MiR‐138‐5p Suppresses Glucose Metabolism to Inhibit the Progression of Nasopharyngeal Carcinoma by Targeting the CENPN/PI3K/Akt/mTOR Signaling

ABSTRACT

Nasopharyngeal cancer (NPC) is a malignant tumor that originates from mucosal epithelial cells of the nasopharynx. As an emerging cancer‐associated microRNA, miR‐138‐5p has been shown to modulate various malignant cell malignancies. However, its precise biological function and regulatory mechanism in NPC remain to be further elucidated. Human NPC cell lines (HK‐1 and C666‐1) were used for in vitro study, and nude mouse xenograft models were used for in vivo investigations. A bioinformatic method was utilized for the prediction of RNA expression patterns and molecular interactions. RT‐qPCR was performed for the detection of miR‐138‐5p and CENPN expression. Luciferase activity detection and RNA pulldown experiments were used for verification of gene interaction. Flow cytometry and colony formation assays were conducted to evaluate cell cycle distribution, proliferative capacity, and apoptosis rate. Furthermore, extracellular acidification rate, lactate generation, and glucose uptake were assessed to explore the regulatory effects of miR‐138‐5p and CENPN on glycolysis in NPC cells. Protein levels of CENPN and key molecules associated with cell cycle, apoptosis, glycolysis, and the PI3K signaling pathway were quantified by western blot analysis. CENPN expression is analyzed to be upregulated in head and neck cancer samples, and its high expression is associated with unfavorable clinical prognosis. In NPC cells, miR‐138‐5p was markedly downregulated and was verified to directly target the 3′ untranslated region of CENPN, thereby negatively regulating its expression. Functionally, overexpressing miR‐138‐5p obstructed malignant phenotype, glycolysis, and tumor growth of NPC cells, where these inhibitory effects were significantly counteracted by enforced CENPN expression. Moreover, miR‐138‐5p overexpression inactivated the PI3K/Akt/mTOR signaling by reducing CENPN expression in NPC cells. In conlusion, miR‐138‐5p inhibits cell growth, glycolysis, and tumorigenesis in NPC by directly targeting CENPN, thereby inhibiting the PI3K/Akt/mTOR signaling pathway. These findings highlight the potential utility of miR‐138‐5p as both a diagnostic indicator and a therapeutic intervention target for NPC management.

1. Introduction

Nasopharyngeal cancer (NPC) is a specific subtype of head and neck squamous carcinoma (HNSC), arising from malignant transformation of nasopharyngeal mucosal epithelial cells [1, 2]. Over recent years, accompanied by rapid advances in therapeutic strategies and profound changes in dietary habits and lifestyles, the incidence and mortality rate of NPC have decreased substantially [3]. This favorable tendency has been attributed to multiple factors, including improved diagnostic methods, more efficacious therapeutic regimens, and alterations in general lifestyle and dietary preferences [4, 5]. Nevertheless, a considerable number of NPC patients still present with an unfavorable prognosis even after receiving combination therapy [6, 7]. Hence, further exploration of key target genes and underlying molecular mechanisms driving NPC progression remains of great clinical significance.

MicroRNAs (miRNAs), a cluster of small noncoding RNAs, exert negative regulation on mRNA post‐transcriptionally by binding to its 3'UTR [8, 9], thereby leading to repression of translation or degradation of mRNA [10]. Previously, miRNAs can regulate over one‐third of mRNAs in humans, and aberrantly expressed miRNAs have been widely reported in NPC. These dysregulated miRNAs participate in the regulation of their target mRNAs and mediate malignant phenotypes of cancer cells, including tumor proliferation and apoptosis resistance [11, 12]. For instance, upregulation of miR‐106a‐5p inhibits autophagy and expedites malignancy of NPC cells [12]. In addition, downregulation of miR‐19a‐3p prominently restrains the proliferative ability and invasion of NPC cells [13]. Importantly, miR‐138‐5p has been demonstrated to impede cell growth and motion while promoting apoptosis in laryngeal carcinoma cells by negatively regulating CLN5 [14]. Nevertheless, the biological function of miR‐138‐5p in NPC is still unclear.

