You are here

  • Home
  • Archive
  • Volume 11, Issue 8
  • Inhibition of tumor intrinsic BANF1 activates antitumor immune responses via cGAS-STING and enhances the efficacy of PD-1 blockade

Article Text

Article menu
Basic tumor immunology
Original research
Inhibition of tumor intrinsic BANF1 activates antitumor immune responses via cGAS-STING and enhances the efficacy of PD-1 blockade
  1. Minglei Wang1,2,
  2. Yiheng Huang1,
  3. Minxin Chen1,
  4. Weiyan Wang3,
  5. Fei Wu1,
  6. Tao Zhong1,
  7. Xiaozheng Chen1,
  8. Fei Wang1,
  9. Yang Li1,
  10. Jinming Yu1,2,4,
  11. Meng Wu1 and
  12. Dawei Chen1,2
  1. 1Department of Radiation Oncology and Shandong Provincial Key Laboratory of Radiation Oncology, Shandong Cancer Hospital and Institute Shandong First Medical University and Shandong Academy of Medical Sciences, Jinan, Shandong, China
  2. 2Department of Oncology, Shandong University Cancer Center, Jinan, Shandong, China
  3. 3School of Basic Medical Sciences, Shandong First Medical University, Jinan, Shandong, China
  4. 4Research Unit of Radiation Oncology, Chinese Academy of Medical Sciences, Jinan, Shandong, China
  1. Correspondence to Dr Dawei Chen; dave0505{at}yeah.net; Dr Jinming Yu; sdyujinming{at}163.com; Dr Meng Wu; wumeng7777{at}163.com

Abstract

Background BANF1 is well known as a natural opponent of cyclic GMP-AMP synthase (cGAS) activity on genomic self-DNA. However, the roles of BANF1 in tumor immunity remain unclear. Here, we investigate the possible impact of BANF1 on antitumor immunity and response to immunotherapy.

Methods The Cancer Genome Atlas public data were analyzed to evaluate the relevance of the expression of BANF1, patients’ survival and immune cell infiltration. We monitored tumor growth and explored the antitumor efficacy of targeting tumor-intrinsic BANF1 in combination with anti-programmed cell death protein-1 (PD-1) in MC38 or B16F10 tumor models in both immunocompetent and immunodeficient mice. Flow cytometry, immunofluorescence and T cells depletion experiments were used to validate the role of BANF1 in tumor immune microenvironment reprogramming. RNA sequencing was then used to interrogate the mechanisms how BANF1 regulated antitumor immunity.

Results We show that upregulated expression of BANF1 in tumor tissues is significantly associated with poor survival and is negatively correlated with immune cell infiltration. Deficiency of BANF1 in tumor cells markedly antagonizes tumor growth in immunocompetent but not immunocompromised mice, and enhances the response to immunotherapy in murine models of melanoma and colon cancer. In the immunotherapy clinical cohort, patients with high BANF1 expression had a worse prognosis. Mechanistically, BANF1 knockout activates antitumor immune responses mediated by cGAS-synthase-stimulator of interferon genes (cGAS-STING) pathway, resulting in an immune-activating tumor microenvironment including increased CD8+ T cell infiltration and decreased myeloid-derived suppressor cell enrichment.

Conclusions BANF1 is a key regulator of antitumor immunity mediated by cGAS-STING pathway. Therefore, our study provides a rational that targeting BANF1 is a potent strategy for enhancing immunotherapy for cancer with BANF1 upregulation.

  • Immune Checkpoint Inhibitors
  • Lymphocytes, Tumor-Infiltrating
  • Melanoma
  • Tumor Microenvironment
  • CD8-Positive T-Lymphocytes

Data availability statement

Data are available on reasonable request.

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See http://creativecommons.org/licenses/by-nc/4.0/.

https://doi.org/10.1136/jitc-2023-007035

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    WHAT IS ALREADY KNOWN ON THIS TOPIC

    • Activation of cyclic GMP-AMP synthase-stimulator of interferon genes (cGAS-STING) in cancer cells functions in restricting tumor growth, and recruiting immune cells for tumor clearance by promoting the expression of inflammatory genes.

    • Despite BANF1 has been identified as a natural opponent of cGAS activity on genomic self-DNA, the understanding of BANF1 in tumor immunity has not been elucidated yet.

    WHAT THIS STUDY ADDS

    • Upregulated BANF1 is significantly associated with poor prognosis and less T cell infiltration.

    • BANF1 knockout antagonized tumorigenesis in immunocompetent but not immunocompromised mice and re-expression of BANF1 reversed this inhibitory effect.

    • BANF1 knockout reshaped an immune-activating tumor microenvironment, including increased CD8+ T cell infiltration and decreased myeloid-derived suppressor cell enrichment.

    • BANF1 knockout activated antitumor response mediated by cGAS-STING pathway.

    • Combining BANF1 knockout with anti-PD-1 improved therapeutic benefit as compared with anti-PD-1 alone.

    HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

    • Targeting BANF1 may synergize with immune checkpoint blockade for patients with solid cancer with BANF1 upregulation.

    Introduction

    Immune checkpoint inhibitors (ICIs) have revolutionized the treatment pattern of various cancers, including melanoma and non-small cell lung cancer (NSCLC).1–3 However, there are still a considerable proportion (40%–60%) of patients who do not respond or do not achieve durable responses to these drugs.4 Thus, it is urgent clinical need to improve the efficacy of ICIs by investigating novel combination strategy to extend their use to a larger number of patients and tumor types.

    One of the major causes of non-responsiveness to ICIs is insufficient tumor infiltration of immune cells, especially effector CD8+ T cells.5 6 Innate immune response is found to be vital for CD8+ T cell infiltration.7 The cyclic GMP-AMP synthase (cGAS)-stimulator of interferon genes (STING) signaling pathway is a key regulator of innate and adaptive immunity by promoting the release of type I interferons (IFNs) and inflammatory cytokines.8–10 Thus, accumulating studies have focused on the regulation of this pathway. Targeting epigenetic modifier PRMT5, knockout of deubiquitinase OTUD5, DNA methyltransferase inhibition, Mn2+ administration, and inhibition of DNA damage checkpoint have been all shown to induce antitumor immunity via cGAS-STING pathway activation.11–15 These evidences support the notion that control of cGAS-STING pathway, may overcome resistance to ICIs.

    Barrier-to-autointegration factor 1 (BAF or BANF1) is a small (10 kDa), highly conserved and abundant DNA-binding protein.16–18 By binding to double-stranded DNA (dsDNA), histones and other partners, BANF1 is involved in multiple cellular processes, including innate immunity, postmitotic nuclear reformation, repair of nuclear envelope rupture, and the DNA damage and repair response.19–22 Loss of BANF1 is lethal during embryogenesis in Caenorhabditis elegans and Drosophila melanogaster.18 23 In mouse and human embryonic stem cells, deletion of BANF1 impairs the pluripotent state.24 As for human tumors, increased expression of BANF1 have been implicated poor prognoses in certain cancers.25 26 But, the role of upregulated BANF1 in cancer is not yet clear. As an essential DNA-binding protein, BANF1 could rapidly binds to cytosolic dsDNA of foreign or endogenous origin. Interestingly, recently studies provide evidence demonstrating that BANF1 restricts cGAS-STING pathway to suppress innate immune responses.27 28 These facts drive us to explore whether elevated BANF1 in tumor cells plays a part in regulating antitumor immunity.

