Long non-coding RNA (lncRNA) CDKN2A-DT promotes proliferation, migration, and invasion of oral squamous cell carcinoma via the PI3K/AKT/mTOR axis

Introduction

Background

Oral squamous cell carcinoma (OSCC) accounts for more than 90% of all oral cancer cases worldwide (1). Approximately 400,000 people worldwide have been affected (2). Given its high incidence and mortality rates, OSCC represents a major public health concern. Current standard treatment primarily involves surgical resection combined with systemic anticancer therapy. However, patient long-term survival outcomes have not been significantly improved, and treatment is frequently limited by adverse effects and the development of drug resistance (3). Therefore, a comprehensive understanding of the molecular mechanisms driving the initiation and progression of OSCC is essential for identifying early diagnostic biomarkers and improving therapeutic strategies (4).

Among non-coding RNAs, long non-coding RNAs (lncRNAs) constitute a large family of RNA molecules exceeding 200 nucleotides in length that lack apparent protein-coding potential (5). Accumulating evidence demonstrates that multiple lncRNAs exhibit aberrant expression patterns in OSCC, showing significant correlations with patients’ pathological characteristics and prognosis, thereby suggesting their potential as promising targets for early diagnosis and targeted therapy (6).

Rationale and knowledge gap

CDKN2A-DT (cyclin-dependent kinase inhibitor 2A divergent transcript), also known as CDKN2A-AS1 or C9orf53, is a lncRNA comprising 1,600 nucleotides. Its genomic locus is located on human chromosome 9 and exhibits structural polymorphism (7). Current evidence indicates that CDKN2A-DT activates the BMP-SMAD signaling pathway by directly interacting with the SOSTDC1 protein, thereby promoting epithelial ovarian cancer (EOC) progression (8). In contrast, in head and neck squamous cell carcinoma (HNSCC), low expression levels of CDKN2A-DT are associated with enhanced tumor cell proliferation (9), suggesting a tissue-specific functional role. Furthermore, this lncRNA has been implicated in the pathogenesis and development of both lung squamous cell carcinoma and lung adenocarcinoma (10). Notably, the biological role and underlying molecular mechanism of CDKN2A-DT in OSCC remain poorly understood.

Objective

This study aims to systematically investigate the effects of CDKN2A-DT on the proliferative, migratory, and invasive capacities of OSCC cells, as well as its underlying molecular mechanisms. We present this article in accordance with the MDAR and ARRIVE reporting checklists (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1-2790/rc).

Methods

Analysis of public databases and clinical samples

The Cancer Genome Atlas (TCGA) data acquisition and processing

Raw transcriptome sequencing data and clinical information of OSCC were downloaded from TCGA database. Perl was used to extract 403 OSCC cases with complete clinical data (371 cancer tissues and 32 paired adjacent non-tumor tissues). Strawberry Perl program was applied to organize clinical information and survival time. Patients were divided into high and low expression groups according to the median expression level of CDKN2A-DT. R software (version 4.0.1) with the survival package was used to analyze the association between CDKN2A-DT expression level and clinical characteristics as well as prognosis. Gene Set Enrichment Analysis (GSEA) was performed to predict related signaling pathways.

Collection of clinical samples

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The clinical study was approved by the Ethics Committee of the Stomatological Hospital Affiliated to Guangxi Medical University (approval No. 2022079), and informed consent was obtained from all individual participants. Tumor tissues and paired adjacent normal tissues were collected from 23 OSCC patients who underwent surgical treatment at the Stomatological Hospital Affiliated to Guangxi Medical University. All specimens were pathologically confirmed, and patients did not receive radiotherapy or chemotherapy before surgery. Immediately after excision, specimens were placed in RNA protective solution (Solarbio) and stored in liquid nitrogen at −80 ℃. Total RNA was extracted from tissues using TRIzol (Invitrogen), and reverse-transcribed into complementary DNA (cDNA) with ReverTra Ace qPCR RT Kit (Takara). Quantitative real-time PCR (qRT-PCR) was performed on a real-time PCR instrument (Invitrogen) using SYBR Master Mixture (Takara) with the following reaction conditions: pre-denaturation at 95 ℃ for two minutes, followed by 40 cycles (95 ℃ for 15 seconds, 60 ℃ for 30 seconds, 72 ℃ for 30 seconds). The primers were procured from Sangon Biotech (Shanghai, China). Primer sequences were as follows: glyceraldehyde-3-phosphate dehydrogenase (GAPDH) forward 5'-TGTTCGTCATGGGTGTGAAC-3', reverse 5'-ATGGCATGGACTGTGGTCAT-3'; CDKN2A-DT forward 5'-TCTTGAGACACGGCTTTTGATA-3' reverse 5'-CATCGTTTCTGACACTGTTGTTT-3'. The 2^−ΔΔCt method was used to calculate the relative mRNA expression level, and the difference in CDKN2A-DT expression between cancer and adjacent tissues was compared.

