Abstract
Programmed death ligand 1 (PD-L1) is a protein expressed in hepatocellular carcinoma (HCC) that drives immune evasion by binding to programmed death receptor 1 (PD-1) on activated T cells. Understanding PD-L1 regulation is essential to understand the immunosuppressive microenvironment for antitumor immunity. We screened ribonucleic acid (RNA)-binding motif proteins (RBMs). RBM30 can enhance PD-L1 expression in HCC cells. In this study, we found that high RBM30 expression in tumor tissues can drive HCC tumor immune evasion and accelerate disease progression via increased PD-L1 transcription. We conducted multiple molecular and high-throughput assays to elucidate the intrinsic molecular mechanisms by which RBM30 upregulates PD-L1 expression in HCC. RBM30 binds to DNA near the transcriptional start site of STAT1 and recruits DOT1L to promote H3K79me3 enrichment, enhancing its accessibility to upregulate STAT1 transcription, consequently activating the PD-L1 transcription. This enhances PD-L1 expression to facilitate immune evasion. These findings reveal the vital role of RBM30 in HCC immune evasion.
Similar content being viewed by others
Introduction
For decades, the treatment of hepatocellular carcinoma (HCC) has been based on surgical resection [1], supplemented with interventional therapy, radiofrequency ablation, and microwave therapy [2]. Chemotherapy with adriamycin and 5-fluorouracil has shown limited effectiveness [3], and the introduction of the targeted drug sorafenib in recent years has revolutionized drug therapy [4]. In recent years, the emergence of immunotherapeutic drugs targeting programmed death receptor 1 (PD-1) and programmed death receptor ligand 1 (PD-L1) has markedly improved the efficacy of drug therapy for HCC [5], ushering in a new era for liver cancer treatment [6].
PD-L1 is a cell membrane protein that plays an important regulatory role in the immune system [7]. PD-L1 is primarily expressed on the surface of tumor cells, inflammatory cells, and some normal tissues [8]. When tumor cells express a substantial amount of PD-L1, they can bind to the PD-1 receptor and activate the PD-1 signaling pathway [9], thereby inhibiting the function of activated T cells [10]. This inhibitory effect allows tumor cells to evade detection by the immune system [11]. Therefore, antibody therapy targeting PD-1/PD-L1 has become a new strategy for cancer treatment [12]. By inhibiting the binding of PD-L1 to PD-1, the activation and functioning of T cells can be restored and immune system effectiveness against tumor cells be enhanced [13]. In recent years, PD-1/PD-L1 inhibitors have instigated a breakthrough in the treatment of advanced HCC [14]. Nivolumab [15] and pembrolizumab [16] have now been approved for the clinical treatment of HCC, offering novel treatment options for patients. Therefore, further studies on the regulatory mechanism of PD-L1 in HCC will provide insights for improving the efficacy of tumor immunotherapy in HCC.
PD-L1 expression is regulated by multiple factors, including transcriptional, post-transcriptional, and post-translational regulatory pathways [17]. Among them, the signal transducer and activator of transcription 1 (STAT1) signaling pathway is an important pathway that induces PD-L1 expression [18]. In the tumor microenvironment, cytokine interferon-γ (IFN-γ) regulates PD-L1 expression by activating the STAT1 signaling pathway [19]. The binding of IFN-γ to its receptor activates a member of the Janus kinase (JAK) family on the receptor, which phosphorylates the tyrosine residue of STAT1 to form a dimer [20]. Dimerized STAT1 enters the nucleus and binds to a specific deoxyribonucleic acid (DNA) sequence, thereby activating the gene expression of PD-L1 [21]. Therefore, further exploration of potential PD-L1 regulators, especially key proteins that affect the STAT1 signaling pathway, could help select patients and overcome resistance to PD-1/PD-L1 therapy.
Disruptor of telomeric silencing 1-like histone lysine methyl-transferase (DOT1L), the only methyl-transferase responsible for the mono-methylation, di-methylation, and tri-methylation of histone 3 lysine 79 (H3K79) [22], is involved in many disease-related cellular processes, such as cell cycle progression and DNA damage response [23]. Previous studies have demonstrated that normal hematopoiesis is a developmental process that is markedly dependent on DOT1L [24], that knockdown of DOT1L leads to impaired blood lineage formation, and that aberrant DOT1L activity is associated with hematopoietic malignancies [25]. The genetic or pharmacological inhibition of DOT1L leads to defects in leukemic cell transformation and maintenance [26]. However, few studies have explored the role of DOT1L in solid tumors, and there are no reports on whether DOT1L regulates intracellular STAT1 or PD-L1 expression.
In our previous studies [27], we investigated angiogenesis in HCC, an important pathological feature, and found that the ribonucleic acid (RNA)-binding motif protein (RBM) family members RBM4 [28], RBM23 [29] and RBM28 [30] can promote neovascularization in HCC. In this study, we continued to screen RBM family members and found that high RBM30 expression in HCC promoted PD-L1 expression. Mechanically, RBM30 directly binds to STAT1 DNA and recruit DOT1L to enhance the H3K79me3 enrichment of this DNA region. This, in turn, increases the openness of the region, leading to the transcription of STAT1 and promoting HCC immune evasion. These findings not only deepen our understanding of the pathogenesis of HCC, especially its tumor immune evasion mechanisms, but also provide a theoretical basis for the selection of future HCC clinical immunotherapy targets.
Materials and methods
Clinical tissue samples
Three cohorts of HCC tissue samples were collected from The Affiliated Taizhou People’s Hospital of Nanjing Medical University. The first cohort consisted of 24 HCC tissues and 24 matched non-HCC tissues, serving as controls, for Western blot analysis. The second cohort consisted of 90 HCC and 90 matched non-HCC tissues for IHC staining assay. Before surgery, none of the patients received chemotherapy or radiation therapy. All patients in this study provided written informed consents. The experimental protocol was approved by the Institutional Ethics Review Board of The Affiliated Taizhou People’s Hospital of Nanjing Medical University (2022-008-01). To validate the authenticity of the specimens, a pathological analysis was conducted.
T cell-mediated tumor cell killing assay
Human PBMCs (peripheral blood mononuclear cells) are separated from the whole blood by a density gradient centrifugation method using Ficoll-Paque media. T cells were activated by treating PBMC with anti-CD3 antibody (100 ng/ml) (300437, Biolegend, San Diego, California, USA), anti-CD28 antibody (100 ng/ml) (302933, Biolegend, San Diego, California, USA) and IL-2 (10 ng/ml) for 48 h, then stained with anti-CD3 (E-AB-F1001D, Elabscience, Wuhan, China) and anti-CD8 (E-AB-F1110C, Elabscience, Wuhan, China), subsequently sorted by FACS, and isolated CD8+ T cells were cultured in RPMI-1640 supplemented with 10% FBS.
5 × 105 of cancer cells were seeded in a 24-well plate. 24 h later, 5 × 106 activated CD8+ T cells (10:1) were seeded and co-cultured with the indicated cancer cells for additional 24 h. Then, cells were washed twice with 1 × PBS to discard T cells and suspended dead cancer cells.
Cell culture
HCC cell lines (SNU-398, Huh-7, HepG2, PLC/PRF/5, Lm3, Hep3B, and MHCC-97H) and human normal hepatocyte lines (THLE-2) were obtained from Shanghai Fuheng Biotechnology Co., Ltd. HepG2, Huh7, MHCC97-H, and Lm3 HCC cell lines were cultured in Dulbecco’s modified Eagle medium, whereas SNU-398 was cultured in Roswell Park Memorial Institute 1640 medium. The PLC/PRF/5 cell line was cultured in minimum essential medium, and Hep3B cells were cultured in Eagle’s minimal essential medium. The media were supplemented with 10% fetal bovine serum, 100 U/mL penicillin, and 100 μg/mL streptomycin. All cell lines were qualified via short tandem repeats (STR) for identification and Mycoplasma testing. All cells are routinely stored in a serum-free cryopreservation medium (C40100, NCM Biotech, Suzhou, China) at −80 °C in a freezer.
Construction of gene-edited mice and chemical-induced mouse models
The gene-edited mice (Rbm30−/− and Rbm30KI/KI) was constructed by Adeno-associated virus provided by Azenta Life Sciences (pGWAA9-Rbm30 and pGWAA9-U6-sgRbm30). In this study, 40 3-week-old male mice were purchased from Jiangsu Hanjiang Biological Company and randomly divided into 4 groups, with 10 mice in each group. After one week of adaptive feeding, the above AAV viral vectors were injected via the tail vein to achieve gene editing. Furthermore, to construct the mouse spontaneous liver cancer model, both wild-type and gene-edited groups received intraperitoneal injections of diethylnitrosamine 2 mg/kg on the 14th day after birth, and CCl4 5 µL/g (CCl4/olive oil at a ratio of 25:75) twice a week for 16 weeks starting from the third week. In the 20th week, the mice were euthanized, and liver tissues were collected for subsequent experiments. This study employed immunohistochemical staining for the detection of RBM30, PD-L1, STAT1, and p-STAT1; flow cytometry to quantify the proportion of CD8+ T cells in each group; multiplex immunofluorescence to identify various immune cell populations within the tissue; and weighing analysis to evaluate the ratio of liver weight to total body weight. Through these comprehensive experimental approaches, the functional role of RBM30 in the progression of HCC was systematically investigated.
Animal experiments were approved by the Experimental Center of Jiangsu Hanjiang Biotechnology Co., LTD (HJSW-23050302), and the animals were raised according to animal welfare laws.
Subcutaneous tumor formation in mice
Four-week-old male C57 black 6 mice were utilized for tumor formation experiments. Subcutaneous injection of 2 × 106 HCC cells per mouse was performed, and after 4 weeks, the mice were euthanized to assess the tumor burden, with subsequent hematoxylin and eosin (HE) and IHC analysis. For the InVivoMAb anti-mouse PD-1 (CD279) (BE0146, BioXcell, West Lebanon, New Hampshire, USA) combination therapy in mice, subcutaneous injections of 1 × 106 HCC cells per mouse were performed, and the mice were injected intra-peritoneally with vehicle or anti-PD-1 (200 µg) as indicated on days 18, 21, 24, and 27. Twelve weeks after injecting tumor cells, the mice were euthanized to assess the tumor burden, with subsequent HE and IHC analysis.