    Here, we integrate analyses of both The Cancer Genome Atlas (TCGA) pan-cancer cohort and murine tumor models, and find that genetic deficiency of the BAF gene BANF1 in melanoma and colorectal cancer cells reprograms the tumor microenvironment (TME) into a hot inflamed T cell-infiltrated tumor. Our data show that such reprograming is mechanistically related to the innate immune response activated by cGAS-STING pathway. Furthermore, we provide evidence that targeting BANF1 can suppress tumor progression and sensitize preclinical models to anti-PD-1 antibody treatment.

    Methods

    Cell culture and treatment

    The melanoma B16F10 cell line was purchased from American Type Culture Collection. MC38 cell line was generously provided from Liufu Deng’s lab in Shanghai Jiao Tong University School of Pharmacy. Cells were cultured in 1640 or DMEM with 10% fetal bovine serum (Gibco), 100 U/mL penicillin and 0.1 mg/mL streptomycin (Sangon biotech) at 37°C in humidified 5% CO2 atmosphere. All cell lines were tested to be mycoplasma free. cGAS knockout (KO) MC38 and STING KO MC38 were from Liufu Deng’s lab. Similar to cGAS KO MC38 and STING KO MC38 cell lines, BANF1 KO MC38 cells and BANF1 KO B16F10 cell were generated using CRISPR/Cas9 system. Two single guided RNA (sgRNA) sequences targeting murine Banf1 were as follows: #2, ATGAAGACCTCTTCCGAGAA; #4, AAAGCAGTCCCGGGACTGCT, which were designed by Zhang lab (https://www.zlab.bio/CRISPOR). Briefly, cells were infected with either BANF1 control or BANF1 KO lentiviral particles purchased from Shanghai Genechem. Then these cells were selected by puromycin in 96-well plates and identified by western blot. H-151 was purchased from MCE and administrated to cells according to the manufacturer’ s recommended protocols.

    Animal experiments

    Wild-type C57BL/6 mice and nude mice (BALB/cAnN.Cg-Foxn1nu/CrlNarl) were from Beijing HFK bioscience, China. A volume of 100 µL of 2.0×106 MC38 cells or 3.5×105 B16F10 cells were injected s.c. into the right hind flank of 6-week-old wild-type C57BL/6 or nude mice. Tumor growth was evaluated by measuring with Vernier calipers every 3–4 days, and tumor volume was calculated using the formula: V=L×W2×0.52, where L is the length and W is the width of the tumor. For CD8+ T cells depletion experiments, mice were treated with anti-CD8 antibody (Bio X Cell, #BE0061) or control IgG (200 µg per mouse on days 3, 6, 10, 13, 17 and 20). For ICI experiments, mice were grafted with B16F10 or MC38 cells and treated with 200 µg anti-PD-1 (Bio X cell, #BE0146) or control IgG intraperitoneally two to three times at the indicated date as shown in Figures. All mice were maintained in the Specific-Pathogen-Free Animal Research Center of Shandong Cancer Hospital.

    Database analysis

    TCGA pan-cancer gene expression data and clinical data were obtained from UCSC Xena (https://xenabrowser.net). BANF1 expression was analyzed with TIMER2.0. The following analyses were done by R software: Package survival was used for prognostic analysis and the best cut-off was calculated by package Maxstat. Package edgeR was used for differential gene analysis and the pathway enrichment was analyzed by gene set enrichment analysis (GSEA) with Hallmark gene sets. Package ESTIMATE was used to infer the immune infiltration score of each sample. The correlation between BANF1 and ImmuneScore was calculated by the method of pearson.

    RNA sequencing

    Total RNA from BANF1 KO MC38 and BANF1 control MC38 cells was extracted using TRIzol Reagent (Sigma, #T9424). The quality of RNA samples was determined with the Agilent 2100 Bioanayzer (Agilent Technologies). Only samples passing quality control were used for RNA sequencing and data analysis (CapitalBio Technology, Beijing, China). Data analysis was performed using R software (V.4.0.1). The RNA-seq data will be shared on reasonable request to the corresponding author.

    Flow cytometry analysis

    Tumor were harvested on indicated days, dissociated into small pieces, digested with collagenase I and DNase I for 30 min at 37°C, and filtered to obtain the single-cell suspensions. The cells were then counted and blocked with Fc block (anti-mouse CD16/32, BD, #553141) for 15 min on ice. The samples were first stained for surface markers followed by intracellular staining according to the manufacturer’s recommended protocol. The following antibodies were used: BV785/786 anti-mouse CD45.2 (Biolegend, #109839), BV605 anti-mouse Ly6C (Biolegend, #128035), BV711 Ly6G (Biolegend, #127643), PE/Cy7 anti-mouse CD11b (Biolegend, #101216), BV421 anti-mouse F4/80 (Biolegend, #123137), APC anti-mouse CD11c (Biolegend, #117310), PerCP/Cy5.5 anti-mouse MHC II (Biolegend, #107625), FITC anti-mouse CD86 (Biolegend, #553691), PE anti-mouse CD206 (Biolegend, #141706), AF700 anti-mouse CD8 (Biolegend, #100730), BV650 anti-mouse CD4 (Biolegend, #100545), BV510 anti-mouse CD3 (Biolegend, #100233), BV510 anti-mouse Gr1(Biolegend, #108457), APC anti-mouse NK1.1 (BD, #550627), BV650 anti-mouse IFNgamma (Biolegend, #505831), BV711 anti-mouse PD-1 (Biolegend, #135231) and live/dead (Fixable Viability Stain 780, BD, #565388). The samples were determined with flow cytometry and the data were analyzed with FlowJo software.

    Immunofluorescence staining

    Tumor cells seeded on sterile glass coverslips or tumor tissue sections were fixed with 4% paraformaldehyde, blocked in donkey serum and incubated with primary antibodies specific for CD8α (1:200, Cell Signaling Technology, #85336), or dsDNA (1:100, Santa Cruz, #sc-58749). Then the cells/tissues were incubated with fluorescence-conjugated secondary antibodies (1:1000, Invitrogen) and mounted with fluorescent mounting medium containing DAPI (Vector). Cell images were captured with a confocal microscope (Zeiss).

    Immunohistochemistry

    Immunohistochemistry was performed for BANF1 on the tissue chip by using anti-BANF1 antibody (1:100, SIGMA, #HPA039242). The multiorgan tissue chip was obtained from Shanghai Outdo Biotech (#HOrgC120PG04, Shanghai, China, IRB reference no: YB M-05-01), which contained 11 types common tumors, with 2–6 cases each: thyroid cancer 3 cases, esophageal cancer 4 cases, gastric cancer 4 cases, colon cancer 6 cases, rectal cancer 4 cases, liver cancer 6 cases, pancreatic cancer 5 cases, lung squamous carcinoma 4 cases, lung adenocarcinoma 6 cases, breast cancer 6 cases, kidney cancer 5 cases. All patients had been pathologically diagnosed.

