Abstract
Background
Aerobic glycolysis is a hallmark of cancers including liver hepatocellular carcinoma (LIHC). RNA m5C methylation is involved in LIHC progression. However, the effect of a m5C writer, NSUN5, on glycolysis in LIHC remains not known. The present study aimed to investigate the effect of NSUN5 on glycolysis in LIHC and the molecular mechanism.
Methods
NSUN5 and EFNA3 expression data were acquired from The Cancer Genome Atlas database. Cell viability and glycolysis were evaluated. Tumor growth was evaluated using the xenograft tumor model. The effect of NSUN5 on EFNA3 m5C methylation was evaluated using methylated RNA immunoprecipitation and dual-luciferase reporter assay.
Results
We found that NSUN5 and EFNA3 expression was increased in LIHC and related to poor survival. Knocking down NSUN5 inhibited LIHC cell viability and glycolysis in vitro, and inhibited tumor growth and glycolysis in vivo. Moreover, the expression of NSUN5 was positively correlated with that of EFNA3. NSUN5 stabilized EFNA3 by promoting m5C modification of EFNA3. Additionally, overexpression of EFNA3 reversed the inhibition of cell viability and glycolysis induced by NSUN5 silence.
Conclusion
Silencing of NSUN5 decelerates LIHC progression by inhibiting glycolysis mediated by EFNA3 with m5C modification, highlighting the potential of NSUN5 as a therapeutic target for LIHC.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12885-025-14714-8.
Keywords: Liver hepatocellular carcinoma, NSUN5, Glycolysis, M5C methylation, EFNA3
Introduction
Liver cancer is a common fatal cancer worldwide. According to epidemiological data, in 2024, the number of new cases of liver cancer in China exceeded 240,000 and the number of deaths exceeded 200,000 [1]. Liver hepatocellular carcinoma (LIHC) is the main type of liver cancer, accounting for 80% of primary liver cancer cases [2]. Most patients with LIHC are diagnosed at an advanced stage, and the current conventional chemotherapy and radiotherapy have limited efficacy in the treatment of advanced LIHC, leading to poor prognosis [3, 4]. Hence, further elucidation of the pathophysiology of HCC is instrumental in exploring potential therapeutic strategies for LIHC.
Glycolysis is a significant hallmark of tumors [5]. Despite the hypoxic environment prevalent in most tumors, they tend to consume glucose and produce lactic acid through glycolysis even under hypoxia or aerobic conditions, thereby acquiring energy [6]. Aerobic glycolysis is initially identified in HCC, which plays a role in controlling LIHC cell malignant phenotypes, such as proliferation, invasiveness, metastasis, drug resistance, and immune evasion [7]. However, the regulatory mechanisms of glycolysis in LIHC remain largely unknown.
5-methylcytosine (m5C) methylation is a common post-transcriptional modification in eukaryotes, particularly in mammals, and is involved in a variety of physiological and pathological processes [8]. Aberrant m5C methylation is closely linked to the occurrence and development of cancers [9]. The processes of m5C modification are mediated by methyltransferases, demethylases, and binding proteins [10]. NSUN5 is a m5C methyltransferase that methylates cytosine into m5C. Previous studies have indicated that NSUN5 functions as an oncogene in LIHC by facilitating cell proliferation, migration, and invasion [11, 12]. However, its impact on glycolysis remains largely unknown. Consequently, comprehending the regulation of NSUN5 on glycolysis in LIHC may offer new perspectives on its pathogenesis.
In the present study, we aimed to investigate the role of NSUN5 in LIHC, particularly its impact on glycolysis. Moreover, we identified the downstream factor of NSUN5. This study will reveal the regulatory mechanism of glycolysis in LIHC and provide theoretical support for NSUN5 as a therapeutic target for LIHC.
Materials and methods
Bioinformatic analysis
The information on tumor and normal tissue samples of patients with LIHC was downloaded from The Cancer Genome Atlas (TCGA) database (https://tcga-data.nci.nih.gov/tcga/). Differentially expressed genes were analyzed using the R “limma” package. Genes that met condition|log2(Fold Change)|>1 and P < 0.05 were regarded as differentially expressed genes between tumor and normal groups. Results were visualized using the volcano plot. According to previous studies [13, 14], a total of 18 m5C-related regulators were selected, including 11 m5C writers (NOP2, NSUN2-7, DNMT1, TRDMT1, DNMT3A, and DNMT3B), 4 erasers (TET1-3 and ALKBH5), and 3 readers (ALYREF, YBX1, and YTHDF2). The expression profile of these 18 regulators in the tumor and normal groups was selected and visualized using the heatmap. The expression of NSUN5 and EFNA3 in different tumor stages and grades was analyzed using the R “ggplot2” package. Survival analysis was evaluated using the R “survival” package, and the survival curve was estimated using the Kaplan-Meier method. Correlation analysis was performed using the R “Cor” package. Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment pathway analysis was performed by the R “clusterProfiler” package. M5C-modified sites in EFNA3 mRNA were predicted using the RNAm5Cfinder tool (http://www.rnanut.net/rnam5cfinder/). The potential sites with random forest probability score ≥ 0.25 were selected.