Expression vector and plasmid construction
pcDNA3.1-Puro-C-3Flag and pCMV-Geneticin (G418)-C-Myc were used for constructing full-length and truncated RBM30 and DOT1L, respectively, in HCC cells. The plasmid vectors pET-28a and pGEX-6P-1 were used to construct purification vectors for the DOT1L and RBM30 proteins. For gene knockout and knockdown, lenti-clustered regularly interspaced short palindromic repeats (CRISPR) V2-Blast and pLKO.1-hydro plasmid vectors were employed, respectively. All constructed plasmids were validated via DNA sequencing and used for subsequent experiments, for which they were transfected in HCC cells using a Lipo8000 transfection reagent (C0533, Beyotime, Shanghai, China) following the manufacturer’s instructions. The transfected cells were selected and maintained in the presence of hygromycin (200 μg/mL), puromycin (1–2 μg/mL), G418 (1 mg/mL), or blasticidin (10 μg/mL).
Dual-luciferase reporter gene assays
pRL-TK (D2760, Beyotime) and pSTAT1-TA-Luc (D4400, Beyotime) reporter gene plasmids were transferred into constructed stable HCC cell lines. After 48 h, a dual-luciferase reporter assay kit (DL101, Vazyme, Nanjing, China) was used to detect fluorescence with a multifunctional microplate detector (Synergy HTX, BioTek, VT, USA). Furthermore, the activity of the STAT1 signaling pathway in each HCC cell line was assessed using the luciferase/Renilla fluorescence ratio.
qPCR assays
Total RNA was extracted using the RNA-easy isolation reagent (R701, Vazyme) and subjected to the HiScript III 1st Strand copy DNA Synthesis Kit (+genomic DNA wiper) (R312, Vazyme). qPCR was performed using an AceQ Universal SYBR qPCR Master Mix (Q511, Vazyme) following the manufacturer’s instructions. All the data were normalized to the housekeeping gene β-actin, and quantitative measures were calculated using the 2−∆∆CT method.
Western blotting assay
A radioimmunoprecipitation assay buffer (P0013C, Beyotime) containing 1 mmol/L phenylmethylsulfonyl fluoride (PMSF) was used to lyse cell samples and fresh clinical tissue samples, followed by centrifugation at 12,000 × g revolutions per minute (rpm) for 10 min at 4 °C. The supernatant with a 5X sodium dodecyl sulfate (SDS) loading buffer (WB2001, NCM Biotech, Suzhou, China) were boiled for 10 min. Subsequently, proteins were separated using SDS-polyacrylamide gel electrophoresis (PAGE) (SLE020, SmartPAGE™ Precast Protein Gel Plus 15 Wells, Changzhou, China) and transferred to polyvinylidene difluoride (PVDF) membranes (10600023, GE, Boston, USA), which were then blocked using 5% skim milk and probed with primary antibodies overnight at 4 °C. On the second day, the PVDF membranes were immersed with corresponding secondary antibodies (horseradish peroxidase (HRP)–labeled goat anti-rabbit immunoglobulin G (IgG) (two heavy (H) + two light (L) chains); A0208, Beyotime) or HRP-labeled goat anti-mouse IgG (H + L) (A0216, Beyotime) for 1 h and imaged using super-sensitive enhanced chemiluminescent substrate (BL520B, Biosharp, Beijing, China).
Primary antibodies used in the Western blotting assay were as follows: anti-flag (1:5000, F1804, Sigma), anti-DOT1L (1:2000, 77087, CST), anti-RBM30 (1:500 15412-1-AP, Proteintech), anti-PD-L1 (1:5,000, 66248-1-Ig, Proteintech), anti-GAPDH (1:20,000, 66004-1-Ig, Proteintech), and anti-myc-tag (1:5000, 16286-1-AP, Proteintech); anti-RBM4B (1:500, NBP1-80469, Novus), anti-phospho-JAK1-Y1022 (1:1000, NB100-82005, Novus), anti-JAK1 (1:1000, A11963, Abclonal), anti-JAK2 (1:1000, A19629, Abclonal), anti-phospho-JAK2-Y1007/1008 (1:1000, AP0531, Abclonal), anti-STAT1 (1:1000, A12075, Abclonal), anti-phospho-STAT1-Y701 (1:1000, AP0135, Abclonal), anti-PIAS1 (1:1000, A4744, Abclonal), and anti-SOCS1 (1:1000, A20838, Abclonal).
Co-IP assay
For Co-IP, as stated in the previous publication [31], cells were lysed with cell lysis buffer for Western blot and IP (P0013, Beyotime) supplemented with 1 mmol/L PMSF, then 50 µL A/G agarose (sc-2003, Santa Cruz) was added to pre-clean the cell lysate. After centrifugation at 12,000 rpm and 4 °C for 10 min the supernatant was collected and incubated with agarose coated with Flag-tag (F1804, Sigma), myc-tag (16286-1-AP, Proteintech), or normal IgG overnight at 4 °C. After extensive washing with IP buffer (20 mM Tris (pH 7.5)), 150 mM NaCl, 1% Triton X-100), interacting proteins were eluted (P9801/P9805, Beyotime), separated with SDS-PAGE, and detected via immunoblotting. The interacting proteins were subjected to mass spectrometry analysis by Shanghai Bioprofile Technology Company Ltd.
Pulldown assay
The target plasmids were transformed into BL21 (DE3) Escherichia coli expression strains, and the above bacteria were grown in 200 mL of Luria-Bertani (LB) medium at 37 °C until reaching an optical density at a wavelength of 600 nm of 0.4–0.6. They were then induced with 0.1 mM isopropyl β-d-1-thiogalactopyranoside and incubated for 16 h at 16 °C.
The bacterial pellet was collected and re-suspended in lysis buffer (1 mM dithiothreitol, 1 mg/mL lysozyme, 1 mM PMSF in phosphate-buffered saline (PBS)). The suspension was then lysed on ice for 30 min. The bacterial lysate was then sonicated for 20 min and centrifuged at 4 °C at 12,000 rpm for 15 min. The DOT1L and RBM30 recombinant proteins were obtained using glutathione beads (#SA008025, Smart-Lifesciences) and Ni NTA beads (#SA004025, Smart-Lifesciences) as per the manufacturer’s instructions, and were then dialyzed overnight at 4 °C using a 1.4 kD dialysis bag.
Two purified proteins were incubated overnight in 300 mL of incubation buffer. They were then incubated with 30 mL of Glutathione Beads or Ni NTA beads for 2 h at 4 °C, followed by washing with 1% tris-buffered saline -Tween 20. Finally, the proteins immobilized on the beads were eluted using 2× loading buffer by heating at 100 °C for 10 min and subsequently analyzed using a Western blotting assay.
Immunocytochemistry assay
The constructed HCC cells were seeded on glass coverslips and cultured with culture medium.
Following the removal of PBS, 4% paraformaldehyde (P0099, Beyotime) was added for 15 min at 37 °C to fix the cells. Subsequently, the cells were washed with PBS thrice. Next, the fixed cells were permeabilized with 0.5% Triton X-100 (ST795, Beyotime) in PBS for 20 min at 37 °C, followed by another three washes with PBS. The permeabilized cells were then blocked with 5% bovine serum albumin (BSA) in PBS for 10 min at 37 °C and incubated with corresponding primary antibody (Anti-Flag (1:200, F1804, Sigma) and Anti-myc-tag (1:200, 16286-1-AP, Proteintech)) in 5% BSA solution at 4 °C overnight. After being washed with 0.05% PBS-Tween 20 thrice, cells were incubated with the corresponding secondary antibody CoraLite488 (SA00013-2, Proteintech) or CoraLite594 (SA00013-3, Proteintech) at 37 °C for 10 min. After three washes with 0.05% PBS-Tween 20, the nuclei were counterstained with Hoechst (C1025, Beyotime). Leica SP8 confocal microscopy was used for the observation of stained cells.
IHC staining
For IHC staining, tissue sections were dewaxed with xylene, hydrated with graded ethanol, and the endogenous peroxidase activity was blocked with 0.3% hydrogen peroxide. The antigen was retrieved by microwave heating in sodium citrate solution (pH 8.0). BSA (5%) was used for blocking. The slide was incubated with primary antibodies at 4 °C overnight, washed with 0.05% PBS-Tween 20, and incubated with the corresponding HRP-conjugated secondary antibody at room temperature for 1 h. 3,3′-Diaminobenzidine was applied for color development, and hematoxylin was used to counterstain the nucleus.
The primary antibodies used in IHC staining assay were as follows: Anti-RBM30 (1:200, NB1-80469, Novus), Anti-p-JAK1-Y1022 (1:100, NB100-82005, Novus), Anti-JAK1 (1:100, A11963, Abclonal), Anti-STAT1 (1:100, A12075, Abclonal), Anti-p-STAT1-Y701 (1:100, AP0135, Abclonal), Anti-PD-L1 (1:10,000, 66248-1-Ig, Proteintech), and Anti-Kiel-67 (1:5000, 27309-1-AP, Proteintech).
RNA immunoprecipitation(RIP) assays and RIP-seq
Briefly, after formaldehyde crosslinking, 25 million cells per sample were harvested and washed with PBS. Cells were lysed with 1 mL of the RIP assay (RIPA) buffer (P0013C, Beyotime) and 2000 U/mL ribonuclease inhibitor (R0102, Beyotime). To fragment RNA, the lysate was subjected to sonication for 30 cycles with 30 s on and 30 s off.
Meanwhile, the input samples were prepared by saving 5% of lysate and adding 1 mL RNA-easy isolation reagent (R701, Vazyme). The residual sample lysate was then centrifuged at 12,000 rpm for 15 min and subjected to IP using 50 µL A/G agarose (sc-2003, Santa Cruz) coated with normal mouse IgG (A7028, Beyotime) or Anti-Flag (F1804, Sigma), which were immuno-precipitated with rotation at 4 °C overnight. Afterward, beads were washed thrice with a cold RIPA buffer, and then immuno-precipitated beads were digested with proteinase K in the RIPA buffer supplemented with 1% SDS and 1.2 mg/mL proteinase K (ST532, Beyotime) and incubated for 1 h at 55 °C with shaking at 1200 rpm. Moreover, 1 mL RNA-easy isolation reagent was added. Finally, RNA was extracted from both input and immuno-precipitated RNA, including DNase I digestion, and subsequently used for qPCR and Next-Generation Sequencing, the RIP-seq databases were built by VAHTS Universal V10 RNA-seq Library Prep Kit for Illumina (NR606, Vazyme Biotech Co., Ltd), which were sequenced by Nanjing Jiangbei New Area Biophamaceutical Public Service Platform Co., Ltd.