    RT-PCR analysis and western blot

    Real-time quantitative PCR (RT-PCR) and western bolt were performed as described previously.29 Primer sequences were listed in online supplemental table 2. The following antibodies were used for western blot: BANF1 (Santa Cruz, #sc-166324), cGAS (Cell Signaling Technology, #31659), STING (Cell Signaling Technology, #13647), p-STING (Cell Signaling Technology, #72971), IRF3 (Cell Signaling Technology, #4302), p-IRF3 (Cell Signaling Technology, #29047), TBK1 (Cell Signaling Technology, #3013), p-TBK1 (Cell Signaling Technology, #5483), Actin (A5441, Sigma-Aldrich).

    Supplemental material

    Statistical analysis

    Statistical analyses were conducted using GraphPad Prism software (V.9.0.0). For comparison of means of two groups, unpaired Student’s t-test was applied. Two-way ANOVA with Tukey’s test was used for tumor growth analysis. For survival curves, p values were calculated using the log-rank (Mantel-Cox) test. All results are shown as mean±SEM (error bars). Data shown are representative of three experiments. Not significant (ns) p>0.05, *p<0.05, **p<0.01, ***p<0.001.

    Results

    BANF1 expression is inversely associated with an antitumor immune signature

    To investigate the relevance of BANF1 to tumor immunity, we first assessed BANF1 expression in TCGA pan-cancers cohort, and found that BANF1 is higher expressed in 15 tumors compared with their corresponding normal samples, such as colon adenocarcinoma (COAD), lung adenocarcinoma (LUAD), rectum adenocarcinoma (READ) and liver hepatocellular carcinoma (LIHC) (figure 1A and online supplemental figure S1A). And there was no difference in six tumors, including kidney chromophobe (KICH), kidney renal clear cell carcinoma (KIRC), pancreatic adenocarcinoma (PAAD), pheochromocytoma and paraganglioma (PCPG), prostate adenocarcinoma (PRAD) and cervical squamous cell carcinoma and endocervical adenocarcinoma (CESC), and another 12 subtypes had no paired normal tissues and could not be compared. Consistently, the enhanced expression of BANF1 was confirmed at the protein level by IHC staining in multiple tumor types, such as COAD, LUAD, LIHC, READ and THCA (figure 1B). Higher BANF1 expression was always positively correlated with poor prognosis (figure 1C and online supplemental figure S1B,C).

    Supplemental material

    Figure 1
    Figure 1

    BANF1 expression is associated with poor survival and reduced immune cell infiltration. (A) Box plots of BANF1 mRNA expression between tumor and adjacent normal tissues in The Cancer Genome Atlas (TCGA) pan-cancer cohorts from The TIMER2.0. Data were analyzed by the Wilcoxon test. *p<0.05; *** p<0.001. (B) Representative immunohistochemistry (IHC) staining images for BANF1 protein expression in tumor and normal tissue of the given cancer patients. (C) Association of BANF1 mRNA level with overall survival (OS) in GBMLGG （Lower grade glioma and Glioblastoma）, HNSC （Head and Neck Cancer）, LIHC, and LUAD cohort from TCGA datasets. (D) BANF1 expression in LUAD cohort (n=513) based on TCGA datasets. Inset shows comparison between low and high BANF1 expression cohorts (top and bottom 100 samples). Data were analyzed by the Wilcoxon test. ***p<0.001. (E) GSEA analysis of all ranked differentially expressed genes in high and low BANF1 group (as showed in D) using Hallmark gene sets. NES, normalized enrichment score. Data were analyzed by one-tailed Fisher’s exact test for multiple comparisons (F) Representative gene sets associated with Inflammatory responses enriched in BANF1 low group. Data were analyzed by one-tailed Fisher’s exact test. (G) Correlation among BANF1 expression with immune cell infiltration score in TCGA pan-cancer cohort. LIHC, liver hepatocellular carcinoma; LUAD, lung adenocarcinoma.

    Then we investigated whether BANF1 expression in tumor cells was associated with particular gene sets (figure 1D). GSEA showed that lung cancer cells with low BANF1 expression exhibited enriched expression of immune-associated signatures, such as Inflammatory response, Allograft rejection, TNFA signaling via NFKB (figure 1E,F). Analysis of COAD and skin cutaneous melanoma (SKCM) cohorts also confirmed enrichment of an immune gene signature in tumors with low BANF1 expression (online supplemental figure S1D–I). In addition to this, we further explored immune cell infiltration according to the ESTIMATE algorithm, and found that BANF1 expression is negatively correlated with Immune score in 15 of 33 TCGA cancer types, especially in COAD (pearson r=−0.14, p<0.02), LUAD (pearson r=−0.28, p<0.001), and SKCM (pearson r=−0.16, p<0.001) (figure 1G). These results supported a potential role of BANF1 in tumor immunity.

    BANF1 knockout decreases tumor growth in syngeneic mice

    In order to assess the function of BANF1 in antitumor immunity, we knocked out the BANF1 gene in two malignant murine cancer cell lines (B16F10 and MC38) using CRISPR-Cas9 technology, respectively (figure 2A). Interestingly, colon cancer model (MC38 allograft model) revealed that tumor growth in immunocompetent C57BL/6 mice and in immunodeficient nude mice show marked differences. In C57BL/6 mice, BANF1 deficiency significantly inhibited tumor growth and tumor formation rate (20%–40% in BANF1 KO group vs 100% in BANF1 control group at day 28; figure 2B). While in nude mice, tumor growth showed slight inhibition in BANF1 KO group, tumor formation rate was not affected (figure 2C). Tumor growth suppression by targeting BANF1 was also confirmed in melanoma cancer model (figure 2D–F). All these suggested that compromised tumor growth of BANF1 absent tumor cells requires an intact immune system.

    Figure 2
    Figure 2

    BANF1 knockout decreases tumor growth in syngeneic mice. (A) Immunoblots showing the sgRNA knockout of BANF1 or sgRNA control in MC38 colon cancer cells (A upper panel) and B16F10 melanoma cells (A down panel). Actin was used as a loading control. (B, C) Growth curves of BANF1 KO (sgBanf1#2 and #4) and BANF1 control (sgNeg) MC38 tumors in C57BL/6 mice (B) and in nude mice (C). The ratio in B represents the tumor formation rate at day 30 (n=4–6 replicates /group). (D–F) Growth curves and tumor weight in grams of BANF1 KO (sgBanf1#2 and #4) and BANF1 control (sgNeg) B16F10 tumors in C57BL/6 mice (D and E) and in nude mice (F). (G) Immunoblots showing BANF1 expression in BANF1 control (sgNeg), BANF1 KO (sgB2) and BANF1 reexpression (sgB2B) MC38 cells. Actin was used as a loading control. (H) Growth curves of BANF1 control (sgNeg), BANF1 KO (sgB2) and BANF1 reexpression (sgB2B) MC38 tumors in C57BL/6 mice. (I) Rechallenge growth curves of BANF1 control (sgNeg) MC38 tumors in treatment naïve C57BL/6 mice and mice that were tumor-free in the first challenge. Statistical significance for tumor growth was calculated by two-way ANOVA with Tukey’s test and Bar graphs were analyzed by unpaired two-tailed Student’s t-test. All results are shown as mean±SEM (error bars). Data shown are representative of three experiments. Not significant (ns) p>0.05, *p <0.05, **p<0.01, ***p<0.001. ANOVA, analysis of variance.

