026micro;L. Single-cell RNA sequencing (scRNA-seq) libraries were prepared using a DNBelab C4 scRNA-seq kit (MGI) with droplet-based mRNA capture. The workflow consisted of cDNA synthesis, PCR amplification, fragment treatment, end repair, A-tailing, and adapter ligation. The resulting libraries were loaded onto DNB nanoarrays and sequenced on the DNBSEQ-T7 platform using Combinatorial Probe-Anchor Synthesis (cPAS).\u003c/p\u003e \u003cp\u003eFor quality control, Fastp and Seurat were applied using the following criteria: \u0026ge;3 cells per gene and \u0026ge;\u0026thinsp;200 genes per cell. Data integration was performed using Canonical Correlation Analysis (CCA) for batch correction, followed by dimensionality reduction via t-SNE and UMAP. Differential gene expression analysis was conducted using the edgeR package, and tissue-specific markers were identified.\u003c/p\u003e\n\u003ch3\u003eThe comet assay\u003c/h3\u003e\n\u003cp\u003eThe comet assays were conducted using the Beyotime Comet Assay Kit (C2041S) following the manufacturer's instructions. Low-melting agarose was heated in water at 70\u0026ndash;80\u0026deg;C for 10 min, and subsequently cooled slowly in a 37\u0026deg;C water bath for \u0026ge;\u0026thinsp;20 min to ensure complete dissolution. Cell suspensions (1\u0026times;10⁶ cells/mL) were mixed with molten agarose at a 1:7.5 (v/v) ratio, and 70 \u0026micro;L of the mixture was pipetted onto comet slides. Slides were refrigerated at 4\u0026deg;C for 10 min to solidify the agarose matrix, followed by incubation in lysis buffer at 4\u0026deg;C for 2 h or overnight to lyse cells. The lysis buffer was pre-chilled at 4\u0026deg;C for \u0026ge;\u0026thinsp;20 min prior to use. After neutralization with the neutralization buffer for 30 min, slides were subjected to horizontal electrophoresis at 25 V for 30 min. Subsequently, a DNA precipitation solution was applied to remove proteins and salts. Samples were stained with propidium iodide (PI) for 20 min in the dark and visualized under an epifluorescence microscope (Olympus). Tail moment analysis was quantified using the Comet Assay Software Project (CASP), with appropriate negative/positive controls included.\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eImmunohistochemistry\u003c/h2\u003e \u003cp\u003eWe constructed a tissue microarray (TMA) from CRC tissue blocks and performed immunohistochemistry (IHC) to assess the expression and localization of specific proteins. IHC staining was conducted using a standardized protocol. Initially, tissue sections were deparaffinized and rehydrated under gentle conditions. Antigen retrieval was achieved by boiling the sections in 1\u0026times;EDTA buffer pH 9.0 for 2.5 min to unmask intracellular antigens. To minimize non-specific binding, sections were blocked with 5% bovine serum albumin (BSA) for 1 h. Subsequently, primary antibodies were applied, including RMI2 rabbit antibody (ab122685, Abcam; 1:300 dilution), γH2AX rabbit antibody (9718, Cell Signaling; 1:200 dilution), CD8A antibody (ZM-0508, ZSGB-bio), IFN-γ antibody (AF5183, Affinity; 1:100 dilution), GZMB antibody (ab255598, Abcam; 1:300 dilution) and Ki-67 antibody (34330, Cell Signaling; 1:200 dilution). After an overnight incubation at 4\u0026deg;C, the secondary antibodies were applied for 1 hour at room temperature. Finally, the staining was visualized using 3,3'-diaminobenzidine (DAB) for 30 s to develop the signal.\u003c/p\u003e \u003cp\u003eMultiplex immunofluorescence analysis was performed using 4% paraformaldehyde-fixed, paraffin-embedded tissue sections mounted on poly-L-lysine-coated slides. The protocol incorporated antigen retrieval via EDTA buffer (pH 8.0) microwave treatment (100\u0026deg;C, 2.5 min), followed by BSA blocking (5%, 1 h, 4\u0026deg;C). Primary antibodies against CD8A (ZM-0508, ZSGB-bio) and RMI2 were incubated overnight at 4\u0026deg;C in 1% BSA-PBS, detected with Alexa Fluor-conjugated secondary antibodies, and the signal was amplified through TSA amplification (PANOVUE, RM-2759). Imaging was performed using the TissueFAXS Cytometry System (TissueGnostics), generating quantitative metrics including cell density (cells/mm\u0026sup2;), and nuclear morphology indices (area, circularity).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eRNA sequencing\u003c/h2\u003e \u003cp\u003eFor RNA-seq analysis, we first performed read filtering and alignment using STAR (version 2.0.2), followed by transcript assembly with StringTie2 (version 1.3.5) to construct the transcriptome. For differential gene expression (DGE) analysis, we utilized the edgeR package to identify genes with a log2 fold change\u0026thinsp;\u0026gt;\u0026thinsp;0.5 and a false discovery rate (FDR)\u0026thinsp;\u0026lt;\u0026thinsp;0.05. To minimize the risk of false positives in our population-level RNA-seq study, we conducted a Wilcoxon rank-sum test on TPM-normalized read counts. Genes with a p-value\u0026thinsp;\u0026gt;\u0026thinsp;0.05 in this test were excluded from our final DEG list. To uncover broader biological implications, we performed functional annotation using Gene Set Enrichment Analysis (GSEA), focusing on pathways with an adjusted FDR indicating statistical significance (Table S5).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eWestern Blotting\u003c/h2\u003e \u003cp\u003eAfter two washes with cold PBS, cells were lysed using cell lysis buffer (P0013, Beyotime) on ice for 20 min. Lysates were centrifuged at 14,000 \u0026times;g for 15 min at 4\u0026deg;C to clarify. The lysate was then heated in gel loading buffer for 10 min and resolved by 11% SDS-PAGE. Proteins were transferred onto a 0.45 \u0026micro;m Immobilon-P PVDF membrane (Millipore) using standard Western blotting techniques. After blocking with PBS containing 5% non-fat milk and 0.1% Tween-20, primary antibodies specific to the target proteins were applied and incubated overnight at 4\u0026deg;C. HRP-conjugated secondary antibodies (W4021 for mouse and W4011 for rabbit, Promega) were then used to detect the target proteins. A high-sensitivity ECL substrate was applied to visualize the bands, which were detected using a MiniChemi imaging system (SageCreation, Beijing).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eFlow Cytometry\u003c/h2\u003e \u003cp\u003eIn our in vivo analysis of tumor-infiltrating lymphocytes (TILs), tumor tissues were harvested and mechanically dissociated. The tissue fragments were then enzymatically digested using a solution containing 0.5 mg/mL collagenase type IV and 200 U/mL DNase I, followed by incubation at 37\u0026deg;C for 1 hour. This enzymatic treatment effectively disrupted the extracellular matrix, enabling the release of individual cells. The resulting cell suspension was filtered through a 70-\u0026micro;m mesh to achieve a homogeneous single-cell suspension. Red blood cells were subsequently lysed using ACK lysis buffer (Solarbio) for 5 minutes at 4\u0026deg;C. Prior to analysis, the cells were incubated with a panel of surface-specific antibodies for 30 minutes on ice, shielded from light to minimize non-specific binding. To differentiate viable from non-viable cells, we employed the Live/Dead Fixable Aqua dye (Thermo Fisher Scientific).\u003c/p\u003e \u003cp\u003eFor intracellular staining, cells were first stimulated with the Leukocyte Activation Cocktail (BD Biosciences) for 4 to 6 hours. This was followed by a dual staining protocol targeting both surface and intracellular markers. The Intracellular Fixation \u0026amp; Permeabilization Buffer Set (88-8824-00, eBioscience) was used according to the manufacturer\u0026rsquo;s guidelines to ensure optimal fixation and permeabilization of the cells.