Cell culture and lentiviral transduction

Cell line culture

The human OSCC SAS and SCC-9 cell lines were provided by Kebai Biotechnology Co., Ltd. (Nanjing, China). SAS and SCC-9 cells were cultured in DMEM and DMEM/F12 media (Solarbio), respectively, supplemented with 10% fetal bovine serum (Gibco) and dual antibiotics (Servicebio), under conditions of 37 ℃, 5% CO₂, and saturated humidity.

CDKN2A-DT silencing and overexpression

To investigate the biological functions of CDKN2A-DT in OSCC cells, we established CDKN2A-DT knockdown and overexpression models in SAS and SCC-9 cells using lentiviral vectors. For the knockdown of CDKN2A-DT expression, target small interfering RNA (siRNA) oligonucleotide sequences (Table 1) and negative control siRNA (sequence: TTCTCCGAACGTGTCACGT) were cloned into the lentiviral expression vector pGV493-GFP (GeneChem, Shanghai). To overexpress CDKN2A-DT, the full-length cfa-CDKN2A-DT (NC_000009.12) sequence was cloned into the GV367 vector (GeneChem, Shanghai). After the vectors were packaged into lentiviral particles, they were used to transduce SAS and SCC-9 cells. Cells were seeded in 6-well plates at a density of 1×10⁵ cells/well. When the confluency reached 70%, lentivirus was added at multiplicity of infection (MOI) of 10, 20, 50, and 100. Fluorescence microscopy was used for observation after 48 hours, and cells with the optimal MOI [50] were selected for subsequent experiments. Experimental groups included: wild type (WT) group, siCDKN2A-DT group, siCDKN2A-DT-negative control (NC) group, CDKN2A-DT-overexpression (OE) group, and CDKN2A-DT-OE-NC group. Total RNA was extracted from cells using TRIzol (Invitrogen), and qRT-PCR was performed after reverse transcription (procedure as same as clinical tissues) to verify the efficiency of CDKN2A-DT knockdown and overexpression.

Cell function assays

Proliferation ability detection

Cell Counting Kit-8 (CCK-8) proliferation assay

OSCC cells were seeded in 96-well plates at a density of 5×10³ cells/well (n=5). After 48 hours of culture, 10 µL of CCK-8 solution (Biosharp) was added to each well, followed by incubation at 37 ℃ for two hours. The absorbance at 450 nm was measured using a microplate reader (Infinite MFlex, TECAN).

Colony formation assay

Uniformly dispersed cells were seeded in 6-well plates at a density of 5×10² cells/well (n=3) and cultured at 37 ℃ for two weeks. Cells were fixed with 4% paraformaldehyde (Servicebio) for 20 minutes and stained with 0.5% crystal violet (Servicebio) for 30 minutes. The colony formation rate was calculated as (number of colonies/number of seeded cells).

Adhesion ability detection

Matrigel (Corning) was melted overnight at 4 ℃, and 10 µL was added to pre-cooled 96-well plates using a pre-cooled pipette, followed by incubation at 37 ℃ for one hour to solidify. Then, 2×10⁴ cells/well (n=5) were added, and incubated with serum-free medium for one hour. Unattached cells were washed away with phosphate-buffered saline (PBS) (Solarbio). 10 µL of CCK-8 solution and 90 µL of serum-free medium were added to each well, and the absorbance at 450 nm was measured after incubation at 37 ℃ for two hours.

Migration and invasion ability detection

Transwell migration assay

Treated cells were resuspended in serum-free medium to a density of 1×10⁵ cells/mL. 500 µL of cell suspension was added to the upper chamber of an 8 µm Transwell insert (Corning) in a 24-well plate (n=3), and 200 µL of complete medium was added to the lower chamber. After 48 hours of incubation, non-migrated cells in the upper chamber were wiped off with a cotton swab. Cells were fixed with 4% paraformaldehyde for 20 minutes and stained with 0.5% crystal violet for 30 minutes. Migrated cells were counted under an inverted microscope (100×, 5 fields per sample).