    In order to rule out a potential off-target effect of BANF1-sgRNA, we rescued the BANF1 expression in the BANF1 KO cells by expressing the sgRNA-mut BANF1. As showed in figure 2G,H, re-expression of BANF1 reversed tumor growth inhibition induced by BANF1 KO in MC38 model. In further experiments, we found that those mice, which have achieved tumor-free in BANF1 KO group after initial inoculation, could rejecte the rechallenge with BANF1 wild-type tumor cells (figure 2I), implying that targeting tumor-intrinsic BANF1 could inhibits tumor growth by inducing antitumor immune memory.

    BANF1 knockout enhances the infiltration of antitumor immune effector cells

    We next investigated whether the BANF1-dependent antitumor activity was associated with the antitumor immune response. We analyzed the immune cells infiltration of tumors in allograft model 12 days post tumor inoculation by flow cytometry. In melanoma cancer model, we evaluated T cells (CD4+, or CD8+), NK cells (CD3-NK1.1+), myeloid-derived suppressor cell (MDSC) (CD11b+Gr1+), conventional dendritic cells (cDCs, CD11b+CD11c+), and tumor associated macrophage cells (TAM, CD11b+F4/80hi). Total CD45+ immune cells infiltration in BANF1 KO (sgBanf1) and BANF1 control (sgNeg) group had no significant difference, but BANF1 KO tumors demonstrated increased CD8+ T and CD4+ T cells enrichment and less CD11b+Gr1+ MDSC infiltration (figure 3A). While in MC38 colon cancer model (figure 3B), it showed more CD8+ T cells, less CD11b+Ly6chi MDSCs and less cDCs enriched in BANF1 KO MC38 tumors. No significant difference in the infiltration of CD4+ T cells and TAM was observed between BANF1 KO tumor and control. And the increased infiltration of CD8+ T cells into BANF1 KO tumor was also confirmed by immunofluorescence (figure 3C,D). Furthermore, to assess the impact of BANF1 KO on cytotoxic effects of CD8+ T cells, flow cytometric analysis was performed and showed that the frequency of IFN-γ positive T cells was increased and the frequency of PDCD1 positive T cells was decreased in BANF1 KO tumors (figure 3E,F), implying that BANF1 KO led to enhanced CD8+ T cell activation. In summary, tumor cell intrinsic BANF1 absence lead to boosted antitumor immune responses with increased effector CD8+ T cells infiltration and activation.

    Figure 3
    Figure 3

    BANF1 knockout evokes CD8+ T cell-dependent antitumor immune responses. (A, B) Infiltration of immune cells in BANF1 control (sgNeg) and BANF1 KO (sgBanf1) B16F10 (A) or MC38 (B) tumors grafted into C57BL/6 mice 12 days post-tumor challenge by flow cytometry using indicated markers. (C, D) Infiltration of CD8+ immune cells in BANF1 control (sgNeg) and BANF1 KO (sgBanf1) MC38 tumors 12 days after grafting into C57BL/6 mice was evaluated by immunofluorescence (C) and quantified with ImageJ (D). Data for each tumor 638 (n=5/group) are shown with means and SEM indicated. DAPI, 4′,6-diamidino-2- phenylindole. Scale bar=20 μm. (E) Representative flow cytometry analysis (left) and quantification (right) of the percent of CD8+ IFN- γ + cells among CD8+ T cells infiltrating B16F10 tumors, collected at day 12. (F) Representative flow cytometry analysis (left) and quantification (right) of the percent of CD8+ PDCD1+ cells among CD8+ T cells infiltrating B16F10 tumors, collected at day 12. (G, H) Experimental scheme (G top) and growth curves of BANF1 KO (sgBanf1) and BANF1 control (sgNeg) B16F10 tumors (G) or MC38 (H) in wild-type C57BL/6 mice administered with control or neutralizing antibodies against CD8a (200 μg per mouse, twice per week, starting 3 days after tumor inoculation, n=4–6 replicates /group). Bar graphs were analyzed by unpaired two-tailed Student’s t-test and tumor growth by two-way ANOVA with Tukey’s test. All results are shown as mean±SEM (error bars). Data shown are representative of three experiments. *p<0.05, **p<0.01, ***p<0.001. ANOVA, analysis of variance.

    To substantiate the contribution of increased infiltration of effector CD8+ T cells on the degree of tumor growth, we monitored tumor growth of BANF1 KO tumors after depletion of CD8+ T cells. BANF1 KO tumors grew faster in mice administrated with neutralizing antibodies to CD8+ T cells (anti-CD8a) compared with those mice without (figure 3G,H). These observations indicated that depletion of tumor-derived BANF1 could induce immune cell infiltration into TME, and effector CD8+ T cells are the major immune component controlling the growth of BANF1 KO tumors.

    BANF1 knockout activates innate immune response

    To explore the underlying mechanism of BANF1 KO-induced infiltration of CD8+ T cells, we performed RNA-seq analysis on BANF1 KO and BANF1 control MC38 cells. In particular, differential gene expression analysis revealed that 2906 upregulated genes and 2278 downregulated genes were identified (p<0.05, logFC≥1; figure 4A; online supplemental table 1). GSEA was performed to functionally annotate the expression differences, the IFNα and IFNγ innate immune response pathways were two of the top enriched pathways triggered by BANF1 KO (figure 4B,C). Among significant differential genes in BANF1 KO cells, we observed increased expression of interferon stimulated genes (ISGs) overlapped in IFNα and IFNγ pathways (figure 4D). Validation by RT-qPCR demonstrated that multiple ISGs including Rsad2, Mx1, Isg15, Irf7, Cxcl10, and Ifnb1 were expressed at least ten-fold higher levels in BANF1 KO compared with BANF1 control tumor cells (figure 4E). Cxcl9, Cxcl10, and Cxcl11 were critical chemokines for effector T cell recruitment into tumor.30 31 As we all know, the protein kinase Tbk1 and transcription factor Irf3 are major effectors of the interferon pathway and have been reported to regulate the expression of chemokines such as Cxcl10 and Cxcl11 and other ISGs. Immunoblotting revealed that BANF1 KO promoted the phosphorylation of Tbk1 and Irf3 at protein level, implying activation of interferon pathway in BANF1 KO tumor cells (figure 4F). These data suggested that BANF1 deficiency in tumor cells may facilitate T cells infiltration by inducing innate immune response and affecting chemokine and cytokine expression.