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eImmunofluorescence\u003c/h2\u003e \u003cp\u003eStable cell lines were plated into 15 mm glass-bottom cell culture dishes (801002, NEST). The following day, the cells were fixed with 4% paraformaldehyde for 15 minutes, permeabilized with 0.5% Triton X-100 for 15 minutes, and blocked with goat serum for 30 minutes at room temperature (RT). Between each step, the cells were washed twice with phosphate-buffered saline (PBS). Subsequently, the cells were incubated with primary antibodies overnight at 4\u0026deg;C. After washing, the cells were incubated with secondary antibodies for 1 hour at RT. Nuclei were counterstained with Hoechst 33342 for 2 minutes, followed by three washes with PBS. Finally, the cells were mounted using an antifade mounting medium (Invitrogen). Imaging was performed using a confocal laser scanning microscope (ZEISS, LSM880, ZEN2.6, 63\u0026times; oil immersion lens).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eELISA\u003c/h2\u003e \u003cp\u003eHuman IFN-β (Thermo Fisher Scientific, 414101) and 2\u0026rsquo;3\u0026rsquo;-cGAMP (Cayman Chemical, 501700) ELISAs were performed according to the manufacturer\u0026rsquo;s instructions. Conditioned media collected from cells cultured for 72 hours after seeding (for IFN-β), and cell lysates prepared for cGAMP analysis were processed accordingly. The results represent the average of three replicates from at least two independent experiments, ensuring statistical robustness.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eQuantitative RT-PCR\u003c/h2\u003e \u003cp\u003eTotal RNA was isolated from cell lysates using the RaPure Total RNA Kit (Magen) according to the manufacturer's instructions. Equal amounts of RNA were reverse-transcribed into cDNA using the HiScript II Q RT SuperMix for qPCR (Vazyme). The synthesized cDNA served as a template for subsequent quantitative PCR analysis, which was performed using the ChamQ Universal SYBR qPCR Master Mix (Vazyme) and gene-specific primers. Expression levels were normalized against GAPDH, a commonly used housekeeping gene for normalization. Details of the specific primers are provided in Table \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eProliferation assay\u003c/h2\u003e \u003cp\u003eBriefly, DLD1, HCT116 and SW480 were seeded at a density of 3,000 cells per well in a 96-well microplate. The culture plate was subsequently transferred to the Incucyte\u0026reg; Live-Cell Analysis System (Sartorius, Germany) for continuous monitoring. The system was configured to acquire phase-contrast images at 2-hour intervals under standard culture conditions (37\u0026deg;C, 5% CO2). Real-time cell proliferation kinetics were quantitatively assessed through automated image analysis, measuring temporal changes in cellular confluence (%) using integrated image processing algorithms. Post-experimental data analysis included generation of proliferation curves and calculation of growth parameters, specifically population doubling time (PDT), through proprietary Incucyte software analysis modules.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eColony formation assay\u003c/h2\u003e \u003cp\u003eBriefly, DLD1, HCT116, and SW480 cells were seeded into 6-well plates at a density of 500 cells per well, in triplicate. The cells were cultured under standard conditions for no fewer than 12 days to allow sufficient time for proliferation and colony formation. After the culture period, cells were gently washed twice with PBS to remove excess medium and debris. To preserve cell morphology and their attachment to the plate, cells were fixed using ethanol for approximately 30 minutes. Following fixation, cells were stained with a 1% solution of methyl violet in PBS for 60 minutes to facilitate colony visualization. After staining, plates were washed with PBS to remove excess dye. Finally, colonies formed by each cell line were counted.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003eStatistical analysis\u003c/h2\u003e \u003cp\u003eAll statistical analyses were performed using GraphPad Prism 8 (La Jolla, CA). Data are presented as mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard error of the mean (SEM). Inter-group correlations in tissue microarray gene expression profiles were quantified using Spearman's rank correlation coefficient. For longitudinal tumor growth dynamics, two-way ANOVA with Tukey-Kramer post hoc comparisons was applied to assess time-dependent differences across multiple treatment cohorts. Kaplan-Meier survival curves were compared using the log-rank test. Between-group comparisons for normally distributed data met parametric assumptions and were evaluated using unpaired Student's t-tests (two-tailed). Comparative analysis of three or more independent groups was conducted using one-way ANOVA for multiple comparisons. Statistical significance was defined as p\u0026thinsp;\u0026lt;\u0026thinsp;0.05.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003eHigh RMI2 expression is associated with poor clinical outcomes in CRC\u003c/h2\u003e \u003cp\u003eTo investigate whether RMI2 is associated with the clinical and pathological features of CRC progression, we assessed RMI2 protein expression by immunohistochemistry (IHC) using a human CRC tissue microarray (TMA) \u003cb\u003e(Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e)\u003c/b\u003e. Consistent with public transcriptomic data from The Cancer Genome Atlas (TCGA), CRC tissues exhibited significantly higher RMI2 staining intensity compared to matched normal tissues \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA, \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB and S1A\u003cb\u003e)\u003c/b\u003e. Notably, elevated RMI2 expression was observed in tumors with deeper invasion, advanced tumor stage, and lymph node metastasis \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eC-E\u003cb\u003e)\u003c/b\u003e. Furthermore, high RMI2 expression was correlated with poor prognosis in the overall cohort and patient subgroups, as well as across other cancer types, based on TCGA database analysis \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eF-H, and S1B\u003cb\u003e)\u003c/b\u003e. RMI2 expression was also significantly higher in metastatic tissues than in matched primary CRC tissues (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eI and \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eJ). These findings suggest that RMI2 may serve as a potential biomarker of CRC progression.\u003c/p\u003e \u003cp\u003eNext, we investigated whether RMI2 expression is associated with the tumor immune microenvironment. We focused on a cohort of patients with refractory MSS CRC who received anti-PD-1 therapy or combination immunotherapy \u003cb\u003e(Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eC and table \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e)\u003c/b\u003e. Computed tomography (CT) imaging revealed a higher response rate to anti-PD-1 therapy in liver metastases of patients with low RMI2 expression than in those with high RMI2 expression. Notably, the objective response rate (ORR) was significantly higher in the RMI2-low group than in the RMI2-high group (28.56% vs. 11.54%, P\u0026thinsp;=\u0026thinsp;0.0039) \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eK and \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eL\u003cb\u003e)\u003c/b\u003e. The RMI2-low group also exhibited a significantly longer median progression-free survival (PFS) compared with the RMI2-high group (10 months versus 2.7 months, P\u0026thinsp;=\u0026thinsp;0.0012) \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eM\u003cb\u003e)\u003c/b\u003e. Given that tumor-infiltrating T lymphocytes (T-TILs) are the primary immune cell population responsible for tumor cell recognition and elimination, we further analyzed the correlation between intratumoral T-TIL infiltration and RMI2 expression using data from TCGA database. We found that RMI2 expression was negatively correlated with T-TIL abundance across 32 cancer types, including colon adenocarcinoma (COAD) \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eN and \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eO\u003cb\u003e)\u003c/b\u003e. Additionally, we evaluated the prognostic impact of RMI2 expression in a clinical immunotherapy cohort using Kaplan-Meier Plotter analysis. Among pan-cancer patients treated with anti-PD-1 therapy, those in the RMI2-high group showed significantly shorter overall survival (OS) and PFS than those in the RMI2-low group \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eP\u003cb\u003e)\u003c/b\u003e. Collectively, these findings indicate that high RMI2 expression is associated with poor clinical outcomes in CRC and may serve as a predictive biomarker for the efficacy of ICIs.