Centromere protein N (CENPN), also named ICEN32, belongs to the centromere protein family [15]. The CENPN gene is situated on chromosome 16q23.2 and is widely conserved in eukaryotic chromosomal DNA [16, 17]. Emerging evidence has implicated CENPN as an oncogenic driver in NPC. For example, CENPN inhibits the CREB‐VAMP8 signaling axis, thereby suppressing autophagy and enhancing paclitaxel resistance in NPC cells [17]. In addition, silencing of CENPN could enhance the radiosensitivity of NPC cells by suppressing AKT/mTOR signaling [18]. Nevertheless, the binding between miR‐138‐5p and CENPN in NPC remains uncharacterized.

In the current study, we proposed a hypothesis that miR‐138‐5p suppresses cell growth and glycolysis in NPC by targeting CENPN and inactivating the PI3K/Akt/mTOR signaling.

2. Materials and Methods

2.1. Cells

All cell lines utilized in the current work were obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). Cell culture was performed in a humidified atmosphere (5% CO₂, 37°C). NP69 is a normal nasopharyngeal cell line, while C666‐1 and HK‐1 are NPC cells. NP69 were incubated in keratinocyte/serum‐free medium (COBIOER, Nanjing, China) containing bovine pituitary extract (Absin, Shanghai, China). The two NPC cell lines were maintained in RPMI‐1640 medium (Sigma Aldrich, St. Louis, USA) supplemented with 10% fetal bovine serum (Ausbian, Australia) and 1% penicillin−streptomycin (Sigma Aldrich).

2.2. Plasmid Transfection

For miRNA upregulation, miR‐138‐5p and miR‐136‐5p mimics purchased from HANBIO (Shanghai, China) were used. The pcDNA‐CENPN vectors (pcDNA‐CENPN) from Zeye Biotechnology (Shanghai, China) were used to overexpress CENPN. C666‐1 and HK‐1 cells were seeded to reach 60%−70% confluency and transfected with the above plasmids and their negative controls (NC mimics, empty vectors) to C666‐1 and HK‐1 cells were performed utilizing lipofectamine 3000 reagent (Invitrogen, Carlsbad, USA). Transfection was conducted for 24 h, followed by gene expression detection using reverse transcription quantitative polymerase chain reaction (RT‐qPCR) to confirm the transfection efficacy.

2.3. RT‐qPCR

RNA extraction from NP69 or NPC cells was achieved utilizing the TRIzol reagent (Absin, Shanghai, China). To synthesize cDNA, a PrimeScript RT Kit (Takara, Dalian, China) was used. Next, PCR was performed using SYBR Green qPCR reagents (Invitrogen) on a CFX96 Touch sequence detection system (Bio‐Rad, Hercules, USA). Primer sequences are shown in Supporting Information S1: Table 1. The RNA level was calculated with the $2^{-\mathrm{\Delta \Delta }{C}_{\mathrm{t}}}$ method and normalized to GAPDH (for CENPN) or U6 (for miRNA).

2.4. Western Blot Analysis

NPC cells, NP69 cells, or mouse tumor samples were lysed using RIPA buffer (Yeasen, Shanghai, China) supplemented with protease and phosphatase inhibitors. Protein concentration was assessed using a BCA protein assay kit (Beyotime). Next, the samples were loaded onto PVDF membranes (Millipore, Billerica, MA, USA) after separation using SDS–Polyacrylamide Gel Electrophoresis (Beyotime). Membranes were blocked with 5% defatted milk powder in TBST at 26°C for 1.5 h and subjected to overnight incubation at 4°C with primary antibodies (Supporting Information S1: Table 2). Following TBST washing, membranes were further probed with horseradish peroxidase secondary antibodies at 26°C for 2 h. The visualization of blots was achieved utilizing ECL substrate (BioRad, Hercules, USA). The intensities of protein bands were quantified using ImageJ software.

