RBM14 as a novel epigenetic-activated tumor oncogene is implicated in the reprogramming of glycolysis in lung cancer

Gene expression profiles were obtained from The Cancer Genome Atlas (TCGA) database (https://​cancergenome.​nih.​gov/​). Protein expression profiles were obtained from Clinical Proteomic Tumor Analysis Consortium (CPTAC) database (https://​proteomic.​datacommons.​cancer.​gov/​pdc/​). A summary of clinical data is shown in Supplementary Table S1-2.

Human bronchial epithelial (HBE) cells and lung cancer cell lines (A549, PC9, H1299, and H1650) were obtained from ATCC. All cells were cultured in RPMI-1640 medium (Hyclone, Logan, USA) supplemented with 10% FBS (Gibco, Grand Island, USA) and antibiotics (100 units/ml penicillin and 100 μg/ml streptomycin) at 37°C in a humidified atmosphere (95% air, 5% CO₂).

The specific siRNA of RBM14 and YY1 overexpressed vector were ordered from Sangon Biotech (Shanghai, China). The sequence of si-RBM14 is 5′-GCATTCTGGCCATAGAGCTCGTATT-3′. The sequence of the scrambled siRNA is 5′-TGCTGACTCCAAAGCTCTG-3′. After being incubated overnight, the siRNAs or plasmids were transfected by using lipofectamine 2000 reagent (ThermoFisher Scientific, Waltham, USA). Forty-eight hours after transfection, cells were harvested for further analysis.

Total RNA was extracted using TRIzol® (Thermo Fisher Scientific). PrimeScript RT kit was used to reverse the mRNA into cDNA, and the SYBR-Green PCR Master One-Mix kit (TransGen, Biotech, Co., Ltd.) was used for quantitative real-time PCR. β-actin was used as an internal control. The sense and anti-sense primers were listed in Table 1.

A Simple ChIP Enzymatic Chromatin IP Kit (#9002, Cell Signaling Technology, Danvers, USA) was used to evaluate the accumulation of YY1, EP300, H3K9ac, and H3K27ac in RBM14 promoter according to the manufacturer’s instructions. Briefly, after being crosslinked with EBM-2 containing 1% formaldehyde, the crosslinked cells were collected in a lysis buffer containing 1% PMSF. Chromatin was digested by micrococcal nuclease, and 2% of aliquots of lysate were used as input control. Lysates were incubated with 3 μg primary antibody or normal rabbit IgG, followed by immunoprecipitation with protein G agarose beads and incubation at 4 °C overnight with gentle shaking. DNA crosslink was reversed by the addition of 5 mol/L NaCl and Proteinase K at 65 °C for 2 h. Immunoprecipitated DNA was purified and amplified by PCR using specific primers. The IgG (ab172730, Abcam), anti-EP300 (ab275378, Abcam), anti-YY1 (ab109228, Abcam), anti-H3K9ac antibody (ab32129, Abcam), and anti-H3K27ac antibody (ab4729, Abcam) were used. Immunoprecipitated DNAs were analyzed by qPCR. Primer sequences targeting RBM14 promoter (−91~+7; ch11: 66616629-66616727, named RBM14-promoter) were as follows: sense, 5′-CATTCCTGAGGAGGACTGCC-3′ and anti-sense, 5′-TCTTCATTTTGTCGCCGCAG-3′.

Cells were lysed by RIPA buffer (P0013C, Beyotime, Jiangsu, China), and the protein concentration was measured using a BCA kit (P0010, Beyotime). The protein was separated using SDS-PAGE and transferred onto a PVDF membrane (ISEQ00010, EMD Millipore). After blocking with 5% skimmed milk, the membrane was incubated with primary antibodies against RBM14 (ab70636, Abcam), YY1 (#63227, Cell Signaling Technology), p300 (#54062, Cell Signaling Technology), and β-actin (ab179467, Abcam) over-night at 4°C. After washing, the secondary antibody (Cell Signaling Technology) was added for 1 h of incubation. Protein signals were analyzed using an enhanced chemiluminescence substrate reagent kit (P0018M, Beyotime).

Cells were lysed with IP buffer (1 mM EDTA, 20 mM HEPES, 150 mM NaCl, 0.05% sodium deoxycholate, and 0.05% NP-40) added protease inhibitors and phosphatase inhibitor cocktails. After centrifugation at 12,000 rpm for 10 min, the cell lysates were incubated with indicated antibody and protein A/G PLUS agarose beads (#9863, #37478, Cell Signaling Technology) overnight at 4 °C. After washing, the beads were lysed with loading buffer and boiled at 100 °C for 10 min. Then, the protein was detected by Western blotting.