Chromatin immunoprecipitation (ChIP) assays and ChIP-seq
This assay was performed using the Sonication ChIP Kit (RK20258, Abclonal) according to the manufacturer’s instructions. Briefly, cells were cross-linked with 37% formaldehyde, pelleted, and re-suspended in a lysis buffer. The cells were sonicated and centrifuged to remove the insoluble material. The supernatants were collected and incubated at 4 °C overnight with indicated antibodies (anti-Flag-tag, AE092, Abclonal) and protein A/G magnetic beads (RM02915, Abclonal). The beads were washed, and the precipitated chromatin complexes were collected, purified, and de-crosslinked. The precipitated DNA fragments were quantified using RT-PCR analysis or applied for Next-Generation Sequencing. the ChIP-seq databasesm were built by Scale ssDNA-seq Lib Prep Kit for Illumina V2 (RK20228, Abclonal), which were sequenced by Nanjing Jiangbei New Area Biophamaceutical Public Service Platform Co., Ltd.
FC-cut-tag, ATAC-seq, mRNA-seq and TMT-MS
FC-cut-tag assay was performed using NovoNGS® CUT&Tag 4.0 High-Sensitivity Kit (N259-YH01) according to the manufacturer’s instructions (anti-STAT1, A19563Abclonal, anti-p-STAT1(Y701), 9167, CST). which were sequenced by Nanjing Jiangbei New Area Biophamaceutical Public Service Platform Co., Ltd.
ATAC-seq assay was performed using Hyperactive ATAC-Seq Library Prep Kit for Illumina (TD711, Vazyme Biotech Co.,Ltd) according to the manufacturer’s instructions. which were sequenced by Nanjing Jiangbei New Area Biopharmaceutical Public Service Platform Co., Ltd. mRNA-seq was commissioned by Shanghai Majorbio Bio-pharm Biotechnology Co., Ltd. (Shanghai, China).
TMT-MS was commissioned by Shanghai Biotree Biomedical Technology Co, Ltd.
Statistical analysis
All cytology assays (including FACS assays and qPCR assays) were performed in triplicate. Animal experiments involved five independent replicates. The GraphPad Prism 9.5 software was used to assess the statistical significance of group differences. Student’s t test or one-way analysis of variance was employed to measure significance. Kaplan–Meier analysis and log-rank tests were employed for overall survival assessment. RBM30 staining intensity correlations with PD-L1, STAT1, and p-STAT1 were quantified using Image-Pro Plus6 and analyzed using the Chi-square test (Fisher’s exact test).
Results
RBM30 upregulates PD-L1 expression in HCC
Immunosuppression characterizes the tumor microenvironment of HCC, in which PD-L1 plays an important role in inhibiting T cell activation and promoting HCC immune evasion [32]. To investigate whether RBMs could regulate PD-L1 expression and accelerate the progression of HCC, we constructed the overexpression plasmids of RBM1–48 using the pcDNA3.1-Puro-C-3Flag vector and then established the overexpression cell lines in Huh-7 cells. Subsequent detection revealed that, only RBM12 and RBM30 overexpression led to the dramatical increase of PD-L1 (Fig. 1A). In subsequent investigations, we identified RBM30 as the subject of study and delved into how RBM30 regulates PD-L1 expression (the study on the regulation of PD-L1 expression by RBM12 is not discussed here [33]). We analyzed RBM30 levels in different human HCC cell lines, revealing that the expression of RBM30 was the lowest in Huh-7 and Hep3B cells and the highest in Lm3 and MHCC-97H cells (Fig. S1A). Based on the differential expression of RBM30 in various HCC cell lines, we performed RBM30 knockout in Lm3 and MHCC-97H cells, while establishing stable overexpressing cell lines in Huh-7 and Hep3B cells. Western blotting assays and qPCR assays found that knocking out RBM30 reduced the total PD-L1 content in Lm3 and MHCC-97H cells (Fig. 1B). Meanwhile, flow cytometry analysis (FACS) confirmed that the PD-L1 positive HCC populations were decreased due to the lack of RBM30 (Fig. 1C). Conversely, overexpression of RBM30 led to a significant increase in total and surface PD-L1 expression in HCC cells (Fig. S1B, C), and the presence of the RBM30–PD-L1 regulatory axis was confirmed in mouse hepatoma cells (H22 and Hepa1-6) (Fig. S1D, E), indicating that RBM30 does regulate the expression of PD-L1 in HCC. As CD8+ T cells are central to tumor killing, and the anti-tumor immunity of CD8+ T cells was inhibited by the expression of PD-L1 in the tumor microenvironment, we co-cultured the established stable HCC cells with activated CD8+ T cells to verify the cytotoxic activity of T cells against HCC after changing the expression of RBM30. Results showed that HCC cells, after knocking out RBM30, are sensitive to T cell mediated killing, and we obtained contrary results in RBM30 overexpressed HCC cells (Fig. 1D). In addition, the cytotoxic T cell activities (IFNg and granzyme B) of CD8+ T cells are enhanced in co-cultures of RBM30-deficient cells compared with co-cultures with control cells, and vice versa (Fig. 1E). In subsequent animal experiments to explore the role of RBM30 in vivo, after being induced by DEN and CCL4 (Fig. 1F), the number of tumors in the RBM30-knockout (KO) group (compared with that in wild-type mice) significantly declined along with a reduction in tumor size. Moreover, immunohistochemical (IHC) staining confirmed a significant decrease in the proliferation rate of HCC in RBM30−/− mice and demonstrated an evident decrease in the content of PD-L1. In contrast, Rbm30KI/KI mice exhibited an increase in liver tumor count and size, accompanied by a notable elevation in PD-L1 expression (Fig. 1G, H). FACS analysis of induced tumors revealed comparable percentages of CD4+ T cells in the immune cell population between KO/KI and control tumors. However, Rbm30-KO tumors exhibited a significantly higher proportion of CD8+ T cells compared to WT tumors, while the KI group showed a significant reduction in this subset (Fig. 1I). Likewise, further examination of tumor tissues by means of Multiplex Immunohistochemical assays (mIHC) disclosed that the highly expressed RBM30 in HCC tumor tissues influenced the content of CD8+ T cells (CD45+ CD3+CD8+ cells) in the microenvironment (Fig. 1J). Taken together, we found that RBM30 enhanced HCC immune evasion by increasing the expression of PD-L1.
A RBM1–48 were overexpressed in Huh-7 cells and verified using Western blotting assays. Further detection revealed that only when RBM12/RBM30 was overexpressed, the PD-L1 expression was increased. B Knockout of RBM30 reduced the protein expression and messenger RNA (mRNA) content of PD-L1 in Lm3 cells. The PD-L1 level in Hep3B cells was noticeably increased after the overexpression of RBM30 (n = 3, ***P < 0.0001). C Fluorescence-Activated Cell Sorting (FACS) assays were applied to detect the surface PD-L1 expression of HCC cells, demonstrating that elimination of RBM30 reduced the expression of PD-L1 on the membrane surface of HepG2 cells, whereas overexpression of RBM30 in Hep3B cells upregulated the content of PD-L1 on the membrane surface of Hep3B cells (n = 3, ***P < 0.0001). D Representative images of HCC cells after T cell-mediated killing assay. Cells were co-cultured with activated CD8+ T cells for 24 h. Activated T cells were washed away and the remaining cancer cells were photographed (n = 3), 100 µm. E Activated CD8+ T cells were harvested for RT-PCR assay to measure the IFN-γ and granzyme B expression. F The schematic diagram illustrates the utilization of DEN + CCL4 for chemical induction modeling. G Induced liver tissue specimens of Rbm30−/− mice exhibited reduced tumor size and PD-L1 content, whereas Rbm30KI/KI mice showed significantly accelerated tumor progression, 500/100 µm. H The liver weight of mice in different groups after modeling was calculated (n = 5, ***P < 0.0001). I FACS analysis of intratumor CD4+ and CD8+ T cells in induced liver tumors as shown in (D), the percentage of CD4+ T cells among the immune cell population was similar in the KO/KI and control tumors, the percentage of CD8+ T cells was significantly higher in Rbm30-KO tumors than WT tumors, and the proportion was significantly reduced in KI group (n = 5, ***P < 0.0001). J mIHC assays disclosed that the highly expressed RBM30 in HCC tumor tissues influenced the content of CD8+T cells (CD45+CD3+CD8+ cells) in the microenvironment.
High RBM30 expression in HCC is negatively associated with patient prognosis
To further investigate the role of RBM30 in HCC progression in vivo, we subcutaneously injected the H22-sg-Rbm30 cells and corresponding control cells into immunocompetent C57BL/6 mice and immunodeficient nude mice, respectively (Fig. S2A). The proliferation of Rbm30-KO group was significantly inhibited in C57BL/6 mice compared to nude mice, confirming that RBM30 may enhance the malignant progression of HCC by promoting immune evasion (Fig. S2A).
To further elucidate the pivotal role of RBM30 in HCC progression, we conducted additional investigations utilizing fresh tissue samples, paraffin sections, and integrated analyses with public databases. We initially utilized 24 freshly obtained clinical samples and observed elevated levels of RBM30 in tumor tissues compared to adjacent non-cancerous tissues (Fig. 2A), and the publicly accessible TCGA database also validated that the expression level of RBM30 in HCC tumor tissue is significantly elevated compared to adjacent tissue, Moreover, the corresponding ROC curve indicates that RBM30 may serve as a robust prognostic indicator for HCC (AUC = 0.896) (Fig. 2B). Utilizing pre-prepared tissue microarrays, we proceeded with further investigation by examining the expression level of RBM30 protein and its correlation with relevant pathological data from the patients. The results revealed that the protein levels of RBM30 in HCC are significantly associated with the extent of tumor size, and clinical staging (Fig. 2C). In subsequent immunohistochemistry staining experiments, we observed a significant negative correlation between high RBM30 expression in HCC and patient prognosis (Fig. 2D), as well as a significant positive correlation with the level of PD-L1 protein in tumor tissue (Fig. 2E), meanwhile, the results of this immunohistochemical experiment were confirmed in the TCGA database (Fig. 2F). Taken together, the results demonstrated that high RBM30 expression in HCC could drive HCC immune escape by increasing the PD-L1 expression in HCC.