2.5. Luciferase Activity Assay

The 3′ UTR sequence of CENPN containing binding site of miR‐138‐5p was obtained and mutated, followed by insertion into pmirGLO reporter (Promega, Madison, USA) to establish CENPN‐wild type (WT) or CENPN‐mutant (MUT). Then, miR‐138‐5p mimics or NC mimics were co‐transfected with WT or MUT reporter plasmids into NPC cells for 2 days. Subsequently, the dual luciferase reporter assay system (Promega) was used to measure luciferase activities; firefly activities were normalized to those of Renilla.

2.6. CCK‐8 Assay

HK‐1 and C666‐1 cells were inoculated into 96‐well plates at a density of 5000 cells/well in triplicate. After cell adhesion, cells were transfected and cultured at 37°C in a 5% CO₂ incubator for 24, 48, or 72 h. Next, each well was supplemented with 10 μL of CCK‐8 solution (MedChem Express, Monmouth Junction, USA) for another 2 h of cell incubation. An iMark microplate reader (BioRad) was used to record the absorbance at 450 nm. Cell viability was calculated as the relative absorbance compared with the control group.

2.7. RNA Pull Down Assay

The RNA Protein Pull Down Kit (Thermo Scientific) was employed to assess the interaction between miR‐138‐5p and CENPN. Briefly, WT or MUT miR‐138‐5p sequences were synthesized and biotin‐labeled to generate bio‐miR‐138‐5p‐WT or bio‐miR‐138‐5p‐MUT. Subsequently, biotinylated miR‐138‐5p‐WT, miR‐138‐5p‐MUT, or negative control (bio‐NC) was transfected into NPC cells. Cell lysates were prepared and incubated with streptavidin‐conjugated magnetic beads for 2 h. The bound RNA−protein complexes were then eluted using biotin elution buffer, and the enrichment of CENPN in the eluates was quantified by RT‐qPCR.

2.8. Cell Cycle Assay

Flow cytometry was utilized to assess cell cycle arrest as reported previously [19]. In brief, after transfection and trypsinization, NPC cells were harvested and treated with RNase or propidium iodide (PI) at 25°C for 15 min. After that, a Gallios flow cytometer (Beckman Coulter, Miami, USA) was employed for measurement of cell distribution at each stage.

2.9. Apoptosis Assay

To measure NPC cell apoptotic rate, an AnnexinV‐Fluorescein Isothiocyanate (FITC) Apoptosis Detection Kit (Vazyme) was used for cell staining and subjected to flow cytometry analysis. After trypsinization, cells were rinsed with ice‐cold medium supplemented with Hank's Balanced Salt Solution (Sigma Aldrich). Following centrifugation, the cells suspended in the binding buffer were treated with Annexin V‐FITC (5 µL) for 15 min and PI (5 µL) for 15 min at 26°C without light exposure. The apoptotic rate was subsequently measured using a flow cytometer (Beckman Coulter).

2.10. Detection of Cell Proliferation

C666‐1 and HK‐1 cells were inoculated into six‐well plates (500 cells/well) and transfected with the indicated vectors. After culturing for 2 weeks under standard conditions, cells were rinsed in 4% paraformaldehyde following PBS washing. Subsequently, the colonies (> 50 cells) were counted manually after 0.5% crystal violet staining (Yuanye Biotechnology, Shanghai, China).

2.11. Measurement of Lactate Generation and Glucose Uptake

Glucose Assay Kit (ab65333) and Lactate Assay Kit (ab65331) (all from Abcam) were used for the measurement of glucose and lactate levels in NPC cells, respectively. According to the user guide, the cell medium was treated with reaction mixtures for 30 min at 26°C, followed by determination of absorbance at 450 nm utilizing a microplate reader (BioRad).

2.12. Metabolic Assay

The extracellular acidification rate (ECAR) in NPC cells was assessed utilizing a Seahorse XF96 Analyzer (Agilent Technology, Santa Clara, USA). In brief, after cells were seeded in specific culture plates, ECAR measurements were conducted under normal conditions or in context of 2‐deoxyglucose (100 mM), oligomycin (5 μM), and glucose (10 mM) purchased from Medchem Express (Monmouth Junction, USA). The experimental protocol comprised a mixing phase of 3 min, followed by a waiting period of 3 min, and concluded with a measurement phase lasting an additional 3 min. ECAR values are recorded as milli‐pH per min.