\u003c/p\u003e \u003cdiv id=\"Sec23\" class=\"Section3\"\u003e \u003ch2\u003eRMI2 loss suppresses tumor growth and modulates the immune tumor microenvironment\u003c/h2\u003e \u003cp\u003eNext, we investigated the functional role of RMI2 in CRC. RMI2 expression was generally higher in CRC cells than in the human intestinal epithelial cell line CCD18-Co \u003cb\u003e(Fig. \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003eA)\u003c/b\u003e. To assess the biological effect of RMI2 on CRC cells, we knocked down RMI2 in HCT116, DLD1 and SW480 cells \u003cb\u003e(Fig. \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003eB)\u003c/b\u003e. RMI2 knockdown significantly inhibited the proliferation and migration of these cell lines (\u003cb\u003eFig. \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003eB-D\u003c/b\u003e). Conversely, lentivirus-mediated RMI2 overexpression (OE) promoted cell proliferation (\u003cb\u003eFig. \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003eE and 2F\u003c/b\u003e), indicating that RMI2 contributes to CRC progression. In vivo, control (NC) or RMI2 knockout (RMI2 sg1/2) MC38 cells were subcutaneously implanted into immunocompetent C57BL/6J and immunodeficient BALB/c nude mice. RMI2 knockout markedly suppressed tumor growth in both immunocompetent and immunodeficient mice \u003cb\u003e(Fig. \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003eA-D)\u003c/b\u003e. Notably, RMI2 knockout resulted in an approximately 78% reduction in tumor volume when MC38 cells were implanted in C57BL/6J mice, whereas the inhibition rate was approximately 60% in immunodeficient BALB/c nude mice \u003cb\u003e(Fig. \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003eE)\u003c/b\u003e. Furthermore, overexpression of RMI2 in C57BL/6J mice promoted tumor growth by approximately 60%, whereas only about 40% tumor growth promotion was observed in immunodeficient mice \u003cb\u003e(Fig. \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003eF-J)\u003c/b\u003e.\u003c/p\u003e \u003cp\u003eTo investigate the impact of RMI2 depletion on the immune microenvironment of MSS CRC, we performed single-cell RNA sequencing (scRNA-seq) to profile the transcriptomic changes in sgRMI2 versus NC CT26 syngeneic tumors \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA and \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB\u003cb\u003e)\u003c/b\u003e. After rigorous quality control, 42,999 high-quality single-cell transcriptomes were obtained from both tumor groups. Notably, inferCNV analysis confirmed that cancer cells exhibited copy number instability, whereas immune populations displayed genomic stability \u003cb\u003e(Fig. S4A)\u003c/b\u003e. Unsupervised clustering combined with uniform manifold approximation and projection (UMAP) visualization revealed distinct cellular populations \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC and S4B\u003cb\u003e)\u003c/b\u003e. Gene Ontology (GO) enrichment analysis using the Kyoto Encyclopedia of Genes and Genomes (KEGG) database revealed significant upregulation of type I/II interferon response pathways and inflammatory response-related signaling pathways in sgRMI2-derived cancer cells \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD and \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE\u003cb\u003e)\u003c/b\u003e. Furthermore, marked expansion of CD8⁺ T cell and natural killer (NK) cell clusters was observed in sgRMI2 tumors \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC and S4C\u003cb\u003e)\u003c/b\u003e. To further dissect the functional differences between the two groups, we performed separate GO enrichment analyses of CD8⁺ T cells and NK cells. In sgRMI2-derived CD8⁺ T cells, multiple pathways associated with T cell activation, proliferation, cytotoxicity, antigen receptor-mediated signaling, and cytokine production were significantly enriched \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eF\u003cb\u003e)\u003c/b\u003e. Similarly, NK cells from sgRMI2 tumors showed prominent enrichment of pathways related to the inflammatory response, cytotoxic activity, and bacterial response \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eG\u003cb\u003e)\u003c/b\u003e. These data demonstrate that high RMI2 expression promotes tumor cell growth and contributes to an immunosuppressive microenvironment.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec24\" class=\"Section2\"\u003e \u003ch2\u003e\u003cb\u003eLoss of RMI2 enhances infiltration and activation of cytotoxic CD8⁺ T cells\u003c/b\u003e\u003c/h2\u003e \u003cp\u003eConsistent with scRNA-seq, RMI2 knockout in MC38 cells significantly increased the infiltration of cytotoxic CD8⁺ T cells and natural killer (NK) cells into the tumor microenvironment, while reducing the proportion of immunosuppressive Ly6G⁺ myeloid-derived suppressor cells (MDSCs). However, the overall abundance of CD45⁺ leukocytes, dendritic cells, and CD4⁺ T cells within the tumors remained unchanged upon RMI2 deletion \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA, \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB and S5A\u003cb\u003e)\u003c/b\u003e. To identify the key immune effector cells mediating tumor growth inhibition under RMI2 knockout models, we selectively depleted CD8⁺ T cells or NK cells in vivo using anti-CD8 or anti-NK1.1 neutralizing antibodies, respectively. Tumor growth was significantly restored upon CD8⁺ T cell depletion, whereas NK cell depletion had no significant effect on tumor progression \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC and \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD\u003cb\u003e)\u003c/b\u003e.\u003c/p\u003e \u003cp\u003eMoreover, RMI2 knockout led to a significant increase in the proportion of effector CD8⁺ T cells, including those expressing granzyme B (GZMB), interferon-gamma (IFN-γ), and proliferation marker Ki67 \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE and S5B\u003cb\u003e)\u003c/b\u003e. IHC staining confirmed that RMI2 knockout tumors exhibited increased infiltration of CD8⁺ T cells, along with elevated expression of GZMB and IFN-γ \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF and \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eG\u003cb\u003e)\u003c/b\u003e. Analysis of refractory MSS CRC patient tissues (cohort 3) and MSS CRC patient (cohort 1) further revealed a significant negative correlation between RMI2 expression and CD8⁺ T cell abundance \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eH and S5C\u003cb\u003e)\u003c/b\u003e. Similarly, RMI2 mRNA expression was negatively correlated with the levels of CD3, and CD8A T cell markers in CRC samples from TCGA database \u003cb\u003e(Fig. S5D)\u003c/b\u003e. In addition, in vitro co-culture experiments demonstrated that CD8⁺ T cells exhibited enhanced cytotoxic activity and functional activation when co-cultured with RMI2 knockdown CT26 cells \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eI-K\u003cb\u003e)\u003c/b\u003e. Collectively, these findings indicate that RMI2 suppresses the recruitment and activation of cytotoxic CD8⁺ T cells.