Cell culture
Normal liver epithelial cells (THLE3) and LIHC cell lines (Hep3B, SNU449, Focus, and HA22T) were purchased from the American Type Culture Collection (Manassas, VA, USA). LIHC cell lines including Huh7 and HCCLM3 were purchased from Procell (Wuhan, China). Hep3B and Huh7 were maintained in Dulbecco’s modified Eagle’s medium (DMEM; Gibco, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (Gibco). HCCLM3 cells were maintained in DMEM supplemented with 20% FBS. SNU449, Focus, and HA22T/VGH cells were cultured in Roswell Park Memorial Institute (RPMI)−1640 (Gibco) supplemented with 10% FBS. THLE3 cells were cultured in bronchial epithelial cell growth medium (BEGM; Lonza, Basel, Switzerland) supplemented with 10% FBS. The culture condition was 37 °C, 5% CO2.
Cell transfection
NSUN5 short hairpin RNA (shNSUN5), short hairpin RNA negative control (shNC), EFNA3 overexpression plasmids (oeEFNA3), and empty vector (oeNC) were acquired from GenScript (Nanjing, China). The plasmids were transfected into SNU449 and Huh7 cells using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA). Briefly, the cells were seeded into 6-well plates one day before transfection. Lipofectamine 2000 was diluted in 250 µl Opti-MEM (Thermo Fisher Scientific, Waltham, MA, USA) and was left for 5 min, while 2 µL plasmids were diluted with another 250 µl Opti-MEM. They were mixed and incubated with the cells for 48 h.
Cell viability analysis
The cell counting kit-8 (CCK-8; Dojingdo, Tokyo, Japan) was used to assess cell viability. Cells were placed in 96-well plates at a concentration of 2 × 103 cells per well and cultured for 24 h. Afterward, 10 µL CCK-8 solution was introduced for incubation for 2 h, and the absorbance at 450 nm was measured using a microplate reader.
Glucose uptake analysis
Glucose uptake was detected using a glucose uptake assay kit (Abcam, Cambridge, MA, USA). SNU449 and Huh7 cells were placed in 96-well plates and starved in the serum-free medium overnight. After incubating with 100 µL KRPH buffer containing 2% BSA for 40 min, the cells were stimulated with 1 µM insulin for 20 min, followed by incubation with 10 mM 2-DG for 20 min. Then, cells were lysed using 80 µL extraction buffer and heated at 85 °C for 40 min. After incubating with the neutralization buffer, the supernatant was collected and diluted using the detection buffer. The sample was incubated with reaction mix A at 37 °C for 1 h and mixed with reaction mix B. The absorbance was detected at 412 nm using a microplate reader.
Lactate production analysis
Lactate production was measured using a lactate assay kit (Cell biolabs, San Diego, CA, USA). The culture medium of SNU449 and Huh7 cells was collected and added to 96-well microtiter plates at a volume of 50 µL. After incubating with 50 µL reaction mix at 37 °C for 45 min in the dark, the absorbance was detected at 540 nm with a spectrophotometric microplate reader.
Seahorse analysis
Extracellular acidification rate (ECAR) was detected using the Seahorse XFe96 analyzer (Seahorse Bioscience, Billerica, MA, USA). SNU449 and Huh7 cells were seeded in 96-well XF cell culture microplates and cultured for 24 h. The cells were successively exposed to 10 mM glucose, 1 µM oligomycin, and 100 mM 2-deoxyglucose (2-DG). Data were analyzed using the Seahorse Wave Desktop software.