    Supplemental material

    Figure 4
    Figure 4

    BANF1 knockout activates innate immune responses. (A) Volcano plot of log2 fold change for genes significantly upregulated (red, right) or downregulated (green, left) in BANF1 KO MC38 cells, compared with BANF1 control cells. (B) GSEA analysis of all ranked differentially expressed genes in BANF1 KO (sgBanf1) versus BANF1 control (sgNeg) group (as showed in A) using Hallmark gene sets. Data were analyzed by one-tailed Fisher’s exact test for multiple comparisons. (C) Representative gene sets associated with IFN responses enriched in BANF1 KO (sgBanf1) group. Data were analyzed by one-tailed Fisher’s exact test. (D) Heatmap showing expression of the shared genes among IFN-α and IFN-γ responses in BANF1 KO (sgBanf1) versus BANF1 control (sgNeg) group. (E) Validation of interferon stimulating genes (ISGs) by RT-qPCR in BANF1-KO (sg#2 and #4) compared to BANF1 control (sgNeg) MC38 cells (n=3/group). (F) Immunoblots showing the activity of interferon response in BANF1 control (sgNeg) and BANF1 KO (sg#2 and #4) MC38 cells. Actin was used as a loading control. Bar graphs were analyzed by unpaired two-tailed Student’s t-test. All results are shown as mean±SEM (error bars). Data shown are representative of three experiments. ** p<0.01, ***p<0.001. NES, Normalized Enrichment Score.

    cGAS-STING axis is indispensable for the immune activation in BANF1 KO tumor cells

    Recently studies found that BANF1 can restrict cGAS-STING pathway to prevent innate immune activation in antiviral immunity.22 The cGAS-STING signaling pathway, which induces production of type I interferon and ISGs, is a key regulator of innate and adaptive immunity.8–10 Based on our RNA-seq results, we wondered whether antitumor immune response induced by BANF1 KO depends on cGAS-STING pathway. To evaluate this hypothesis, we first pharmacologically inhibited STING activity using antagonist H-151, and found that treatment with the STING inhibitor diminished the activation of Sting and Tbk1 in BANF1 KO tumor cells (figure 5A). Meanwhile, H-151 also significantly reversed the upregulation of ISGs at mRNA levels in BANF1 KO cells (figure 5B), suggesting interferon response evoked by BANF1 KO partly depends on Sting-Tbk1-Irf3 axis.

    Figure 5
    Figure 5

    cGAS-STING axis is indispensable for the immune activation in BANF1 KO tumor cells. (A) Effects of H-151 on cGAS-STING signaling in BANF1 KO and BANF1 control MC38 cells measured by the induction of TBK1 and STING phosphorylation. (B) The expression of Irf1, Irf3, Irf4, Irf7, and Irf8 mRNA by RT-qPCR in control or BANF1 KO MC38 cells treated with H-151 for 0 hour, 4 hours, and 24 hours. (C) Immunoblots showing TBK1 677 and STING phosphorylation change in control, BANF1 KO, BANF1/STING double KO (sgSB) and BANF1/cGAS double KO (sgCB) MC38 cells. (D) RT-qPCR analysis of selected ISGs, including Rsad2, ISG15, Cxcl11, and Cxcl10, mRNA levels in control, BANF1 KO, BANF1/STING double KO (sgSB) and BANF1/cGAS double KO (sgCB) MC38 cells (n=3/group). (E) Representative immunofluorescence images of dsDNA antibody labeling (green) (upper panel) and quantitation analysis (lower panel) in control (sgNeg), BANF1 KO (sgBanf1) and BANF1 reexpression (sgBanf1+B) MC38 cells. The results of quantitative analysis are shown in the histogram. Scale bar=10 μ m. (F) Growth curves of BANF1 control (sgNeg), BANF1 KO (sgBanf1), BANF1/STING double KO (sgSB) and BANF1/cGAS double KO (sgCB) MC38 cells in C57BL/6 mice (upper panel, n=6/group), and representative images of tumor sizes at day 21 (lower panel). Bar graphs were analyzed by unpaired two-tailed Student’s t-test and tumor growth by two-way ANOVA with Tukey’s test. All results are shown as 690 mean±SEM (error bars). Data shown are representative of three experiments. Not significant (ns) p>0.05, *p<0.05, ** p<0.01, ***p<0.001. ANOVA, analysis of variance.

    To further corroborate these results, we then constructed Banf1/Cgas (#sgCB), or Banf1/Sting1 (#sgSB) double knockout MC38 cells, and found that double KO effectively reversed BANF1 KO induced upregulation of pTbk1 (figure 5C). Inactivation of the Cgas or Sting gene could also greatly suppress the upregulation of ISGs, like Rsad2, Isg15, Cxcl10, and Cxcl11(figure 5D), indicating that cGAS could be a vital element in immune activation in BANF1 KO cells. Usually, cGAS is activated by interacting with cytosolic double-stranded (dsDNA) in a sequence-independent manner. Indeed, we observed higher levels of dsDNA accumulation in the cytosol of BANF1 KO cells (figure 5E).

    Further, we tested whether activation of cGAS-STING pathway contributed to the antitumor effects of BANF1 KO tumor cells. Indeed, cGAS KO or Sting KO completely abrogated BANF1 KO-mediated inhibition of tumor growth and tumor formation rate in vivo to a level comparable to that of the control (figure 5F). In all these experiments using both genetic and pharmacological inhibition implied BANF1 KO-induced antitumor immunity depends on activation of cGAS-STING pathway.

    BANF1 KO sensitizes cancer cells to PD-1 blockade treatment

    Based on our findings that the importance of BANF1 knockout in driving antitumor immunity, we tested whether deletion of BANF1 was sufficient to enhance the efficacy of checkpoint blockade by assessing the tumor growth and mice survival. The treatment schedules of MC38 and B16F10 tumor-bearing mice were illustrated in figure 6A,D. Our results indicated that BANF1 KO alone or anti-PD-1 treatment alone decreased the tumor growth to some extent, but didn’t significantly prolong the survival of tumor-bearing mice. However, combining BANF1 KO with anti-PD-1 improved their therapeutic benefit as compared with anti-PD-1 alone or BANF1 KO alone. Such improvement was translated into a remarkable delay in the tumor growth and large survival benefit (figure 6B,C).

    Figure 6
    Figure 6

    BANF1 KO sensitizes cancer cells to PD-1 blockade treatment. (A, B) Experimental scheme (A) and tumor growth curves (B) of BANF1 KO (sgBanf1) and BANF1 control (sgNeg) B16F10 tumors in C57BL/6 mice administered with control or antibody against PD-1 (n=4–6 replicates /group). (C) Survival curves of mice showed in A, p values were calculated using the log-rank (Mantel-Cox) test. (D, E) Experimental scheme (D) and tumor growth curves (E) of BANF1 KO (sgBanf1) and BANF1 control (sgNeg) MC38 tumors in C57BL/6 mice administered with control or antibody against PD-1 (n=4–6 replicates /group). (F) Rechallenge growth curves of BANF1 control (sgNeg) MC38 tumors in treatment naïve C57BL/6 mice and mice that were inoculate with BANF1 KO cells and tumor-free after anti-PD-1 treatment in the first challenge (E) Tumor growth were analyzed by two way ANOVA with Tukey’s test. All results are shown as mean±SEM (error bars). Data shown are representative of three experiments. *p<0.05, **p<0.01, ***p<0.001. ANOVA, analysis of variance.