\u003c/p\u003e \u003cdiv id=\"Sec25\" class=\"Section3\"\u003e \u003ch2\u003eLoss of RMI2 leads to elevate DNA damage and impair homologous recombination\u003c/h2\u003e \u003cp\u003eAs previously reported, RMI2 deficiency results in increased DNA damage, leading to genomic instability\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. A higher tumor mutational burden, driven by an increased frequency of somatic nonsynonymous coding mutations, generates a greater abundance of tumor-specific mutant peptides, which may enhance tumor immunogenicity and render these cancers more responsive to immune checkpoint therapies\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e. Therefore, we sought to determine whether RMI2 plays a role in regulating DNA damage in CRC cells. Western blot and immunofluorescence staining showed increased expression and formation of γH2AX foci, a marker of unrepaired DNA lesions, in RMI2 knockdown cells \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA-C\u003cb\u003e)\u003c/b\u003e. Consistently, the comet assay demonstrated significantly elevated levels of DNA damage following RMI2 knockdown \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD and \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eE\u003cb\u003e)\u003c/b\u003e. In addition, RMI2 knockdown increased apoptosis \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eF and S6A\u003cb\u003e)\u003c/b\u003e. IHC staining of CRC tissues further revealed a significant negative correlation between RMI2 expression and the abundance of γH2AX-positive cells \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eG\u003cb\u003e)\u003c/b\u003e. Given that most CRCs harbor loss-of-function mutations in TP53, a key regulator of genomic stability\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e, RMI2 has emerged as a promising therapeutic target that selectively exacerbate DNA damage in these tumors.\u003c/p\u003e \u003cp\u003eNext, we assessed the impact of RMI2 on the DNA damage repair pathways. RMI2 knockdown significantly impaired homologous recombination (HR) repair efficiency, but it had no significant effect on non-homologous end joining (NHEJ) \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eH and \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eI\u003cb\u003e)\u003c/b\u003e. Furthermore, the formation of BRCA1 and RAD51 foci, two critical mediators of the HR repair process \u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e, was markedly reduced in RMI2-deficient cells compared to controls \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eJ, \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eK, S6C and S6D\u003cb\u003e)\u003c/b\u003e. Co-immunoprecipitation assays further demonstrated that RMI2 knockdown significantly decreased the endogenous interaction between BRCA1 and RAD51 \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eL and S6E\u003cb\u003e)\u003c/b\u003e. Collectively, these findings indicate that RMI2 is essential for maintaining genomic stability by supporting homologous recombination-mediated DNA repair, and that its loss leads to the accumulation of unrepaired DNA damage and impaired HR efficiency.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec26\" class=\"Section3\"\u003e \u003ch2\u003eRMI2 deficiency increases cytosolic DNA and activates the cGAS-STING pathway\u003c/h2\u003e \u003cp\u003eNext, we stained cytosolic double-stranded DNA (dsDNA) using PicoGreen and observed a significantly higher proportion of cytosolic DNA-positive cells in RMI2 knockdown cells \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA and \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB\u003cb\u003e)\u003c/b\u003e. The percentage of micronuclei also increased in the RMI2 knockdown cells \u003cb\u003e(Fig. S7A and S7B)\u003c/b\u003e. These findings indicated that RMI2 deficiency promotes the release of nuclear DNA into the cytoplasm. To further explore the underlying mechanisms, we performed RNA sequencing (RNA-seq) of CRC samples from TCGA. Consistent with our scRNA-seq data, gene set enrichment analysis (GSEA) revealed significant upregulation of immune-associated gene sets such as IFN-α response, IFN-γ response, and inflammatory response in RMI2-deficient cells \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC and \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD\u003cb\u003e)\u003c/b\u003e. It is well established that cytoplasmic DNA resulting from DNA damage can activate innate immune responses, including antimicrobial and antitumor immunity\u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e. Next, we investigated whether RMI2 modulates the activity of the cGAS-STING signaling pathway. The baseline phosphorylation levels of endogenous STING, TBK1 and interferon regulatory factor 3 (IRF3) were higher in the RMI2 knockdown cells \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eE\u003cb\u003e)\u003c/b\u003e. Moreover, the proportion of cells with detectable cytosolic DNA accumulation was significantly higher in RMI2 knockdown cells compared to the control and RMI2-rescued cells \u003cb\u003e(Fig. S7C)\u003c/b\u003e.\u003c/p\u003e \u003cp\u003eSince some CRC cell lines such as SW480 and DLD1 express STING, whereas RKO robustly silence STING, we first determined the consequences of RMI2 knockdown across these different cell lines \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eF\u003cb\u003e)\u003c/b\u003e. RMI2 knockdown increased IFN-β secretion in both SW480 and DLD1 cells, but not in RKO cells, which lack functional STING \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eG\u003cb\u003e)\u003c/b\u003e. Consistently, the expression of canonical STING target genes, including IFN-β, IFN-γ, CXCL10, ISG15, and IRF7, were increased in RMI2 knockdown cells, whereas their expression in RMI2-rescued cells remained comparable to that in control cells \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eH, \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eI and S7D\u003cb\u003e)\u003c/b\u003e. To directly assess whether the STING pathway mediates enhanced IFN signaling induced by RMI2 depletion, we generated double-knockout cells lacking both RMI2 and STING in SW480 and CT26 models. As expected, STING knockout completely abolished RMI2 knockdown-induced IFN-β secretion, as well as the upregulation of STING target genes, including IFN-β, IFN-γ, CXCL10, ISG15, and IRF7, in both cell lines \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eJ, \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eK and S7E\u003cb\u003e)\u003c/b\u003e. Furthermore, STING deficiency abrogated the inhibitory effect of RMI2 knockout on tumor growth in CT26 cells \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eL-N\u003cb\u003e)\u003c/b\u003e. The proportions of effector T cells, including GZMB⁺ CD8⁺ T cells and IFN-γ⁺ CD8⁺ T cells, differed significantly between the sgRMI2 and double-knockout groups, suggesting that the tumor microenvironment in the double-knockout group exhibited a more immunosuppressive phenotype \u003cb\u003e(Fig. S7F and S7G)\u003c/b\u003e. In addition, overexpression of RAD51 attenuated RMI2 knockdown-induced γH2AX foci formation and reduced the expression of STING target genes, including IFN-β, IFN-γ, CXCL10, ISG15, and IRF7 \u003cb\u003e(Fig. S8A-C)\u003c/b\u003e. Collectively, these findings demonstrate that RMI2 suppresses cytosolic DNA sensing and IFN signaling in cancer cells by inhibiting the cGAS-STING pathway, and that STING is required for the full immunostimulatory effects of RMI2 loss.