RNA isolation and quantitative real-time PCR (qPCR)
Total RNA was isolated from LIHC cells and tumors using TRIzol reagent (Invitrogen). After evaluating RNA concentration and purification, cDNA first chain was synthesized using the PrimeScript II reverse transcriptase (Takara, Beijing, China). mRNA expression was detected using qPCR using the TB Green Premix Ex Taq II FAST qPCR (Takara, Beijing, China). β-actin served as the normalization. The results were calculated using the 2−ΔΔCT method. Specific primer sequences used for qPCR were as follows: NSUN5 F: 5’-CGCTACCATGAGGTCCACTAC-3’, R: 5’-GCATCTCGCACCACGTCTT-3’; HK2 F: 5’-TTGACCAGGAGATTGACATGGG-3’, R: 5’-CAACCGCATCAGGACCTCA-3’; PKM2 F: 5’-ATGTCGAAGCCCCATAGTGAA-3’, R: 5’-TGGGTGGTGAATCAATGTCCA-3’; EFNA3 F: 5’-TGGGAAGCGGAGAAGCC-3’, R: 5’-CAGCAGCAGCAGCGGAG-3’; β-actin F: 5’-CATGTACGTTGCTATCCAGGC-3’, R: 5’-CTCCTTAATGTCACGCACGAT-3’.
Xenograft tumor model establishment
BALB/c nude mice (4-6-week-old, male, 18–20 g) were purchased from Vital River (Beijing, China). The mice were randomly divided into two groups: adeno-associated viruses (AAV) carrying shNC (AAV-shNC) and AAV carrying shNSUN5 (AAV-shNSUN5). Each group contained five mice. AAV-shNSUN5 and AAV-shNC were transfected into Huh7 cells. The transfected cells were suspended in PBS and were subcutaneously injected into the right back of mice (5 × 106 cells). Tumor volume was measured every 7 days and calculated using the (width2 × length)/2 formula. After 4 weeks of cell injection, all mice were sacrificed by inhaling isoflurane. Tumors were collected, imaged, and weighed. Moreover, tumors were fixed in 10% formalin for further use.
Immunohistochemistry (IHC)
Tumors were embedded in paraffin and sectioned at 5-µm thickness. After deparaffinating and rehydrating, the sections were boiled with citrate buffer for antigen retrieval and incubated with 3% hydrogen peroxidase to remove endogenous peroxidases. The sections were incubated with anti-ki-67 at 4 °C overnight, and subsequently incubated with horseradish peroxidase-conjugated secondary antibody at 37 °C for 30 min. The diaminobenzidine (DAB) substrate kit (Roche, Basel, Switzerland) was used for color development.
Western blotting
Tumor tissues were lysed using radio immunoprecipitation assay (RIPA) buffer (Beyotime, Shanghai, China) supplemented with protease inhibitor cocktail on ice and centrifuged at 12,000 g for 10 min to collect the supernatant. After detecting protein concentration, 30 µg protein of each sample was separated using sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred onto polyvinylidene fluoride membrane. Following blocking with 5% skim milk, the membranes were incubated with primary antibodies against NSUN5, EFNA3, HK2, PKM2, and ACTB at 4 °C overnight, and then incubated with secondary antibody at room temperature for 1 h. Finally, bands were visualized using the BeyoECL Plus reagent (Beyotime).
Methylated RNA immunoprecipitation (MeRIP)
EFNA3 m5C levels were measured using the m5C MeRIP kit (BersinBio, Guangzhou, China). Total RNA was isolated using TRIzol reagent. RNA was broken using ultrasound. The fragmented RNA was incubated with 4 µg m5C antibody at 4 °C for 4 h. The Protein A/G magnetic beads were further added and incubated for 1 h. After washing, the beads were collected, and RNA was isolated. EFNA3 expression was measured using qPCR.
Dual-luciferase reporter assay
The wild-type (WT) fragments of EFNA3 and the mutated (MUT) EFNA3 containing the potential m5C site were inserted into pGL3-basic firefly luciferase reporter vectors (Promega, Madison, WI, USA). SNU449 and Huh7 cells were co-transfected with WT or MUT reporter plasmids, shNC or shNSUN5, and pRL-TK Renilla luciferase reporter vectors (Promega) using Lipofectamine 2000. After 48 h, the luciferase activity was measured using the dual-luciferase assay system (Promega).
RNA stability analysis
SNU449 and Huh7 cells in 6-well plates and cultured for about 80% cell confluence. WT EFNA3 plasmids or MUT EFNA3 plasmids at the 92 m5C site and shNC or shNSUN5 were co-transfected into SNU449 and Huh7 cells. Then the cells were exposed to 2 µg/mL actinomycin D (Sigma-Aldrich, St. Louis, MO, USA) for 0, 4, 8, and 12 h. EFNA3 expression was detected using qPCR.
Statistical analysis
Bioinformatics data were analyzed using the R software. Experimental data were analyzed using GraphPad Prism 8 software. Data were shown as mean ± standard deviation. The comparison was analyzed using the unpaired Student’s t-test or one-way ANOVA. P < 0.05 were considered statistically significant.