    In addition, the potentiation of the anti-PD-1 efficacy was also confirmed in MC38 cancer model (figure 6E). Strikingly, mice with complete responses after the initial challenge fully resisted tumor rechallenge (figure 6F), indicating that BANF1 KO combined with ICI induced efficient long-term memory immune responses and a durable cure of the animals.

    Human BANF1 expression is negatively associated with CD8+ T cells infiltration and prognosis

    Having uncovered the importance of BANF1 in tumor immunity, we sought to assess the translational applicability of our findings in human cancers. First, we correlated BANF1 expression with a variety of tumor-infiltrating immune cells using multiplex immunofluorescence staining (mIF) of human NSCLC tissue chips. Interestingly, samples with higher BANF positive cells showed a lower CD8+ T cells infiltration and a lower PD-L1 positive cell proportion (figure 7A, online supplemental figure S2). Further correlation analysis showed that the proportion of BANF1 positive cells was significantly negatively correlated with immune cells (CD45+), especially CD45+CD8+ T cells (figure 7B). Then, we analyzed the correlation of BANF1 expression and clinical immune checkpoint blockade (ICB) response. Higher Tumor Immune Dysfunction and Exclusion (TIDE) prediction score represents a higher potential for immune evasion, which suggests that the patients were less likely to benefit from ICB therapy.32 In TCGA-SKCM cohort, we observed that the BANF1-low subgroup had a lower TIDE score and higher levels of PD-L1 (CD274) than the BANF1-low subgroup, implying that BANF1-low patients could benefit more from ICB therapy (figure 7C). Aside, we assessed the prognostic value of BANF1 expression in the clinical cohort with immunotherapy with Kaplan-Meier Plotter.33 As shown in figure 7D, for pan-cancer patients treated with anti-PD-1 therapy, BANF1-high group patients had shorter overall survival and progression-free survival than BANF1-low group patients, especially in melanoma patients treated with anti-PD1 therapy. These results suggested that BANF1 expression may correlated with tumor immunosuppressive microenvironment and poor prognosis for immunotherapy.

    Supplemental material

    Figure 7
    Figure 7

    Human BANF1 expression is negatively associated with CD8+ T cell infiltration and prognosis. (A) Percentages of CD45+CD8+ T cells and percentages of CD274 positive cells in tumor region in NSCLC tumor tissues of 140 patients (BANF1 Low: <50% of tumor cells are positive (blue points); BANF1 High: ≥50% of tumor cells are negative (red points)). (B) Scatter plots showing the correlation of BANF1 positive cell proportion with tumor infiltrating immune cells, including CD45+ immune cells, CD45+CD8+ T cells. (C) Bar graph showing TIDE score, Exclusion score and CD274 between BANF1-low and BANF1-high group in TCGA-SKCM cohort. (D) Overall survival and progression-free survival analysis of patients based on relative expression of BANF1 in the Kaplan-Meier Plotter Immunotherapy cohort. The upper panel showing all tumor type samples with anti-PD-1 treatment and the lower panel showing melanoma samples with anti PD-1 treatment (Pembrolizumab). *p<0.05, *** p<0.001.

    Discussion

    In the current study, we demonstrated that genetic deletion of BANF1 in cancer cells leads to cGAS-STING-mediated innate immune activation, which, in turn, results in the expression of multiple ISGs and inflammatory chemokines such as CXCL10/11, recruiting CD8+ T cells into the TME. Importantly, combining BANF1 ablation and anti-PD-1 antibody, resulted in enhanced antitumor efficacy. Our findings addressed the unmet clinical needs to develop innovative strategies to improve the efficacy of ICIs.

    BANF1 overexpression has been described in many different cancer types, including gastric, cervical, esophageal, and breast cancer.25 26 34 35 In this study, we confirmed the prevalence of high expression of BANF1 in multiple tumors. Elevated BANF1 was reported be associated with poor prognosis, cell proliferation and migration. However, its role in tumor immunity has not yet been described. Here, we observed that BANF1 upregulation correlates significantly inversely with immune infiltration. Subsequent in vivo studies revealed that deletion of BANF1 in cancer cells induces increased CD8+ T cells infiltration in TME, resulting in strong antitumor activity. These finding implied that in human tumors, elevated BANF1 may promotes an immunosuppressive microenvironment by decreasing T cell infiltration. Indeed, this notion was supported by TIDE analysis that BANF1 high patients have higher TIDE score, especially exclusion score, that is, low T cell infiltration, but not T cell dysfunction.

    An important finding of this study is to identify that tumor-derived BANF1 play a key role in antitumor immunity. In our two mouse models, we found consistent results that ablation of BANF1 results in inflamed TME by enhancing T cell infiltration and reducing MDSC infiltration. Our findings revealed that targeting BANF1 inhibits tumor growth in immunocompetent but not immunodeficient mice, while CD8+ T cells-clearing eliminates this inhibitory effect. Furthermore, BANF1 deletion induced the activation of IFN response and the production of inflammatory chemokines, which elucidates the mechanism of T cell infiltration. Therefore, CD8+ T cell mediated antitumor immunity is the major contributor of tumor growth delay in our models. This is in line with the prevailing view that boosting CD8+ T cells infiltration is an effective approach to induce antitumor immunity and overcome primary resistance to ICB. Although we also found an immunosuppressive MDSC population change, we have not yet explored them in detail, given their heterogeneity. These will require further investigation.

    As an dsDNA sensor, BANF1 can bind to nuclear genomic DNA and cytosolic dsDNA based on its dynamic position. Recent study reported that loss of BANF1 in mouse microglial cells results in increased levels of cytosolic dsDNA and an enhanced innate immune response.27 Consistent with this fact, we also showed that BANF1 deletion results in accumulation of cytosolic dsDNA and the activation of cGAS-STING pathway-mediated innate immune responses in cancer cells. Although cGAS-STING pathway is a double-edged sword for tumors, more and more evidences showed that this pathway is critical to antitumor immunity. Therefore, many factors that regulating this pathway have been developed as potential targets used for tumor-therapy.11 15 36–38 Here, we demonstrated that the antitumor efficacy induced by targeting BANF1 is almost offset by cGAS/STING knockout. It indicates that the high expression of BANF1 in tumors may be a biomarker of the advantaged population for STING agonists. On the other hand, our findings provided strong support for better understanding of the regulatory mechanism of how tumor-intrinsic cGAS signaling is suppressed to facilitate tumorigenesis by escaping immune surveillance. In addition to development of STING agonists, BANF1 can be used as a potential target for immunocombination therapy. It will take more research to further explore the detailed mechanism of BANF1 in tumor immunity.