2.13. Animal Experiments

BALB/c nude mice (4 weeks old; female, 18 ± 1.5 g) were obtained from Cavens Laboratory Animal Co. Ltd (Changzhou, China) and kept under specific pathogen‐free conditions at 22 ± 2°C, 50 ± 5% humidity, with a 12 h light/dark cycle and free access to sterile food and water. All experimental procedures involving animals were reviewed and approved by the Animal Care and Use Committee of Hubei Provincial Hospital of Traditional Chinese Medicine. The animals were assigned into three experimental groups (n = 5 mice) using computer‐generated randomization (1): NC mimics (2), miR‐138‐5p mimics, and (3) miR‐138‐5p mimics + pcDNA‐CENPN groups. The NC mimics group served as the baseline control. Sample size was determined based on power analysis and consistent with previous studies involving xenograft assays [20, 21, 22]. Throughout the study, the 3Rs principles (replacement, reduction, and refinement) were rigorously applied to minimize animal use, distress, and procedural burden.

On Day 0, each mouse received a single subcutaneous injection of 100 μL PBS containing 5 × 10⁶ C666‐1 cells stably transfected with the NC mimics, miR‐138‐5p mimics, or miR‐138‐5p mimics + pcDNA‐CENPN into the right flank. Four days after injections, tumor volume was measured every 2 days by two independent observers who were blinded to the group allocation utilizing the formula: volume = length × width² × 0.5. Fourteen days following injection, all animals were humanely euthanized by cervical dislocation and confirmed by the absence of breathing and heartbeat for 30 s. Subsequently, tumors were promptly excised, photographed, and weighed. All animal handling, housing, and euthanasia procedures strictly adhered to the guide for the care and use of laboratory animals. During the study, mice were excluded if they showed severe illness or tumor regression before the end point; no mice were excluded in the study.

2.14. Statistical Analysis

For comparison between two groups, differences were assessed utilizing Student's t‐test. One‐way analysis of variance followed by a Bonferroni post hoc test was utilized for comparisons among more than two groups. All analyses were performed with GraphPad Prism 8.0. Prior to t‐test and ANOVA, the normality and homogeneity of variance assumptions were verified. Only data satisfying these assumptions were subjected to the parametric tests stated above. Non‐normally distributed data were analyzed using Mann−Whitney U test or Kruskal−Wallis test followed by Dunn's post hoc correction. The data are shown as the mean ± standard deviation. The value of p < 0.05 was considered statistically significant.

3. Result

3.1. High CENPN Level in HNSC Tissues and NPC Cells

According to UALCAN website, CENPN presents a higher level in 520 HNSC tissues than that in 44 normal tissues (Figure 1A). Additionally, data from the Kaplan−Meier Plotter website demonstrate that upregulated CENPN predicts poor prognosis of HNSC patients (Figure 1B). Experimental data confirmed that mRNA and protein expression of CENPN were upregulated in C666‐1 and HK‐1 cells (Figure 1C−E, p < 0.001). These findings demonstrate the upregulation of CENPN in HNSC tissues and NPC cells.

Figure 1.

CENPN expression is upregulated in HNSC tissues and NPC cells. (A) The prediction of CENPN level in 520 HNSC tissues and 44 normal tissues using the UALCAN website (https://ualcan.path.uab.edu/index.html). (B) The prediction of the relationship between CENPN level and survival probability of HNSC patients based on the Kaplan−Meier plotter database (https://kmplot.com/analysis/). (C–E) The measurement of CENPN mRNA and protein levels in NP69, HK‐1, and C666‐1 cells using RT‐qPCR and western blot analysis. n = 3. **p < 0.001.