Results
NSUN5 is highly expressed in LIHC
To investigate the clinical significance of NSUN5 in LIHC, we performed bioinformatic analysis to analyze NSUN5 expression using TCGA database. The volcano plot showed 1630 upregulated genes and 1895 downregulated genes between tumor and normal groups (Fig. 1A). Then, we chose m5C methylation-related genes to make a heatmap. NSUN5 expression was significantly higher in tumor tissues than that in normal tissues (Fig. 1B). The box plot further showed that NSUN5 expression was increased in primary tumor tissues, compared with the normal group (Fig. 1C). Furthermore, NSUN5 expression was analyzed across different tumor stages and grades. As the tumor progressed from stage 1 to stage 3, the expression of NSUN5 gradually increased, but it showed downregulation in stage 4 compared to stage 3 (Fig. 1D). Across different tumor grades, the expression of NSUN5 gradually increased from grade 1 to grade 3 but gradually decreased from grade 3 to grade 4 (Fig. 1E). The relationship between NSUN5 expression and survival was also acquired from TCGA database. As shown in Fig. 1F, high NSUN5 expression was associated low survival rate, compared with low NSUN5 expression (p = 0.00012). In addition, several LIHC cell lines were cultured, and we detected NSUN5 expression in these cells. The results of qPCR showed that as compared with the THLE3 cells, NSUN5 expression was elevated in Hep3B, SNU449, Huh7, HCCLM3, and HA22T cells, particularly SNU449 and Huh7 cells, while NSUN5 expression had no significance between THLE3 and Focus cells (Fig. 1G). The results indicate that NSUN5 expression is upregulated in LIHC and associated with poor prognosis.
Fig. 1.
NSUN5 is highly expressed in LIHC. (A) The volcano plot showed differentially expressed genes in LIHC tumor tissues and adjacent normal tissues from TCGA database. The red dots indicated upregulated genes, while the blue dots indicated downregulated genes. (B) The heatmap showed the expression patterns of m5C methylation-related enzymes from differentially expressed genes in tumor and normal tissues. Blue: low expression; red: high expression. (C) The box plot showed the expression of NSUN5 in the primary tumor and normal samples. (D) The box plot showed the expression of NSUN5 at different cancer stages (normal, stages 1–4). (E) The box plot showed the expression of NSUN5 in different tumor grades (normal, grades 1–4). (F) The survival curve showed the survival rate of patients with LIHC in NSUN5 high and low/medium groups. (G) The expression of NSUN5 in normal liver epithelial cells (THLE3) and LIHC cell lines (Hep3B, SNU449, Huh7, HCCLM3, Focus, and HA22T). n = 3. *P < 0.05. **P < 0.01. ns: no significance
Knocking down NSUN5 inhibits glycolysis in LIHC cells
Since NSUN5 was the highest in the two cell lines SNU449 and Huh7, they were selected for functional experiments. To investigate the effect of NSUN5 on cellular behaviors, shNSUN5 and shNC were transfected into SNU449 and Huh7 cells. Following shNSUN5 transfection, the expression of NSUN5 was decreased (Fig. 2A). Cell viability was suppressed by NSUN5 knockdown, compared with the shNC group (Fig. 2B). Next, we evaluated glycolysis. We found that silencing of NSUN5 reduced glucose uptake, lactate production, and ECAR (Fig. 2C-E). In addition, the expression of glycolysis-related genes was detected by qPCR. The results showed that knocking down NSUN5 induced the downregulation of HK2 and PKM2 expression (Fig. 2F and G). Together, silencing of NSUN5 inhibits cell viability and glycolysis in LIHC cells.
Fig. 2.
Knocking down NSUN5 inhibits glycolysis in LIHC cells. (A) Following shNSUN5 and shNC transfection, qPCR was performed to determine NSUN5 expression. SNU449 and Huh7 cells were transfected with shNC and shNSUN5, and cell phenotype was evaluated. (B) Cell viability was measured using CCK-8. (C) Glucose uptake was detected using a glucose uptake assay kit. (D) Lactate production was measured using a lactate assay kit. (E) ECAR was measured using Seahorse analysis. (F) HK2 and (G) PKM2 mRNA expression was detected using qPCR. n = 3. **P < 0.01
NSUN5 promotes m5C methylation of EFNA3
To investigate the downstream factor of NSUN5, we performed correlation analysis, and found 2082 positively related genes and 1721 negatively related genes (Fig. 3A). These correlated genes were enriched in several pathways, including glycolysis/gluconeogenesis (Fig. 3B). The specific correlation values between NSUN5 and the genes that involved in glycolysis are shown in Fig. 3C. We chose EFNA3, which had the highest positive correlation with NSUN5, for the follow-up study. EFNA3 expression was predicted to be higher in LIHC primary tumor tissues than that in normal tissues, according to TCGA data (Fig. 3D). Moreover, EFNA3 expression was increased from stage 1 to stage 3 and was reduced from stage 3 to stage 4 (Fig. 3E). Besides, EFNA3 expression was higher in high-grade tumors (grade 3 and grade 4) than that in low-grade tumors (grade 1 and grade 2) (Fig. 3F). Patients with high EFNA3 expression showed low survival than those with low EFNA3 expression (p = 0.00023; Fig. 3G). The expression of EFNA3 in SNU449 and Huh7 cell lines was increased, compared with the THLE3 group (Fig. 3H).
