Targeting MORC2 activates transposable element-mediated viral mimicry and potentiates immune checkpoint blockade in triple-negative breast cancer Authors: Fang-Lin Zhang, Shao-Ying Yang, Yin-Ling Zhang, Qian Zhao, Wen-Xiao Yang, Lisa Andriani, Jia-Yang Cai, Min-Ying Huang, Xin Hu, Zhi-Min Shao, A-Yong Cao, Da-Qiang Li Published: 2026 ABSTRACT AbstractImmune checkpoint blockade (ICB) has shown limited efficacy in triple-negative breast cancer (TNBC), highlighting the need to elucidate mechanisms of immune evasion and identify novel therapeutic targets. Here, we identify MORC family CW-type zinc finger 2 (MORC2), an ATP-dependent chromatin remodeler, as a key epigenetic suppressor of antitumor immunity in TNBC. MORC2 is significantly upregulated in TNBC and correlates with an immunosuppressive tumor microenvironment and poor response to ICB. Genetic ablation of MORC2 inhibited tumor growth in immunocompetent but not immunodeficient mice, accompanied by enhanced CD8+ T cell cytotoxicity and M1-like macrophage polarization. Mechanistically, MORC2 cooperated with the SET domain bifurcated histone lysine methyltransferase 1 (SETDB1) to deposit histone H3 lysine 9 trimethylation (H3K9me3) repressive marks at transposable elements (TEs), thereby silencing their expression and suppressing viral mimicry through inhibition of endogenous nucleic acid-sensing pathways and interferon responses. Furthermore, SETDB1 methylated MORC2 at residues K234 and K643 to enhance its stability, thus establishing a positive feedback loop that reinforces epigenetic silencing. Importantly, therapeutic targeting of MORC2 using antisense oligonucleotides (ASOs) synergized with anti-PD-1 therapy to suppress tumor growth in mouse TNBC models. Clinically, MORC2 expression inversely correlated with CD8+ T cell infiltration and activation in human TNBC samples. Collectively, our findings establish MORC2 as a novel epigenetic immune checkpoint and highlight its therapeutic potential to overcome ICB resistance in TNBC.Graphical Abstract Supplementary InformationThe online version contains supplementary material available at 10.1186/s12943-026-02679-6.Keywords: Triple-negative breast cancer, Immunotherapy, Chromatin remodeling, Transcriptional repression, Tumor immune microenvironment Significance Resistance to ICB poses a major clinical challenge in TNBC. This study identifies MORC2 as a key epigenetic suppressor of tumor immunogenicity by silencing transposable elements and suppressing interferon signaling, and highlights MORC2 as a promising therapeutic target to overcome ICB resistance in TNBC. Introduction Triple-negative breast cancer (TNBC), defined by the absence of estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2) expression, is the most aggressive subtype of breast cancer [1]. It is characterized by a higher incidence in younger women, a pronounced tendency for early recurrence and distant metastasis, and consequently, significantly worse survival outcomes compared to other subtypes [1, 2]. This subtype of breast cancer accounts for approximately 15–20% of all breast cancer cases. These tumors are generally unresponsive to conventional endocrine and HER2-targeted therapies, leading to a lack of specific therapeutic options. As a result, chemotherapy remains the cornerstone of systemic therapy for both early-stage and advanced TNBC patients [3]. Nevertheless, its therapeutic efficacy is unsatisfactory, yielding suboptimal outcomes with a median survival of 9–12 months and a five-year survival rate below 30%. This critical situation underscores an unmet medical need and highlights the imperative to develop more effective and optimized combination therapeutic strategies. In recent years, advances in immunotherapy have brought a paradigm shift in the treatment of TNBC, largely exemplified by the clinical introduction of immune checkpoint blockade (ICB) targeting the programmed cell death protein 1/programmed cell death ligand 1 (PD-1/PD-L1) axis and cytotoxic T-lymphocyte protein 4 (CTLA-4) [3]. Excitingly, multiple recent landmark clinical trials, such as IMpassion031 [4], IMpassion130 [5], KEYNOTE-355 [6], KEYNOTE-522 [7, 8], CamRelief [9], and TORCHLIGHT [10], have demonstrated that adding anti-PD-1 or PD-L1 monoclonal antibodies to conventional chemotherapy significantly improves pathological complete response rates and prolongs survival in patients with TNBC. However, only a small subset (approximately 20–30%) of patients with TNBC benefits from ICB therapy [11]. Its efficacy is substantially impeded by low response rates, frequent drug resistance, the complex tumor immune microenvironment, and the absence of reliable predictive biomarkers [12]. To address these challenges, it is imperative to decipher the mechanisms of therapeutic resistance, identify predictive biomarkers for precise patient stratification, and discover novel therapeutic targets to enhance ICB efficacy in TNBC. In eukaryotic cells, genomic DNA is wrapped around a histone octamer (composed of H2A, H2B, H3, and H4) to form a highly condensed structure called chromatin [13]. This compact organization creates a steric barrier that limits the access of DNA-binding proteins to DNA, thereby impeding essential DNA-templated processes such as gene transcription and DNA damage repair [14, 15]. To enable these functions, chromatin must undergo dynamic structural changes in a process known as chromatin remodeling [13, 15]. These changes are primarily carried out by ATP-dependent chromatin remodelers, which utilize energy from ATP hydrolysis to alter nucleosome positioning and overall chromatin architecture [13, 16]. Growing evidence shows that dysregulation of chromatin remodelers not only drives tumorigenesis and affects therapeutic responses but also creates direct vulnerabilities for therapeutic targeting [15, 17, 18]. Moreover, chromatin remodelers are emerging as pivotal players in shaping the tumor immune microenvironment and dictating immunotherapy efficacy [19–24]. Given the multitude and functional importance of chromatin remodelers, in-depth investigation of their functions and molecular mechanisms in antitumor immunity is pivotal for uncovering novel immunotherapy targets in TNBC. MORC family CW-type zinc finger 2 (MORC2) is a member of the microrchidia (MORC) family of nuclear proteins [25]. Protein structure analysis reveals that it contains a conserved ATPase domain at its N-terminus, a chromo-like domain at the C-terminus, and a CW-type zinc finger domain (CW-ZF) (also referred to as a derived plant homeodomain, PHD-X) in the middle region [25–28]. Based on the known functions of other proteins that contain these key domains, it is suggested that MORC2 may be involved in epigenetic regulation, particularly in chromatin remodeling. Indeed, our studies and others recently demonstrated that MORC2 is a novel ATP-dependent chromatin remodeler, and plays a crucial role in chromatin accessibility, gene transcription, and DNA repair [29–32]. In addition, our recent series of studies have shown that MORC2 is a novel oncoprotein that promotes breast cancer progression and confers resistance to endocrine therapy, chemotherapeutic drugs, and ionizing radiation [28, 29, 33–39]. Nevertheless, its functional role in cancer immunotherapy, especially in TNBC, remains largely unexplored. In this study, we demonstrate that MORC2 is upregulated in TNBC and functions as a critical epigenetic immune suppressor to drive tumor immune evasion. Mechanistically, MORC2 collaborates with the SET domain bifurcated histone lysine methyltransferase 1 (SETDB1) to establish a repressive chromatin state, silencing transposable elements (TEs) and suppressing the viral mimicry response. Antisense oligonucleotide (ASO)-mediated targeting of MORC2 synergizes with PD-1 blockade to promote tumor regression in TNBC mouse models. Collectively, our work provides the first evidence for targeting MORC2 to potentiate immunotherapy via activation of viral mimicry and highlights MORC2 as a promising therapeutic target for TNBC. Materials and methods Cell culture and reagents The human TNBC cell lines (BT-549 and MDA-MB-231), mouse breast cancer cell lines (4T1, 66 cl-4, and AT-3), and human embryonic kidney 293T (HEK293T) cell line were sourced from the Cell Bank of Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China) and Fudan University Shanghai Cancer Center (FUSCC). Upon receipt, the cells were expanded, cryopreserved in multiple aliquots, and used in experiments within six months. Authentication of all cell lines was performed using short tandem repeat (STR) profiling. Mycoplasma contamination was routinely monitored using a PCR-based assay [38]. Cell cultures were maintained in Dulbecco’s Modified Eagle Medium (DMEM; BasalMedia, #L110) supplemented with 10% fetal bovine serum (FBS; ExCell Biol, #FSP500) and 1% penicillin-streptomycin solution (BasalMedia, #S110B). The detailed information of the chemical reagents used in this study is provided in Supplementary Table S1. The multi-omics datasets and clinical samples Transcriptomic and proteomic datasets of all breast cancer types from the Chinese Breast Cancer Genome Atlas (CBCGA) Project cohort [40] and transcriptomic dataset of TNBC from the FUSCC-TNBC Project cohort [41] have been described previously. Data from The Cancer Genome Atlas (TCGA) [42] and the Molecular Taxonomy of Breast Cancer International Consortium (METABRIC) [43] were obtained from the Genomic Data Commons Data Portal (https://portal.gdc.cancer.gov/) and cBioPortal (http://cbioportal.org), respectively. Proteomic dataset of TNBC was sourced from the Clinical Proteomic Tumor Analysis Consortium (CPTAC) [44]. TNBC specimens and matched normal breast tissues were collected from patients undergoing surgery at FUSCC, who had not received any prior treatment. The study protocol was reviewed and approved by the FUSCC Ethics Committee (#NSFC-25-0303-92), and written informed consent was obtained from all participants prior to surgery. Cell proliferation and colony formation assays For cell viability assays using the IncuCyte system, cells were plated in 24-well plates. Following overnight incubation at 37 °C, cell density was monitored daily for 3 days using the IncuCyte software (Essen BioScience) to assess cell proliferation rate. For cell viability assessment using the CCK-8 assays, cells were seeded in 96-well plates. Following an appropriate incubation period, the cells were treated with CCK-8 solution (Yeasen, #40203ES92). For colony formation assays, cells were seeded at a density of 500 cells per well in 6-well plates, with each condition performed in triplicate. After 10 to 14 days of culture, colonies were fixed using methanol and subsequently stained with 0.2% crystal violet for visualization. Xenograft mouse models and treatment All in vivo experiments were conducted in strict accordance with the protocol approved by the Animal Experiments Committee of FUSCC (#FUSCC-IACUC-2024058). To