Wound healing assay

Treated cells were seeded in 6-well plates at a density of 2×10⁴ cells/well (n=3) and cultured until 70–80% confluency. Linear scratches were made using a sterile 200 µL pipette tip, and cells were cultured in serum-free medium. Images were captured at 0, 24, and 48 hours under an inverted microscope, and quantitative analysis was performed using ImageJ software.

Invasion assay

10 µL of melted Matrigel was added to the upper chamber of Transwell inserts and solidified at 37 ℃ for one hour. 200 µL of cell suspension (2×10⁵ cells/mL) was added to the upper chamber (n=3), and 750 µL of complete medium was added to the lower chamber. After 48 hours of incubation, cells were fixed, stained, and counted.

Western blot

To elucidate the mechanisms by which CDKN2A-DT regulates OSCC development, we analyzed the PI3K/AKT/mTOR pathway, given its well-established significance in OSCC progression. Proteins were extracted from each group and quantified using a bicinchoninic acid (BCA) protein quantification kit (Beyotime). After separation by 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), proteins were transferred to polyvinylidene fluoride (PVDF) membranes (300 mA, 150 minutes). Membranes were blocked with 5% non-fat milk (Servicebio) at room temperature for one hour, and then incubated with primary antibodies against GAPDH, AKT, p-AKT, PI3K P85α, mTOR, p-mTOR, S6, p-S6, matrix metalloproteinase-2 (MMP-2), Paxillin, and p-Paxillin (Bioworld, China) at 4 ℃ overnight. Then, membranes were incubated with horseradish peroxidase (HRP)-labeled secondary antibodies (Beyotime) at room temperature for one hour, and developed with enhanced chemiluminescence (ECL) plus reagent (Beyotime). GAPDH was used as an internal reference. Protein visualization was performed using ECL reagent and the BIO-RAD imaging platform.

Nude mouse xenograft model constructed with SAS cells stably knocked down/overexpressed CDKN2A-DT

A protocol was prepared before the study without registration. All animal experiments were performed under a project license (No. 202305001) granted by Animal Ethics Committee of Guangxi Medical University, in compliance with institutional guidelines for the care and use of animals. Four-week-old BALB/c nude mice were procured from the Laboratory Animal Center of Guangxi Medical University, with 15 males and 15 females. Mice were included if they were: (I) specific pathogen-free (SPF); (II) without visible signs of disease or abnormal behavior. Prior to the commencement of the experiment, 30 mice were randomly assigned to five groups (WT group, siCDKN2A-DT group, siCDKN2A-DT-NC group, CDKN2A-DT-OE group, and CDKN2A-DT-OE-NC group), with six mice in each group. Random assignment was carried out using a computer - generated random number table to ensure that the allocation of each mouse among the groups was completely random. Grouping should ensure that the weight differences among the mice in each group are minimized to reduce the potential impact of weight differences on tumor growth. This study was approved by the Animal Ethics Committee of Guangxi Medical University (No. 202305001). The treated mice were housed in individually ventilated cages (IVCs) within animal rooms maintained under SPF conditions. 1×10⁷ SAS cells were subcutaneously injected into the right upper limb. From the 7th day after inoculation (recorded as measurement day 0), the long diameter (L) and short diameter (W) of tumors were measured with a caliper every two days, and the volume was calculated according to the formula V=(L×W²)/2. Ten days after measurement, mice were sacrificed by cervical dislocation. Mice dying during the experimental procedure were excluded from further analysis, as their incomplete data could not contribute to the final results. Tumor tissues were collected, weighed, fixed with 4% paraformaldehyde, and subjected to hematoxylin-eosin (HE) staining. During the experiment, neither the injection personnel nor the personnel measuring the tumor size and volume were aware of the specific treatment groups to which the mice were assigned. Moreover, the measurement of tumor size and volume was carried out by different researchers.