    The finding that tumors with BANF1 deletion exhibit increased expression of both type I IFN and the immune checkpoint ligand, PD-L1(CD274), provided a rationale for enhancing the effect of PD-1 blockade by targeting BANF1. Our data showed combined treatment with anti-PD-1 and BANF1 ablation is therapeutically superior to treatment with either one alone in melanoma and colon cancer models. Future work will be needed to explore the potential synergism of BANF1 inhibition and other ICIs or with other treatment.

    In summary, our finding supported the notion that BANF1 modulates immune escape by blocking effector immune cell infiltration. Targeting BANF1 combined with immune checkpoint inhibition may offer potential new opportunities for cancer immunotherapy. It is tempting to speculate that such combination therapy will benefit multiple cancer types and extend their use to non-responder patients.

    Supplemental material

    Data availability statement

    Data are available on reasonable request.

    Ethics statements

    Patient consent for publication

    Ethics approval

    All experiments were approved by the Animal Research Ethics Committee of Shandong Cancer Hospital (#2022004016).

    Acknowledgments

    The authors appreciate members of our laboratories for their kind suggestions. We would like to thank Prof. Liufu Deng for his kind donation of cell lines.

    References

    1. ↵
      1. de Miguel M,
      2. Calvo E
      . Clinical challenges of immune Checkpoint inhibitors. Cancer Cell 2020;38:326–33. doi:10.1016/j.ccell.2020.07.004
      OpenUrlCrossRefPubMed
    2. ↵
      1. Larkin J,
      2. Chiarion-Sileni V,
      3. Gonzalez R, et al
      . Combined Nivolumab and Ipilimumab or monotherapy in untreated Melanoma. N Engl J Med 2015;373:23–34. doi:10.1056/NEJMoa1504030
      OpenUrlCrossRefPubMed
    3. ↵
      1. Doroshow DB,
      2. Sanmamed MF,
      3. Hastings K, et al
      . Immunotherapy in non-small cell lung cancer: facts and hopes. Clin Cancer Res 2019;25:4592–602. doi:10.1158/1078-0432.CCR-18-1538
      OpenUrlAbstract/FREE Full Text
    4. ↵
      1. Smyth MJ,
      2. Ngiow SF,
      3. Ribas A, et al
      . Combination cancer Immunotherapies tailored to the tumour Microenvironment. Nat Rev Clin Oncol 2016;13:143–58. doi:10.1038/nrclinonc.2015.209
      OpenUrlCrossRefPubMed
    5. ↵
      1. Topalian SL,
      2. Drake CG,
      3. Pardoll DM
      . Immune Checkpoint blockade: a common denominator approach to cancer therapy. Cancer Cell 2015;27:450–61. doi:10.1016/j.ccell.2015.03.001
      OpenUrlCrossRefPubMed
    6. ↵
      1. Duan Q,
      2. Zhang H,
      3. Zheng J, et al
      . Turning cold into hot: firing up the tumor Microenvironment. Trends Cancer 2020;6:605–18. doi:10.1016/j.trecan.2020.02.022
      OpenUrl
    7. ↵
      1. Stäger S,
      2. Kaye PM
      . Cd8+ T-cell priming regulated by Cytokines of the innate immune system. Trends Mol Med 2004;10:366–71. doi:10.1016/j.molmed.2004.06.003
      OpenUrlCrossRefPubMedWeb of Science
    8. ↵
      1. Kwon J,
      2. Bakhoum SF
      . The cytosolic DNA-sensing cGAS-STING pathway in cancer. Cancer Discov 2020;10:26–39. doi:10.1158/2159-8290.CD-19-0761
      OpenUrlAbstract/FREE Full Text
    9. ↵
      1. Zheng J,
      2. Mo J,
      3. Zhu T, et al
      . Comprehensive elaboration of the cGAS-STING signaling axis in cancer development and Immunotherapy. Mol Cancer 2020;19:133. doi:10.1186/s12943-020-01250-1
    10. ↵
      1. Fuertes MB,
      2. Kacha AK,
      3. Kline J, et al
      . Host type I IFN signals are required for antitumor Cd8+ T cell responses through Cd8{Alpha}+ Dendritic cells. J Exp Med 2011;208:2005–16. doi:10.1084/jem.20101159
      OpenUrlAbstract/FREE Full Text
    11. ↵
      1. Kim H,
      2. Kim H,
      3. Feng Y, et al
      . Prmt5 control of cGAS/STING and Nlrc5 pathways defines Melanoma response to antitumor immunity. Sci Transl Med 2020;12:eaaz5683. doi:10.1126/scitranslmed.aaz5683
    12. ↵
      1. Guo Y,
      2. Jiang F,
      3. Kong L, et al
      . Otud5 promotes innate antiviral and antitumor immunity through Deubiquitinating and stabilizing STING. Cell Mol Immunol 2021;18:1945–55. doi:10.1038/s41423-020-00531-5
      OpenUrl
    13. ↵
      1. Wu S-Y,
      2. Xiao Y,
      3. Wei J-L, et al
      . MYC suppresses STING-dependent innate immunity by Transcriptionally Upregulating Dnmt1 in triple-negative breast cancer. J Immunother Cancer 2021;9:e002528. doi:10.1136/jitc-2021-002528
    14. ↵
      1. Lv M,
      2. Chen M,
      3. Zhang R, et al
      . Manganese is critical for antitumor immune responses via cGAS-STING and improves the efficacy of clinical Immunotherapy. Cell Res 2020;30:966–79. doi:10.1038/s41422-020-00395-4
      OpenUrlPubMed
    15. ↵
      1. Wang L,
      2. Yang L,
      3. Wang C, et al
      . Inhibition of the ATM/Chk2 axis promotes cGAS/STING signaling in Arid1A-deficient tumors. J Clin Invest 2020;130:5951–66. doi:10.1172/JCI130445
      OpenUrlPubMed
    16. ↵
      1. Sears RM,
      2. Roux KJ
      . Diverse cellular functions of barrier-to-Autointegration factor and its roles in disease. J Cell Sci 2020;133:jcs246546. doi:10.1242/jcs.246546
    17. ↵
      1. Halfmann CT,
      2. Roux KJ
      . Barrier-to-Autointegration factor: a first Responder for repair of nuclear Ruptures. Cell Cycle 2021;20:647–60. doi:10.1080/15384101.2021.1892320
      OpenUrl
    18. ↵
      1. Zheng R,
      2. Ghirlando R,
      3. Lee MS, et al
      . Barrier-to-autointegration factor (BAF) bridges DNA in a discrete, higher-order nucleoprotein complex. Proc Natl Acad Sci U S A 2000;97:8997–9002. doi:10.1073/pnas.150240197
      OpenUrlAbstract/FREE Full Text
    19. ↵
      1. Wiebe MS,
      2. Traktman P