Fig. 3.
NSUN5 promotes m5C methylation of EFNA3. (A) The scatter plot showed the correlated genes with NSUN5. (B) The bubble plot showed the KEGG pathway enrichment analysis results of NSUN5-related genes. (C) The matrix heatmap showed the correlation between NSUN5 and genes that enriched in the glycolysis/gluconeogenesis pathway. (D) The box plot showed the expression of EFNA3 in LIHC primary tumor samples and normal samples. (E) The box plot showed the expression of EFNA3 at different cancer stages (normal, stages 1–4). (F) The box plot showed the expression of EFNA3 in different tumor grades (normal, grades 1–4). (G) The survival curve showed the survival rate of patients with LIHC in EFNA3 high and low/medium groups. (H) The expression of NSUN5 in normal liver epithelial cells (THLE3) and LIHC cell lines (SNU449 and Huh7). (I) After NSUN5 knockdown, EFNA3 expression was detected using qPCR. (J) EFNA3 expression was measured using MeRIP after NSUN5 knockdown. (K) Possible m5C modified sites in EFNA3. (L) The m5C sites in EFNA3 were verified using dual-luciferase reporter assay. (M) The effect of NSUN5 knockdown on EFNA3 mRNA stability when EFNA3 was wild type or mutated at site 92. n = 3. *P < 0.05. **P < 0.01. ns: no significance
Subsequently, the regulation of NSUN5 on m5C modification of EFNA3 was identified. Knocking down NSUN5 decreased EFNA3 mRNA expression in LIHC cells (Fig. 3I), and concurrently reduced m5C levels of EFNA3 (Fig. 3J). m5C methylation sites in EFNA3 were predicted, and the results showed that m5C modification might occur at sites 92 and 341 in the EFNA3 mRNA sequence (Fig. 3K). The sites were verified using dual-luciferase reporter assay. The results showed that WT reporter plasmids containing site 92 reduced the luciferase activity, while WT reporter plasmids containing site 341 did not affect the luciferase activity (Fig. 3L), suggesting site 92 was the m5C modified site in EFNA3. Furthermore, the stability of EFNA3 was evaluated. We found that NSUN5 knockdown reduced the mRNA stability of EFNA3.However, when EFNA3 mutated at site 92, NSUN5 could not change the stability of EFNA3 mRNA (Fig. 3M). In summary, EFNA3 expression is elevated in LIHC, and NSUN5 stabilizes EFNA3 by promoting its m5C modification at site 92.
Knocking down NSUN5 inhibits glycolysis in LIHC cells by reducing EFNA3 expression
The effect of EFNA3 on glycolysis mediated by NSUN5 was evaluated. SNU449 and Huh7 cells were transfected with oeNC and oeEFNA3, and EFNA3 expression was elevated in the oeEFNA3 group (Fig. 4A). Knocking down NSUN5 suppressed LIHC cell viability, which was counteracted by EFNA3 overexpression (Fig. 4B). The inhibition of glucose uptake, lactate production, and ECAR that caused by NSUN5 knockdown was reversed by EFNA3 overexpression (Fig. 4C-F). Additionally, silencing of NSUN5 decreased HK2 and PKM2 expression, while EFNA3 overexpression abrogated the effect induced by NSUN5 silence (Fig. 4G and H). Taken together, silencing of NSUN5 inhibits glycolysis in LIHC cells via downregulating EFNA3 expression.
Fig. 4.