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec27\" class=\"Section3\"\u003e \u003ch2\u003eRMI2 deficiency sensitizes the tumor response to anti-PD-1 therapy\u003c/h2\u003e \u003cp\u003eWe further investigated whether RMI2 deficiency enhanced the therapeutic efficacy of anti-PD-1 blockade. In the CT26 orthotopic tumor model, RMI2-deficient tumors exhibited significantly reduced tumor volumes and prolonged survival following treatment with anti-PD-1 antibodies \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA and \u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB\u003cb\u003e)\u003c/b\u003e. Similarly, in subcutaneous tumor models established with either MC38 or CT26 cells, RMI2-deficient tumors treated with anti-PD-1 antibodies showed markedly suppressed tumor growth and reduced tumor weight \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eD, \u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eE, S9A and S9B\u003cb\u003e)\u003c/b\u003e. Flow cytometry and histological analysis revealed that the proportions of key immune cell populations, including NK cells, CD4⁺ T cells, and CD8⁺ T cells, were significantly increased in RMI2-deficient tumors upon anti-PD-1 treatment \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eF and S9C-E\u003cb\u003e)\u003c/b\u003e. Furthermore, the frequency of effector CD8⁺ T cells expressing GZMB, IFN-γ, and Ki67 was significantly increased elevated in RMI2-deficient tumors following anti-PD-1 therapy \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eG and S9F-H\u003cb\u003e)\u003c/b\u003e. IHC staining confirmed that combined RMI2 deficiency and anti-PD-1 treatment led to increased expression of GZMB, and IFN-γ in tumor tissues \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eH\u003cb\u003e)\u003c/b\u003e. Consistent with these findings, quantitative PCR analysis revealed significant upregulation of TNF-α, GZMB, and IFN-γ mRNA levels in RMI2-deficient tumors treated with anti-PD-1 antibodies \u003cb\u003e(Fig. S9I)\u003c/b\u003e. Together, these results demonstrate that RMI2 deficiency enhances the responsiveness of colorectal cancer to anti-PD-1 immunotherapy, suggesting that targeting RMI2 may represent a promising strategy to improve the efficacy of checkpoint blockade.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec28\" class=\"Section2\"\u003e \u003ch2\u003eRMI2 deficiency sensitizes the tumor response to chemotherapy\u003c/h2\u003e \u003cp\u003eGiven that both 5-fluorouracil (5-FU) and ionizing radiation (IR) induce DNA damage\u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e, we investigated whether RMI2 deficiency could enhance the efficacy of these treatments. Notably, western blot and immunofluorescence analyses revealed that RMI2-deficient cells exhibited increased expression and formation of γH2AX foci, and the cGAS-STING signaling pathway upon treatment with either IR or 5-FU \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eA, S10A and S10B\u003cb\u003e).\u003c/b\u003e Consistent with this, the formation of γH2AX foci and the proportion of cytosolic DNA positivity were higher in RMI2 deficiency cells treated with 5-FU \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eB, \u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eC and S10C\u003cb\u003e)\u003c/b\u003e. In addition, 5-FU treatment induced a marked increase in IFN-β secretion in RMI2-deficient cells \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eD and S10D\u003cb\u003e)\u003c/b\u003e. The mRNA levels of key interferon-stimulated genes (ISGs), including IFN-β, IFN-γ, CXCL10, ISG15, and IRF7, were also significantly upregulated following 5-FU treatment, with effects further potentiated by RMI2 loss \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eE and S10E\u003cb\u003e)\u003c/b\u003e. To determine whether RMI2 deficiency enhances the sensitivity of CRC tumors to 5-FU chemotherapy, we evaluated tumor growth in subcutaneous models using CT26 cells. RMI2-deficient tumors exhibited significantly reduced growth and tumor weight following 5-FU treatment \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eF and \u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eG\u003cb\u003e)\u003c/b\u003e. Importantly, the abundance of infiltrating immune cells, including NK cells, CD4⁺ T cells, and CD8⁺ T cells, was significantly higher in RMI2-deficient tumors treated with 5-FU \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eH and \u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eI\u003cb\u003e)\u003c/b\u003e. Furthermore, the proportions of effector CD8⁺ T cells expressing GZMB, IFN-γ, and Ki67 were also significantly increased \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eJ\u003cb\u003e)\u003c/b\u003e. IHC validated that RMI2-deficient tumors treated with 5-FU displayed elevated expression of GZMB, and IFN-γ \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eK and \u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eL\u003cb\u003e)\u003c/b\u003e. Consistent with these findings, the mRNA levels of TNF-α, GZMB, and IFN-γ were also increased in these tumors \u003cb\u003e(Fig. S10F)\u003c/b\u003e. In a clinical cohort receiving neoadjuvant 5-FU-based chemotherapy, low RMI2 expression correlated with higher treatment response rates, an inverse association between RMI2 negativity and CD8\u003csup\u003e+\u003c/sup\u003e T-cell infiltration, and significantly prolonged overall survival \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eM-O, and S10G\u003cb\u003e)\u003c/b\u003e. Collectively, these findings demonstrate that RMI2 deficiency sensitizes CRC tumors to 5-FU chemotherapy and enhances the associated antitumor immune response, suggesting its potential as a therapeutic target to improve chemotherapeutic efficacy and immune activation.\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eTumor cells employ multiple strategies to evade antitumor immunity, including the suppression of cytosolic nucleic acid sensing pathways\u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e, \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e. In this study, we demonstrate that RMI2 deficiency disrupts homologous recombination-mediated DNA repair, leads to the accumulation of cytosolic double-stranded DNA (dsDNA), and activates intrinsic antitumor and immune signaling pathways \u003cb\u003e(Fig. S11)\u003c/b\u003e. Collectively, these alterations compromise the genomic stability and enhance the immunogenicity of tumor cells. Notably, these tumor-intrinsic changes significantly improve the efficacy of ICB therapy, particularly in immunologically cold tumors such as MSS CRC.\u003c/p\u003e \u003cp\u003eThrough comprehensive analysis of clinical data, we identified RMI2 expression as an independent prognostic biomarker in patients with MSS CRC patients receiving immune checkpoint inhibitors. High RMI2 expression was correlated with poorer therapeutic responses and significantly shorter progression-free survival (PFS; HR\u0026thinsp;=\u0026thinsp;3.045, 95% CI: 1.569\u0026ndash;5.909; P\u0026thinsp;=\u0026thinsp;0.0012). Specifically, patients with low RMI2 expression exhibited a median PFS nearly threefold longer than that of patients with high RMI2 expression (10 months vs. 2.7 months). These findings highlight RMI2 as a potential predictive biomarker for ICI response and underscore its role in shaping the tumor immune microenvironment.\u003c/p\u003e \u003cp\u003eA key factor contributing to the limited efficacy of immunotherapy in cold tumors is the lack of an inflamed, immunogenic tumor microenvironment, which is often characterized by reduced infiltration of cytotoxic CD8⁺ T cells and defective type I interferon signaling\u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e. Our data from immunocompetent mouse models reveal that RMI2 deficiency not only inhibits tumor growth but also enhances the recruitment and activation of multiple immune cell populations, most notably CD8⁺ T cells. This shift significantly sensitizes CRC tumors to anti-PD-1 therapy. Mechanistically, RMI2 loss activates the cGAS-STING pathway, leading to increased phosphorylation of STING, TBK1, and IRF3; elevated secretion of IFN-β, and upregulation of downstream interferon-stimulated genes (ISGs), including ISG15, CXCL10, and IRF7. Furthermore, RMI2 depletion selectively enhanced the infiltration of CD8⁺ T cells into the tumor microenvironment, although no significant increase was observed in macrophages. These findings indicate that RMI2 deficiency transforms the tumor immune landscape from immunosuppressive to immunostimulatory.