A The results of Western blotting assays showed that in fresh clinical samples, RBM30 protein expression in tumor tissues was significantly increased compared with that in paracancerous tissues (n = 24). B The TCGA database confirms that RBM30 expression is significantly higher in HCC tumor tissue compared to adjacent tissue, and ROC curve indicates that RBM30 may serve as a robust prognostic indicator for HCC (AUC = 0.896). C Correlation analysis between RBM30 expression levels in HCC tissues and clinicopathological parameters. D Immunohistochemical staining assays using tissue microarray (TMA) revealed that the higher the expression of RBM30 in HCC, the worse the prognosis of patients (P = 0.001), 500/100 µm. E Immunohistochemical staining assays using TMA demonstrated a positive correlation between RBM30 content and PD-L1 protein expression in tumor tissues (R = 0.3286, P = 0.0031), 500/100 µm. F Analysis of The Cancer Genome Atlas database showed that the expression of RBM30 in HCC tumor tissues was negatively correlated with the prognosis of patients. Further analysis showed that there was a significant positive correlation between RBM30 content and PD-L1 expression in HCC (R = 0.202, ***P < 0.001). G mRNA-seq and TMT-MS are applied to reveal the underlying mechanisms through which RBM30 operates in depth. H The conjoint analysis results of mRNA-seq and TMT-MS revealed that RBM30 facilitates HCC immune evasion by augmenting intracellular STAT1 levels, thereby promoting the expression of PD-L1. I The RNA immunoprecipitation sequencing (RIP-seq) assays (anti-RBM30-Flag) revealed a widespread distribution of RBM30 across diverse RNA types. J The results from GO and KEGG analyses of RIP-seq indicated that RBM30 plays an extensive role in various biological processes within HCC through multiple signaling pathways. K The results of the integrated analysis of RIP-seq, mRNA-seq, and TMT-MS revealed a significant overlap between the binding targets and differentially expressed genes among these three datasets. However, no discernible association was observed between RBM30 and STA1 mRNA as well as PD-L1 mRNA.
To explore how RBM30 regulates PD-L1 expression in HCC, transcriptome messenger RNA sequencing (mRNA-seq) and tandem mass tagging mass spectrometry (TMT-MS) were performed in Hep3B-ov-RBM30 and corresponding control cells, respectively. The conjoint analysis results revealed that the differentially expressed genes were highly correlated between mRNA-seq and TMT-MS (Fig. 2G). Further analysis confirmed that RBM30 was involved in the regulation of tumor immunity in HCC by upregulating the expression of PD-L1, moreover, among the above differentially expressed genes, we found that RBM30 greatly increased the intracellular expression of STAT1, which is the key transcription factor that plays a crucial role in modulating intracellular PD-L1 levels (Fig. 2H).
As RBM30 acts as an RNA-binding protein, RNA immunoprecipitation sequencing (RIP-seq) assays (anti-RBM30-Flag) were conducted to investigate the interaction between RBM30 and mRNA in HCC cells (Figs. 2I and S2B). The findings demonstrated that RBM30 exhibited extensive distribution across diverse RNA types, with a predominant presence in mRNA encoding proteins. Furthermore, the results of GO and KEGG analyses indicated that RBM30 participated in various biological processes in HCC through multiple signaling pathways (Fig. 2J). The subsequent investigation revealed that RBM30 has the potential to bind to a substantial number of mRNAs, and the findings from mRNA-seq and TMT-MS experiments exhibited significant overlap. Interestingly, STAT1/PD-L1 mRNA was not found among the potential binding targets of RBM30, and genes known to directly regulate STAT1 mRNA expression were identified (Fig. 2K), indicating that high expression of RBM30 in HCC might drive HCC immune evasion, independent of its functionality as an RNA-binding protein.
RBM30 upregulates PD-L1 expression in HCC via the STAT1 signaling pathway
To further explore the underlying mechanism by which RBM30 promotes STAT1 expression in HCC, we analyzed the protein sequence of RBM30 in the NCBI database. Besides the two RNA recognition motif domains in the N-terminal region (which mediate RNA binding), the 161–177 amino acid sequence of RBM30 contained a zinc finger structure, which suggested that RBM30 might bind to chromatin DNA to drive HCC tumor immune evasion (Fig. S3A). Chromatin immunoprecipitation (ChIP) assays were performed against RBM30 (anti-RBM30-Flag) in Hep3B cells, followed by next-generation sequencing (ChIP-seq) (Fig. 3A). The results showed that RBM30 exhibited extensive binding affinity towards genomic DNA, and the ChIP peak-calling program model-based analysis of the ChIP-seq data exhibited that RBM30 could potentially be involved in the modulation of multiple signaling pathways within HCC (Fig. 3B). Further analysis on the distribution position of RBM30 on genomic DNA revealed that RBM30 locates predominantly in proximity to the transcriptional initiation and termination sites of genes (Fig. 3C), which indicated that RBM30 potentially promotes immune evasion in HCC by influencing gene transcription.
A Chromatin immunoprecipitation (ChIP) assays were performed against RBM30 (anti-RBM30-Flag) in Hep3B cells, followed by next-generation sequencing (ChIP-seq). B GO enrichment analysis and KEGG pathway enrichment analysis revealed involvement of RMB30 in a diverse range of biological processes in HCC. C Upon analyzing the distribution disparities of RBM30 across various chromatin elements, we observed a significant occupation of ASH2L-K312R within the proximal promoter region. D ATAC-seq demonstrated that extensive changes in chromatin accessibility were observed after RBM30 overexpression, accompanied by extensive changes in accessible chromatin regions. E In comparison to corresponding control cells, the analysis results of ATAC-seq suggest a significant reduction in open chromatin within the proximal promoter region of Hep3B-ov-RBM30 cells. F By conducting joint analysis of ChIP-seq and ATAC-seq results above revealed that there was a certain degree of correlation between the two. G Functional enrichment analysis and signaling pathway analysis on the overlapping regions revealed that RBM30 affects the JAK-STAT signaling pathway mediated by IFN-stimulated PD-L1 expression. H Integrated analysis using ChIP-seq, ATAC-seq, and mRNA-seq revealed that RBM30 exerts regulatory control over chromatin accessibility by directly binds to genomic DNA, thereby influencing the transcriptional regulation of specific genes. I IGV visualization results demonstrated the direct binding of RBM30 to the transcription start site (TSS) of VEGFA DNA.
We employed Assay for Transposase-Accessible Chromatin using sequencing (ATAC-seq) to conduct further exploration (Fig. 3D). By analyzing the differences in the position of genomic open regions after overexpression of RBM30, we discovered that the results were highly detailed compared with those of the aforementioned ChIP-seq, and a significant change occurred in the openness of the TSS region (Fig. 3E). By conducting joint analysis of ChIP-seq and ATAC-seq results above (Fig. 3F), we found that there was a certain degree of correlation between the two. After conducting functional enrichment analysis and signaling pathway analysis on the overlapping regions, we found that RBM30 is likely to affect the JAK-STAT signaling pathway mediated by IFN-stimulated PD-L1 expression (Fig. 3G).
Integrated ChIP-seq, ATAC-seq, and mRNA-seq analysis demonstrated that RBM30 regulates chromatin accessibility through direct genomic DNA binding, influencing transcriptional regulation of specific genes including STAT1 (Fig. 3H). Visualization results confirmed the direct binding of RBM30 to the STAT1 DNA region (transcription start site, TSS), thereby enhancing STAT1 DNA accessibility and promoting increased STAT1 mRNA expression in HCC (Fig. 3I).
From visualization results, we discovered that although the overexpression of RBM30 led to an increase in the content of PD-L1 mRNA along with an enhancement in the openness of the chromatin-related regions of the gene, no significant binding of RBM30 to the mRNA or DNA of PD-L1 was detected. This indicates that RBM30 does not directly regulate the content of PD-L1 (Fig.S3B). Moreover, among several crucial proteins that can mediate immunosuppression in the intra-tumoral microenvironment, no direct regulation by RBM30 was identified (Fig.S3C).
RBM30 upregulates STAT1 to facilitate PD-L1 expression
The findings obtained from high-throughput sequencing were further verified by ChIP-qPCR assays. It was found that RBM30 did directly bind to the DNA in the STAT1 promoter region (Fig. 4A). Based on the above findings, we hypothesized that RBM30 might promote immune evasion in HCC by upregulating PD-L1 expression via the transcriptional upregulation of STAT1. Dual-luciferase reporter gene assays confirmed that RBM30 knockout reduced the activity of the STAT1 signaling pathway in HCC, which was accompanied by a decrease in the intracellular STAT1 mRNA content, verified by qPCR assays in Lm3 and MHCC-97H cells. In contrast, after the overexpression of RBM30 in HCC cells, the activity of the STAT1 signaling pathway (Figs. 4B and S4A) and the STAT1 mRNA content significantly increased (Figs. 4C and S4B). Moreover, Western blotting assays revealed that the regulation of RBM30 expression only affected the intracellular levels of STAT1 and its active form phosphorylated (p)-STAT1 but exerted no significant effect on the levels of its upstream kinases (JAK1 and JAK2) or the internal regulatory proteins of the STAT1 signaling pathway (Protein inhibitor of activated STAT 1 [PIAS1] and Suppressor of cytokine signaling 1 [SOCS1]) (Figs. 4D and S4C). In the subsequent cut-tag experiment (Fig. 4E), we demonstrated that RBM30 overexpression markedly increased the enrichment of STAT1 and p-STAT1 within the genomic DNA region of PD-L1 (Fig. 4F). Taken together, the above experiments demonstrated that high RBM30 expression in HCC can affect the STAT1 signaling pathway, verifying that the STAT1 signaling pathway mediated RBM30 to enhance the expression of PD-L1.
A ChIP-qPCR assays confirmed that there is an interaction between RBM30 and the STAT1 DNA region. B Dual-luciferase reporter gene assays revealed that the activity of the STAT1 signaling pathway was significantly decreased in Lm3 cells after knocking out RBM30, and vice versa in Hep3B cells (n = 3, ***P < 0.0001). C qPCR assays demonstrated that the STAT1 mRNA content was dramatically increased in Hep3B cells after RBM30 overexpression, and vice versa in Lm3 cells (n = 3, ***P < 0.0001). D Western blotting assays indicated that RBM30 knockout resulted in a decrease in the expression of both STAT1 protein and its phosphorylated active form, p-STAT1, in HCC cells, as well as vice versa. However, this manipulation did not have any impact on the levels of other crucial proteins within the STAT1 signaling pathway or endogenous negative regulatory proteins. E The distribution of STAT1 and p-STAT1 on the chromatin genome in the constructed Hep3B-ov-RBM30 and control cell lines was examined using the Cut-Tag approach. F The visualization results of IGV indicated that the enrichment degree of STAT1 and p-STAT1 in the PD-L1 region was significantly enhanced after overexpression of RBM30. G Western blotting assays and qPCR assays were conducted in Lm3-sg-RBM30 cells, in which STAT1-wild type (WT), STAT1 Y701D (phosphomimetic mutant form), and STAT1 Y701F (dominant-negative mutant form) plasmids were transfected, showing that there was no significant rebound in the level of PD-L1 in HCC cells caused by RBM30 knockdown when transfected with inactivated STAT1 Y701F plasmid (n = 3, ***P < 0.0001). H FACS assays were conducted in Lm3-sg-RBM30 cells, in which STAT1-WT, STAT1 Y701D, and STAT1 Y701F plasmids were transfected, revealing that the RBM30 knockout-induced decrease in the expression of PD-L1 on Lm3 cell membranes was restored only when STAT1-WT/STAT1 Y701D plasmids were transferred (n = 3, ***P < 0.0001). I Western blotting and qPCR assays confirmed that blocking the STAT1 signaling pathway or interfering with STAT1 rescued the increased expression of PD-L1 caused by RBM30 overexpression in Hep3B cells (n = 3, ***P < 0.0001, ns indicates no statistical significance). J FACS assays revealed that RBM30 relies on the STAT1 signaling pathway to promote an increase in surface PD-L1 expression in HCC cells (n = 3, ***P < 0.0001, ns indicates no statistical significance).