Statistical analysis

Statistical analysis was performed using SPSS version 27.0 (IBM; Chicago, IL, USA) and R software (version 4.0.1) with the survival package. Data are presented as means ± standard deviation (SD). Quantitative data were tested for normality and homogeneity of variance. Between-group comparisons were performed using Student’s t-test for two groups and one-way analysis of variance (ANOVA) for multiple groups. Tukey’s post hoc test was used when variances were homogeneous, while the Games-Howell test was adopted for heterogeneous variances. The prognostic value of clinical features and CDKN2A-DT expression for overall survival (OS) was assessed using univariate and multivariate Cox proportional hazards regression analyses. Hazard ratios (HRs) with 95% confidence intervals (CIs) and P values were calculated.

Results

Elevated CDKN2A-DT expression correlates with poor prognosis in OSCC patients

Data analysis from the TCGA database indicated a significant elevation of CDKN2A-DT expression in OSCC tissues compared to normal tissues (Figure 1A,1B). These results were corroborated by our qRT-PCR analysis of clinically obtained OSCC tumor tissues and adjacent non-tumor tissues, which also demonstrated significantly higher CDKN2A-DT expression in the tumor tissues compared to the adjacent non-tumor tissues (Figure 1C).

Figure 1 Expression of CDKN2A-DT in Oral Squamous Cell Carcinoma (OSCC) and its Prognostic Significance Based on The Cancer Genome Atlas (TCGA) Databases. (A) The expression levels of CDKN2A-DT in OSCC and normal tissues from TCGA databases were analyzed using a two-sample t-test. (B) The expression levels of CDKN2A-DT in OSCC and normal tissues from TCGA databases were analyzed using a matched t-test. (C) CDKN2A-DT expression was significantly higher in tumor tissues than in adjacent non-tumor tissues. (D) The expression levels of CDKN2A-DT in males and females from TCGA databases were analyzed. (E) The expression levels of CDKN2A-DT in different pathological stages of OSCC from TCGA databases were analyzed. (F) The expression levels of CDKN2A-DT in different T stages from TCGA databases were analyzed. (G) The expression levels of CDKN2A-DT in different age groups from TCGA databases were analyzed. (H)The expression levels of CDKN2A-DT in HPV-negative and HPV-positive groups from TCGA databases were analyzed. (I) The expression levels of CDKN2A-DT in different N stages from TCGA databases were analyzed. (J)Multivariate Cox proportional hazards analysis of prognostic factors in OSCC was conducted. (K) GSEA predicted that the high-expression phenotype of CDKN2A-DT was enriched in gene sets associated with eight pathways, including apoptosis, autophagy regulation, p53 signaling transduction, and MAPK signaling transduction. CI, confidence interval; GSEA, Gene Set Enrichment Analysis; HPV, human papillomavirus; OSCC, oral squamous cell carcinoma; TCGA, The Cancer Genome Atlas.

These findings indicate distinct expression levels of CDKN2A-DT between OSCC and normal tissues. Notably, CDKN2A-DT expression was significantly higher in male OSCC patients compared to female patients (Figure 1D), and elevated in the G3 pathological stage relative to other stages (Figure 1E). Additionally, expression levels were significantly higher in the T1 stage compared to other T stages (Figure 1F). However, no significant correlations were observed with age, human papillomavirus (HPV) infection and N stage (Figure 1G-1I). These results suggest that CDKN2A-DT expression is strongly correlated with gender, histologic grade, and T stage.

Univariate Cox analysis identified N stage (HR =1.810, P<0.001), T stage (HR =1.532, P=0.005), and CDKN2A-DT expression (HR =0.965, P=0.04) as significant prognostic indicators for OSCC. These factors were then included in Cox multivariate analysis, which confirmed that N stage (HR =1.880, P=0.003) and CDKN2A-DT expression (HR =0.961, P=0.03) were independent predictors of patient OS (Figure 1J, Table 2).

Finally, GSEA predicted that the high expression phenotype of CDKN2A-DT is enriched in gene sets associated with eight pathways, including apoptosis, autophagy regulation, p53 signal transduction, and MAPK signal transduction (Figure 1K). These findings suggest that CDKN2A-DT may play a regulatory role in the pathogenesis and progression of OSCC through these signaling pathways.