3.2. MiR‐138‐5p Lowered CENPN Level by Binding to Its 3′ UTR

To predict the potential miRNAs that have binding sites on 3'UTR of CENPN, the ENCORI website and miRDB website are used. As shown in Figure 2A, the Venn diagram reveals that five overlapped miRNAs are screened out. Moreover, RT‐qPCR analysis revealed that miR‐512‐3p and miR‐223‐3p are upregulated in NPC cells (Figure 2B), while miR‐136‐5p and miR‐138‐5p levels were diminished in HK‐1 and C666‐1 cells, and miR‐302e displayed no significant dysregulation in these cells (Figure 2C). Hence, miR‐138‐5p and miR‐136‐5p were selected for the following assays. The two miRNAs were overexpressed in NPC cell lines by transfection of specific miRNA mimics (Figure 2D,E). Furthermore, overexpressed miR‐136‐5p and miR‐138‐5p noticeably repressed CENPN mRNA level, and miR‐138‐5p upregulation had a more significant inhibitive impact on CENPN mRNA level than miR‐136‐5p mimics (Figure 2F,G). Therefore, miR‐138‐5p is chosen for further exploration. As shown by western blot analysis, miR‐138‐5p mimics effectively lessened CENPN protein expression (Figure 2H,I). The binding sequences between miR‐138‐5p and CENPN 3'UTR were predicted using Targetscan website (Figure 2J). Moreover, overexpression of miR‐138‐5p repressed the luciferase activities in HK‐1 and C666‐1 cells transfected with CENPN‐WT 3'UTR instead of CENPN‐MUT 3'UTR (Figure 2K,L). In addition, CENPN level was markedly increased in bio‐miR‐138‐5p WT group compared to bio‐NC group, whereas no significant CENPN enrichment was discovered in bio‐miR‐138‐5p MUT group (Figure 2M). Taken together, these findings confirmed the binding relation of miR‐138‐5p and CENPN.

Figure 2.

MiR‐138‐5p inhibits CENPN level by targeting its 3'UTR. (A) The Venn diagram shows the overlapping upstream miRNAs of CENPN predicted by miRDB (https://mirdb.org/) and ENCORI (https://rnasysu.com/encori/) websites. (B, C) The measurement of miR‐512‐3p, miR‐136‐5p, miR‐223‐3p, miR‐138‐5p, and miR‐302e expression in NP69 and NPC cells (HK‐1, C666‐1) using RT‐qPCR. (D, E) The overexpression efficiency of miR‐136‐5p and miR‐138‐5p was evaluated using RT‐qPCR in HK‐1 and C666‐1 cells. (F, G) RT‐qPCR was performed to measure the effect of miR‐136‐5p mimics and miR‐13 8‐5p mimics on CENPN mRNA expression in HK‐1 and C666‐1 cells. (H, I) Western blot analysis was conducted to assess the effect of miR‐138‐5p mimics on CENPN protein level in HK‐1 and C666‐1 cells. (J) The binding site between miR‐138‐5p and CENPN 3'UTR was predicted from TargetScan. (K–M) Luciferase reporter assay and RNA pull‐down assays were conducted to validate the interaction between miR‐138‐5p and CENPN in HK‐1 and C666‐1 cells. n = 3. *p < 0.05, ***p < 0.001.

3.3. MiR‐138‐5p Repressed NPC Cell Malignancy by Targeting CENPN

We then assessed the roles of miR‐138‐5p and CENPN on malignant phenotype of NPC cells. First, overexpression of CENPN reversed the miR‐138‐5p mimics‐mediated downregulation of CENPN but did not affect the miR‐138‐5p level (Figure 3A–C). Moreover, the inhibitive impact of miR‐138‐5p mimics on cell viability was restored by CENPN upregulation (Figure 3D,E). Consistently, miR‐138‐5p upregulation weakened the proliferative capability of NPC cells, and CENPN overexpression reversed this result (Figure 3F,G). Moreover, miR‐138‐5p mimics induced cell cycle arrest in G0/GI stage, and CENPN overexpression counteracted this trend (Figure 3H−J). Furthermore, the increase in cell apoptosis rate caused by miR‐138‐5p upregulation was neutralized by CENPN overexpression (Figure 3K,L). In addition, miR‐138‐5p mimics lowered Cyclin D1 and Bcl‐2 protein levels but enhanced p21, p27, and Bax protein levels in NPC cells, whereas CENPN overexpression reversed these results (Figure 3M−P). In summary, overexpressing miR‐138‐5p can inhibit NPC cell malignancy via downregulation of CENPN.