Knocking down NSUN5 inhibits glycolysis in LIHC cells by reducing EFNA3 expression. (A) Following oeEFNA3 and oeNC transfection, qPCR was performed to determine EFNA3 expression. SNU449 and Huh7 cells were transfected with shNC, shNSUN5, shNSUN5 + oeNC, and shNSUN5 + oeEFNA3, and cell phenotype was evaluated. (B) Cell viability was measured using CCK-8. (C) Glucose uptake was detected using a glucose uptake assay kit. (D) Lactate production was measured using a lactate assay kit. (E, F) ECAR was measured using Seahorse analysis. (G) HK2 and (H) PKM2 mRNA expression was detected using qPCR. n = 3. **P < 0.01
Silencing of NSUN5 reduces xenograft tumor growth and glycolysis by reducing EFNA3 expression
The effect of NSUN5 on tumor growth in vivo was investigated in a xenograft model. NSUN5 expression was decreased in tumors of the AAV-shNSUN5 group, compared with the AAV-shNC group (Fig. 5A and C). EFNA3 mRNA and protein levels were also lower in the AAV-shNSUN5 group than that in the AAV-shNC group (Fig. 5B and C). The images of tumor-bearing mice and isolated tumors showed that knockdown of NSUN5 reduced tumor size (Fig. 5D and E). Tumor weight and volume were lower in the AAV-shNSUN5 group than that in the AAV-shNC group (Fig. 5F and G). The expression of proliferation-related factor ki-67 was reduced after NSUN5 knockdown (Fig. 5H). Glycolysis was also analyzed in the tumors. We found that knockdown of NSUN5 induced the downregulation of HK2 and PKM2 expression (Fig. 5I and J). Together, silencing of NSUN5 inhibits tumor growth by decreasing EFNA3 expression via inhibiting glycolysis.
Fig. 5.
Silencing of NSUN5 reduces xenograft tumor growth and glycolysis by reducing EFNA3 expression. The xenograft mouse model was generated by injecting Huh cells transfected with AAV-shNC and AAV-shNSUN5. (A) The expression of NSUN5 in the tumors was detected using qPCR. (B) The expression of EFNA3 in the tumors was detected using qPCR. (C) Protein levels of NSUN5 and EFNA3 were measured using western blotting in tumors. (D) Images of tumor-bearing mice. (E) Images of tumors removed from mice. (F) Tumor weight was quantified. (G) The changes in tumor volume after model establishment. (H) Representative images of IHC staining of ki-67 expression in tumors. (I) HK2 and PKM2 mRNA levels in tumors were measured using qPCR. (J) Protein levels of HK2 and PKM2 in tumors were measured using western blotting. n = 5. **P < 0.01
Discussion
In the present study, we explored the role of NSUN5 in LIHC, particularly focusing on its regulation of glycolysis and the underlying molecular mechanism. We found that NSUN5 promoted glycolysis in LIHC cells by m5C modification of EFNA3, thereby accelerating LIHC progression.
Growing evidence has reported that NSUN5 is highly expressed in cancers and facilitates tumor progression, such as esophageal cancer, glioblastoma, and gastric cancer [15, 16]. Similar to these studies, it has also been reported that NSUN5 can promote the progress of LIHC. Han et al. [11] have revealed that NSUN5 promotes LIHC cell migration, invasion, and epithelial-mesenchymal transition in vitro, and accelerates tumor formation in vivo. Gu et al. [12] have indicated that NSUN5 promotes cell proliferation and tumor growth in LIHC. In the present study, the results indicated that NSUN5 expression was higher in the tumor tissues than that in normal tissues, which was related to tumor stages, grade, and poor survival, consistent with previous findings [17, 18]. We also found that knocking down NSUN5 inhibited tumor growth, consistent with previous studies [11, 12].
However, whether NSUN5 can regulate glycolytic phenotypes in LIHC remains unexplored. Glycolysis is the main way for tumor cells to obtain energy, and inhibiting glycolysis may be an effective strategy for the clinical treatment of LIHC [19]. Currently, there is only one study that elucidates a potential link between NSUN5 and glycolysis, which found that NSUN5 participated in the Warburg effect regulated by ENO3 in clear cell renal cell carcinoma [20]. Herein, we found that knocking down NSUN5 inhibited glucose uptake, lactate production, ECAR, and downregulated HK2 and PKM2 expression, suggesting the inhibition of glycolysis. The findings demonstrate that NSUN5 acts as an oncogene in LIHC by promoting glycolysis.