assess tumorigenic potential, mycoplasma-free cells were inoculated into the mammary fat pads of 5- to 6-week-old female immunocompetent BALB/c or C57BL/6 mice, or immunodeficient BALB/c nude mice. All antisense oligonucleotides (ASOs) were designed and synthesized by RiboBio (Guangzhou, China). Briefly, the ASOs employ a 5-10-5 gapmer design with phosphorothioate backbone modifications, which mediates RNase H-dependent cleavage of the target Morc2a mRNA while inherently limiting off-target activity. Their knockdown efficiency was evaluated through in vitro delivery at varying concentrations. Targeting sequences for ASOs are outlined in Supplementary Table S2. The most effective ASO sequence was selected for in vivo delivery optimization and employed in subsequent in vivo studies. ASOs were administered via intratumoral (i.t.) injection at a dose of 5 nmol (dissolved in PBS) three times per week. To evaluate the response to immunotherapy, mice were treated with either an isotype control or an anti-PD-1 antibody (10 mg/kg) via intraperitoneal (i.p.) injection. Tumor volume was measured three times per week and calculated using the formula (Length × Width2)/2. Growth curves were subsequently plotted based on the derived volumes. The health status of the mice was closely monitored throughout the study. Humane endpoints were defined as a 10% body weight loss, the development of cachexia, or when the tumor’s maximum diameter exceeded 15 mm or its volume surpassed 2000 mm³. At the end of the experiment, mice were euthanized, and tumor tissues were collected for subsequent analysis. Hematoxylin-eosin (HE) staining was then performed by Servicebio (Shanghai, China) to confirm tumor histology. Guide RNAs (gRNAs), short hairpin RNAs (shRNAs), and DNA constructs Stable knockout (KO) cell lines were generated using lentivirus-mediated CRISPR-Cas9 technology. For each target gene, two guide RNA (gRNA) oligos were annealed and cloned into the BsmB1-digested lentiCRISPR-v2 vector (Addgene, #52961). Targeting sequences for sgRNAs are outlined in Supplementary Table S3. TNBC cells were transfected with the lentiCRISPR-v2 lentiviral vector encoding the respective sgRNAs, generating two biological replicates. After transfection, cells were selected with puromycin for 7–10 days. The resulting KO colonies were expanded and their KO status was confirmed by immunoblotting assays. For gene silencing, short hairpin RNAs (shRNAs) were designed and synthesized by Sangon Biotech (Shanghai, China). The shRNAs were annealed and cloned into the pLKO.1 lentiviral vector (Addgene, #10878). After sequence verification, lentiviral particles were packaged and transfected into target cells. Stable cell lines were then selected using puromycin or neomycin for 7–10 days as described previously [38]. The efficiency of gene silencing was confirmed by immunoblotting assays. Targeting sequences for shRNAs are presented in Supplementary Table S4. The cDNAs of target genes were amplified via PCR and subcloned into pCDH-CMV-EF1-Puro lentiviral vector (System Biosciences, #CD510B-1) or pLVX-neo-IRES vector (Biofeng, #632181) to generate Flag- and HA-tagged expression constructs. Truncation mutants of MORC2 and SETDB1 were cloned into pCDH vectors. Site-directed mutagenesis of MORC2 and SETDB1 was carried out using a ClonExpress Ultra One Step Cloning kit (Vazyme Biotech, #C117-01), followed by treatment with the methylation-sensitive restriction enzyme DpnI (New England Biolabs, #R0176S) to eliminate the parental template. All constructs were sequence-verified by Sangon Biotech (Shanghai, China). Detailed information of DNA constructs is provided in Supplementary Table S5. Immunoblotting and co-immunoprecipitation (co-IP) assays For immunoblotting assays, proteins were isolated from cultured cells and tissue samples using RIPA buffer containing phosphatase and protease inhibitor cocktails (Bimake; #B15003 and #B14002, respectively). Equal protein quantities were separated by SDS-PAGE and transferred to PVDF membranes (Meck Millipore, #IPVH00010). Following blocking with 5% bovine serum albumin (BSA), the membranes were incubated with specific primary antibodies. Protein bands were subsequently detected using the ChemiStar High-sig ECL kit (Tanon, #180-5001E). For co-immunoprecipitation (co-IP) assays, cells were lysed with NP40 buffer. Equal amounts of lysates were incubated with the designated antibodies overnight at 4 °C. Immunoprecipitated proteins were washed three times with NP40 buffer and analyzed by immunoblotting as described above. Antibody information is provided in Supplementary Table S6. Flow cytometry analysis Tissue samples were harvested and immediately placed in ice-cold RPMI-1640 medium, supplemented with 10% FBS and 1% penicillin-streptomycin solution. Tumor specimens were finely minced into small fragments. This was followed by enzymatic digestion with a cocktail containing collagenase type I (Sigma-Aldrich, #SCR103), hyaluronidase (Yeasen, #20426ES80), and dispase II (Yeasen, #40104ES80). The digestion process was carried out at 37 °C in a constant temperature shaker set to 100 rpm for 40 min. Following digestion, the resulting cell suspensions were sequentially filtered through 100 μm and 40 μm cell strainers (JETBIOFIL, #CSS013100 and #CSS013040, respectively). Then, the single-cell suspensions were treated with Red Blood Cell Lysis Buffer (BioLegend, #420302) to eliminate red blood cells. After that, the cells were resuspended and counted to achieve a concentration of 1 × 106 cells per sample. Fixable Viability Stain 780 (BD Biosciences, #565388) was applied to label dead cells. After this staining, the cells were washed and incubated with fluorescently labeled antibodies targeting surface proteins at a 1:100 dilution in Cell Staining Buffer (BD Pharmingen, #554656) for 30 min at 4 °C. For intracellular staining, the Fixation/Permeabilization Kit (BD Pharmingen, #554714) was used for fixation and permeabilization, after which cells were incubated with the specified antibodies in the dark for 30 min. Sample analysis was conducted using the Beckman Coulter CytoFlex S flow cytometer, and the data were processed and analyzed using FlowJo software. Detailed information on the antibodies used in this assay is provided in Supplementary Table S7. Representative flow cytometry gating plots are provided in Supplementary Figure S1. Multiplex immunohistochemistry (mIHC) staining The human breast cancer tissue microarray was obtained from Shanghai OUTDO Biotech (#HBreD180Bc01-2). All samples were derived from patients with pathologically confirmed breast cancer, and none had received neoadjuvant chemotherapy or radiotherapy before surgery. After excluding detached spots from the tissue microarray, a total of 168 viable spots (25 normal and 143 TNBC specimens) were subjected to staining. Tumor staging was assessed according to the established tumor-node-metastasis (TNM) staging system. As described previously [45], mIHC analysis was performed using the following antibodies, including MORC2 (Affinity, #AF6501), SETDB1 (Proteintech, #11231-1-AP), IFN-γ (Abcam, #abs119966), and GZMB (Abcam, #ab255598). All staining procedures were conducted in accordance with the manufacturers’ protocols and were carried out by BaiQianDu Biotech (Wuhan, China). Whole-slide fluorescence images were acquired using a 3DHISTECH Pannoramic SCAN slide scanner and subsequently analyzed using SlideViewer software. Patients were stratified into high- and low-expression groups based on the median fluorescence intensity across the cohort. Immunofluorescent (IF) staining Cells were fixed with 4% paraformaldehyde for 30 min and permeabilized with 0.5% Triton X-100 for 10 min. Following this, they were blocked with 5% BSA and then incubated with primary antibodies overnight at 4 °C. After thorough washing, the cells were stained with Alexa Fluor 555 (CST, #4409S or #4413S) or Alexa Fluor 488 (CST, #4408S or #4412S), and counterstained with DAPI (Abcam, #ab104139). For J2 antibody (SCICONS, #10010200) against double-stranded RNA (dsRNA) staining, cells were either treated or not treated with 20 U/mL of RNase III (MCE, #HY-KE7056) for 30 min at 37 °C after permeabilized. For dsDNA staining (Abcam, #ab27156), cells were either treated or not treated with 30 µM of the reverse transcriptase inhibitor Lamivudine (Selleck, #S1706) for 48 h before proceeding with the staining protocol. Immunofluorescent images were captured using a Leica SP5 confocal microscope and analyzed using ImageJ software. The analysis was performed blinded to the experimental conditions. RNA-sequencing (RNA-seq) To ensure cell culture quality, mycoplasma contamination was eliminated using Myc-AwayTM (Yeasen Biotech, #40607ES08). Total RNA was subsequently extracted from cultured cells using TRIzol reagent (Invitrogen, #15596018) following the manufacturer’s protocol. RNA-Seq library preparation and sequencing were conducted on the Illumina NovaSeq platform (Illumina, USA) at Honsunbio Biotech (Shanghai, China) using NovaSeq Reagent Kits for high-throughput sequencing. For data analysis, RNA-seq reads were aligned to the reference genome using STAR (v2.7.11a) [46]. Quantification was performed using TEtranscripts (v2.0.3) [47], a comprehensive tool specifically designed for analyzing both gene and transposable element (TE) expression. Briefly, TE analysis was conducted using species-specific GTF annotation files (GRCm38_Ensembl_rmsk_TE.gtf and GRCh38_Ensembl_rmsk_TE.gtf for mouse and human, respectively), obtained from the Hammell Lab’s software repository (https://hammelllab.labsites.cshl.edu/software/#TEtranscripts) [47]. To detect broad, distributed de-repression signals across TE families rather than identify a small set of individually significant loci, we used edgeR with TMM normalization to identify differentially expressed TEs defined by a raw p-value < 0.05. The quantitative results are provided in the Supplementary Excel file. The differential gene expression analysis was performed using the DESeq2 program [48]. Genes meeting the criteria of a p-value < 0.05 and an absolute fold change ≥ 1.5 were identified as differentially expressed genes (DEGs). Gene Set Enrichment Analysis (GSEA) was conducted to elucidate the biological significance. Single-cell RNA sequencing (scRNA-seq) Mouse tumor samples were dissociated using the Mouse Tumor Dissociation Kit (Miltenyi Biotec, #130-096-730), following the manufacturer’s protocol. scRNA-seq was performed by the experienced team at NovelBio (Shanghai, China) using the 10× Genomic Chromium Single Cell Platform. The scRNA-seq data were analyzed using the NovelBrain Cloud Analysis Platform (www.novelbrain.com), a system developed by NovelBio, which produced a consolidated expression matrix. We used a standard single-cell analysis workflow for quantitative analysis. A total of 18,095 cells were filtered based on quality criteria. Cells with more than 200 expressed genes and