CDKN2A-DT regulates proliferation, adhesion, migration, and invasion of OSCC cells

Upon establishing lentiviral vector-mediated CDKN2A-DT knockdown and overexpression models in SAS and SCC-9 cell lines, qRT-PCR validation demonstrated a significant downregulation of CDKN2A-DT expression in the siRNA group compared to the control group, whereas a marked upregulation was observed in the overexpression group. Subsequent experiments focused on the siRNA-1 group and its negative control, as well as the OE group and its negative control, to assess the impact of CDKN2A-DT expression on the biological characteristics of OSCC cells (Figure 2A,2B). We then analyzed the functional effects of CDKN2A-DT on the proliferation, adhesion, migration, and invasion activities of OSCC cells.

Figure 2 Analysis of the effects of siRNA and CDKN2A-DT-OE lentivirus on CDKN2A-DT expression in SCC-9 and SAS cells. (A) Comparison of the inhibitory efficiency of three different siRNA plasmids on CDKN2A-DT expression in SCC-9 and SAS cells. (B) Effect of CDKN2A-DT-OE plasmids on the upregulation of CDKN2A-DT expression in SCC-9 and SAS cells. (A) SAS: siRNA-1 vs. siRNA-NC, P=0.004; siRNA-2 vs. siRNA-NC, P=0.001; SCC-9: siRNA-2 vs. siRNA-NC, P=0.008. **, P<0.01; ***, P<0.001. NC, negative control; OE, overexpression; siRNA, small interfering RNA.

CCK-8 and colony formation assays demonstrated that CDKN2A-DT significantly promotes the proliferation of OSCC cells (Figure 3A,3B). The adhesion test revealed that CDKN2A-DT significantly enhances adhesion activity in OSCC cells (Figure 3C). Additionally, we investigated whether CDKN2A-DT is involved in the metastasis of SAS and SCC-9 cells. The Transwell migration and wound healing assays indicated that overexpression of CDKN2A-DT significantly increases the migration ability of SAS and SCC-9 cells, while down-regulation of CDKN2A-DT expression had the opposite effect (Figure 4A,4B). Furthermore, in the transmembrane invasion experiment, we observed that the number of invasive cells in CDKN2A-DT knockdown cells was significantly lower than in the control group (Figure 4C). In summary, these results suggest that CDKN2A-DT promotes the carcinogenesis of OSCC cells in vitro.

Figure 3 CDKN2A-DT impacts the proliferation and adhesion of OSCC cells. (A,B) SCC-9 and SAS cell proliferation were measured via the CCK-8 assay (A) and colony formation assay (stained with 0.5% crystal violet) (B). CDKN2A-DT overexpression enhanced cellular proliferation, while silencing impaired such proliferation. (C) SCC-9 and SAS cell adhesion were measured via the adhesion assay. CDKN2A-DT overexpression enhanced adhesion activity, whereas its silencing had the opposite effect. (A) SAS WT vs. OE, P=0.03; SAS OE vs. OE-NC, P=0.04. (B) SAS WT vs. OE, P=0.003. (C) SCC-9 siRNA vs. siRNA-NC, P=0.006. *, P<0.05; **, P<0.01; ***, P<0.001. NC, negative control; OE, overexpression; OSCC, oral squamous cell carcinoma; siRNA, small interfering RNA; WT, wild type.

Figure 4 CDKN2A-DT plays a vital role in the migration and invasion of OSCC. The migration assay (stained with 0.5% crystal violet) (A) and wound healing assays (B) were used to verify the migration capability of SAS and SCC-9 cells, while the invasion assay (stained with 0.5% crystal violet) (C) was used to confirm their invasive potential. These experiments demonstrate that CDKN2A-DT overexpression can promote the migration and invasion of OSCC cells, which are critical steps in the malignant progression and distant metastasis of tumors. (A) SCC-9 WT vs. OE, P=0.007; SCC-9 OE vs. OE-NC, P=0.01. (B) SCC-9 WT vs. OE, P=0.001; SCC-9 siRNA vs. WT, P=0.004; SCC-9 siRNA vs. siRNA-NC, P=0.01. (C) SCC-9 WT vs. OE, P=0.007; SCC-9 OE vs. OE-NC, P=0.01. *, P<0.05; **, P<0.01; ***, P<0.001. NC, negative control; OE, overexpression; OSCC, oral squamous cell carcinoma; siRNA, small interfering RNA; WT, wild type.