We further undertook rescue experiments in various cell types. In Lm3/MHCC-97H cells, we introduced the STAT1-wild type (WT) plasmid into RBM30-KO cells, as well as the STAT1-Y701D (a phosphorylated mimic) and STAT1-Y701F (dominant-negative mutant). The results showed that the total and surface PD-L1 expression recovered when the WT and Y701D STAT1 but not Y701F were overexpressed in the Lm3/MHCC-97H-sg-RBM30 cells (Figs. 4G, H and S4D, E). However, in Huh-7/Hep3B cells with STAT1 expression increased owing to RBM30 overexpression, when the elevated expression of STAT1 was knocked down or the RBM30-overexpression cells were treated with STAT1 inhibitors, PD-L1 expression was reduced to varying degree, which was confirmed via qPCR, Western blotting, and FACS assays. Similar results were also found in mouse HCC cells (Fig. 4I, J and F, G), demonstrating that RBM30 can bind to the STAT1 DNA region, increasing the STAT1 content in HCC, which activated the STAT1 signaling pathway to increase the PD-L1 content in HCC.
RBM30 enhances STAT1 transcription by recruiting DOT1L to promote HCC immune evasion
However, it remains unclear how RBM30, as an atypical transcription factor, promotes the openness of the STAT1 DNA region. Thus, we employed co-immunoprecipitation (Co-IP) and LC-MS/MS to identify potential RBM30 binding proteins. In consideration of the transcriptional regulation role of RBM30, we focused on searching for epigenetic modifiers that interact with RBM30 and regulate chromatin DNA accessibility. DOT1L, a universally expressed methyl-transferase responsible for the methylation of histone H3 at the lysine 79 site (H3K79me1/2/3) [34], was found, which was identified by Co-IP assays (Figs. 5A and S5A, B) and the interaction between RBM30 and DOT1L was further confirmed using pulldown assays. Moreover, immunocytochemistry assays confirmed that RBM30 colocalized with DOT1L in the nucleus (Fig. 5B, C), indicating that RBM30 can directly bind to DOT1L in the nucleus to form heterodimers.
A A protein with a molecular weight of approximately 200 KDa was found to show potential interaction with RBM30, and it was identified by mass spectrometry as DOT1L. CoIP assays validated the interaction between RBM30 and DOT1L. B Immunofluorescence assays demonstrated that RBM30 and DOT1L were co-localized in the nucleus of HCC cells. C Pulldown assays revealed that RBM30 binds to DOT1L. D Western blotting assays and qPCR assays revealed that STAT1 signaling pathway enhanced by RBM30 overexpression was significantly reversed after reducing DOT1L expression or inhibiting DOT1L function in HCC cells, accompanied by a significant reduction in the total intracellular PD-L1 protein and mRNA levels (n = 3, ***P < 0.0001, ns indicates no statistical significance). E FACS assays confirmed that reducing DOT1L expression or inhibiting DOT1L function reduced the increased surface PD-L1 expression caused by the overexpression of RBM30 in HCC cells (n = 3, ***P < 0.0001, ns indicates no statistical significance). F ChIP-seq against DOT1L and H3K79me3 conducted in Hep3B-ov-RBM30 and control cells revealed alterations in the enrichment regions of DOT1L and H3K79me3 on chromatin following the overexpression of RBM30. G RBM30-DOT1L complex regulates the expression of multiple genes in HCC. H The changes in the sequences of DOT1L and H3K79me3 binding peaks following RBM30 overexpression demonstrated a substantial enhancement in their enrichment within the specific STAT1 DNA region where RBM30 binds. This led to an increased accessibility of chromatin at the STAT1 DNA region, resulting in an upregulation of STAT1 expression. I ATAC-seq revealed that, by combining DOT1L shRNA and small molecule inhibitors and performing, the enhanced openness of the genomic STAT1 region due to RBM30 overexpression decreased after reducing the protein content of DOT1L or inhibiting the function of DOT1L in cells.
Previous studies have shown that DOT1L, which belongs to the lysine methyltransferase subfamily, is a histone methyl-transferase that can methylate the “Lys-79” of histone H3 during transcription, and DOT1L can bind to DNA at the promoter region and participate in regulating the spatial structure of chromatin DNA [35]. Meanwhile, in this study, RBM30 was found to form a heterodimer with DOT1L, according to multiple omics conjoint analysis, binding to the STAT1 DNA on chromatin to enhance the accessibility of this chromatin region. Building upon these theoretical foundations and experimental findings, ChIP-seq against DOT1L and H3K79me3 were conducted in Hep3B-ov-RBM30 and control cells, respectively, the results revealed that following the overexpression of RBM30, there were varying degrees of alterations observed in the enrichment regions of DOT1L and H3K79me3 on DNA (Figs. 5D and S5C). Furthermore, by integrating the aforementioned findings of high-throughput sequencing (Fig. 5E), we found that the alterations in the sequences of DOT1L and H3K79me3 binding peaks before and after RBM30 overexpression in Hep3B cells indicated a significant increase in the enrichment of these binding peaks within the STAT1 DNA region where RBM30 binds, thereby enhancing accessibility, which confirmed that RBM30 really interacted with DOT1L to form a transcriptional regulatory complex, ultimately enhancing chromatin accessibility at the STAT1 DNA region and upregulating STAT1 expression (Fig. 5F).
To confirm the role of DOT1L in RBM30-enhanced STAT1 signaling, we used small interfering RNA targeting DOT1L and specific inhibitors in RBM30-overexpressing HCC cells. We found that the RBM30 overexpression-induced enhancement of the STAT1 signaling pathway was significantly reversed after reducing DOT1L expression or inhibiting DOT1L function in HCC cells, accompanied by a significant reduction in the total intracellular PD-L1 protein and mRNA levels (Figs. 5G and S5D), and the same decline of PD-L1 expression on the cell membrane surface was observed in FACS assays (Figs. 5H and S5E). Meanwhile, in murine cells, the elevated cytoplasmic and cell membrane Pd-l1 levels resulting from Rbm30 overexpression were attenuated by Dot1l knockdown (Fig. S5F, G).
Further investigations by combining DOT1L shRNA and small molecule inhibitors and performing ATAC-seq showed that the enhanced openness of the genomic STAT1 region due to RBM30 overexpression decreased significantly after reducing the protein content of DOT1L or inhibiting the function of DOT1L in the cells (Fig. 5I), indicating that RBM30 was dependent on DOT1L to activate the STAT1 signaling pathway and promote the increase of PD-L1 content in HCC.
RBM30 forms a heterodimer via its C-terminus interacting with the C-terminal domain of DOT1L
The aforementioned studies have demonstrated that RBM30 relies on the formation of heterodimers with DOT1L to augment the transcriptional regulation of STAT1. The Co-IP assays were applied to explore the protein domains that RBM30 and DOT1L dependance with each other.
To this end, we constructed different truncated plasmids of RBM30 and DOT1L respectively (Figs. 6A and S6A), which demonstrated that the binding of RBM30 to DOT1L is dependent on amino acids 268–359 protein–protein interaction domain (P-PIM) (Figs. 5B and S6B). Coincidentally, other results also revealed that DOT1L relied on its C-terminal domain to interact with RBM30 (Fig.S6C, D), indicating that RBM30 formed a heterodimer via its C-terminus to interact with the C-terminal domain of DOT1L.
A Construction Strategy of RBM30 Short Form. B Co-IP assays demonstrated that RBM30 uses its protein–protein interaction domain (P-PIM) (amino acids 268–359) to bind to DOT1L. C Construction Strategy for DOT1L Short Form. D Co-IP assays demonstrated that DOT1L interacts with RBM30 via its C-terminal domain (amino acids 1156–1537). E The ChIP-qPCR assays revealed that RBM30 relies on its protein-DNA binding domain (P-DBM, amino acids 161–177) for binding with STAT1 DNA. The absence of P-DBM or P-PIM domains rendered RBM30 unable to recruit DOT1L to cluster near the transcription start site of STAT1, thereby impeding H3K79me3 enrichment in this region. F ATAC-seq analysis indicated that the absence of either the P-DBM or P-PIM domain in RBM30 does not result in a significant enhancement of chromatin accessibility at the genomic STAT1 region. G Dual-luciferase reporter gene assays and qPCR assays revealed that RBM30 relies on the P-DBM and P-PIM domains to enhance the activity of the STAT1 signaling pathway (n = 3, ***P < 0.0001, ns indicates no statistical significance). H The Western blotting and qPCR assays demonstrated the pivotal role of the P-DBM and P-PIM domains in enhancing intracellular STAT1 content and activating the STAT1 signaling pathway by RBM30, augmenting intracellular PD-L1 expression. (n = 3, ***P < 0.001, ns indicates no statistical significance). I FACS assays confirmed that RBM30 increased HCC surface PD-L1 expression only when both the P-DBM and P-PIM domains were present at the same time (n = 3, ***P < 0.0001, ns indicates no statistical significance).
We conducted further experiments to verify the roles of the P-PIM domains in RBM30 in promoting PD-L1 expression in HCC. Firstly, ChIP-qPCR assays (anti-Flag) were conducted on HCC cells containing Flag-tagged full-length and truncated RBM30 constructs. The results indicated that direct binding of RBM30 to the STAT1 DNA region was not observed when overexpressing the RBM30-JD3 truncated plasmid lacking the zinc finger domain (161-177 amino acid sequence, i.e., P-DIM). Subsequent ChIP-qPCR assays (Anti-DOT1L/H3K79me3) further showed that in the absence of P-DIM or P-PIM in RBM30, DOT1L and H3K79me3 were unable to enhance enrichment within the STAT1 DNA region (Figs. 6E and S6E). In the subsequent ATAC-seq analysis, we observed that the absence of either the P-DBM or P-PIM domain in RBM30 does not result in a significant enhancement of chromatin accessibility at the genomic STAT1 region (Fig. 6F). This finding suggests that these domains are critical for RBM30’s interaction with chromatin and its recruitment of DOT1L, thereby facilitating increased chromatin openness at the STAT1 DNA locus.