mitochondrial UMI rates below 20% were retained, and mitochondrial genes were removed from the expression data. Using Seurat (version 5), the data were normalized, scaled, and regressed. Batch effects across samples were corrected using the Harmony algorithm. Principal component analysis (PCA) was performed on the top 2,000 highly variable genes. For visualization, Uniform Manifold Approximation and Projection (UMAP) [49] was applied for cell clustering. Cell type annotation was performed based on canonical marker genes for each annotated population. DEGs between clusters were identified with Seurat’s FindMarkers function, applying thresholds of adjusted p-value < 0.05 and log2 fold change > 0.25. GSEA was conducted on DEGs to explore associated biological pathways and processes. CUT&RUN assays and data analysis CUT&RUN assays were performed using the CUT&RUN Kit (Vazyme, #HD102-01) with anti-MORC2 antibody (Affinity, #AF6501) and anti-SETDB1 antibody (Proteintech, #11231-1-AP). Raw reads were trimmed with Cutadapt and aligned to the human reference genome (GRCh38) using Bowtie2 with parameters --very-sensitive-local --no-mixed --no-discordant -X 700. Reads with a MAPQ score < 10 or those mapping to the mitochondrial genome were discarded. Signal tracks were generated using deepTools bamCoverage with RPKM normalization and a bin size of 10 bp. Peaks were called using MACS3 with the parameters --broad --nomodel --extsize 200 and IgG as the control. Peaks with a signalValue > 2 were retained for downstream analysis. Co-occupied peaks were defined as the overlap between MORC2 and SETDB1 peak sets using BEDTools. The overlap of these peaks with RepeatMasker-annotated transposable element (TE) regions was quantified using BEDTools intersect. Signal heatmaps and aggregate profiles were generated using deepTools computeMatrix in reference-point mode (± 3 kb, bin size = 50 bp), followed by plotHeatmap and plotProfile. RNA extraction and real-time quantitative PCR (RT-qPCR) Total RNA was isolated from cultured cells using TRIzol reagent (Invitrogen, #15596018). The extracted RNA was reverse-transcribed into cDNA using HiScript III RT SuperMix (Vazyme, #R323-01). Quantitative PCR (qPCR) was carried out with Chamq Universal SYBR qPCR Master Mix (Vazyme, #Q711-03), and relative gene expression levels were determined using the 2−ΔΔCT method. The qPCR primers are listed in Supplementary Table S8. Chromatin immunoprecipitation (ChIP) assays The ChIP assays were conducted with the SimpleChip Enzymatic Chromatin IP Kit (Magnetic Beads) (Cell Signaling Technology, #9003S), adhering to the provided protocol. Initially, cells were subjected to fixation with 1% formaldehyde at ambient temperature, followed by the addition of a 10× glycine solution to neutralize any residual formaldehyde. Subsequent to sonication, chromatin fragments were immunoprecipitated employing an anti-H3K9me3 antibody, with normal IgG serving as the negative control. The protein-DNA complexes were then isolated using protein A/G magnetic beads. Both the immunoprecipitated and input DNA were subsequently purified and analyzed via qPCR to assess enrichment levels. The primer sequences utilized in the ChIP assays are detailed in Supplementary Table S9. Statistical analyses All quantitative data are presented as the mean ± standard deviation (SD). In the boxplots, the lower and upper limits represent the minimum and maximum values respectively, while the centerline denotes the median. Statistical analyses were carried out using GraphPad Prism and R software. The specific statistical methods used for each experiment are indicated in the figure legends. Statistical significance was defined as a p-value of less than 0.05 (*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001; “ns” indicates no significance). Results MORC2 is a potential negative predictor of response to ICB therapy in cancer While recent studies have shed light on the roles of MORC2 in development and cancer [40, 50–52], its function in modulating cancer immunity has yet to be elucidated. Intriguingly, a recent in vivo CRISPR-Cas9 screen targeting 936 chromatin regulators revealed Morc2a, the mouse paralog of human MORC2, as a potential negative regulator of tumor-cell response to ICB therapy in a mouse model of Lewis lung carcinoma [53] (Supplementary Figure S2A). To further validate this finding, we compared MORC2 expression levels between responders (Res) and non-responders (non-Res) to ICB therapy in multiple publicly available datasets from the Immune Checkpoint Blockade Therapy Atlas (ICBatlas; https://guolab.wchscu.cn/ICBatlas/) [54]. This analysis incorporated the cohorts from non-small cell lung cancer (NSCLC) [55], gastric cancer [56], and melanoma [57] (Supplementary Figure S2B). Consistently, we found that lower MORC2 expression levels were associated with improved response to ICB therapy in all three cohorts (Supplementary Figure S2C). These results suggest that MORC2 may influence the efficacy of ICB therapy in cancer. Given that the MORC gene family also includes three other members (MORC1, MORC3, and MORC4) in addition to MORC2 [25], we therefore extended our analysis to evaluate their expression levels in relation to immunotherapy response in these cohorts. In contrast to MORC2, high expression of MORC1 was positively associated with enhanced sensitivity to ICB therapy in the melanoma cohort [57] (Supplementary Figure S2D). In addition, MORC1 expression was not detected in either NSCLC [55] or gastric cancer [56]. This absence of detection may be attributed to its tissue-specific expression pattern, as MORC1 is predominantly expressed in male germ cells [58]. Furthermore, MORC3 expression showed no significant association with ICB response across all three cohorts (Supplementary Figure S2E). Lastly, elevated MORC4 expression was positively linked to improved ICB response in the gastric cancer [56], but not in the NSCLC [55] and melanoma [57] cohorts (Supplementary Figure S2F). Furthermore, a pan-cancer anti-PD-1 survival analysis using KMplot confirmed that MORC2 showed a strong and consistent negative association with immunotherapy outcomes, whereas MORC4 exhibited only a borderline association (Supplementary Figure S2G). In our recent study, TNBC was classified into four transcriptome-based subtypes, including immunomodulatory (IM), basal-like immune-suppressed (BLIS), luminal androgen receptor (LAR), and mesenchymal-like (MES) [41]. The subsequent FUTURE clinical trial based on this classification further demonstrated that the IM subtype exhibits relative sensitivity to anti-PD-1 therapy [59]. To investigate the potential involvement of MORC gene family members in this setting, we compared the mRNA expression levels of MORC1-4 between the IM subtype and the combined non-IM subtypes (BLIS, LAR, and MES) using the FUSCC-TNBC transcriptomic dataset [41]. We found that MORC2 and MORC4 expression were significantly downregulated in the IM subtype relative to the non-IM subtypes (Supplementary Figure S3A-B). In contrast, MORC1 expression was undetectable in this dataset, and MORC3 levels showed no significant difference between the two subgroups (Supplementary Figure S3C-D). Accumulating evidence shows that higher levels of tumor-infiltrating lymphocytes (TILs) are associated not only with better prognosis but also with enhanced response to ICB therapy in patients with TNBC [60, 61]. Intriguingly, analysis of the same FUSCC-TNBC dataset [41] further demonstrated that higher expression levels of MORC2 and MORC4 were associated with reduced infiltration of both intratumoral (iTILs) and stromal (sTILs) TILs (Supplementary Figure S3E-H). In contrast, MORC3 expression showed no significant association with either iTILs or sTILs levels (Supplementary Figure S3I-J). In immunotherapy-treated TNBC cohorts, MORC2 expression was significantly reduced following treatment in the GSE225078 [62] and GSE260989 [63] datasets (Supplementary Figure S4A-B). In the GSE241876 [64] dataset, the low-MORC2 group achieved a higher objective response rate than the high-MORC2 group (Supplementary Figure S4C). Kaplan-Meier analysis of the PAM50-defined basal-like subtype showed that high MORC2 expression was associated with worse overall survival in both microarray and RNA-sequencing cohorts (Supplementary Figure S4D-E). Furthermore, scRNA-seq analysis of 32 TNBC samples [65] revealed that MORC2 was predominantly enriched in epithelial tumor cells compared to immune subsets (Supplementary Figure S4F-G). Taken together, our integrated analysis reveals that, in contrast to other MORC family members, MORC2 expression levels are consistently associated with reduced sensitivity to ICB therapy across all analyzed datasets. Nevertheless, the role of MORC2 in shaping the cancer immune landscape and its underlying molecular mechanisms remain elusive. MORC2 depletion suppresses tumor growth in TNBC in an immune microenvironment-dependent manner Clinically, immunotherapy is more commonly utilized for TNBC than for other subtypes. This preference is driven by the distinct immunologic features of TNBC, including its higher tumor mutation burden, greater immunogenicity, elevated PD-L1 expression, and abundant TILs [12, 66]. To assess the potential of MORC2 as an immunotherapy target in TNBC, we first analyzed its expression levels in normal breast tissues, non-TNBC (including ER/PR-positive and HER2-positive subtypes), and TNBC samples. This analysis was conducted using the transcriptomic datasets from the internal CBCGA (Chinese Breast Cancer Genome Atlas) [40] as well as the external TCGA [42] and METABRIC [43] cohorts. The results showed that MORC2 mRNA levels were consistently elevated in TNBC compared to non-TNBC and normal breast tissues across three independent cohorts (Fig. 1A-C). In contrast, the mRNA levels of MORC1, MORC3, and MORC4 did not show a significant increase in TNBC relative to normal breast tissues and non-TNBC specimens in these datasets (Supplementary Figure S5A-I). The upregulation of MORC2 mRNA levels in TNBC was further validated in the FUSCC-TNBC transcriptomic dataset [41], which comprised 88 pairs of TNBC specimens and their matched adjacent normal breast tissues (Fig. 1D). This pattern was not recapitulated for MORC1, MORC3, or MORC4 in the same cohort (Supplementary Figure S5J-L). MORC2 is upregulated in TNBC, and its depletion suppresses tumor growth in a manner that critically depends on the immune microenvironment. A-C Expression profiling of MORC2 in normal breast tissues, non-TNBC (including ER/PR-positive and HER2-positive subtypes), and TNBC samples based on transcriptomic datasets from the CBCGA (A), TCGA (B), and METABRIC (C) cohorts. D Expression profile of MORC2 in the transcriptomic dataset from the FUSCC-TNBC cohort. E-F Expression profiling of MORC2 across normal breast tissues, non-TNBC, and TNBC samples in proteomic datasets from the CBCGA (E) and CPTAC (F) cohorts. G-H Immunoblotting analysis