      . Poxviral B1 kinase overcomes barrier to Autointegration factor, a host defense against virus replication. Cell Host Microbe 2007;1:187–97. doi:10.1016/j.chom.2007.03.007
      OpenUrlCrossRefPubMedWeb of Science
    20. ↵
      1. Samwer M,
      2. Schneider MWG,
      3. Hoefler R, et al
      . DNA cross-bridging shapes a single nucleus from a set of mitotic Chromosomes. Cell 2017;170:956–72. doi:10.1016/j.cell.2017.07.038
      OpenUrlCrossRefPubMed
    21. ↵
      1. Halfmann CT,
      2. Sears RM,
      3. Katiyar A, et al
      . Repair of nuclear Ruptures requires barrier-to-Autointegration factor. J Cell Biol 2019;218:2136–49. doi:10.1083/jcb.201901116
      OpenUrlAbstract/FREE Full Text
    22. ↵
      1. Bolderson E,
      2. Burgess JT,
      3. Li J, et al
      . Barrier-to-Autointegration factor 1 (Banf1) regulates poly [ADP-ribose] polymerase 1 (Parp1) activity following oxidative DNA damage. Nat Commun 2019;10:5501. doi:10.1038/s41467-019-13167-5
    23. ↵
      1. Furukawa K,
      2. Sugiyama S,
      3. Osouda S, et al
      . Barrier-to-Autointegration factor plays crucial roles in cell cycle progression and nuclear organization in Drosophila. J Cell Sci 2003;116:3811–23. doi:10.1242/jcs.00682
      OpenUrlAbstract/FREE Full Text
    24. ↵
      1. Cox JL,
      2. Mallanna SK,
      3. Ormsbee BD, et al
      . Banf1 is required to maintain the self-renewal of both Mouse and human embryonic stem cells. J Cell Sci 2011;124(Pt 15):2654–65. doi:10.1242/jcs.083238
      OpenUrlAbstract/FREE Full Text
    25. ↵
      1. Li J,
      2. Hu B,
      3. Fang L, et al
      . Barrier-to-Autointegration factor 1: A novel biomarker for gastric cancer. Oncol Lett 2018;16:6488–94. doi:10.3892/ol.2018.9432
      OpenUrlCrossRef
    26. ↵
      1. Li J,
      2. Wang T,
      3. Pei L, et al
      . Expression of Vrk1 and the downstream gene Banf1 in Esophageal cancer. Biomedicine & Pharmacotherapy 2017;89:1086–91. doi:10.1016/j.biopha.2017.02.095
      OpenUrl
    27. ↵
      1. Ma H,
      2. Qian W,
      3. Bambouskova M, et al
      . Barrier-to-Autointegration factor 1 protects against a basal cGAS-STING response. mBio 2020;11:e00136-20. doi:10.1128/mBio.00136-20
    28. ↵
      1. Guey B,
      2. Wischnewski M,
      3. Decout A, et al
      . BAF restricts cGAS on nuclear DNA to prevent innate immune activation. Science 2020;369:823–8. doi:10.1126/science.aaw6421
      OpenUrlAbstract/FREE Full Text
    29. ↵
      1. Yang Y,
      2. Wu M,
      3. Cao D, et al
      . Zbp1-MLKL Necroptotic signaling potentiates radiation-induced antitumor immunity via Intratumoral STING pathway activation. Sci Adv 2021;7:eabf6290. doi:10.1126/sciadv.abf6290
    30. ↵
      1. Tokunaga R,
      2. Zhang W,
      3. Naseem M, et al
      . Cxcl9, Cxcl10, Cxcl11/Cxcr3 axis for immune activation - A target for novel cancer therapy. Cancer Treat Rev 2018;63:40–7. doi:10.1016/j.ctrv.2017.11.007
      OpenUrlCrossRefPubMed
    31. ↵
      1. Russo E,
      2. Santoni A,
      3. Bernardini G
      . Tumor inhibition or tumor promotion? the duplicity of Cxcr3 in cancer. J Leukoc Biol 2020;108:673–85. doi:10.1002/JLB.5MR0320-205R
      OpenUrl
    32. ↵
      1. Jiang P,
      2. Gu S,
      3. Pan D, et al
      . Signatures of T cell dysfunction and exclusion predict cancer Immunotherapy response. Nat Med 2018;24:1550–8. doi:10.1038/s41591-018-0136-1
      OpenUrlCrossRefPubMed
    33. ↵
      1. Kovács SA,
      2. Győrffy B
      . Transcriptomic Datasets of cancer patients treated with immune-Checkpoint inhibitors: a systematic review. J Transl Med 2022;20:249. doi:10.1186/s12967-022-03409-4
    34. ↵
      1. Zhang G
      . Expression and Prognostic significance of Banf1 in triple-negative breast cancer. Cancer Manag Res 2020;12:145–50. doi:10.2147/CMAR.S229022
      OpenUrlCrossRef
    35. ↵
      1. Mao L,
      2. Zhang Y,
      3. Mo W, et al
      . Banf1 is downregulated by Irf1-regulated microRNA-203 in Cervical cancer. PLoS ONE 2015;10:e0117035. doi:10.1371/journal.pone.0117035
    36. ↵
      1. Wu S,
      2. Zhang Q,
      3. Zhang F, et al
      . Her2 recruits Akt1 to disrupt STING signalling and suppress antiviral defence and Antitumour immunity. Nat Cell Biol 2019;21:1027–40. doi:10.1038/s41556-019-0352-z
      OpenUrlCrossRefPubMed
    37. ↵
      1. Wei B,
      2. Xu L,
      3. Guo W, et al
      . Shp2-mediated inhibition of DNA repair contributes to cGAS-STING activation and chemotherapeutic sensitivity in colon cancer. Cancer Res 2021;81:3215–28. doi:10.1158/0008-5472.CAN-20-3738
      OpenUrlAbstract/FREE Full Text
    38. ↵
      1. Kumar S,
      2. Zeng Z,
      3. Bagati A, et al
      . Carm1 inhibition enables Immunotherapy of resistant tumors by dual action on tumor cells and T cells. Cancer Discov 2021;11:2050–71. doi:10.1158/2159-8290.CD-20-1144
      OpenUrlAbstract/FREE Full Text

    Supplementary materials

    Footnotes

    • MW and YH contributed equally.

    • Contributors JY and MWu conceived and designed the study. FWu performed the bioinformatic analysis. MWang, YH, MC, WW, TZ, XC, FWang, and YL performed the experiments. MWang and DC wrote the original draft. All authors reviewed the manuscript. DC acts as guarantor for the overall content.

    • Funding This work was supported by the Academic Promotion Program of Shandong First Medical University (2019ZL002), Research Unit of Radiation Oncology, Chinese Academy of Medical Sciences (2019RU071), the foundation of National Natural Science Foundation of China (82172676, 31900649, 81902608, and 82030082), the foundation of Natural Science Foundation of Shandong (ZR201911040452，ZR2021YQ52, ZR2020LZL016), and the Postdoctoral Innovation Project of Shandong Province to MWang.

    • Competing interests None declared.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.