Dual-luciferase reporter gene assays and qPCR assays revealed that only the co-presence of the P-DBM and P-PIM domains in RBM30 could significantly increase intracellular STAT1 mRNA levels and enhance the activity of the STAT1 signaling pathway (Figs. 6G and S6F). Through Western blotting and qPCR assays, we not only confirmed the crucial role of P-PIM and P-DIM in RBM30 for enhancing STAT1 protein expression and activating the STAT1 signaling pathway (indicated by p-STAT1) in HCC cells, but also discovered that the absence of these key domains rendered RBM30 incapable of promoting intracellular PD-L1 content increase (Figs. 6H and S6G). Finally, our FACS assays revealed that the P-DBM and P-PIM domains are the crucial structural components of RBM30 responsible for promoting an increase in PD-L1 content on the cell membrane (Figs. 6I and S6H). These findings demonstrated that RBM30 relied on the P-DBM domain for binding to STAT1 DNA and recruited DOT1L through the P-PIM domain to form a transcriptional regulatory complex, and this complex promoted the enrichment of H3K79me3 in the STAT1 DNA region to enhance its accessibility and facilitate the transcription of STAT1.
RBM30 and DOT1L synergistically enhances PD-L1 expression in HCC
The above study revealed a significant upregulation of RBM30 in hepatocellular carcinoma, which leads to an augmentation in STAT1 levels and subsequent activation of the STAT1 signaling pathway through forming a heterodimer with DOT1L, ultimately resulting in elevated PD-L1 expression in HCC cells. Next, we conducted further validation of this regulatory mechanism in vivo. Analyzing the liver samples obtained from the gene-edited mice (Rbm30−/− and Rbm30KI/K) used in the above experiments, we found that compared with that in the wild-type mice, the expression of Stat1 and p-Stat1 in the Rbm30−/− mice was significantly reduced without influencing the content and activation of upstream key kinases (p-Jak1/2), which was consistent with the results obtained at the cellular level. Conversely, the liver samples from Rbm30-overexpressing mice exhibited increases in their Stat1 and p-Stat1 contents (Fig. 7A). Furthermore, IHC staining using tissue microarrays indicated a positive correlation between RBM30 content and the protein expression of STAT1 (Fig. 7B) and p-STAT1 (Fig. 7C) in tumor tissues, which indicates that the RBM30/DOT1L-STAT1-PD-L1 regulatory axis possesses significant clinical significance. To further explore the role of the RBM30/DOT1L heterodimer in the regulation of PD-L1 in vivo, we conducted subcutaneous tumor formation experiments in mice for an in-depth examination. The overexpression of Rbm30 resulted in increased HCC proliferation, which was significantly reduced upon knocking down Dot1l (Fig. 7D). The subsequent IHC staining unveiled that Rbm30-overexpression tumors exhibited an upregulation of Stat1 protein expression, thereby augmenting the activity of the Stat1 signaling pathway (as indicated by p-Stat1), consequently fostering the expression of Pd-l1, and all these effects were reversed when Dot1l interference was applied, underscoring the dependence of Rbm30-induced expression on Dot1l (Fig. 7E). When the implanted tumors were analyzed using FACS, we observed that the proportion of CD4+ T cells within the immune cell population was consistent across all queues of implanted tumors. However, there was a significant decrease in the percentage of CD8+ T cells in Rbm30-overexpressing tumors compared to control tumors. Conversely, when Dot1l was knocked down, this proportion was reversed (Fig. 7F). Similar results in the IHC staining and FACS of knockout groups were obtained (Fig. S7A, B).
A Analysis of the liver samples obtained from the gene-edited mice (Rbm30−/− and Rbm30KI/K). Immunohistochemical (IHC) staining revealed that in the Rbm30−/− group mice, the expression of Stat1 and p-Stat1 was reduced compared with that in wild-type mice without influencing the content and activation of upstream key phosphorylated Jak1/2. Conversely, the liver samples from RBM30-overexpressing mice (RBM30KI/KI) exhibited an increase in Stat1 and p-Stat1 content. B Immunohistochemical staining assays using TMA displayed a positive correlation between RBM30 content and the protein expression of STAT1 in tumor tissues (R = 0.6671, ***p < 0.001). C Immunohistochemical staining assays using TMA demonstrated a positive correlation between RBM30 content and the protein expression of p-STAT1 in tumor tissues (R = 0.4591, ***p < 0.001). D Overexpression of Rbm30 resulted in increased HCC proliferation, which was significantly decelerated upon knocking down Dot1l content (n = 5, ***P < 0.0001, ns indicates no statistical significance). E IHC staining revealed that Rbm30-overexpressing tumors upregulated Stat1 protein expression, leading to enhanced activity of the Stat1 signaling pathway (indicated by p-Stat1), thus promoting the expression of Pd-l1, and these effects were reversed when Dot1l interference was applied. F The FACS assays revealed a consistent proportion of CD4+ T cells within the immune cell population across all queues of implanted tumors. However, there was a significant reduction in the percentage of CD8+ T cells observed in Rbm30-overexpressing tumors compared to control tumors. Conversely, knocking down Dot1l reversed this proportion (n = 5, ***P < 0.0001, ns indicates no statistical significance).
In conclusion, these findings validate the role of RBM30 in promoting the malignant progression of HCC through activation of the STAT1 signaling pathway, underscoring the clinical significance of the RBM30/DOT1L-STAT1-PD-L1 regulatory axis.
RBM30 knockout enhances the efficacy of HCC immunotherapy
In our previous investigation, we discovered that the knockout of RBM30 resulted in a reduction of PD-L1 expression in hepatocellular carcinoma cells. Additionally, through animal experiments, we observed a significant decrease in HCC tumor volumes in mice with RBM30 knockout cells compared to the control group. Over the past few years, tumor immunotherapy, particularly anti-PD-1/PD-L1 therapy, has achieved remarkable advancements in HCC treatment. Therefore, we conducted further experiments to explore the potential application of RBM30 knockout as a therapeutic approach for HCC (Fig. 8A). The results revealed that simultaneous knockdown of RBM30 and administration of MAb anti-mouse PD-1 (CD279) significantly inhibited HCC tumor volumes compared to using monoclonal antibodies alone or RBM30 knocking down alone (Fig. 8B), which was further confirmed by IHC assays (Fig. 8C). FACS assays further revealed that joint knockout RBM30 treatment and application of anti-PD-1 can significantly improve the proportion of CD8+ T cells without influencing the percent of CD4+ T cells in HCC compared with single drug therapy (Fig. 8D). Take together, this in vivo animal experiment demonstrated that knocking out RBM30 greatly enhanced the therapeutic efficacy of anti-PD-1/PD-L1 treatment in HCC and highlighted the potential value of targeting RBM30 for HCC treatment.
A Schematic diagram demonstrated the plan to explore the application potential of RBM30 knockout in HCC treatment. B The application of InVivoMAb anti-mouse PD-1 (CD279) combined with Rbm30 knockout significantly inhibited HCC tumor volume compared with that seen with a monoclonal antibody or Rbm30 knockout applied separately. C Further IHC assays demonstrated that the application of InVivoMAb anti-mouse PD-1 (CD279) combined with Rbm30 knockout did ulteriorly inhibit the progression of HCC (indicated by Ki-67). D FACS analysis of intratumoral T cells in different treatment groups revealed that the combination therapy of RBM30 knockout and anti-PD-1 application significantly increased the proportion of CD8+ T cells without affecting the percentage of CD4+ T cells in HCC, compared to single drug therapy (n = 5, ***P < 0.0001). E The pattern exhibited in this study is that RBM30, which is highly expressed in HCC, binds to the transcription start region of the STAT1 DNA on chromatin and recruits DOT1L, an important epigenetic protein, to promote H3K79me3 enrichment in these regions, heightening the accessibility of the STAT1 DNA in chromatin, which upregulates STAT1 content in HCC, enhancing the activity of the STAT1 signaling pathway and increasing PD-L1 in HCC to drive tumor immune evasion.
We have discovered that the highly expressed RBM30 in HCC tumor tissue possessed the remarkable ability to bind to the initiation site of STAT1 DNA transcription, thereby recruiting DOT1L to form heterodimers. This unique interaction promoted the enrichment of H3K79me3 near the initiation site of STAT1 DNA transcription, ultimately enhancing the accessibility and facilitating efficient transcription of STAT1. Consequently, it significantly amplified the activity of the STAT1 signaling pathway, leading to an elevated expression level of PD-L1 in HCC tumors (Fig. 8E). This intricate mechanism contributes to an immune evasion phenomenon within hepatocellular carcinoma and accelerates disease progression.
Discussion
Primary liver cancer, especially HCC, is a serious threat to the life and health of Chinese people [36]. Owing to the insidious onset of HCC, less than 30% of patients are suitable for radical treatment at the time of initial diagnosis [37]. Systemic antitumor therapy plays an important role in the treatment of advanced HCC [38]. In recent years, immunotherapy for HCC, represented by anti-PD-1/PD-L1 [39], has developed rapidly. It is quickly becoming the first- and second-line treatment options for HCC as it can control the progression of the disease and prolong the survival time of patients [40]. Previous studies have shown that PD-L1 expression on the surface of HCC is significantly upregulated and is closely related to tumor invasiveness and postoperative recurrence [41]. Therefore, an in-depth exploration of the regulatory mechanism of PD-L1 in HCC will provide a potential theoretical basis for the selection of future anti-PD-1/PD-L1 therapeutic adjuvant targets [42]. In the present study, we found that RBM30 could increase the PD-L1 content in HCC by enhancing STAT1 transcription.