of MORC2 protein levels in 14 paired TNBC and adjacent normal tissues (G). Quantification of relative MORC2 protein levels using ImageJ software is shown in (H). N, normal; T, tumor. I-J mIHC analysis of MORC2 protein expression levels in a tissue microarray containing 143 TNBC specimens and 25 normal breast tissues. The same TMA sections were used for multiplex immunofluorescence staining of SETDB1 (Fig. 6A). Representative mIHC images and quantification of relative MORC2 expression levels using ImageJ software are shown in (I) and J, respectively. K-P Negative control (NC), Morc2a-KO, and Morc2a-KO plus Morc2a (rescue) murine TNBC cells were inoculated into the mammary fat pads of 5-week-old syngeneic BALB/c or C57BL/6 mice. 4T1 (n = 6) and 66cl-4 (n = 7) cells were injected into BALB/c mice, and AT-3 cells (n = 8) were injected into C57BL/6 mice. Mice were euthanized at the experimental endpoint, and syngeneic tumors were excised for weight and volume measurement. Tumor volumes and weights from 4T1, 66cl-4, and AT-3 tumor-bearing mice are presented in panels (K-L), M-N, and O-P, respectively. Notes: Statistical analyses were performed using the Kruskal-Wallis test (A–C, E, F), Wilcoxon test (D, J), paired Student's t-test (H), two-way repeated measures ANOVA followed by Tukey's post-hoc test (K, M, O), one-way ANOVA followed by Tukey's post-hoc test (L), and unpaired Student's t-test (N, P). *, **, ***, and **** indicate statistical significance at the p < 0.05, p < 0.01, p < 0.001, and p < 0.0001 levels, respectively Corroborating the transcriptomic data, MORC2 protein levels were also elevated in TNBC compared to normal and non-TNBC tissues in both CBCGA [40] and CPTAC [44] proteomic datasets (Fig. 1E-F, respectively). This pattern was unique to MORC2, as the protein levels of MORC1, MORC3, and MORC4 showed no consistent increase in TNBC within the same cohorts (Supplementary Figure S6A-F). To further validate these findings, we collected 14 pairs of primary TNBC specimens and matched adjacent normal tissues to assess MORC2 protein expression levels by immunoblotting. The results demonstrated significant up-regulation of MORC2 in TNBC tissues compared to their normal counterparts (Fig. 1G-H). Consistently, mIHC analysis of a tissue microarray containing 143 TNBC specimens and 25 normal breast tissues demonstrated significantly stronger MORC2 staining in TNBC compared to normal controls (Fig. 1I-J). Overall, these results suggest that MORC2 is significantly upregulated in TNBC tissues. Based on these findings, we next sought to determine whether the tumor-suppressive effects of MORC2 loss are mediated by the tumor immune microenvironment. To this end, we generated Morc2a-knockout (KO) 4T1 murine breast cancer cells using CRISPR-Cas9 technology targeting the mouse ortholog of human MORC2. Then, Morc2a was reintroduced into knockout cells to validate on-target specificity (Supplementary Figure S7A). Cell proliferation and colony formation assays confirmed that Morc2a depletion did not significantly affect cell growth in vitro (Supplementary Figure S7B-D). Consistent with these findings, Morc2a knockout did not significantly affect cell growth in vitro in 66 cl-4 and AT-3 cell models (Supplementary Figure S7E-H). We next orthotopically implanted these cells into immunodeficient BALB/c nude mice, and found that Morc2a deficiency did not significantly alter tumor growth (Supplementary Figure S7I-L). Next, we assessed the impact of Morc2a deficiency on tumor growth in an immunologically intact setting by orthotopically engrafting these cells into immunocompetent mice. Remarkably, Morc2a KO substantially attenuated tumor growth, leading to significantly reduced tumor volume and weight compared with negative control (NC) cells (Fig. 1K-L). To confirm on-target specificity in the 4T1 model, re-expression of Morc2a in Morc2a-knockout cells partially restored tumor growth (Fig. 1K-L), confirming that the observed phenotype is specifically attributable to Morc2a loss. Consistently, Morc2a knockout had significantly inhibitory effect on tumor growth in the syngeneic 66 cl-4 and AT-3 models (Fig. 1M-P). Taken together, these results demonstrate that MORC2 is upregulated in TNBC tissues and its depletion suppresses tumor growth in a manner critically dependent on an intact tumor immune microenvironment in TNBC. MORC2 deletion elicits robust anti-tumor immunity by remodeling the tumor immune microenvironment To elucidate the potential immune-regulatory mechanisms of MORC2 in TNBC, we conducted scRNA-seq on syngeneic tumors derived from control and Morc2a-KO 4T1 cells in the immunocompetent BALB/c mice. Through dimensionality reduction and clustering with the uniform manifold approximation and projection (UMAP) method [49], we identified 18 distinct cell clusters. These clusters were subsequently annotated based on integrated marker gene expression, revealing 9 major cell types such as epithelial cells, fibroblasts, macrophages, neutrophils, and T cells (Fig. 2A-B). Notably, the loss of tumor cell-intrinsic Morc2a profoundly remodeled the tumor immune microenvironment, which was marked by significant changes in the proportions of macrophages, T cells, and neutrophils (Fig. 2C). Given the paradoxical and complex role of neutrophils in anti-tumor immunity in TNBC to date [67], we next focused our analysis on the impact of Morc2a depletion on macrophages and T cells. MORC2 deletion elicits robust anti-tumor immunity by remodeling the tumor immune microenvironment. A Dot plot depicting the mean expression levels of marker genes for the major cell populations identified in this study. B UMAP visualization of major cell populations identified by scRNA-seq in NC and Morc2a-KO 4T1 tumors. C Butterfly lollipop plot showing proportional changes of individual cell types from scRNA-seq data. Each lollipop corresponds to a specific cell type, and its bar length denotes the proportion in NC vs. Morc2a-KO 4T1 tumors. D Bubble plot showing GSEA results of differentially expressed genes in macrophages from NC vs. Morc2a-KO 4T1 tumors. Highlighted key pathways are associated with immune effector activation. E Volcano plot of differentially expressed genes in macrophages identified from scRNA-seq data of NC and Morc2a-KO 4T1 tumors. Genes linked to mouse MHC class I antigen presentation are highlighted. F Ridgeline plot of antigen presentation and co-stimulation activity scores in macrophages, based on scRNA-seq data from NC and Morc2a-KO 4T1 tumors. G-I Flow cytometry analysis of macrophage populations in NC, Morc2a-KO, and rescued 4T1 tumors (n = 6). The proportions of total macrophages (CD11b⁺F4/80⁺), M1 macrophages (iNOS⁺), and M2 macrophages (CD206⁺) are shown in (G), (H), and (I), respectively. J Box plots showing the expression levels of Cd4, Cd8a, Gzmb, and Prf1 in T cells, based on scRNA-seq data from NC and Morc2a-KO 4T1 tumors. K-N Flow cytometry analysis of T cell populations in NC, Morc2a-KO, and rescued 4T1 tumors (n = 6). The proportions of CD4⁺ T cells (K), CD8⁺ T cells (L), IFN-γ⁺CD8⁺ T cells (M), and GzmB⁺CD8⁺ T cells (N) are shown. Notes: Statistical analyses were performed using the Kolmogorov-Smirnov test (F), one-way ANOVA (G-I and K-N), and Wilcoxon test (J). *, **, and *** indicate statistically significant at p < 0.05, p < 0.01, and p < 0.001 level, respectively. no, no significance To delineate the functional changes in macrophages, we performed GSEA using differentially expressed genes identified from a comparison of scRNA-seq profiles of control versus Morc2a-KO tumors. Our analysis found that Morc2a-KO syngeneic tumors exhibited significant enrichment of pathways associated with macrophage activation in anti-tumor immunity, such as the TNF-α/NF-κB (tumor necrosis factor-alpha/ nuclear factor-Kappa B), IFN-γ (interferon-gamma), IFN-α, inflammatory response, and IL-6/JAK/STAT3 (interleukin 6/Janus kinase/signal transducer and activator of transcription 3) signaling pathways (Fig. 2D). Consequently, macrophages in Morc2a-KO tumors exhibited enhanced expression of key mouse MHC-I (major histocompatibility complex class I) molecules, such as H2-K1 (histocompatibility 2, K region locus 1), H2-D1 (histocompatibility 2, D region locus 1), and H2-T23 (histocompatibility 2, T region locus 23) (Fig. 2E). This upregulation was associated with a concomitant enhancement in their capacity for antigen presentation and co-stimulation (Fig. 2F). Supporting our findings, GSEA of bulk RNA-seq datasets from both TCGA [42] and METABRIC [43] cohorts revealed significant enrichment of the antigen processing and presentation pathways in TNBC patients with low MORC2 expression (Supplementary Figure S8A). Consistent with our initial analysis, flow cytometry analysis revealed that the overall proportion of macrophages (CD11b⁺F4/80⁺) among total leukocytes (CD45⁺) was increased in Morc2a-KO tumors (Fig. 2G). Specifically, they harbored a higher proportion of M1 macrophages (marked by inducible nitric oxide synthase-positive, iNOS⁺) and a lower proportion of M2 macrophages (CD206⁺) compared to controls (Figs. 2H-I). Notably, re-expression of Morc2a in knockout cells reversed macrophage polarization to control levels. Collectively, these results suggest that MORC2 deletion elicits a potent anti-tumor immune response by reprogramming macrophages from an anti-inflammatory M2 to a pro-inflammatory M1 phenotype, a key mechanism in anti-tumor immunity [68, 69]. Next, we analyzed the functional impact of MORC2 deletion on T cell-mediated immunity in TNBC. It is well established that cytotoxic T lymphocytes (CTLs) are activated upon T cell receptor (TCR) recognition of tumor antigens presented by MHC-I molecules, and then eliminate tumor cells primarily through the perforin-granzyme pathway [70]. In line with this mechanism, our scRNA-seq analysis revealed a marked upregulation of critical CTL genes in Morc2a-KO tumors. These included Cd8a (CD8 subunit alpha), which is essential for T cell recognition and activation, as well as the key effector molecules Prf1 (perforin-1) and Gzmb (granzyme B). In contrast, no such upregulation was observed for CD4 following Morc2a depletion (Fig. 2J). Consistently, flow cytometry analysis revealed that although the proportion of CD4⁺ T cells within the total functionally mature T cell population (CD3⁺) remained unchanged following Morc2 depletion (Fig. 2K), tumors lacking Morc2a exhibited a marked increase in the overall percentage of CD8⁺ T cells (Fig. 2L). This was accompanied by elevated frequencies of IFN-γ⁺ and GzmB⁺ cytotoxic subsets (Figs. 2M-N). The on-target specificity was confirmed by the restoration of all these phenotypes upon Morc2a re-expression in Morc2a-KO cells. To validate these findings in human subjects, we interrogated RNA-seq datasets from the TCGA [42] and METABRIC [43] cohorts. Spearman correlation analysis revealed that MORC2 expression was significantly negatively correlated with the well-established