The transfected cells (1 × 10⁴ per well) were cultured in 96 well plates for 72 h of incubation. Then, 10 μl Cell Counting Kit (CCK)-8 solution (C0038, Beyotime) was added to each well and incubated at 37°C for 2 h. The absorbance was detected at 450 nm using a microplate reader.

Apoptotic cells were detected using an Annexin V-fluorescein isothiocyanate (FITC) Apoptosis Detection kit (#556547, BD Biosciences) according to the manufacturer’s instructions. Briefly, cells were re-suspended in 600 μl binding buffer. Then, 5 μl Annexin V/FITC and 5 μl propidium iodide (PI) were added. After being stained in dark for 15 min at room temperature, cells were analyzed by using a FACS Calibur (BD Biosciences).

Forty-eight hours after transfection, PC9 and A549 cells were cultured in the phenol red-free medium for 24 h. The level of glucose and lactate in the culture supernatant was detected by Glucose Assay Kit (#361510, Rsbio, Shanghai, China) and Lactate Assay Kit (#A019–2, Nanjing Jiancheng Bioengineering Institute, China) according to the manufacturer’s instructions. A Seahorse XF Glycolysis Stress Test kit (#103020-100, Agilent Technologies, Santa Clara, USA) was used to evaluate ECAR according to the manufacturer’s instructions. The basal ECAR was measured under the basal condition, followed by the sequential addition to each well of glucose (10 mM), oligomycin (2 mM), and 2-deoxyglucose (100 mM).

All data analyses were performed using GraphPad Prism software (v6.02). Quantitative results are displayed as the mean ± standard error. Two-tailed Student’s t test and one-way analysis of variance (ANOVA) followed by Tukey’s test were used to compare the differences in two groups and multiple groups. Kaplan–Meier survival curve was analyzed with the log-rank test. Each experiment was conducted three times at a minimum. P < 0.05 was considered statistically significant.

We firstly assessed the level of RBM14 in various types of tumor tissues by Tumor Immune Estimation Resource (TIMER, cistrome.shinyapps.io/timer) database and found that RBM14 expression was significantly increased in BLCA, CESC, CHOL, COAD, HNSC, LICH, LUAD, LUSC, PRAD, READ, and STAD (Fig. 1A). Decreased RBM14 level was found in BRCA, KICH, KIRC, KIRP, and THCA. Here, we focused on exploring the function of RBM14 in lung cancer. Correlation analysis between RBM14 expression and prognostic value was assessed by the Kaplan-Meier Plotter database (Fig. 1B,C). A high expression of RBM14 in patients with LUAD was related to poor overall survival. However, in LUSC patients, the expression of RBM14 was uncorrelated to overall survival. Moreover, the protein level of RBM14 also was found to be upregulated in CPTAC LUAC samples (Fig. 1D). Besides, the mRNA and protein levels of RBM14 were significantly enhanced in lung cancer cell lines when compared with its levels in HBE cells (Fig. 1E, F). To reveal the relationship between RBM14 level and clinical-pathological parameters of LUAD patients, the UALCAN database was fully utilized. The data illustrated that RBM14 expression was closely associated with TP53 mutation status and individual cancer stage, whereas there was no relation between RBM14 expression and patient’s gender, patient’s age, patient’s smoking habits, and nodal metastasis status (Fig. S1). These results suggest that RBM14 is upregulated and significantly associated with the prognosis of LUAC.

Epigenetic regulation of promoters plays a very significant role in the expression of downstream genes. Considering that DNA methylation is an important mode of epigenetic modification, we explored the DNA methylation status of RBM14 promoter in the MEXPRESS database (Fig. 2A). The results indicated that methylation at CpG locations 66616247, 66616562, 66617965, 66624432, and 66626628 exhibited a statistically significant relationship with RBM14 expression. Besides, a significant association also was detected between RBM14 expression and copy number alteration. The data from UALCAN database suggested that the promoter methylation level of RBM14 in LUAD was significantly decreased (Fig. 2B). Moreover, the treatment of 5-aza-dC, an inhibitor for DNA methylation, slightly upregulated the level of RBM14 mRNA in four lung cancer cell lines (Fig. 2C), indicating that the expression of RBM14 was regulated by DNA methylation.