STAT1 is a protein that plays a critical regulatory role inside cells and is involved in the regulation of tumor cell proliferation and immune evasion [43]. In recent years, studies have demonstrated that the function and abundance of STAT1 are regulated by various mechanisms [44], and that various post-translational protein modifications are involved in the regulation of STAT1 protein activity [45]. Phosphorylation at the Tyr701 and Ser727 sites plays an important role in STAT1-mediated interferon-α/γ signaling [46]. PIAS1 can bind to STAT1 to inhibit phosphorylation at these sites, exerting a blocking effect [46]. The cyclic adenosine monophosphate response element-binding protein (CREB)-binding protein can directly bind to STAT1 and enhance the expression of interferon-stimulated genes mediated by STAT1 [47]. In addition, the deubiquitinase OTUB2 can enhance the phosphorylation and dimerization of STAT1 by deubiquitinating STAT1 in tongue and esophageal squamous cell carcinoma, promoting phosphoserine synthesis [48]. Furthermore, research has also found that Su(var)3-9, Enhancer-of-zeste and Trithorax domain containing 2 methyltransferase can regulate the expression of antiviral genes mediated by interferon alpha via mono-methylating the lysine at position K525 of STAT1 [49]. In addition, various intracellular regulatory mechanisms affect the expression of STAT1. In psoriasis, methyltransferase 3 affects the differentiation and function of γδT cells and their subsets by regulating the abundance of N⁶-methyladenosine modification in STAT1 mRNA and the stability of STAT1 mRNA [50]. However, micro-RNA miR-1a-3p promoted IFN-γ-mediated bacterial clearance in macrophages by inducing the degradation of long non-coding RNA Sros1 in the cytoplasm, thereby promoting the stability of STAT1 mRNA [51]. However, in this study, we demonstrated that in HCC, RBM30 directly binds to the STAT1 promoter region and recruits DOT1L to enhance H3K79me3 enrichment in this genomic locus, thereby increasing chromatin accessibility and upregulating STAT1 transcription. This discovery not only expands the regulatory mechanisms of the STAT1 signaling pathway but also provides deeper insights into the molecular pathological mechanisms underlying HCC.
Researches shows that the STAT1 signaling pathway can promote the expression of PD-L1, which binds to PD-1 on the surface of T cells, inhibiting the activation and proliferation of T cells and enabling tumor cells to evade immune surveillance by T cells [18]. After PD-L1 binds to PD-1 on the surface of T cells, it transmits inhibitory signals, reducing the activity of T cells and preventing them from effectively recognizing and attacking tumor cells [52]. When the content of PD-L1 in tumor cells is low, this inhibitory effect weakens, and T cells can be activated more effectively, thereby enhancing the immune surveillance and killing ability against tumor cells [53]. Low expression of PD-L1 in tumor cells means that fewer “do not attack me” signals are sent to the immune system, and the recognition and attack of tumor cells by the immune system will not be overly interfered with, which is conducive to the better anti-tumor effect of immune cells [8]. In this study, we demonstrated that RBM30 promotes the expression of PD-L1 via the STAT1 signaling pathway. In conjunction with the findings from our animal experiments, upon knocking out RBM30 in HCC and combining it with anti-PD-1 therapy, we observed that the therapeutic efficacy in this group was significantly enhanced compared to both the control group and the group treated with anti-PD-1 monotherapy. These results suggest that in future clinical applications, targeting RBM30 may serve as a potential strategy to decrease PD-L1 levels in tumors and thereby enhance the efficacy of immunotherapy for HCC.
DOT1L plays an important role in the regulation of gene expression. In recent years, the function of DOT1L, especially in histone modification, has been intensively studied [54]. Studies have shown that DOT1L plays a role in regulating gene transcription in normal cells and can regulate gene expression by methylating H3K79 [55], a specific site on the histone protein. The degree of methylation of the site can affect the structure and organization of chromatin to regulate gene transcription [56]. Further studies have found that DOT1L is overexpressed or mutated in various cancers, including acute myeloid leukemia, breast cancer, and lung cancer [57], accelerating the malignant progression of these diseases [57]. Additionally, the studies on DOT1L have also led to the development of inhibitors targeting this protein [58], and some of them have entered clinical trials, showing therapeutic potential [59]. In this study, we found that the high expression of RBM30 in HCC could recruit DOT1L near the transcription start site of STAT1 DNA and enhance STAT1 transcriptional activity by enhancing H3K79me3 enrichment in this region to promote HCC immune escape. We also found that treatment with DOT1L inhibitors (EPZ-5676 or SGC0946) could reduce the increased PD-L1 content caused by RBM30 overexpression, which provides a theoretical basis for the potential application of DOT1L inhibitors.
In summary, our study revealed that RBM30 formed a heterodimer with DOT1L, which enhanced the openness of STAT1 DNA by promoting H3K79me3 enrichment, thereby advancing the expression of PD-L1 in HCC by enhancing STAT1 transcription, facilitating HCC immune evasion. These findings illuminate the mechanism of RBM30 function in HCC, further enhancing our understanding of the molecular pathogenesis of HCC, and provide a potential basis for the selection of anti-PD-1/PD-L1 targets in future therapeutic strategies.
Conclusions
In this study, we discovered that elevated RBM30 expression in hepatocellular carcinoma (HCC) promotes the recruitment of DOT1L to the STAT1 transcription start site (TSS), leading to increased H3K79me3 enrichment. This enhances STAT1 transcription and activates the STAT1 signaling pathway, resulting in elevated PD-L1 levels in HCC cells and accelerated immune escape. This finding not only deepens our understanding of the molecular pathogenesis of HCC but also provides a potential theoretical foundation for future clinical diagnosis and therapeutic applications.
Data availability
Raw and processed sequencing data have been deposited at SRA database (PRJNA1089826) and are publicly available as of the date of publication. The rest of the data reported in this paper will be shared by the lead contact upon request. This paper does not report original code. Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request. All data generated or analyzed during this study are included in this published article and its supplementary information files, The high-throughput sequencing data generated in this study are publicly available at SRA datebase (PRJNA1089826).
References
Endo Y, Sasaki K, Moazzam Z, Woldesenbet S, Yang J, Araujo Lima H, et al. The impact of a liver transplant program on the outcomes of hepatocellular carcinoma. Ann Surg. 2023;278:230–8.
Brown Z, Tsilimigras D, Ruff S, Mohseni A, Kamel I, Cloyd J, et al. Management of hepatocellular carcinoma: a review. JAMA Surg. 2023;158:410–20.
He M, Huang Y, Du Z, Lai Z, Ouyang H, Shen J. et al. Lenvatinib, Toripalimab plus FOLFOX Chemotherapy in Hepatocellular Carcinoma Patients with Extrahepatic Metastasis: A Biomolecular Exploratory, Phase II Trial (LTSC). Clin Cancer Res. 2023;29:5104–5115. https://doi.org/10.1158/1078-0432.CCR-23-0060.
Zanuso V, Rimassa L, Braconi C. The rapidly evolving landscape of HCC: Selecting the optimal systemic therapy. Hepatology. 2025;81:1365–1386. https://doi.org/10.1097/HEP.0000000000000572.
Yu S. Immunotherapy for hepatocellular carcinoma: recent advances and future targets. Pharmacol Ther. 2023;244:108387.
El Zarif T, Nassar A, Adib E, Fitzgerald B, Huang J, Mouhieddine T, et al. Safety and activity of immune checkpoint inhibitors in people living with HIV and cancer: a real-world report from the cancer therapy using checkpoint inhibitors in people living with HIV-international (CATCH-IT) consortium. J Clin Oncol. 2023;41:3712–23.
Gao X, Xu N, Li Z, Shen L, Ji K, Zheng Z, et al. Safety and antitumour activity of cadonilimab, an anti-PD-1/CTLA-4 bispecific antibody, for patients with advanced solid tumours (COMPASSION-03): a multicentre, open-label, phase 1b/2 trial. Lancet Oncol. 2023;24:1134–46.
Sharma P, Goswami S, Raychaudhuri D, Siddiqui B, Singh P, Nagarajan A, et al. Immune checkpoint therapy-current perspectives and future directions. Cell. 2023;186:1652–69.
Sánchez-Magraner L, Gumuzio J, Miles J, Quimi N, Martínez Del Prado P, Abad-Villar M, et al. Functional engagement of the PD-1/PD-L1 complex but not PD-L1 expression is highly predictive of patient response to immunotherapy in non-small-cell lung cancer. J Clin Oncol. 2023;41:2561–70.
Wang T, Johmura Y, Suzuki N, Omori S, Migita T, Yamaguchi K, et al. Blocking PD-L1-PD-1 improves senescence surveillance and ageing phenotypes. Nature. 2022;611:358–64.
Huang Q, Wu X, Wang Z, Chen X, Wang L, Lu Y, et al. The primordial differentiation of tumor-specific memory CD8 T cells as bona fide responders to PD-1/PD-L1 blockade in draining lymph nodes. Cell. 2022;185:4049–66.e4025.
Callahan M, Chapman P. PD-1 or PD-L1 blockade adds little to combination of BRAF and MEK inhibition in the treatment of BRAF V600-mutated melanoma. J Clin Oncol. 2022;40:1393–5.
Liu Y, Wang L, Song Q, Ali M, Crowe W, Kucera G, et al. Intrapleural nano-immunotherapy promotes innate and adaptive immune responses to enhance anti-PD-L1 therapy for malignant pleural effusion. Nat Nanotechnol. 2022;17:206–16.
Zhao J, Yap D, Chan Y, Tan B, Teo C, Syn N, et al. Low programmed death-ligand 1-expressing subgroup outcomes of first-line immune checkpoint inhibitors in gastric or esophageal adenocarcinoma. J Clin Oncol. 2022;40:392–402.
Kang Y, Chen L, Ryu M, Oh D, Oh S, Chung H, et al. Nivolumab plus chemotherapy versus placebo plus chemotherapy in patients with HER2-negative, untreated, unresectable advanced or recurrent gastric or gastro-oesophageal junction cancer (ATTRACTION-4): a randomised, multicentre, double-blind, placebo-controlled, phase 3 trial. Lancet Oncol. 2022;23:234–47.
Qin S, Chen Z, Fang W, Ren Z, Xu R, Ryoo B, et al. Pembrolizumab versus placebo as second-line therapy in patients from asia with advanced hepatocellular carcinoma: a randomized, double-blind, phase III trial. J Clin Oncol. 2023;41:1434–43.
Cha J, Chan L, Li C, Hsu J, Hung M. Mechanisms controlling PD-L1 expression in cancer. Mol Cell. 2019;76:359–70.
Cerezo M, Guemiri R, Druillennec S, Girault I, Malka-Mahieu H, Shen S, et al. Translational control of tumor immune escape via the eIF4F-STAT1-PD-L1 axis in melanoma. Nat Med. 2018;24:1877–86.
Ladányi A, Tímár J. Immunologic and immunogenomic aspects of tumor progression. Semin Cancer Biol. 2020;60:249–61.
Philips R, Wang Y, Cheon H, Kanno Y, Gadina M, Sartorelli V, et al. The JAK-STAT pathway at 30: much learned, much more to do. Cell. 2022;185:3857–76.