\u003c/p\u003e \u003cp\u003eIn addition to its effects in immunotherapy, we found that RMI2 deficiency potentiates the therapeutic efficacy of conventional DNA-damaging agents. In models treated with IR or 5-FU, RMI2-deficient cells exhibited elevated γH2AX foci formation, which is indicative of increased DNA damage and impaired repair. Consistent with these observations, RMI2 loss enhanced activation of the cGAS-STING pathway following IR or 5-FU treatment, as evidenced by increased STING pathway phosphorylation and IFN-β production. In the context of 5-FU-based chemotherapy, RMI2-deficient tumors demonstrated significantly reduced growth and tumor weight, accompanied by increased infiltration of cytotoxic immune cells, including NK cells, CD4⁺ T cells, and CD8⁺ T cells. Among CD8⁺ T cells, the proportion of effector cells expressing GZMB, and IFN-γ was also significantly elevated. Immunohistochemical analysis further confirmed the increased expression of GZMB and IFN-γ in RMI2-deficient, 5-FU-treated tumors, along with upregulated tumor mRNA levels of TNF-α, GZMB, and IFN-γ. Clinically, in a cohort of patients receiving neoadjuvant 5-FU-based chemotherapy, low RMI2 expression is associated with improved response rates and significantly prolonged overall survival. These data suggest that RMI2 deficiency not only enhances tumor cell intrinsic susceptibility to DNA-damaging agents but also augments antitumor immunity, thereby synergizing with conventional chemotherapy.\u003c/p\u003e \u003cp\u003eGiven the essential role of RMI2 in maintaining genome stability across various cellular compartments\u003csup\u003e\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e, \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e, and its overexpression in multiple cancer types, including CRC, where high RMI2 expression correlates with advanced tumor stage and poor prognosis, we emphasize the need to carefully evaluate the systemic effects and safety profile of RMI2-targeted strategies\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. Notably, RMI2 exhibits relatively low expression in normal tissues, suggesting therapeutically exploitable vulnerability. This differential expression pattern provides a tumor-selective therapeutic window in which RMI2 inhibition preferentially targets rapidly proliferating cancer cells while sparing normal tissues. Mechanistically, RMI2 depletion exerts a dual therapeutic effect: (1) it induces DNA damage and impairs homologous recombination, leading to oncogene addiction-dependent cytotoxicity; and (2) it activates the cGAS-STING pathway, enhances type I interferon signaling and promotes the recruitment of cytotoxic immune cells such as NK cells and CD8⁺ T cells. This dual mechanism not only drives direct tumor cell killing but also synergizes with immune checkpoint blockade to enhance antitumor immunity. Recent research has demonstrated that genomic instability can effectively enhance the efficacy of immunotherapy in MSS colon cancer\u003csup\u003e\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e. This dual-action paradigm represents a promising strategy for improving response rates in immunologically cold tumors and warrants further exploration in combinatorial therapeutic regimens.\u003c/p\u003e \u003cp\u003eHowever, several important questions remain. The precise dose-dependent effects of RMI2 modulation on immune activation, identification of predictive biomarkers for patient stratification, and assessment of potential combinatorial toxicities in broader preclinical models require further systematic investigation. Future studies should also explore the therapeutic potential of combining RMI2-targeted approaches with additional immunotherapies, cytotoxic agents, or STING agonists to maximize the clinical benefit.\u003c/p\u003e \u003cp\u003eIn summary, our study reveals that RMI2 acts as a critical negative regulator of cytosolic DNA sensing and repair. Genetic perturbation of RMI2 disrupts genomic integrity while simultaneously activating innate immune responses, thereby converting immunologically cold tumors into immunologically active (\u0026ldquo;hot\u0026rdquo;) tumors. These tumor-intrinsic changes significantly enhance the responsiveness to immune checkpoint blockade and chemotherapy. Collectively, our findings suggest that RMI2 is a novel and promising therapeutic target for improving the efficacy of immunotherapy in MSS CRC and other immunologically cold malignancies.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eFundings\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the National Science Fund for Distinguished Young Scholars (82425045), National Nature Science Foundation in China (NSFC) (82272800, 82102952,\u0026nbsp;82403266 and 82103080), Noncommunicable Chronic Diseases of National Science and Technology Major Project (2024ZD0520304), and the CSCO-Tongshu Gene Tumor Research Foundation Youth Project (No. Y-tongshu2021/qn-0303).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of interest\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing financial interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll human tissue studies were approved by the Institutional Review Board of The Sixth Affiliated Hospital, Sun Yat-sen University (2022ZSLYEC-002).\u003c/p\u003e\n\u003cp\u003eAll animal experiments were performed in accordance with the Guidelines for the Care and Use of Laboratory Animals and approved by the Animal Research Committee of The Sixth Hospital of Sun Yat-sen University (SYSU-IACUC-2020112701).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe RMI2 gene expression dataset and corresponding clinical survival information were retrieved from the Kaplan-Meier Plotter database (https://kmplot.com). Patients were stratified into high- and low-expression cohorts based on the median RMI2 expression levels, and overall survival differences were analyzed using Kaplan-Meier survival curves with log-rank testing.\u003c/p\u003e\n\u003cp\u003eRaw and processed sc-RNA data have been deposited with the Gene Expression Omnibus (GEO accession GSE 308382).\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eWang S, Zheng R, Li J, Zeng H, Li L, Chen R, \u003cem\u003eet al.\u003c/em\u003e Global, regional, and national lifetime risks of developing and dying from gastrointestinal cancers in 185 countries: a population-based systematic analysis of GLOBOCAN. \u003cem\u003eThe lancet Gastroenterology \u0026amp; hepatology\u003c/em\u003e 2024, 9(3): 229\u0026ndash;237.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAndr\u0026eacute; T, Shiu KK, Kim TW, Jensen BV, Jensen LH, Punt C, \u003cem\u003eet al.\u003c/em\u003e Pembrolizumab in Microsatellite-Instability-High Advanced Colorectal Cancer. \u003cem\u003eThe New England journal of medicine\u003c/em\u003e 2020, 383(23): 2207\u0026ndash;2218.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHu H, Kang L, Zhang J, Wu Z, Wang H, Huang M, \u003cem\u003eet al.\u003c/em\u003e Neoadjuvant PD-1 blockade with toripalimab, with or without celecoxib, in mismatch repair-deficient or microsatellite instability-high, locally advanced, colorectal cancer (PICC): a single-centre, parallel-group, non-comparative, randomised, phase 2 trial. \u003cem\u003eThe lancet Gastroenterology \u0026amp; hepatology\u003c/em\u003e 2022, 7(1): 38\u0026ndash;48.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePelka K, Hofree M, Chen JH, Sarkizova S, Pirl JD, Jorgji V, \u003cem\u003eet al.\u003c/em\u003e Spatially organized multicellular immune hubs in human colorectal cancer. \u003cem\u003eCell\u003c/em\u003e 2021, 184(18): 4734\u0026ndash;4752.e4720.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRoerden M, Spranger S. Cancer immune evasion, immunoediting and intratumour heterogeneity. \u003cem\u003eNature reviews Immunology\u003c/em\u003e 2025.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWestcott PMK, Sacks NJ, Schenkel JM, Ely ZA, Smith O, Hauck H, \u003cem\u003eet al.\u003c/em\u003e Low neoantigen expression and poor T-cell priming underlie early immune escape in colorectal cancer. \u003cem\u003eNature cancer\u003c/em\u003e 2021, 2(10): 1071\u0026ndash;1085.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOchoa de Olza M, Navarro Rodrigo B, Zimmermann S, Coukos G. Turning up the heat on non-immunoreactive tumours: opportunities for clinical development. \u003cem\u003eThe Lancet Oncology\u003c/em\u003e 2020, 21(9): e419-e430.