We further identified that NSUN5 expression was positively related to the expression of EFNA3, which was enriched in the glycolysis/gluconeogenesis pathway. Moreover, we found that NSUN5 stabilizes EFNA3 by facilitating the m5C methylation. EFNA3 is a member of the Ephrin family that is involved in tumor progression. EFNA3 expression is elevated in cancers and regulates cancer cell migration, invasion, proliferation, and apoptosis to facilitate malignant advancement [20, 21]. In LIHC, EFNA3 is a promising prognostic biomarker [22]. We also found that EFNA3 expression was elevated in LIHC and related to poor survival, consistent with this study. Several previous studies have explored the role of EFNA3 in LIHC. Husain et al. [23] have revealed that EFNA3 regulates LIHC cell proliferation, migration, and self-renewal. Additionally, Yu et al. [24] have found that EFNA3 is a target of miR-210 that promotes cisplatin resistance and growth of LIHC cells. However, the impact of EFNA3 on glycolysis remains unknown. In this study, the results indicated that overexpression of EFNA3 reversed the inhibition of glycolysis induced by NSUN5 silence. These findings suggest that NSUN5 promotes glycolysis in LIHC cells by stabilizing EFNA3 via the m5C methylation. However, the role of EFNA3 in vivo is not clear, which will be further studied in our future work.
A major limitation of this study is that we have not yet clearly identified the m5C reader of EFNA3. It remains unknown whether this reader and possible specific antagonists regulate the stability of EFNA3, which will be further studied in our research.
Our research confirmed the specificity of EFNA3 as a target for the glycolytic function regulated by NSUN5 in LIHC through MeRIP, correlation analysis and rescue experiments. Despite this, we thick that NSUN5 may also potentially regulate other metabolic regulators, such as genes enriched in the glycolytic pathway as shown by KEGG results, or key glycolytic enzyme-coding genes HK2 and PKM2. In future research, we will continue to explore the downstream targets of NSUN5-mediated m5C modification in the LIHC glycolysis process and further clarify the regulatory mechanism of NSUN5.
In conclusion, our study demonstrates that NSUN5 is highly expressed in LIHC and related to poor survival. Knocking down NSUN5 inhibits LIHC tumor growth by suppressing glycolysis, which depends on the inhibition of EFNA3 methylation. This study describes the connection between m5C modification and glycolysis in LIHC and provides a reliable basis for NSUN5 as a therapeutic target for LIHC. The clinical translation strategy of this study still faces significant challenges. In future research, we will develop specific inhibitors or small molecules of NSUN5 or EFNA3 and explore their effects on glycolysis in LIHC. In addition, the tumor-targeting property of these substances is enhanced through the encapsulation of lipid nanoparticles. These studies will promote the progress of clinical translation.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Acknowledgements
Not applicable.
Author contributions
All authors participated in the design, interpretation of the studies and analysis of the data and review of the manuscript. Y H drafted the work and revised it critically for important intellectual content; X D, H C, J C and W X were responsible for the acquisition, analysis and interpretation of data for the work; L L made substantial contributions to the conception or design of the work. All authors read and approved the final manuscript.
Funding
The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.
Data availability
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.
Declarations
Ethics approval and consent to participate
This study was approved by the Ethics Committee of Hangzhou TCM Hospital affiliated to Zhejiang Chinese Medical University. All animal experiments should comply with the ARRIVE guidelines. All methods were carried out in accordance with relevant guidelines and regulations.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Wu Y, He S, Cao M, Teng Y, Li Q, Tan N, et al. Comparative analysis of cancer statistics in China and the united States in 2024. Chin Med J (Engl). 2024;137(24):3093–100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Damjanovska S, Alao H, Zebrowski E, Kowal C, Kostadinova L, Davitkov P, et al. During HCV DAA Therapy Plasma Mip1B, IP10, and miRNA Profile Are Distinctly Associated with Subsequent Diagnosis of Hepatocellular Carcinoma: A Pilot Study. Biology (Basel). 2022;11(9):1262. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Li J, Zhou M, Liu F, Xiong C, Wang W, Cao Q, et al. Hepatocellular carcinoma: Intra-arterial delivery of Doxorubicin-loaded Hollow gold nanospheres for photothermal Ablation-Chemoembolization therapy in rats. Radiology. 2016;281(2):427–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Luo GL, Guo BQ, Wu LX, Shen YX, Xie T. Application of Tivantinib for hepatocellular carcinoma: A Meta-Analysis study. Evid Based Complement Alternat Med. 2022;2022:1976788. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Ganapathy-Kanniappan S, Geschwind JF. Tumor Glycolysis as a target for cancer therapy: progress and prospects. Mol Cancer. 2013;12:152. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Paul S, Ghosh S, Kumar S. Tumor glycolysis, an essential sweet tooth of tumor cells. Semin Cancer Biol. 2022;86(Pt 3):1216–30. [DOI] [PubMed] [Google Scholar]