markers of activated CD8⁺ T cells, such as CD8A/B, PRF1, GZMB, and CD247 (Supplementary Figure S8B). Consistently, MORC2 expression showed a significant inverse correlation with the cytolytic activity score, which was calculated based on the transcript levels of two key cytolytic effectors GZMA and PRF1 [71] (Supplementary Figure S8C-D). These results collectively suggest that MORC2 plays a critical role in suppressing anti-tumor T cell responses. Consistent with the above results, analysis of RNA-seq datasets from the TCGA [42] and METABRIC [43] cohorts also revealed that MORC2 expression was inversely correlated with a broad spectrum of key anti-tumor immune molecules, including those involved in macrophage priming and polarization (IRF5 and STAT1), pro-inflammatory responses (IL6 and IL1B), CTL recruitment (CXCL9 and CXCL10), and antigen presentation/activation (CD86 and HLA-DRA) [69, 72] (Supplementary Figure S8E). Collectively, converging evidence from mouse models and human cohorts suggests that MORC2 fosters an immunosuppressive tumor microenvironment in TNBC primarily by impairing the anti-tumor functions of M1-like macrophages and CD8⁺ T cells. MORC2 deficiency unleashes transposable element (TE) expression and elicits an interferon response To explore how tumor cell-intrinsic MORC2 deficiency reshapes the immune microenvironment and enhances anti-tumor immunity, we conducted RNA-seq on both NC and Morc2a‑KO 4T1 cells. Subsequent GSEA revealed a significant enrichment of the interferon response pathway in Morc2a-KO cells (Fig. 3A). RT-qPCR assays further confirmed that Morc2a depletion led to the marked upregulation of a panel of classic interferon-stimulated genes (ISGs), including Cxcl9, Ifit3, Ifi207, Ifitm1, Bst2, and Oas3 (Fig. 3B). The interferon response pathway and its associated ISGs play critical roles in coordinating the anti-tumor activities of both M1-like macrophages and T cells. They directly enhance the phagocytosis, killing capacity, and antigen presentation function of M1-like macrophages, while also directly promoting the activation, proliferation, and cytotoxic activity of T cells [73, 74]. In line with this enhanced interferon response, we observed a concomitant upregulation of a panel of mouse MHC-I-related genes, such as B2m, H2-K1, H2-D1, and H2-Q1, in Morc2a-KO cells (Fig. 3C-D). Furthermore, flow cytometry assays and immunofluorescent staining also demonstrated a significant upregulation of the classical mouse MHC-I molecules H-2Kb/H-2Db in Morc2a-KO cells (Fig. 3E-F). Together, these findings highlight the pivotal role of tumor-intrinsic MORC2 in regulating anti-tumor immunity via the interferon-mediated response. MORC2 deficiency unleashes transposable element (TE) expression and elicits an interferon response. A GSEA of RNA-seq data showing significant enrichment of the interferon response pathway in Morc2a-KO vs. NC 4T1 cells. The normalized enrichment score (NES) and p-value are shown. B RT-qPCR analysis of selected interferon response pathway genes in NC and Morc2a-KO 4T1 cells. C Heatmap showing expression levels of mouse MHC-I complex genes from RNA-seq data of Morc2a-KO vs. NC 4T1 cells. D RT-qPCR analysis of selected genes in the antigen processing and presentation pathway in NC and Morc2a-KO 4T1 cells. E Flow cytometry analysis of mouse MHC-I molecules (H-2Kb/H-2Db) in NC and Morc2a-KO 4T1 cells. Representative histograms (left) and quantification of mean fluorescence intensity (MFI; right) are shown. F Immunofluorescent staining of mouse MHC-I molecules in NC and Morc2a-KO 4T1 cells. Representative images (left) and quantification (right) are shown. Scale bar: 10 μm. G Volcano plot illustrating differentially expressed TEs in Morc2a-KO vs. NC 4T1 cells, based on RNA-seq data. H RT-qPCR validation of the TEs indicated in (G) in NC and Morc2a-KO 4T1 cells. I GSEA of RNA-seq data showing enrichment of the interferon response pathway in MORC2-KO vs. NC MDA-MB-231 cells. The NES and p-value are indicated. J RT-qPCR analysis of selected interferon response pathway genes in NC vs. MORC2-KO MDA-MB-231 cells. K Heatmap depicting expression levels of MHC-I complex genes from RNA-seq data of MORC2-KO vs. NC MDA-MB-231 cells. L RT-qPCR analysis of selected MHC-I complex genes in NC vs. MORC2-KO MDA-MB-231 cells. M Volcano plot showing differentially expressed TEs in MORC2-KO vs. NC MDA-MB-231 cells, based on RNA-seq data. N RT-qPCR validation of the TEs indicated in (M) in NC and MORC2-KO MDA-MB-231 cells. Notes: Statistical analyses were performed using unpaired one-way ANOVA (B, D, H, J, L, and N). *, **, and *** indicate statistically significant at p < 0.05, p < 0.01, and p < 0.001 level, respectively The interferon pathway can be activated by aberrant expression of endogenous nucleic acids derived from transposable elements (TEs) through a mechanism similar to cellular defense against viral infection [75, 76]. Given that MORC2 is a crucial regulator of chromatin compaction and gene silencing [30, 31, 77], we next evaluated TE expression upon MORC2 deletion using TEtranscripts, a computational package designed for quantifying TE expression from RNA-seq data [47]. Indeed, deletion of Morc2a in 4T1 cells significantly altered the landscape of TEs (Fig. 3G). RT-qPCR analysis further demonstrated that the loss of Morc2a robustly induced the transcription of specific TEs, as exemplified by the elevated expression of LTR53-int (long terminal repeat 53-internal sequence), RLTR26D (repetitive long terminal repeat 26D), RLTR9A3B (repetitive long terminal repeat 9A3B) and RLTR6C (repetitive long terminal repeat 6 C) (Fig. 3H). Next, we investigated whether the phenotypes observed in mice could be recapitulated in human cells. To this end, we knocked out MORC2 in the human TNBC cell line MDA-MB-231 and performed RNA-seq. GSEA of the differentially expressed genes revealed a significant enrichment of the interferon signaling pathway upon MORC2 depletion (Fig. 3I). RT-qPCR assays further demonstrated the upregulation of key downstream ISGs, including CXCL9, CXCL10, BST2 (bone marrow stromal cell antigen 2), and IFIT3 (interferon-induced protein with tetratricopeptide repeats 3) (Fig. 3J). Moreover, MORC2 depletion significantly upregulated key genes involved in the human MHC-I antigen presentation pathway, including B2M (β-2-microglobulin), HLA-A (human leukocyte antigen-A), HLA-C, and TAP1 (transporter associated with antigen processing 1) (Fig. 3K-L). Concurrently, MORC2 loss enhanced the transcription of multiple TEs, such as LTR06 (long terminal repeat 06), LTR22C0 (long terminal repeat 22C0), LTR22C2 (long terminal repeat 22C2), and MER74C (medium reiteration-frequency repeat 74 C) (Fig. 3M-N). Subfamily-level analysis of TEs further revealed that ERVK subfamily members represent the most prominently upregulated TE class upon MORC2 loss, serving as the primary drivers of viral mimicry in this context (Supplementary Figure S9A). However, knockdown of two representative ERV elements, LTR22C2 (ERVK) and MER74C (ERVL), produced no significant effects on ISG expression levels (Supplementary Figure S9B-C), suggesting that viral mimicry is driven by the cumulative de-repression of multiple ERV loci rather than by any single element. Consistently, genome-wide analysis confirmed broad, multi-locus de-repression of ERV elements across all four subclasses upon MORC2 loss, whereas LINE deregulation was restricted to evolutionarily young, retrotransposition-competent L1 subfamilies (L1HS, L1PA2, L1PA3) (Supplementary Figure S9D-E), indicating that MORC2 broadly suppresses ERV elements. Collectively, these results from human and mouse models demonstrate a profoundly conserved role for MORC2 in suppressing TE transcription. To further investigate the association between MORC2 expression and interferon pathway activation, we conducted single-sample Gene Set Enrichment Analysis (ssGSEA) using the ISGs signature [78] across eight cancer types with high global prevalence in the TCGA dataset, including TNBC, lung adenocarcinoma (LUAD), lung squamous cell carcinoma (LUSC), stomach adenocarcinoma (STAD), prostate adenocarcinoma (PRAD), liver hepatocellular carcinoma (LIHC), bladder carcinoma (BLCA), and esophageal carcinoma (ESCA). Our data demonstrated that MORC2-low tumors exhibited a significant increase in ISG scores compared to MORC2-high tumors (Supplementary Figure S10A). Furthermore, Spearman correlation analysis revealed a strong inverse relationship between MORC2 expression levels and ISG scores across these cancer types (Supplementary Figure S10B). We next performed GSEA on a pan-cancer cohort from TCGA using the hallmark gene sets to systematically identify pathways and biological processes associated with MORC2 expression levels. Notably, MORC2-low tumors exhibited significant enrichment of immune-related pathways, particularly those linked to IFN-α and IFN-γ response, IL6/JAK/STAT3 signaling, and inflammatory response (Supplementary Figure S10C). Taken together, our findings reveal a central role for MORC2 in suppressing tumor immunity through the inhibition of interferon signaling. MORC2 deficiency triggers interferon responses by activating nucleic acid-sensing pathways Endogenous nucleic acids derived from TEs are primarily generated through two pathways, including bidirectional transcription producing double-stranded RNA (dsRNA) and reverse transcription yielding double-stranded DNA (dsDNA) [79, 80]. These nucleic acids can be recognized by specific pattern recognition receptors. In this context, dsRNA is detected via the RIG-I/MDA5–MAVS (retinoic-acid-inducible gene I/melanoma-differentiation-associated gene 5-mitochondrial antiviral signaling protein) pathway [81], whereas dsDNA is sensed by the cGAS-STING (cyclic GMP-AMP synthase-stimulator of interferon genes) axis [82]. Upon ligand binding, both pathways converge to activate the key downstream signaling molecules TBK1 (TANK-binding kinase 1) and IRF3 (interferon regulatory factor 3). This activation induces a robust interferon response and promotes the subsequent expression of ISGs [75, 76] (Supplementary Figure S11A). We next conducted a comprehensive analysis of the expression levels of these sensors and the activation status of their downstream signaling molecules following MORC2 depletion in MDA-MB-231 and BT-549 cells. Notably, MORC2 ablation enhanced the expression of MDA5 and cGAS proteins, along with the phosphorylation (activation) of their downstream effectors STING, TBK1, and IRF3 (Fig. 4A). Furthermore, these MORC2 depletion-induced signaling events were markedly potentiated by IFN-γ treatment (Fig. 4B). Consistent with these findings, IFN-γ treatment also enhanced MORC2 loss-induced upregulation of ISGs, such as CXCL9, IFIT3, and BST2 (Fig. 4C). These results demonstrate that MORC2 serves as a key regulator of the interferon-modulated nucleic acid-sensing pathway. MORC2 deficiency triggers interferon responses by activating