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXiao D, Dong S, Yang S, Liu Z. CKS2 and RMI2 are two prognostic biomarkers of lung adenocarcinoma. \u003cem\u003ePeerJ\u003c/em\u003e 2020, 8: e10126.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZheng Ms B, Wang Ms H, Wang Ms JX, Liu Ms ZH, Zhang Md P, Zhang Md D. The Clinical Significance of RMI2 in Hepatocellular Carcinoma. \u003cem\u003eTechnology in cancer research \u0026amp; treatment\u003c/em\u003e 2021, 20: 15330338211045496.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWei W, Ying X, Chen L, Sun Q, Lu X, Xia Y, \u003cem\u003eet al.\u003c/em\u003e RecQ mediated genome instability 2 (RMI2): a potential prognostic and immunological biomarker for pan-cancers. \u003cem\u003eAging\u003c/em\u003e 2022, 14(9): 4107\u0026ndash;4136.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang L, Hao C, Han B, Zeng G, Han L, Cao C, \u003cem\u003eet al.\u003c/em\u003e RMI2 is a novel prognostic and predictive biomarker for breast cancer. \u003cem\u003eCancer medicine\u003c/em\u003e 2023, 12(7): 8331\u0026ndash;8350.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhan W, Liu Y, Gao Y, Gong R, Wang W, Zhang R, \u003cem\u003eet al.\u003c/em\u003e RMI2 plays crucial roles in growth and metastasis of lung cancer. \u003cem\u003eSignal transduction and targeted therapy\u003c/em\u003e 2020, 5(1): 188.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eConcannon K, Morris BB, Gay CM, Byers LA. Combining targeted DNA repair inhibition and immune-oncology approaches for enhanced tumor control. \u003cem\u003eMolecular cell\u003c/em\u003e 2023, 83(5): 660\u0026ndash;680.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang D, Elenbaas B, Murugesan K, Shah K, Montesion M, Gounaris I, \u003cem\u003eet al.\u003c/em\u003e Relationship among DDR gene mutations, TMB and PD-L1 in solid tumour genomes identified using clinically actionable biomarker assays. \u003cem\u003eNPJ precision oncology\u003c/em\u003e 2023, 7(1): 103.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGermano G, Lamba S, Rospo G, Barault L, Magr\u0026igrave; A, Maione F, \u003cem\u003eet al.\u003c/em\u003e Inactivation of DNA repair triggers neoantigen generation and impairs tumour growth. \u003cem\u003eNature\u003c/em\u003e 2017, 552(7683): 116\u0026ndash;120.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBen-David U, Amon A. Context is everything: aneuploidy in cancer. \u003cem\u003eNature reviews Genetics\u003c/em\u003e 2020, 21(1): 44\u0026ndash;62.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKuang X, Li J. Chromosome instability and aneuploidy as context-dependent activators or inhibitors of antitumor immunity. \u003cem\u003eFrontiers in immunology\u003c/em\u003e 2022, 13: 895961.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTubbs A, Nussenzweig A. Endogenous DNA Damage as a Source of Genomic Instability in Cancer. \u003cem\u003eCell\u003c/em\u003e 2017, 168(4): 644\u0026ndash;656.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang Z, Zhao J, Wang G, Zhang F, Zhang Z, Zhang F, \u003cem\u003eet al.\u003c/em\u003e Comutations in DNA Damage Response Pathways Serve as Potential Biomarkers for Immune Checkpoint Blockade. \u003cem\u003eCancer research\u003c/em\u003e 2018, 78(22): 6486\u0026ndash;6496.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKeenan TE, Burke KP, Van Allen EM. Genomic correlates of response to immune checkpoint blockade. \u003cem\u003eNature medicine\u003c/em\u003e 2019, 25(3): 389\u0026ndash;402.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang B, Xu P, Ablasser A. Regulation of the cGAS-STING Pathway. \u003cem\u003eAnnual review of immunology\u003c/em\u003e 2025.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen Q, Sun L, Chen ZJ. Regulation and function of the cGAS-STING pathway of cytosolic DNA sensing. \u003cem\u003eNature immunology\u003c/em\u003e 2016, 17(10): 1142\u0026ndash;1149.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFalahat R, Berglund A, Perez-Villarroel P, Putney RM, Hamaidi I, Kim S, \u003cem\u003eet al.\u003c/em\u003e Epigenetic state determines the in vivo efficacy of STING agonist therapy. \u003cem\u003eNature communications\u003c/em\u003e 2023, 14(1): 1573.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKonno H, Yamauchi S, Berglund A, Putney RM, Mul\u0026eacute; JJ, Barber GN. Suppression of STING signaling through epigenetic silencing and missense mutation impedes DNA damage mediated cytokine production. \u003cem\u003eOncogene\u003c/em\u003e 2018, 37(15): 2037\u0026ndash;2051.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShang G, Zhang C, Chen ZJ, Bai XC, Zhang X. Cryo-EM structures of STING reveal its mechanism of activation by cyclic GMP-AMP. \u003cem\u003eNature\u003c/em\u003e 2019, 567(7748): 389\u0026ndash;393.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eErgun SL, Fernandez D, Weiss TM, Li L. STING Polymer Structure Reveals Mechanisms for Activation, Hyperactivation, and Inhibition. \u003cem\u003eCell\u003c/em\u003e 2019, 178(2): 290\u0026ndash;301.e210.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang C, Shang G, Gui X, Zhang X, Bai XC, Chen ZJ. Structural basis of STING binding with and phosphorylation by TBK1. \u003cem\u003eNature\u003c/em\u003e 2019, 567(7748): 394\u0026ndash;398.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLu C, Guan J, Lu S, Jin Q, Rousseau B, Lu T, \u003cem\u003eet al.\u003c/em\u003e DNA Sensing in Mismatch Repair-Deficient Tumor Cells Is Essential for Anti-tumor Immunity. \u003cem\u003eCancer cell\u003c/em\u003e 2021, 39(1): 96\u0026ndash;108.e106.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuan J, Lu C, Jin Q, Lu H, Chen X, Tian L, \u003cem\u003eet al.\u003c/em\u003e MLH1 Deficiency-Triggered DNA Hyperexcision by Exonuclease 1 Activates the cGAS-STING Pathway. \u003cem\u003eCancer cell\u003c/em\u003e 2021, 39(1): 109\u0026ndash;121.e105.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKaneta A, Nakajima S, Okayama H, Matsumoto T, Saito K, Kikuchi T, \u003cem\u003eet al.\u003c/em\u003e Role of the cGAS-STING pathway in regulating the tumor-immune microenvironment in dMMR/MSI colorectal cancer. \u003cem\u003eCancer immunology, immunotherapy: CII\u003c/em\u003e 2022, 71(11): 2765\u0026ndash;2776.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWei B, Xu L, Guo W, Wang Y, Wu J, Li X, \u003cem\u003eet al.\u003c/em\u003e SHP2-Mediated Inhibition of DNA Repair Contributes to cGAS-STING Activation and Chemotherapeutic Sensitivity in Colon Cancer. \u003cem\u003eCancer research\u003c/em\u003e 2021, 81(12): 3215\u0026ndash;3228.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDuan J, Chen T, Li Q, Zhang Y, Lu T, Xue J, \u003cem\u003eet al.\u003c/em\u003e Protein arginine methyltransferase 6 enhances immune checkpoint blockade efficacy via the STING pathway in MMR-proficient colorectal cancer. \u003cem\u003eJournal for immunotherapy of cancer\u003c/em\u003e 2025, 13(3).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSaha S, Ghosh M, Li J, Wen A, Galluzzi L, Martinez LA, \u003cem\u003eet al.\u003c/em\u003e Serine Depletion Promotes Antitumor Immunity by Activating Mitochondrial DNA-Mediated cGAS-STING Signaling. \u003cem\u003eCancer research\u003c/em\u003e 2024, 84(16): 2645\u0026ndash;2659.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRizvi NA, Hellmann MD, Snyder A, Kvistborg P, Makarov V, Havel JJ, \u003cem\u003eet al.\u003c/em\u003e Cancer immunology. Mutational landscape determines sensitivity to PD-1 blockade in non-small cell lung cancer. \u003cem\u003eScience (New York, NY)\u003c/em\u003e 2015, 348(6230): 124\u0026ndash;128.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchumacher TN, Schreiber RD. Neoantigens in cancer immunotherapy. \u003cem\u003eScience (New York, NY)\u003c/em\u003e 2015, 348(6230): 69\u0026ndash;74.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChin KV, Ueda K, Pastan I, Gottesman MM. Modulation of activity of the promoter of the human MDR1 gene by Ras and p53. \u003cem\u003eScience (New York, NY)\u003c/em\u003e 1992, 255(5043): 459\u0026ndash;462.