CDKN2A-DT regulates OSCC development via PI3K/AKT/mTOR signaling pathways

Our findings confirm a close association between CDKN2A-DT and the PI3K/AKT/mTOR signaling pathway (Figures 5,6). Specifically, CDKN2A-DT likely regulates the expression of proteins such as PI3K p85α, AKT, p-AKT, mTOR, p-mTOR, MMP-2, paxillin, p-paxillin, S6, and p-S6 through this pathway. These regulatory effects influence various biological functions of OSCC, including proliferation, migration, and invasion.

Figure 5 PI3K/AKT signaling pathway proteins, including PI3K p85α, AKT, p-AKT, mTOR, and p-mTOR, were assessed via WB in SAS and SCC-9 cells, with ImageJ used for analysis. Analyses were conducted using data from three independent experiments. Comparisons between OE-NC and CDKN2A-DT-OE groups or siRNA-NC and CDKN2A-DT-siRNA groups were performed using ANOVA with Tukey’s post hoc test or Games‑Howell test. (A) SAS: PI3K: WT vs. siRNA, P=0.007; WT vs. OE, P=0.003; OE vs. OE-NC, P=0.02. t-AKT: WT vs. siRNA, P>0.99; WT vs. OE, P>0.99; siRNA vs. siRNA-NC, P=0.792; OE vs. OE-NC, P=0.98. mTOR: WT vs. siRNA, P=0.003; WT vs. OE, P=0.01; OE vs. OE-NC, P=0.02. p-mTOR: WT vs. siRNA, P=0.002. p-AKT/t-AKT: WT vs. OE, P=0.03; siRNA-NC vs. siRNA, P=0.008; OE vs. OE-NC, P=0.03. (B) SCC-9: t-AKT: WT vs. siRNA, P>0.99; WT vs. OE, P=0.34; siRNA vs. siRNA-NC, P>0.99; OE vs. OE-NC, P=0.90. p-mTOR: WT vs. siRNA, P=0.02; WT vs. OE, P=0.05; siRNA vs. siRNA-NC, P=0.02; OE vs. OE-NC, P=0.02. p-AKT/t-AKT: WT vs. OE, P=0.03; OE vs. OE-NC, P=0.008. ns, P>0.05; *, P<0.05; **, P<0.01; ***, P<0.001; n=3 for each independent experiment. ANOVA, analysis of variance; NC, negative control; ns, not significant; OE, overexpression; siRNA, small interfering RNA; WB, Western blot; WT, wild type.

Figure 6 WB analysis was employed to detect changes in the expression of invasion-related protein MMP-2, adhesion-related proteins paxillin and p-paxillin, as well as proliferation-related proteins S6 and p-S6 in SAS and SCC-9 cells following the inhibition or overexpression of CDKN2A-DT. Comparisons between OE-NC and CDKN2A-DT-OE groups or siRNA-NC and CDKN2A-DT-siRNA groups were performed using ANOVA with Tukey’s post hoc test or Games‑Howell test. (A) SAS: p-paxillin: siRNA vs. siRNA-NC, P=0.10. (B) SCC-9: p-paxillin: WT vs. siRNA, P=0.06; WT vs. OE, P=0.38. p-S6: WT vs. siRNA, P=0.005; WT vs. OE, P=0.004; OE vs. OE-NC, P=0.01. ns, P>0.05; *, P<0.05; **, P<0.01; ***, P<0.001; n=3 for each independent experiment. ANOVA, analysis of variance; MMP-2, matrix metalloproteinase-2; NC, negative control; ns, not significant; OE, overexpression; p-paxillin, phosphorylated paxillin; p-S6, phosphorylated S6; siRNA, small interfering RNA; WB, Western blot; WT, wild type.

CDKN2A-DT regulates malignant proliferation in OSCC xenografts

Observations of experimental data from 6 mice per group of various of transplanted tumors in nude mice (Figure 7A) revealed that the weight [(Figure 7B) WT: 18.93±0.15, siRNA: 16±1.23, siRNA-NC: 18.15±0.47, OE: 20.73±0.64, OE-NC: 18.25±0.59, unit: g] and volume [(Figure 7C) WT: 1,062±73.86, siRNA: 483.5±115.2, siRNA-NC: 725.7±89.31, OE: 1,474±277, OE-NC: 765.4±173.2, unit: mm³] of tumors in the high CDKN2A-DT expression group increased more rapidly, significantly surpassing those in the wild-type control and blank control groups. Conversely, the growth of transplanted tumors in the low CDKN2A-DT expression group was slower than that in the control group.