nucleic acid-sensing pathways. A Immunoblotting analysis of dsRNA and dsDNA sensors and the activation status of their downstream signaling molecules in NC and MORC2-KO MDA-MB-231 and BT-549 cells. Protein expression levels were quantified using ImageJ and normalized to β-actin. p-, phosphorylated. B NC and MORC2-KO MDA-MB-231 and BT-549 cells were treated with or without 20 ng/mL IFN-γ for 24 h and then subjected to immunoblotting with the indicated antibodies. C NC and MORC2-KO MDA-MB-231 and BT-549 cells were treated with or without 20 ng/mL IFN-γ for 24 h and then subjected to RT-qPCR. D NC and MORC2-KO MDA-MB-231 and BT-549 cells were permeabilized and then pretreated with or without 20 U/mL RNase III for 30 min at 37 °C, followed by immunofluorescent staining of dsRNA using the anti-J2 antibody. Fluorescence intensity was quantified using ImageJ. E NC and MORC2-KO MDA-MB-231 and BT-549 cells were pretreated with or without 30 µM lamivudine (3TC) for 48 h, followed by immunofluorescent staining of dsDNA. Fluorescence intensity was quantified using ImageJ. F RT-qPCR analysis of selected interferon-stimulated genes in MDA-MB-231 and BT-549 cells under the following conditions: NC, MORC2-KO alone, and MORC2-KO combined with either MDA5 or cGAS knockdown. Notes: Statistical significance for panels (C-F) was determined by one-way ANOVA. *, **, and *** indicate statistically significant at p < 0.05, p < 0.01, and p < 0.001 level, respectively Next, we conducted immunofluorescent staining using the J2 antibody, which specifically recognizes dsRNA structures [78]. Our results revealed a significant increase in cytoplasmic J2 signals in MORC2-KO cells relative to controls, indicating robust accumulation of dsRNA. This increase was partially abolished upon treatment of MORC2-depleted MDA-MB-231 and BT-549 cells with RNase III, an enzyme that specifically degrades dsRNA [83] (Fig. 4D and Supplementary Figure S11B). Consistent with this, flow cytometry assays further demonstrated a significant increase in cytoplasmic dsRNA levels upon MORC2 deletion (Supplementary Figure S11C-D). These data further support our hypothesis that MORC2 inhibition triggers the activation of the dsRNA-sensing pathway. Since MORC2 loss also upregulated the dsDNA-sensing pathway (Fig. 4A), we hypothesized that this might be due to dsDNA generated from the reverse transcription of TEs. To test this, we treated cells with the nucleoside reverse transcriptase inhibitor lamivudine (also known as 3TC) [84] and found that it effectively attenuated the increase in cytosolic DNA abundance induced by MORC2 depletion in MDA-MB-231 and BT-549 cells (Fig. 4E and Supplementary Figure S11E). Additionally, 3TC treatment significantly suppressed ISG induction in MORC2-depleted cells, implicating TE-derived reverse-transcribed DNA intermediates in innate immune activation (Supplementary Figure S11F). This result is consistent with the model that reverse-transcribed dsDNA from TEs contributes to this pathway activation. In support of the above findings, we found that knockdown of the nucleic acid sensors MDA5 (dsRNA) or cGAS (dsDNA) in MORC2-depleted cells significantly attenuated MORC2 loss-induced upregulation of classic ISGs, such as CXCL9, BST2, and IFIT3 (Fig. 4F and Supplementary Figure S11G). Additionally, loss of MORC2 did not alter MORC3 or MORC4 expression levels (Supplementary Figure S11H), suggesting the absence of compensatory transcriptional regulation among MORC family members. Together, these findings establish that MORC2 loss activates the interferon response through the essential engagement of both cytosolic dsRNA and dsDNA sensing pathways. This occurs via the unleashing of TE-derived nucleic acids, which are recognized by these pathways to mount a robust anti-tumor immune response. MORC2 collaborates with the histone methyltransferase SETDB1 to transcriptionally repress TEs Next, we sought to elucidate the mechanism of MORC2-mediated transcriptional repression of TEs and interferon response in TNBC. It has been documented that the chromatin mark histone H3 trimethylation at lysine 9 (H3K9me3) plays a pivotal role in repressing TEs [85] and that histone-lysine N-methyltransferase SETDB1 is the primary enzyme catalyzing the deposition of H3K9me3 [86]. Furthermore, MORC2 is a chromatin remodeling protein that serves as a localized silencer by bridging distal DNA segments to facilitate chromatin compaction and transcriptional repression [30, 87]. Based on these findings, we hypothesized that MORC2 may collaborate with SETDB1 to mediate the transcriptional repression of TEs. Notably, analysis of transcriptomic datasets from TCGA [42] and METABRIC [43] cohorts as well as proteomic datasets from CBCGA [40] and CPTAC [44] revealed that both mRNA and protein levels of SETDB1 were upregulated in TNBC tissues relative to normal samples (Supplementary Figure S12A-D). Moreover, SETDB1 mRNA levels showed a strong negative correlation with the ISG signature score in the TCGA-TNBC dataset [78] (Supplementary Figure S12E-F). These results suggest that SETDB1, similar to MORC2, plays a role in TNBC immunosuppression. To investigate a potential collaboration between MORC2 and SETDB1 in repressing TEs, we first examined their physical interaction. Reciprocal co-IP assays revealed that MORC2 interacted with SETDB1 at the endogenous level in MDA-MB-231 and BT-549 cells (Fig. 5A-B). Furthermore, immunofluorescent staining demonstrated substantial co-localization of the two proteins in both cell lines (Fig. 5C, the yellow area indicated by the blue arrows in the zoomed-in images, and Supplementary Figure S13A). Collectively, these data confirm that MORC2 interacts with SETDB1 in TNBC cells. MORC2 collaborates with the histone methyltransferase SETDB1 to transcriptionally repress TEs. A-B Reciprocal co-immunoprecipitation (co-IP) assays using control IgG, anti-MORC2 (A), or anti-SETDB1 (B) antibody to detect the interaction between endogenous MORC2 and endogenous SETDB1 in MDA-MB-231 and BT-549 cells. C Immunofluorescent analysis showing colocalization (yellow signals) of MORC2 (red) and SETDB1 (green) in MDA-MB-231 and BT-549 cells. Scale bar: 10 μm. D-E Schematic diagram of MORC2 protein truncation mutants (D) and co-IP assays identifying the SETDB1-binding domains of MORC2 (E). F-G Schematic diagram of SETDB1 protein truncation mutants (F) and co-IP assays identifying the MORC2-binding domains of SETDB1 (G). H Integrative Genomics Viewer (IGV) tracks showing H3K9me3 signals at the genomic locus containing LTR22C2 and MER74C, based on analysis of a publicly available ChIP-seq dataset for H3K9me3 in NC and MORC2-KO human chronic myeloid leukemia K562 cells [32]. I-J RT-qPCR analysis of the expression levels of indicated TEs (I) and selected ISGs (J) in MDA-MB-231 and BT-549 cells stably expressing empty vector (pCDH), Flag-MORC2 alone, or Flag-MORC2 combined with SETDB1 knockdown. K ChIP-qPCR analysis of H3K9me3 enrichment at the indicated TE loci in MDA-MB-231 and BT-549 cells. Notes: Statistical significance for panels I-K was analyzed by one-way ANOVA. *, **, and *** indicate statistically significant at p < 0.05, p < 0.01, and p < 0.001 level, respectively Human MORC2 protein contains an N-terminal ATPase domain, a central CW-type zinc finger (CW-ZF) domain, and a C-terminal chromo-like (CL) domain [25–28] (Fig. 5D). The ATPase domain in MORC2 is critical for its chromatin compaction activity [31, 77, 88], while the CW-ZF domain could function as a histone methylation reader [89, 90]. In addition, the C-terminal CL domain exhibits a versatile binding capability and can recognize methylated peptides in both histones and nonhistone proteins [91, 92]. To map the domain of MORC2 responsible for binding SETDB1, we generated three truncation constructs (termed T1-T3; Fig. 5D), and then assessed their interactions by co-IP assays in HEK293T cells. The results showed that the N-terminus of MORC2 containing the ATPase domain was required for its interaction with SETDB1 (Fig. 5E). The human SETDB1 protein contains two adjacent Tudor domains (Tudor 1 and 2) at its N-terminus and a C-terminal SET domain (Fig. 5F). The Tudor domains mediate protein-protein interactions by binding to proteins with dimethylated arginine or lysine residues as well as to methylated histone tails [93]. The C-terminal SET domain, in contrast, is essential for its methyltransferase activity [94]. To identify the MORC2-interacting domains within SETDB1, we generated two truncation mutants (T1-T2; Fig. 5F). Subsequent co-IP assays revealed that the N-terminal region of SETDB1 was essential for its association with MORC2 (Fig. 5G). We next sought to determine whether SETDB1 contributes to MORC2-dependent TE silencing and ISG induction. To investigate this, we first analyzed the publicly available ChIP-seq dataset for H3K9me3 in NC and MORC2-KO human chronic myeloid leukemia K562 cells [32]. The analysis revealed a reduction in H3K9me3 levels at the LTR22C2 and MER74C loci upon MORC2 depletion (Fig. 5H), suggesting that MORC2 and H3K9me3 could be involved in their repression. Based on these clues and our above findings (Fig. 3), we selected LTR22C2 and MER74C as representative TEs to study the silencing mechanisms mediated by the MORC2/SETDB1 pathway. To address this, we first stably knocked down SETDB1 in MORC2-overexpressing MDA-MB-231 and BT-549 cells (Supplementary Figure S13B). RT-qPCR assays demonstrated that SETDB1 knockdown effectively reversed the transcriptional repression of both LTR22C2 and MER74C as well as their downstream ISGs (CXCL9 and IFIT3) induced by MORC2 overexpression (Fig. 5I-J). To further delineate the underlying mechanism for these effects, we examined H3K9me3 enrichment at LTR22C2/MER74C loci via ChIP-qPCR assays. The results showed that MORC2 overexpression enhanced H3K9me3 levels at LTR22C2 and MER74C loci, and this effect was notably abolished upon SETDB1 knockdown (Fig. 5K). CUT&RUN assays revealed that MORC2 and SETDB1 co-occupied a substantial fraction of TE loci across the genome (Supplementary Figure S13C-F). This co-enrichment was further confirmed at representative TE loci (LTR22C2 and MER74C) by CUT&RUN-qPCR assays (Supplementary Figure S13G). Taken together, these findings demonstrate that MORC2 collaborates with SETDB1 to establish H3K9me3 marks at TE loci, leading to their silencing and subsequently inhibiting interferon response. Methylation of MORC2 by SETDB1 enhances its protein stability Given the observed physical interaction between MORC2 and SETDB1 (Fig. 5A-G), we next investigated the possibility of their reciprocal regulation. mIHC staining showed that SETDB1 expression was significantly upregulated in TNBC tissues (Fig. 6A-B) and was positively correlated with MORC2 expression (Fig. 6C-D). Consistent with these findings, analysis of the proteomic datasets from both CBCGA [40] and CPTAC [44] cohorts revealed a significant positive correlation between their