Xie W, Medeiros L, Li S, Tang G, Fan G, Xu J. PD-1/PD-L1 Pathway: A Therapeutic Target in CD30+ Large CellLymphomas. Biomedicines. 2022;10:1587. https://doi.org/10.3390/biomedicines10071587.
Okada Y, Feng Q, Lin Y, Jiang Q, Li Y, Coffield V, et al. hDOT1L links histone methylation to leukemogenesis. Cell. 2005;121:167–78.
Copeland R. Molecular pathways: protein methyltransferases in cancer. Clin Cancer Res. 2013;19:6344–50.
Chen S, Zhao Y, Liu S, Zhang J, Assaraf Y, Cui W, et al. Epigenetic enzyme mutations as mediators of anti-cancer drug resistance. Drug Resist Updates 2022;61:100821.
Deshpande A, Bradner J, Armstrong S. Chromatin modifications as therapeutic targets in MLL-rearranged leukemia. Trends Immunol. 2012;33:563–70.
Saygin C, Carraway H. Emerging therapies for acute myeloid leukemia. J Hematol Oncol. 2017;10:93.
Liu Y, Wu Q, Sun T, Huang J, Han G, Han H. DNAAF5 promotes hepatocellular carcinoma malignant progression by recruiting USP39 to improve PFKL protein stability. Front Oncol. 2022;12. https://doi.org/10.3389/fonc.2022.1032579.
Han H, Lin T, Wang Z, Song J, Fang Z, Zhang J, et al. RNA-binding motif 4 promotes angiogenesis in HCC by selectively activating VEGF-A expression. Pharmacol Res. 2023;187:106593.
Han H, Lin T, Fang Z, Zhou G. RBM23 drives hepatocellular carcinoma by activating NF-κB signaling pathway. BioMed Res Int. 2021;2021:6697476.
Han H, Yuan Y, Li C, Liu L, Yu H, Han G, et al. RNA-binding motif protein 28 enhances angiogenesis by improving STAT3 translation in hepatocellular carcinoma. Cancer Lett. 2024;604. https://doi.org/10.1016/j.canlet.2024.217191.
Zhao W, Lin T, Fang Z, Zhou H, Du Y, Han H. NAT10 promotes the malignant progression of hepatocellular carcinoma through upregulating RelA/p65 acetylation. J Biol Regul Homeost Agents. 2023;37:2935–46.
Pinter M, Jain R, Duda D. The current landscape of immune checkpoint blockade in hepatocellular carcinoma: a review. JAMA Oncol. 2021;7:113–23.
Han H, Shi Q, Zhang Y, Ding M, He X, Liu C. et al. RBM12 drives PD-L1-mediated immune evasion in hepatocellular carcinoma by increasing JAK1 mRNA translation. Oncogene. 2024;43:3062–3077. https://doi.org/10.1038/s41388-024-03140-y.
Willemsen L, Prange K, Neele A, van Roomen C, Gijbels M, Griffith G, et al. DOT1L regulates lipid biosynthesis and inflammatory responses in macrophages and promotes atherosclerotic plaque stability. Cell Rep. 2022;41:111703.
Gilan O, Talarmain L, Bell C, Neville D, Knezevic K, Ferguson D, et al. CRISPR-ChIP reveals selective regulation of H3K79me2 by menin in MLL leukemia. Nat Struct Mol Biol. 2023;30:1592–606.
Sangro B, Chan S, Meyer T, Reig M, El-Khoueiry A, Galle P. Diagnosis and management of toxicities of immune checkpoint inhibitors in hepatocellular carcinoma. J Hepatol. 2020;72:320–41.
Johnson P, Zhou Q, Dao D, Lo Y. Circulating biomarkers in the diagnosis and management of hepatocellular carcinoma. Nat Rev Gastroenterol Hepatol. 2022;19:670–81.
Kanwal F, Singal A. Surveillance for hepatocellular carcinoma: current best practice and future direction. Gastroenterology. 2019;157:54–64.
Pinter M, Scheiner B, Peck-Radosavljevic M. Immunotherapy for advanced hepatocellular carcinoma: a focus on special subgroups. Gut. 2021;70:204–14.
Rimassa L, Finn R, Sangro B. Combination immunotherapy for hepatocellular carcinoma. J Hepatol. 2023;79:506–15.
Shen Y, Hsu C, Jeng Y, Ho M, Ho C, Yeh C, et al. Reliability of a single-region sample to evaluate tumor immune microenvironment in hepatocellular carcinoma. J Hepatol. 2020;72:489–97.
Kang H, Oh J, Chun S, Kim D, Ryu Y, Hwang H, et al. Immunogenomic landscape of hepatocellular carcinoma with immune cell stroma and EBV-positive tumor-infiltrating lymphocytes. J Hepatol. 2019;71:91–103.
Chen Y, Hao X, Sun R, Wei H, Tian Z. Natural killer cell-derived interferon-gamma promotes hepatocellular carcinoma through the epithelial cell adhesion molecule-epithelial-to-mesenchymal transition axis in hepatitis B Virus transgenic mice. Hepatology. 2019;69:1735–50.
Li Y, Zhang C, Martincuks A, Herrmann A, Yu H. STAT proteins in cancer: orchestration of metabolism. Nat Rev Cancer. 2023;23:115–34.
Han H, Zhu W, Lin T, Liu C, Zhai H. N4BP3 promotes angiogenesis in hepatocellular carcinoma by binding with KAT2B. Cancer Sci. 2022;113:3390–404.
Ivashkiv L. IFNγ: signalling, epigenetics and roles in immunity, metabolism, disease and cancer immunotherapy. Nat Rev Immunol. 2018;18:545–58.
Subramani A, Hite M, Garcia S, Maxwell J, Kondee H, Millican G, et al. Regulation of macrophage IFNγ-stimulated gene expression by the transcriptional coregulator CITED1. J Cell Sci. 2023;136.
Chang W, Luo Q, Wu X, Nan Y, Zhao P, Zhang L, et al. OTUB2 exerts tumor-suppressive roles via STAT1-mediated CALML3 activation and increased phosphatidylserine synthesis. Cell Rep. 2022;41:111561.
Chen K, Liu J, Liu S, Xia M, Zhang X, Han D, et al. Methyltransferase SETD2-mediated methylation of STAT1 Is critical for interferon antiviral activity. Cell. 2017;170:492–506.e414.
Xiao Z, Wang S, Tian Y, Lv W, Sheng H, Zhan M, et al. METTL3-mediated m6A methylation orchestrates mRNA stability and dsRNA contents to equilibrate γδ T1 and γδ T17 cells. Cell Rep. 2023;42:112684.
Xu H, Jiang Y, Xu X, Su X, Liu Y, Ma Y. et al.Inducible degradation of lncRNA Sros1 promotes IFN-γ-mediated activation of innate immune responses by stabilizing Stat1 mRNA.Nat Immunol. 2019;20:1621–1630. https://doi.org/10.1038/s41590-019-0542-7.
Jiang Y, Zheng Y, Zhang YW, Kong S, Dong J, Wang F, et al. Reciprocal inhibition between TP63 and STAT1 regulates anti-tumor immune response through interferon-γ signaling in squamous cancer. Nat Commun. 2024;15.
Zietse M, van Leeuwen RWF, Barjesteh van Waalwijk van Doorn-Khosrovani S, de Boer JE, Dupree R, Mathijssen RHJ, et al. Interchangeability of immune checkpoint inhibitors: an urgent need for action. Lancet Oncol. 2024;25:e611–e616. https://doi.org/10.1016/S1470-2045(24)00212-2.
Zhao S, Allis C, Wang G. The language of chromatin modification in human cancers. Nat Rev Cancer. 2021;21:413–30.
Yi Y, Ge S. Targeting the histone H3 lysine 79 methyltransferase DOT1L in MLL-rearranged leukemias. J Hematol Oncol. 2022;15:35.
Campbell R, Tummino P. Cancer epigenetics drug discovery and development: the challenge of hitting the mark. J Clin Invest. 2014;124:64–9.
Alexandrova E, Salvati A, Pecoraro G, Lamberti J, Melone V, Sellitto A, et al. Histone methyltransferase DOT1L as a promising epigenetic target for treatment of solid tumors. Front Genet. 2022;13:864612.
Wood K, Tellier M, Murphy S. DOT1L and H3K79 methylation in transcription and genomic stability. Biomolecules. 2018;8. https://doi.org/10.3390/biom8010011.
Gu X, Hua Y, Yu J, Yang L, Ge S, Jia R, et al. Epigenetic drug library screening reveals targeting DOT1L abrogates NAD synthesis by reprogramming H3K79 methylation in uveal melanoma. J Pharm Anal. 2023;13:24–38.
Acknowledgements
We sincerely thank the central Laboratory of Taizhou People’s Hospital Affiliated to Nanjing Medical University for the help of instruments and equipment. We thank Home for Researchers editorial team (www.home-for-researchers.com) for language editing service and drawing the pattern diagram.
Funding
This project was supported by National Natural Science Foundation of China (82303115), The Youth Fund of Taizhou People’s Hospital Affiliated to Nanjing Medical University (TZKY20240109), Scientific research start-up fund of Taizhou People’s Hospital (QDJJ202106), Taizhou Society Development Project, Jiangsu, China (TS202308), General Project of Jiangsu Provincial Health Commission (H2023030), China Postdoctoral Science Foundation (2024M751162), Top Talent Support Program for young and middle-aged people of Wuxi Health Committee (HB2023060).
Author information
Authors and Affiliations
Contributions
Zhicheng Gong and Hexu Han performed study concept and design; Hexu Han and Chen Gong completed the experimental part of the project; Yue Zhang and Cuixia Liu performed development of methodology and writing, review and revision of the paper; Junxing Huang provided acquisition, analysis and interpretation of data, and statistical analysis; Yifan Wang and Dakun Zhao provided technical and material support. All authors read and approved the final paper.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Ethics approval and consent to participate
The experimental protocol was approved by the Institutional Ethics Review Board of The Affiliated Taizhou People’s Hospital of Nanjing Medical University (2022-008-01). Animal ethical certificates were approved by the Experimental Center of Jiangsu Hanjiang Biotechnology Co., Ltd (HJSW-23050302), and the animals were raised according to animal welfare laws.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Han, H., Gong, C., Zhang, Y. et al. RBM30 recruits DOT1L to activate STAT1 transcription and drive immune evasion in hepatocellular carcinoma. Oncogene 44, 3955–3973 (2025). https://doi.org/10.1038/s41388-025-03550-6
Received:
Revised:
Accepted:
Published:
Issue date:
DOI: https://doi.org/10.1038/s41388-025-03550-6