Figure 7 Overexpression of CDKN2A-DT accelerates the growth of OSCC solid tumors, whereas its inhibition decelerates this growth. (A) Gross observational results of the xenograft tumors following excision. Changes in weight (B) and volume (C) during the growth of OSCC solid tumors. (D) Histological analyses revealed that overexpression of CDKN2A-DT was able to accelerate the growth of OSCC (stained with HE). (B) WT vs. siRNA, P=0.01; WT vs. OE P=0.01; siRNA vs. siRNA-NC, P=0.04; OE vs. OE-NC, P=0.003. (C) WT vs. siRNA, P=0.002; WT vs. OE P=0.20; siRNA vs. siRNA-NC, P=0.09; OE vs. OE-NC, P=0.04. ns, P>0.05; *, P<0.05; **, P<0.01. NC, negative control; ns, not significant; OE, overexpression; OSCC, oral squamous cell carcinoma; siRNA, small interfering RNA; WT, wild type.

Simultaneously, HE staining results (Figure 7D) demonstrated that transplanted tumors in the CDKN2A-DT overexpression group displayed typical features of malignant tumor cell heterogeneity. (I) At the cellular level: tumor cells exhibited marked variation in size and diverse morphological shapes. (II) Nuclear atypia: the nuclei were prominently hyperchromatic with irregular nuclear membranes. (III) Atypical mitotic figures: under high-power magnification, various pathological mitotic figures were observed (Figure 7D, red arrows), indicative of uncontrolled proliferation and aberrant cell division in tumor cells. This observation supports the notion that CDKN2A-DT is significantly upregulated in G3. Conversely, the low expression group of CDKN2A-DT showed inhibited mitosis, no obvious cellular atypia, and resembled normal squamous epithelium.

Discussion

Key findings

This study provides the first systematic investigation into the functional role of lncRNA CDKN2A-DT in OSCC. Our results reveal a significant upregulation of CDKN2A-DT in both clinical OSCC tissue samples and established cell lines. Subsequent in vitro functional assays confirm that CDKN2A-DT plays a critical role in regulating malignant phenotypes in OSCC cells. Furthermore, in vivo experiments using nude mouse models validate the tumor-promoting effects of CDKN2A-DT. Collectively, these findings establish CDKN2A-DT as a key driver in the progression of OSCC. As a bona fide lncRNA, CDKN2A-DT is not expected to encode a protein, and its functions are therefore mediated through RNA-based mechanisms. In this study, we focused on its regulatory effect on downstream protein expression.

Strengths and limitations

The present study employs a comprehensive and well-integrated experimental design, establishing a rigorous “clinical-basic-translational” research continuum spanning from analyses of TCGA database and clinical samples, to cellular functional assays, mechanistic investigations, and validation via xenograft mouse models. A dual genetic strategy involving both knockdown and overexpression was utilized to systematically evaluate the regulatory roles of CDKN2A-DT in proliferation, adhesion, migration, and invasion across two cell lines (SAS and SCC-9), thereby strengthening the robustness of the findings. This work is the first to elucidate the molecular mechanism by which CDKN2A-DT activates the PI3K/AKT/mTOR signaling axis via upregulation of the PI3K regulatory subunit p85α, and further confirms its role as an independent prognostic factor for OS in OSCC patients, demonstrating both theoretical novelty and clinical relevance. The research adheres to strict ethical standards and maintains rigorous analytical approaches throughout. However, our study has several limitations. First, while the clinical relevance of CDKN2A-DT was validated in our in-house cohort, the sample size for qRT-PCR analysis was relatively limited (n=23). Given this modest sample size, this cohort should be regarded as preliminary supporting evidence, and the conclusions drawn solely from it require further validation in larger, independent cohorts. It is important to emphasize, however, that the central conclusions of this study—namely, the oncogenic role of CDKN2A-DT and its mechanism via the PI3K/AKT/mTOR pathway—are primarily supported by robust multi-omics data from public databases (e.g., TCGA) and comprehensive functional data from both in vitro and in vivo experiments. The in-house cohort serves as an additional translational bridge, but the weight of evidence lies with the mechanistic and multi-dataset analyses. Second, although our in vitro data suggest that CDKN2A-DT regulates the PI3K/AKT/mTOR pathway by upregulating p85α, we did not validate this axis by immunohistochemistry (IHC) in clinical tissues. Owing to the limited quantity and quality of available specimens, IHC could not be performed in the present study. Future studies with larger and better-characterized clinical samples will be needed to verify the correlation between CDKN2A-DT and downstream proteins such as p85α. Third, the study did not directly assess the impact of CDKN2A-DT on the levels of the PI3K catalytic subunit p110. However, existing evidence suggests that increased expression of the PI3K regulatory subunit p85α may enhance the lipid kinase activity of the PI3K complex, thereby facilitating downstream AKT phosphorylation (11). For instance, in hepatocellular carcinoma, carbohydrate response element binding protein (ChREBP) sustains persistent activation of the PI3K/AKT pathway by upregulating PI3K p85α expression, consequently promoting tumor progression (11). Lastly, it remains unclear whether CDKN2A-DT regulates PI3K p85α expression at the transcriptional, post-transcriptional, or post-translational level. Future studies employing techniques such as (I) qPCR and promoter luciferase assays to assess transcriptional regulation; (II) mRNA stability assays and 3' untranslated region (3'UTR) luciferase reporters to evaluate post-transcriptional mechanisms. Additionally, RNA pull-down and RNA immunoprecipitation (RIP) will help identify direct molecular interactors that mediate these effects.