expression levels in TNBC (Fig. 6E-F). These results suggest a positive relationship between those two proteins. SETDB1-mediated methylation enhances MORC2 protein stability. A-D mIHC analysis of MORC2 and SETDB1 expression in a TNBC tissue microarray comprising 143 TNBC specimens and 25 normal breast tissues. A Representative mIHC images of SETDB1. B Quantification of SETDB1 expression using ImageJ with statistical comparison (Wilcoxon test). **, p < 0.01. C Representative images of SETDB1 staining in TNBC tissues with high or low MORC2 expression. D Spearman correlation coefficient and p-value between MORC2 and SETDB1 expression are shown. Scale bar: 20 μm. E-F Spearman correlation analysis of MORC2 and SETDB1 expression across normal breast tissues, non-TNBC, and TNBC tissues, based on proteomic datasets from the CBCGA (E) and CPTAC (F) cohorts. G Immunoblotting analysis of MORC2 expression levels in NC and SETDB1-KO MDA-MB-231 and BT-549 cells. H NC and SETDB1-KO MDA-MB-231 and BT-549 cells were treated with or without 100 µg/mL cycloheximide (CHX) for the indicated durations and subjected to immunoblotting with the indicated antibodies. Relative MORC2 protein levels were quantified using ImageJ. I-L HEK293T cells were transfected with the indicated plasmids for 48 h and then subjected to sequential co-IP and immunoblotting with the indicated antibodies. M MDA-MB-231 and BT-549 cells were treated with increasing doses of the SETDB1 inhibitor mithramycin A (MMA) for 24 h, followed by immunoblotting with the indicated antibodies. N MDA-MB-231 and BT-549 cells stably expressing wild-type (WT) or 2KR-MORC2 were treated with or without 100 µg/mL CHX for the indicated times and subjected to immunoblotting with the indicated antibodies. Relative MORC2 protein levels were quantified using ImageJ We next investigated the mutual regulatory relationship between these two proteins. Our data showed that while MORC2 knockout did not alter SETDB1 levels (Supplementary Figure S14A), SETDB1 depletion substantially reduced MORC2 protein expression without altering its transcript levels (Fig. 6G and Supplementary Figure S14B). These results indicate that SETDB1 regulates MORC2 at the post-transcriptional level. In support of this notion, cycloheximide (CHX) chase assays demonstrated that SETDB1 depletion significantly shortened the half-life of the MORC2 protein (Fig. 6H and Supplementary Figure S14C), suggesting that SETDB1 enhances MORC2 protein stability. Given the ability of the histone methyltransferase SETDB1 to methylate non-histone substrates such as Akt and p53 [95–97], we next investigated its potential in regulating MORC2 methylation. As shown in Fig. 6I, knockdown of SETDB1 led to a significant reduction in the methylation levels of MORC2. Conversely, ectopic expression of wild-type SETDB1 significantly enhanced MORC2 methylation levels, in contrast to its catalytically inactive mutants (H1224K or C1226A) [94, 98] (Fig. 6J). To identify SETDB1-dependent methylation sites on MORC2, we immunoprecipitated MORC2 from HEK293T cells and analyzed it by mass spectrometry. Using this approach, we identified two potential methylation sites on lysine residues K234 and K643 of MORC2, which are localized in its N-terminal ATPase domain and central CW-ZF domain, respectively (Supplementary Figure S14D). Two additional potential methylation sites (K683 and K986) in MORC2 were also predicted using the PhosphoSitePlus (http://www.phosphosite.org) database and the computational tool GPS-MCP (Group-based Prediction System-Methyl-group Specific Predictor) (http://msp.biocuckoo.org/), respectively (Supplementary Figure S14E). To validate these candidate methylation sites in MORC2, we individually mutated each lysine (K) to arginine (R) as described previously [95, 96], and evaluated its methylation levels in the presence of SETDB1. As shown in Fig. 6K, individual K234R or K643R mutation partially reduced MORC2 methylation compared to the other mutants. Consistently, the double mutant (K234R/K643R, termed 2KR) exhibited a more pronounced, significant reduction in MORC2 methylation (Fig. 6L, compare lanes 3 and 2). Moreover, treatment with the SETDB1 inhibitor mithramycin A (MMA) [96] substantially abrogated SETDB1-induced methylation of WT MORC2 but produced no evident effect on the 2KR mutant (Fig. 6L, compare lanes 5 and 4). These results collectively identify K234 and K643 as the major methylation sites in MORC2. Importantly, both sites are highly conserved across species (Supplementary Figure S14F). These results demonstrate that methylation of MORC2 at K234 and K643 is mediated by SETDB1. We next investigated whether methylation of MORC2 by SETDB1 contributes to its protein stability. As shown in Supplementary Figure S14G, wild-type SETDB1 enhanced MORC2 protein expression levels. In contrast, catalytically deficient SETDB1 mutants (H1224K and C1226A) [94, 98] had no significant effect. This result indicates that regulation of MORC2 by SETDB1 depends on its methyltransferase activity. In support of this notion, the SETDB1 inhibitor MMA [96] decreased MORC2 protein expression levels in a dose-dependent manner (Fig. 6M). These findings suggest that SETDB1 regulates MORC2 protein expression, at least in part, via its methyltransferase activity. CHX chase assays further demonstrated that the 2KR (K234R/K643R) mutant had a significantly shorter half-life than WT MORC2, indicating that SETDB1-mediated methylation enhances MORC2 protein stability (Fig. 6N and Supplementary Figure S14H). Moreover, reintroduction of WT MORC2, but not the 2KR mutant, into MORC2-knockout cells suppressed both TE activation and ISG induction (Supplementary Figure S14I-J). Collectively, our findings demonstrate that SETDB1 interacts with MORC2 to methylate it, thereby promoting MORC2 protein stability, which is required for its immunosuppressive function. Targeting MORC2 with antisense oligonucleotides (ASOs) potentiates the efficacy of ICB therapy in TNBC Given the established role of MORC2 in suppressing interferon response and antitumor immunity (Figs. 2, 3 and 4), we hypothesized that genetic or pharmacological inhibition of MORC2 could act synergistically with immunotherapy to improve outcomes in TNBC. To translate this concept into a therapeutic strategy, we investigated whether targeting MORC2 with antisense oligonucleotides (ASOs) could enhance the response to ICB in TNBC mouse models. To this end, we first evaluated three Morc2a-specific ASOs (termed ASO-Morc2a) in 4T1 cells. All three significantly reduced Morc2a expression at a concentration of 50 nM. Among them, ASO #3 exhibited the highest potency and was consequently selected for subsequent animal experiments (Supplementary Figure S15A). Next, we evaluated the effects of ASO-Morc2a on cell proliferation using colony formation assays. The results showed that ASO-mediated knockdown of Morc2a did not produce a discernible suppressive effect on the growth of 4T1 cells in vitro (Supplementary Figure S15B-D). We therefore proceeded to explore its therapeutic potential in vivo by administering a chemically modified ASO with high stability and bioavailability to 4T1 tumor-bearing BALB/c mice (Fig. 7A). The results demonstrated that ASO-Morc2a administration potentiated the efficacy of anti-PD-1 therapy in this model (Fig. 7B-C). ASO-mediated Morc2a silencing was well tolerated in mice, with no significant body weight loss observed (Supplementary Figure S15E). Robust on-target knockdown was confirmed in harvested tumor tissues (Supplementary Figure S15F-H). qPCR analysis further revealed significant upregulation of ISGs, including Cxcl9 and Ifit3, in ASO-treated tumors (Supplementary Figure S15I). Flow cytometric analysis showed that ASO plus anti-PD-1 combination therapy shifted macrophage polarization toward the M1 phenotype (Fig. 7D-F) and significantly increased CD8⁺ T cell infiltration, with elevated IFN-γ⁺ and GzmB⁺ cytotoxic subsets (Fig. 7G-I), indicating broad tumor microenvironment remodeling toward a pro-inflammatory, anti-tumor state. These findings collectively suggest that targeting MORC2 effectively sensitizes TNBC tumors to anti-PD-1 therapy. Targeting MORC2 sensitizes TNBC to immunotherapy. A Schematic of the in vivo treatment regimen. A total of 3 × 10⁵ 4T1 cells were inoculated into the mammary fat pad of 5-week-old female BALB/c mice. When tumor volumes reached 50–100 mm³, mice received intratumoral (i.t.) injections of ASO-Morc2a (5 nmol) or PBS (50 µL) three times per week for a total of six doses, combined with intraperitoneal (i.p.) administration of anti-PD-1 antibody (10 mg/kg) or isotype control at the indicated time points (n = 6 per group). B-C Treatment efficacy was assessed by tumor volume (B) and final tumor weight (C). D-F Flow cytometry analysis of macrophage populations in control, anti-PD-1, and anti-PD-1 plus ASO tumors (n = 6). The proportions of total macrophages (CD11b⁺F4/80⁺), M1 macrophages (iNOS⁺), and M2 macrophages (CD206⁺) are shown in (D), E, and F, respectively. G-I Flow cytometry analysis of T cell populations in control, anti-PD-1, and anti-PD-1 plus ASO tumors (n = 6). The proportions of CD8⁺ T cells (G), IFN-γ⁺CD8⁺ T cells (H), and GzmB⁺CD8⁺ T cells (I) are shown. J Percentages of CD8⁺ T cells in the tissue microarray containing 143 TNBC tissues, stratified by MORC2 expression levels (MORC2ˡᵒʷ vs. MORC2ʰⁱᵍʰ based on median relative fluorescence intensity). K Spearman correlation between MORC2 expression and CD8⁺ T cell infiltration in the TNBC tissue microarray. L-M Representative mIHC images of MORC2 and GZMB co-staining in the TNBC tissue microarray (L). Relative expression levels quantified by ImageJ are shown in (M). Scale bar: 20 μm. N-O Representative mIHC images of MORC2 and IFN-γ co-staining in the TNBC tissue microarray (N). Relative expression levels quantified by ImageJ are shown in (O). Scale bar: 20 μm. P-Q Spearman correlation analysis between MORC2 and GZMB (P), and between MORC2 and IFN-γ (Q) expression in the TNBC tissue microarray. R Proposed working model. MORC2 collaborates with SETDB1 to transcriptionally suppress TEs, thereby maintaining an immune-suppressive TME. In contrast, therapeutic ablation of MORC2 via ASOs derepresses TEs, leading to the accumulation of endogenous nucleic acids. These nucleic acids activate pattern recognition receptor pathways, triggering a robust interferon response and subsequent anti-tumor immunity, which ultimately sensitizes tumors to immune checkpoint blockade therapy. Notes: P-values were calculated two-way repeated measures ANOVA followed by Tukey's post-hoc test (B), one-way ANOVA (C-I), two-tailed Student's t-test (J, M, O), and Spearman's correlation analysis (K, P, Q). *, **, and *** indicate statistically significant at p < 0.05, p < 0.01, and p < 0.001 level, respectively. ns, no significance MORC2 expression levels are inversely correlated with CD8⁺ T cell infiltration and activation in clinical TNBC samples