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrannon AR, Vakiani E, Sylvester BE, Scott SN, McDermott G, Shah RH, \u003cem\u003eet al.\u003c/em\u003e Comparative sequencing analysis reveals high genomic concordance between matched primary and metastatic colorectal cancer lesions. \u003cem\u003eGenome biology\u003c/em\u003e 2014, 15(8): 454.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDaboussi F, Dumay A, Delac\u0026ocirc;te F, Lopez BS. DNA double-strand break repair signalling: the case of RAD51 post-translational regulation. \u003cem\u003eCellular signalling\u003c/em\u003e 2002, 14(12): 969\u0026ndash;975.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLord CJ, Ashworth A. The DNA damage response and cancer therapy. \u003cem\u003eNature\u003c/em\u003e 2012, 481(7381): 287\u0026ndash;294.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eH\u0026auml;rtlova A, Erttmann SF, Raffi FA, Schmalz AM, Resch U, Anugula S, \u003cem\u003eet al.\u003c/em\u003e DNA damage primes the type I interferon system via the cytosolic DNA sensor STING to promote anti-microbial innate immunity. \u003cem\u003eImmunity\u003c/em\u003e 2015, 42(2): 332\u0026ndash;343.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShah MA, Schwartz GK. Cell cycle-mediated drug resistance: an emerging concept in cancer therapy. \u003cem\u003eClinical cancer research: an official journal of the American Association for Cancer Research\u003c/em\u003e 2001, 7(8): 2168\u0026ndash;2181.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSharma RA, Plummer R, Stock JK, Greenhalgh TA, Ataman O, Kelly S, \u003cem\u003eet al.\u003c/em\u003e Clinical development of new drug-radiotherapy combinations. \u003cem\u003eNature reviews Clinical oncology\u003c/em\u003e 2016, 13(10): 627\u0026ndash;642.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZou SS, Qiao Y, Zhu S, Gao B, Yang N, Liu YJ, \u003cem\u003eet al.\u003c/em\u003e Intrinsic strategies for the evasion of cGAS-STING signaling-mediated immune surveillance in human cancer: How therapy can overcome them. \u003cem\u003ePharmacological research\u003c/em\u003e 2021, 166: 105514.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRoulois D, Loo Yau H, Singhania R, Wang Y, Danesh A, Shen SY, \u003cem\u003eet al.\u003c/em\u003e DNA-Demethylating Agents Target Colorectal Cancer Cells by Inducing Viral Mimicry by Endogenous Transcripts. \u003cem\u003eCell\u003c/em\u003e 2015, 162(5): 961\u0026ndash;973.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJhunjhunwala S, Hammer C, Delamarre L. Antigen presentation in cancer: insights into tumour immunogenicity and immune evasion. \u003cem\u003eNature reviews Cancer\u003c/em\u003e 2021, 21(5): 298\u0026ndash;312.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSingh TR, Ali AM, Busygina V, Raynard S, Fan Q, Du CH, \u003cem\u003eet al.\u003c/em\u003e BLAP18/RMI2, a novel OB-fold-containing protein, is an essential component of the Bloom helicase-double Holliday junction dissolvasome. \u003cem\u003eGenes \u0026amp; development\u003c/em\u003e 2008, 22(20): 2856\u0026ndash;2868.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXu D, Guo R, Sobeck A, Bachrati CZ, Yang J, Enomoto T, \u003cem\u003eet al.\u003c/em\u003e RMI, a new OB-fold complex essential for Bloom syndrome protein to maintain genome stability. \u003cem\u003eGenes \u0026amp; development\u003c/em\u003e 2008, 22(20): 2843\u0026ndash;2855.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTan PB, Verschoor YL, van den Berg JG, Balduzzi S, Kok NFM, Ijsselsteijn ME, \u003cem\u003eet al.\u003c/em\u003e Neoadjuvant immunotherapy in mismatch-repair-proficient colon cancers. \u003cem\u003eNature\u003c/em\u003e 2025.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"cell-death-and-disease","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"cddis","sideBox":"Learn more about [Cell Death \u0026 Disease](http://www.nature.com/cddis/)","snPcode":"41419","submissionUrl":"https://mts-cddis.nature.com/cgi-bin/main.plex","title":"Cell Death \u0026 Disease","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"RMI2, cGAS-STING, colorectal cancer, DNA damage, tumor microenvironment","lastPublishedDoi":"10.21203/rs.3.rs-8706982/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8706982/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eMicrosatellite-stable (MSS) colorectal cancers (CRCs) are largely refractory to immune checkpoint inhibitors (ICIs) due to immunologically \"cold\" tumor microenvironments (TMEs) characterized by limited T-cell infiltration. While RecQ-mediated genome instability 2 (RMI2) is known for regulating DNA repair, its role in anti-tumor immunity remains unclear. Here, we demonstrate that RMI2 acts as an adaptive oncoprotein in CRC by suppressing innate immune activation through enhanced homologous recombination repair (HRR). Mechanistically, RMI2 stabilizes BRCA1-RAD51 complexes, accelerates DNA double-strand break repair, and limits cytosolic DNA release. Conversely, RMI2 deficiency impairs HRR, causing cytosolic DNA accumulation, cGAS-STING pathway activation, and type I interferon signaling that boosts anti-tumor immunity. Notably, RMI2 knockout in mice synergizes with PD-1 blockade and fluorouracil to induce robust tumor regression and prolonged survival. These findings uncover a previously unrecognized role for RMI2 in maintaining immune evasion through the coordinated regulation of DNA repair and cGAS-STING-dependent innate immune signaling, positioning RMI2 as a promising therapeutic target to convert MSS CRCs from \"cold\" to \"hot\" tumors and overcome ICI resistance.\u003c/p\u003e","manuscriptTitle":"RMI2 Depletion Recovers cGAS-STING Signaling to Enhance Immunotherapy in Colorectal Cancer","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-19 11:59:06","doi":"10.21203/rs.3.rs-8706982/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"revise","date":"2026-03-10T09:21:47+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"This content is not available.","date":"2026-03-08T05:12:03+00:00","index":2,"fulltext":"This content is not available."},{"type":"editorInvitedReview","content":"This content is not available.","date":"2026-03-01T13:24:15+00:00","index":1,"fulltext":"This content is not available."},{"type":"reviewerAgreed","content":"This content is not available.","date":"2026-02-17T01:46:09+00:00","index":2,"fulltext":"This content is not available."},{"type":"reviewerAgreed","content":"This content is not available.","date":"2026-02-16T16:53:21+00:00","index":1,"fulltext":"This content is not available."},{"type":"reviewersInvited","content":"","date":"2026-02-13T13:53:00+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-01-28T11:55:52+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-01-27T07:14:43+00:00","index":"","fulltext":""},{"type":"submitted","content":"Cell Death \u0026 Disease","date":"2026-01-27T07:14:42+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"cell-death-and-disease","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"cddis","sideBox":"Learn more about [Cell Death \u0026 Disease](http://www.nature.com/cddis/)","snPcode":"41419","submissionUrl":"https://mts-cddis.nature.com/cgi-bin/main.plex","title":"Cell Death \u0026 Disease","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"97e87e4a-3bf2-411d-b283-277b56c78253","owner":[],"postedDate":"February 19th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"in-revision","subjectAreas":[{"id":61894923,"name":"Biological sciences/Cancer/Cancer microenvironment"},{"id":61894924,"name":"Biological sciences/Cell biology/Cell death"}],"tags":[],"updatedAt":"2026-03-10T09:32:53+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-19 11:59:06","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8706982","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8706982","identity":"rs-8706982","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}