Figure 3.

Overexpression of miR‐138‐5p inhibits NPC cell malignancy by targeting CENPN. (A, B) The effect of miR‐138‐5p and/or CENPN overexpression on mRNA and protein levels of CENPN was measured by RT‐qPCR and western blot analysis. (C) The effect of miR‐138‐5p and CENPN on miR‐138‐5p level was evaluated using RT‐qPCR. (D, E) CCK‐8 assays were performed to measure the viability of HK‐1 and C666‐1 cells. (F, G) The measurement of colony formation of HK‐1 and C666‐1 cells. (H–J) Flow cytometry was performed to measure NPC cell cycle progression. (K, L) The measurement of cell apoptosis of HK‐1 and C666‐1 cells through Annexin V‐FITC and PI staining. (M–P) Cyclin D1, p21, p27, Bcl‐2, and Bax protein levels were measured by western blot analysis in HK‐1 and C666‐1 cells. n = 3. ***p < 0.001 versus NC mimics group. ^## p < 0.01, ^### p < 0.001 versus miR‐138‐5p mimics group.

3.4. MiR‐138‐5p Exerted a Suppressive Effect on NPC Glycolysis by Inversely Regulating CENPN

MiR‐138‐5p mimics weakened glycolysis and glycolytic ability in HK‐1 and C666‐1 cells, and CENPN overexpression reversed these trends (Figure 4A,B). Additionally, the inhibitive impact of miR‐138‐5p mimics on glucose consumption and lactate production was neutralized by CENPN overexpression (Figure 4C,D). Meanwhile, miR‐138‐5p mimics reduced GLUT1 and HK2 protein levels, and the trend was counteracted by CENPN overexpression (Figure 4E−H). Collectively, miR‐138‐5p dampens glycolysis in NPC cells via targeting CENPN.

Figure 4.

Overexpression of miR‐138‐5p inhibits glycolysis in NPC cells by targeting CENPN. (A, B) In the context of miR‐138‐5p overexpression or concurrent overexpression with CENPN, the Seahorse XF assay was performed to measure the extracellular acidification rate (ECAR) in HK‐1 and C666‐1 cells. (C, D) Corresponding assay kits were used to measure glucose consumption and lactate production in HK‐1 and C666‐1 cells. (E–H) GLUT1 and HK2 protein levels in HK‐1 and C666‐1 cells were quantified by western blot analysis. n = 3. ***p < 0.001 versus NC mimics group. ^### p < 0.001 versus miR‐138‐5p mimics group.

3.5. MiR‐138‐5p Overexpression Inhibited the PI3K/Akt/mTOR Activity by Targeting CENPN

Based on the results of western blot analysis, overexpressing miR‐138‐5p inhibited the phosphorylated levels of PI3K, Akt, and mTOR in NPC cells, and CENPN overexpression rescued these phosphorylated proteins (Figure 5A−D), implying that miR‐138‐5p overexpression inactivates the PI3K signaling by targeting CENPN.

Figure 5.

MiR‐138‐5p overexpression inactivates the PI3K/Akt/mTOR signaling in NPC cells by targeting CENPN. (A–D) Western blot analysis was performed to quantify protein levels of PI3K, p‐PI3K, Akt, p‐Akt, mTOR, and p‐mTOR in HK‐1 and C666‐1 cells overexpressing miR‐138‐5p and/or pcDNA‐CENPN. n = 3. ***p < 0.001 versus NC mimics group. ^### p < 0.001 versus miR‐138‐5p mimics group.