- 7.Feng J, Li J, Wu L, Yu Q, Ji J, Wu J, et al. Emerging roles and the regulation of aerobic Glycolysis in hepatocellular carcinoma. J Exp Clin Cancer Res. 2020;39(1):126. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Chen H, Ge XL, Zhang ZY, Liu M, Wu RY, Zhang XF, et al. M(5)C regulator-mediated methylation modification patterns and tumor microenvironment infiltration characterization in lung adenocarcinoma. Transl Lung Cancer Res. 2021;10(5):2172–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Cui J, Zheng L, Zhang Y, Xue M. Bioinformatics analysis of DNMT1 expression and its role in head and neck squamous cell carcinoma prognosis. Sci Rep. 2021;11(1):2267. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Wang C, Zhang C, Yang S, Xiang J, Zhou D, Xi X. Identification and validation of m5c-related LncRNA risk model for ovarian cancer. J Ovarian Res. 2023;16(1):96. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Han H, Zhang C, Shi W, Wang J, Zhao W, Du Y, et al. NSUN5 facilitates hepatocellular carcinoma progression by increasing SMAD3 expression. Adv Sci (Weinh). 2025;12(2):e2404083. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Gu X, Li P, Gao X, Ru Y, Xue C, Zhang S, et al. RNA 5-methylcytosine writer NSUN5 promotes hepatocellular carcinoma cell proliferation via a ZBED3-dependent mechanism. Oncogene. 2024;43(9):624–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Nombela P, Miguel-Lopez B, Blanco S. The role of m(6)A, m(5)C and Psi RNA modifications in cancer: novel therapeutic opportunities. Mol Cancer. 2021;20(1):18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Yang L, Ren Z, Yan S, Zhao L, Liu J, Zhao L, et al. Nsun4 and Mettl3 mediated translational reprogramming of Sox9 promotes BMSC chondrogenic differentiation. Commun Biol. 2022;5(1):495. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Cui Y, Hu Z, Zhang C. RNA methyltransferase NSUN5 promotes esophageal Cancer via 5-Methylcytosine modification of METTL1. Mol Carcinog. 2025;64(3):399–409. [DOI] [PubMed] [Google Scholar]
- 16.Su Y, Liu J, Zheng Z, Shi L, Huang W, Huang X, et al. NSUN5-FTH1 Axis inhibits ferroptosis to promote the growth of gastric Cancer cells. Cell Biochem Biophys. 2023;81(3):553–60. [DOI] [PubMed] [Google Scholar]
- 17.Sun GF, Ding H. NOP2-mediated m5C methylation of XPD is associated with hepatocellular carcinoma progression. Neoplasma. 2023;70(3):340–9. [DOI] [PubMed] [Google Scholar]
- 18.Zhang XW, Wu LY, Liu HR, Huang Y, Qi Q, Zhong R, et al. NSUN5 promotes progression and predicts poor prognosis in hepatocellular carcinoma. Oncol Lett. 2022;24(6):439. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Alves AP, Mamede AC, Alves MG, Oliveira PF, Rocha SM, Botelho MF, et al. Glycolysis Inhibition as a strategy for hepatocellular carcinoma treatment?? Curr Cancer Drug Targets. 2019;19(1):26–40. [DOI] [PubMed] [Google Scholar]
- 20.Wang J, Ju HJ, Zhang F, Tian H, Wang WG, Ma YL, et al. A novel NSUN5/ENO3 pathway promotes the Warburg effect and cell growth in clear cell renal cell carcinoma by 5-methylcytosine-stabilized ENO3 mRNA. Am J Transl Res. 2023;15(2):878–95. [PMC free article] [PubMed] [Google Scholar]
- 21.Dong C, Li P, Wu Y, Guo Z, He R. The 1q21.3 region driver gene EFNA3 promotes disease progression via Inhibition of lung adenocarcinoma cell apoptosis. Transl Cancer Res. 2022;11(5):1309–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Lin P, Yang H. EFNA3 is a prognostic biomarker for the overall survival of patients with hepatocellular carcinoma. J Hepatol. 2022;77(3):879–80. [DOI] [PubMed] [Google Scholar]
- 23.Husain A, Chiu YT, Sze KM, Ho DW, Tsui YM, Suarez E, et al. Ephrin-A3/EphA2 axis regulates cellular metabolic plasticity to enhance cancer stemness in hypoxic hepatocellular carcinoma. J Hepatol. 2022;77(2):383–96. [DOI] [PubMed] [Google Scholar]
- 24.Yu H, Shi G. Cisplatin chemotherapy-induced miRNA-210 signaling inhibits hepatocellular carcinoma cell growth. Transl Cancer Res. 2019;8(2):626–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.