Histone acetylation is an alternative mode of epigenetic regulation. EP300 is a histone acetyltransferase that not only possess acetyltransferase activity but is also able to bind to different transcription factors to form different transcriptional regulatory complexes that regulate the expression of target genes. By analyzing the ChIP-Seq data of A549 cells in ENCODE database, we found that there was an abundant accumulation of H3K9ac, H3K7ac, EP300, and YY1 in RBM14 promoter (Fig. S2A). Thus, we wondered whether the YY1-EP300 axis was implicated in the transcription of RBM14. By analyzing the association between RBM14 expression and EP300 and YY1 levels using TIMER database, we found that RBM14 expression was strongly associated with the expression of EP300, CBP, and CTCF in various types of tumor tissues (Fig. S2B). In LUAD tissues, RBM14 expression shows a positive correlation between the expression of EP300 and YY1 (Fig. 3A). LUAD tissues showed higher YY1 mRNA levels, but no significant change in the level of EP300 compared with normal tissues (Fig. 3B). It was worth noting that the protein level of EP300, and YY1 was increased in LUAD tissues (Fig. 3C). Besides, the survival curve showed that EP300 and YY1 cannot serve as an independent prognostic factor for LUAD (Fig. 3D). All these data indicated that the transcription of RBM14 may be regulated by YY1-EP300 axis-mediated histone acetylation.

The results of the ChIP assay also confirmed the accumulation of EP300 and YY1 in the promoter region of RBM14 (Fig. 4A,B). YY1 and EP300 can colocalize to the promoter region of RBM14, indicating potential interactions between YY1 and EP300 protein. To test this hypothesis, the co-IP assay was performed. The data indicated that there could be an interaction between YY1 and EP300 protein (Fig. 4C). Moreover, we found that YY1 overexpression increased the accumulation of YY1, EP300, H3K9ac, and H3K27ac in the promoter region of RBM14 in both PC9 and A549 cells (Fig. 4D-E). YY1 overexpression upregulated the mRNA level of YY1 and RBM14 but did not affect EP300 expression (Fig. 4F, G). These data suggest that YY1 overexpression enhances the recruitment of EP300, leading to the acetylation of H3 in the promoter regions of RBM14 and transcriptional activation of RBM14.

We next explored the role of YY1-RBM14 on lung cancer cell growth and apoptosis. The specific siRNA for RBM14 (si-RBM14) decreased the protein level of RBM14 and did not affect the protein levels of YY1 (Fig. 5A). YY1 overexpression increased the protein level of YY1 and RBM14. Live-cell counting and CCK-8 assay displayed that YY1 overexpression significantly promoted the proliferation, whereas RBM14 knockdown inhibited the proliferation of PC9 and A549 cells (Fig. 5B-C). More apoptotic cells were present in RBM14 knockdown cells and lesser apoptotic cells were observed in YY1 upregulated cells (Fig. 5D). RBM14 knockdown reversed the change of growth and apoptosis in PC9 and A549 cells induced by YY1 overexpression. These results suggested that YY1-induced transcriptional activation of RBM14 is involved in the growth and apoptosis of lung cancer.

The malignant proliferation of tumor cells depends on the energy supply of glycolysis. We then explored the function of the YY1-RBM14 axis on glycolysis. YY1 overexpression increased glucose consumption, lactate production, and ECAR in PC9 and A549 cells (Fig. 6A–D). However, RBM14 knockdown had the opposite effect. RBM14 knockdown rescued the enhanced glucose consumption, lactate production, and ECAR induced by YY1 overexpression. To further confirm the effect of the YY1-RBM14 axis on glycolysis, the glycolysis-related gene expression was assessed after YY1 overexpression or RBM14 knockdown (Fig. 6E–J). YY1 overexpression increased the mRNA level of HK2, PDK1, PKM2, LDHA, GLUT1, and GLUT3. RBM14 knockdown decreased the mRNA level of HK2, PDK1, PKM2, LDHA, GLUT1, and GLUT3 and recovered the effect on glycolysis-related gene expression mediated by YY1 overexpression. These results indicate that YY1 promotes glycolysis of lung cancer by enhancing RBM14 expression.

We also estimated the relationship between the YY1-RBM14 axis and immune infiltration by using TIMER database. High YY1 expression in LUAD tissues increased the infiltration of immune cells, such as CD8+ T cells, neutrophils, and macrophages (Fig. S3A). However, low YY1 expression in LUAD tissues increased the infiltration of B cells and Tregs. Besides, EP300 expression was positively related to the infiltration of CD4+ T cells, CD8+ T cells, Tregs, neutrophils, and macrophages (Fig. S3B). As shown in Fig. S3D, the expression of RBM14 had a positive correlation with the infiltration of CD4+ T cells, Tregs, and neutrophils. However, RBM14 expression did not correlate with the infiltration of CD8+ T cells, B cells, and macrophages. These results indicate a close association between the YY1-RBM14 axis and immune infiltration in LUAD.