3.6. Upregulation of miR‐138‐5p Inhibited Xenograft Tumor Growth in NPC via CENPN/PI3K/Akt/mTOR Signaling

As illustrated in Figure 6A, overexpression of miR‐138‐5p resulted in smaller xenograft tumors, whereas concurrent overexpression of CENPN abolished this suppressive effect. Moreover, the inhibitive effect of miR‐138‐5p mimics on tumor volume and tumor weight was also restored by CENPN overexpression (Figure 6B,C). Moreover, the reduced phosphorylated levels of PI3K, Akt, and mTOR induced by miR‐138‐5p mimics in xenograft tumors were significantly reversed upon CENPN overexpression (Figure 6D,E). Collectively, miR‐138‐5p inhibits tumorigenesis in NPC by regulating the CENPN/PI3K/Akt/mTOR pathway.

Figure 6.

Upregulation of miR‐138‐5p inhibits xenograft tumor growth in NPC by targeting CENPN/PI3K/Akt/mTOR signaling. (A) The images of excised tumors in three experimental groups (miR‐138‐5p mimics, miR‐138‐5p mimics + pcDNA‐CENPN, and NC mimics). (B) Tumor volume was measured every 2 days within 14 days, and a curve was drawn to show tumor growth. (C) Tumor weight was recorded after euthanasia and tumor excision. (D, E) Western blot analysis was performed to measure protein levels of PI3K, p‐PI3K, Akt, p‐Akt, mTOR, and p‐mTOR in xenograft tumors. n = 5. ^### p < 0.001 versus miR‐138‐5p mimics group.

4. Discussion

Emerging studies have demonstrated that CENPN is highly expressed and functions as an oncoprotein in NPC [15, 16]. A previous study reported that FAM225A promotes NPC cell proliferation and metastasis by activating the cGAS‐STING pathway through regulation of CENPN [15]. Moreover, CENPN has been identified to be aberrantly upregulated in NPC samples, and silencing CENPN suppressed glucose metabolism, restrained cell proliferation, induced cell cycle arrest, and promoted apoptosis [16]. In line with these findings, our bioinformatics analysis reveals elevated CENPN expression in HNSC tissues, which was associated with poor patient prognosis. Consistently, our experimental results verified the upregulation of CENPN in NPC cells, and the overexpression of CENPN facilitated cell growth and glycolysis in NPC by positively regulating the PI3K signaling.

Mechanistically, several regulatory mechanisms underlying CENPN upregulation have been documented in NPC. On one hand, CENPN is transcriptionally activated by the transcription factor IRF2 [16]. On the other hand, the long non‐coding RNA FAM225A recruits FUS to stabilize and elevate CENPN expression [16]. Nevertheless, the canonical post‐transcriptional regulatory mechanism by which CENPN is repressed by miRNAs at the translational level remains poorly understood. In our study, miR‐138‐5p was selected for further exploration due to the potent inhibitory effect of miR‐138‐5p mimics on CENPN expression. We subsequently validated the direct binding between miR‐138‐5p and CENPN 3'UTR.

Excessive cell proliferation and division are undoubtedly the most prominent phenotypes of cancer cells [23]. These processes act as critical intrinsic drivers of tumor initiation and progression [24, 25]. Specifically, aberrant cell growth and division disrupt normal cellular homeostasis, enabling cancer cells to proliferate rapidly, invade surrounding tissues, and drive tumor progression and malignant evolution [26, 27]. Accumulating evidence indicates that miR‐138‐5p exerts tumor‐suppressive roles by inhibiting cell proliferation and division in several tumors. For instance, miR‐138‐5p inhibits cell growth and induces cell cycle blockage in colorectal cancer by targeting MCU [28]. In addition, enforced expression of miR‐138‐5p induces G0/G1 phase cell cycle arrest and promotes apoptosis in lung cancer cells [29]. In line with these findings, our results demonstrated that miR‐138‐5p mimics attenuated the proliferative ability of NPC cells, while inducing cell cycle arrest and apoptosis. These phenotypic changes were significantly reversed by CENPN overexpression. Moreover, in vivo xenograft experiments confirmed that CENPN overexpression abolished the inhibitory effect of miR‐138‐5p mimics on tumorigenesis.