To assess the clinical relevance of our findings, we analyzed a tissue microarray comprising 143 clinical TNBC samples to evaluate the relationship between MORC2 expression and T cell infiltration and activation. The results demonstrated that MORC2-low tumors displayed significantly enhanced CD8⁺ T cell infiltration (Fig. 7J-K). Furthermore, the production of GZMB and IFN-γ was markedly elevated in MORC2-low tumors (Fig. 7L-O), and both molecules exhibited a negative correlation with MORC2 expression levels (Fig. 7P-Q). Collectively, these findings suggest that MORC2 ablation enhances the efficacy of anti-PD-1 therapy by promoting CD8⁺ T cell infiltration and activation, highlighting the therapeutic potential of targeting MORC2 to improve immunotherapy outcomes in TNBC. Discussion In this study, we identified MORC2 as a key collaborator with SETDB1 in suppressing TEs and restraining downstream interferon signaling within tumor cells, resulting in tumor immune microenvironment reprogramming and impaired anti-tumor immunity. Furthermore, we showed that ASO-mediated silencing of MORC2 elicits a robust T-cell-mediated anti-tumor immune response and potently enhances the efficacy of anti-PD-1 therapy in preclinical TNBC models. These findings support a promising combinatorial strategy that utilizes ASO-based targeting of MORC2 to overcome immunotherapy resistance in TNBC (Fig. 7R). First, we report that MORC2 is a potential therapeutic target for overcoming immunotherapy resistance in TNBC. Based on its association with reduced TIL infiltration, worse overall survival, and differential expression between IM and non-IM subtypes (Fig. 1 and Supplementary Figures S2-S4), MORC2 may serve as a candidate predictor reflecting tumor microenvironment immunosuppression status. However, formal prospective validation with multivariate modeling will be required before clinical adoption. Additionally, MORC2 deletion suppressed TNBC syngeneic tumor growth in immunocompetent but not immunodeficient mice, and ASO-mediated Morc2a silencing potentiated anti-PD-1 efficacy in the 4T1 model (Figs. 1 and 7), underscoring the immune-dependent nature of MORC2’s tumor-promoting function. ASO-Morc2a selectively reduced Morc2a expression without affecting Morc3 or Morc4 (Supplementary Figure S15F-H), confirming on-target specificity, and no overt systemic toxicity was observed at the doses employed (Supplementary Figure S15E). The overexpression of MORC2 in TNBC relative to normal breast tissue suggests a potential therapeutic window; nonetheless, normal tissue toxicity and comprehensive off-target transcriptomic assessment remain important priorities for future preclinical and clinical translation. Validation in additional models, including patient-derived xenografts with humanized immune systems, will be important for establishing the generalizability of MORC2 targeting across the heterogeneous TNBC disease spectrum. Second, we elucidated the mechanism by which MORC2 modulates anti-tumor immunity. MORC2 deletion derepresses TEs, triggering an interferon response that remodels the tumor immune microenvironment (Fig. 2). Integrative analysis of scRNA-seq and bulk RNA-seq data identified MORC2 as a key regulator of tumor-intrinsic interferon signaling that shapes the immunological landscape of tumor microenvironment in TNBC (Fig. 3). scRNA-seq of public TNBC datasets further confirmed predominant MORC2 enrichment in epithelial tumor cells versus immune subsets (Supplementary Figure S4F-G), suggesting that MORC2 acts primarily through tumor-intrinsic effects; however, potential modulation of MORC2 in immune compartments following systemic ASO delivery cannot be excluded and warrants further investigation. Notably, MORC2 deletion in TNBC cells derepresses TEs, whose reactivation activates both the MDA5-MAVS (dsRNA) and cGAS-STING (dsDNA) nucleic acid-sensing pathways and potently induces interferon responses (Figs. 3 and 4 and Supplementary Figures S9-11). These data provide translational evidence that therapeutic MORC2 inhibition may induce interferon responses to enhance anti-tumor immunity. The phenomenon of epigenetic de-repression of TEs triggering a viral mimicry response has been documented for several chromatin regulators, including DNMT1/3A, EZH2, and LSD1. MORC2 is mechanistically distinct from these regulators in that it collaborates with SETDB1 to mediate H3K9me3 deposition rather than acting through DNA methylation or H3K27me3/H3K4me1 pathways, representing a potentially complementary and non-redundant layer of TE repression. This mechanistic distinction raises the possibility that combined inhibition of MORC2 with DNMT or EZH2 inhibitors could produce additive or synergistic effects on TE de-repression and immune activation. Additionally, we discovered that MORC2 collaborates with SETDB1 to transcriptionally repress TEs via H3K9me3 deposition at TE loci. Although the epigenetic regulation of TE transcription has been well characterized in mammalian germ cells and early embryonic development [99, 100], its functional relevance in tumor cells remains poorly understood. Our study shows that MORC2 interacts with SETDB1 to act as a transcriptional repressor of specific TEs, and that MORC2 depletion reduces SETDB1-dependent H3K9me3 deposition, leading to TE reactivation and subsequent interferon response (Fig. 5 and Supplementary Figures S12-14). The preferential co-occupancy of MORC2 and SETDB1 at TE loci, together with broad multi-locus TE de-repression upon MORC2 loss, raises fundamental questions about the mechanisms governing locus specificity. As neither MORC2 nor SETDB1 possesses intrinsic DNA sequence recognition activity, locus-specific targeting is likely conferred upstream by KRAB-ZFPs and their co-repressor KAP1/TRIM28, which recognize specific TE sequences and recruit the SETDB1-containing repressive machinery. The preferential effect of MORC2 loss on ERV elements—but not LINE elements at the class level—aligns with the known biology of KRAB-ZFP-mediated silencing, which disproportionately targets LTR retrotransposons, particularly ERVK subfamily members, in mammalian cells [101, 102]. The class-level sparing of LINE elements, with upregulation restricted to young retrotransposition-competent L1 elements, further argues that MORC2-dependent silencing is not a general heterochromatin maintenance mechanism but rather a selective repressive function targeting evolutionarily recent, transcriptionally active TE families that retain the greatest immunostimulatory potential. Whether MORC2’s ATPase activity contributes to chromatin remodeling at these loci—for instance, by facilitating nucleosome repositioning to enable SETDB1-mediated H3K9me3 spreading—remains an important question. Third, we demonstrate that SETDB1 induces MORC2 methylation and enhances its protein stability. Although SETDB1 is widely recognized as a histone methyltransferase that catalyzes H3K9me3 [85, 86], growing evidence indicates its capacity to methylate non-histone proteins such as AKT [95, 96], thereby modulating their activity and promoting cancer progression. Our study identifies MORC2 as a novel substrate of SETDB1, showing that SETDB1 interacts with MORC2 and promotes its methylation-dependent stabilization (Fig. 6 and Supplementary Figure S14). These findings introduce an additional level of intricacy to the post-translational modification (PTM) code in the regulation of the biological functions of MORC2. Developing specific antibodies targeting the methylation sites of MORC2 is highly necessary in future research. Given that MORC2 dimerization has recently been shown to maintain its stability and activity by shielding it from chaperone-mediated autophagy (CMA)-dependent lysosomal degradation [103], future biochemical and cellular studies are required to examine whether SETDB1-mediated methylation influences MORC2 dimer formation and, consequently, its stability. Moreover, leveraging the mechanistic understandings discovered in this study, experimentally determining whether SETDB1 inhibitors can counter the phenotypes triggered by high levels of MORC2 expression and subsequently boost the efficacy of immunotherapy in TNBC is of great significance. In summary, our study unveils a pivotal mechanism by which MORC2 suppresses tumor immunogenicity in TNBC. Specifically, we demonstrate that specifically inhibiting MORC2 with ASOs stimulates the transcription of TEs, triggers the activation of interferon signaling, and fosters anti-tumor T-cell responses. Together, these effects restrain TNBC progression and synergize with anti-PD-1 therapy, positioning MORC2 targeting as a promising therapeutic strategy to overcome immune evasion in this aggressive malignancy. Supplementary Information Supplementary Material 1. Supplementary Material 2. Acknowledgements We are grateful to the clinical and basic research teams of the Department of Breast Surgery and the Shanghai Key Laboratory of Breast Cancer at Fudan University Shanghai Cancer Center for their technical and resource support throughout this study. Abbreviations antisense oligonucleotides co-immunoprecipitation histone H3 lysine 9 trimethylation immune checkpoint blockade immunofluorescence MORC family CW-type zinc finger 2 programmed cell death protein 1 real-time quantitative PCR SET domain bifurcated histone lysine methyltransferase 1 short hairpin RNA transposable elements triple-negative breast cancer Authors’ contributions FLZ: Data curation, formal analysis, investigation, writing the manuscript draft, and funding acquisition. SYY, QZ, Lisa, WXY, and JYC: Data analysis and investigation. YLZ: Funding acquisition and technical assistance. XH, ZMS, and AYC: Clinical specimens and research resources. DQL: Funding acquisition, project administration, and manuscript editing. Funding The work in the Li laboratory is supported, in whole or in part, by the National Natural Science Foundation of China (82573892 and 82072918 to FLZ, and 82372633 and 82573776 to DQL). Data availability all data associated with this study are present in the paper or the Supplementary Information. Any additional information required to reanalyze the data reported in this manuscript is available from the corresponding authors upon reasonable request. Declarations Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Fang-Lin Zhang, Shao-Ying Yang, Yin-Ling Zhang and Qian Zhao contributed equally to this work. Contributor Information Fang-Lin Zhang, Email: zhangfanglin555@sina.com. A-Yong Cao, Email: caoayong@fudan.edu.cn. Da-Qiang Li, Email: daqiangli1974@fudan.edu.cn. References Associated Data Supplementary Materials Supplementary Material 1. Supplementary Material 2. Data Availability Statement all data associated with this study are present in the paper or the Supplementary Information. Any additional information required to reanalyze the data reported in this manuscript is available from the corresponding authors upon reasonable request.