Glycolysis constitutes a key metabolic hallmark of cancer [30]. Malignant cells rely on glycolysis to satisfy their energy demands, converting most intracellular glucose into lactate even under aerobic conditions [31]. This metabolic adaptation provides cancer cells with a competitive growth advantage [32]. In response to hypoxia induced by rapid tumor expansion, cancer cells further augment glycolytic flux, increase lactic acid generation, and consequently reduce the extracellular pH value [32]. Published reports indicate that miR‐138‐5p inhibits aerobic glycolysis in colorectal and hepatocellular cancer [33, 34]. In accordance with these findings, our results revealed that miR‐138‐5p mimics significantly inhibited glycolysis in NPC cells, as evidenced by reduced glucose consumption and lactate production. Notably, these inhibitory effects were largely abrogated by CENPN overexpression.

Other miR‐138‐5p target genes documented in previous studies also warrant further investigation. In retinoblastoma, miR‐138‐5p suppresses cell proliferation, migration, and invasion by targeting pyruvate dehydrogenase kinase 1 (PDK1) [35]. Moreover, osteosarcoma cell‐derived exosomal ELFN1‐AS1 acts as a competing endogenous RNA to interact with miR‐138‐5p and miR‐1291, thereby abrogating the suppressive effects of these two miRNAs on cAMP response element‐binding protein 1 (CREB1) and consequently promoting osteosarcoma cell proliferation, migration, and invasion [36]. Notably, PDK1 and CREB1 are recognized as key glycolysis‐related genes in cancer [37, 38, 39]. In NPC, the transcription factor FOXM1 has been shown to activate PDK1 and enhance glycolysis, thereby driving NPC progression and conferring a poor prognosis [40]. Knockdown of CREB1 inhibits glycolysis and promotes apoptosis in gastric cancer [38]. In colorectal cancer, CREB1 acts as a key transcriptional regulator that upregulates glycolytic genes including GLUT1 and HK2 [39]. Although CREB1‐mediated regulation of glycolysis has not yet been reported in NPC, its oncogenic function in this malignancy has been well established. Given that individual miRNAs often regulate multiple downstream targets, we hypothesized that PDK1 and CREB1 may also serve as potential targets of miR‐138‐5p, thereby contributing to glycolysis and tumorigenesis in NPC. Future studies are warranted to further explore the functional roles of the miR‑138‑5p/PDK1 and miR‑138‑5p/CREB1 regulatory axes in NPC.

Mechanistically, accumulating evidence has revealed that CENPN regulates the activation of the PI3K/Akt signaling in several malignancies including lung [41], esophageal [42], and breast cancer [43]. We therefore hypothesized that CENPN may exert a similar regulatory effect on the PI3K/Akt/mTOR signaling in NPC. As anticipated, overexpression of miR‐138‐5p markedly inhibited the phosphorylated levels of PI3K, Akt, and mTOR in vitro and vivo, whereas CENPN upregulation effectively counteracted these inhibitory effects.

In conclusion, our study further elucidated the molecular mechanism underlying the oncogenic role of CENPN in NPC. We demonstrated that miR‐138‐5p significantly attenuated glycolysis and malignant phenotype in NPC cells by downregulating CENPN and inactivating the PI3K/Akt/mTOR pathway. This regulatory axis may represent a promising therapeutic target for the intervention of NPC.

Author Contributions

Hongxia Guan conceived and designed the experiments. Hongxia Guan, Yang Jiang, and Hui Li carried out the experiments. Hongxia Guan, Yang Jiang, and Hui Li analyzed the data. Hongxia Guan, Yang Jiang, and Hui Li drafted the manuscript. All authors agreed to be accountable for all aspects of the work. All authors have read and approved the final manuscript.

Funding

The authors have nothing to report.

Ethics Statement

All animal operations were approved by the Animal Care and Use Committee of Hubei Provincial Hospital of Traditional Chinese Medicine.

Consent

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File

Data Availability Statement

The data sets used or analyzed during the current study are available from the corresponding author on reasonable request.