Comparison with similar research

The PI3K/AKT/mTOR signaling pathway plays a pivotal role in the pathogenesis of OSCC, extensively participating in the regulation of various malignant biological processes including cell proliferation, angiogenesis, invasion and metastasis, autophagy, and epithelial-mesenchymal transition (EMT) (12-14). Consequently, inhibitors targeting this pathway (such as rapamycin, an mTOR inhibitor) have emerged as promising therapeutic strategies (15). In recent years, a growing body of research has revealed that multiple lncRNAs serve as key regulators of the PI3K/AKT/mTOR signaling pathway in OSCC. For instance, lncRNA growth arrest specific 5 (GAS5) suppresses laryngeal cancer cell proliferation and metastasis by modulating the PI3K/AKT/mTOR pathway (16), while downregulation of laminin subunit gamma 2 (LAMC2) inhibits PI3K/AKT/mTOR pathway activation to induce autophagy, thereby suppressing OSCC proliferation, invasion, and metastasis (17). Western blot results from this study indicate that CDKN2A-DT, acting as an RNA regulator, likely amplifies the activation of the PI3K/AKT/mTOR signaling pathway by increasing the protein level of its downstream target, the upstream regulatory subunit p85α rather than altering the total amount of core kinases, thereby driving the multi-dimensional progression of malignant phenotypes in OSCC.

Explanations of findings

From the perspective of signal transduction cascade, the overexpression of the lncRNA CDKN2A-DT serves as a critical initiating event leading to upregulation of PI3K P85α protein expression. As a regulatory subunit of class I PI3K, P85α not only stabilizes the catalytic subunit p110α via its inter-Src homology 2 (iSH2) domain (18), but also regulates membrane localization capacity of the PI3K holoenzyme (19). Consistent with the observed upregulation of p85α protein by CDKN2A-DT, this increase in p85α can then exert its canonical functions. For instance, Cheung et al. (20) demonstrated that P85α binds to phosphatase and tensin homolog (PTEN) and enhances its phosphatase activity, thereby restricting excessive PI3K signaling through this negative feedback mechanism. However, the sustained elevation of p-AKT observed in this study suggests that the overwhelming activation signal mediated by CDKN2A-DT may override or even bypass PTEN-mediated negative feedback inhibition. Notably, the significant increase in p-AKT despite unchanged total AKT levels clearly indicates that CDKN2A-DT regulates AKT phosphorylation activation at the post-translational modification level. Activated p-AKT subsequently functions as a signaling hub: on one hand, it enhances phosphorylation of mTOR and S6 to activate the mTOR complex 1 (mTORC1) pathway; on the other hand, it potentially upregulates MMP-2 expression through non-canonical pathways, reflecting its pleiotropic regulatory mode (21). Ultimately, the synchronized downregulation of key pathway components upon CDKN2A-DT knockdown provides reverse validation of its core driving role in this oncogenic cascade.

Activation of the above signaling pathways enables CDKN2A-DT to orchestrate a series of downstream cellular events that collectively drive the malignant progression of OSCC. In terms of cell proliferation, concurrent upregulation of p-mTOR and p-S6 indicates effective activation of the mTORC1 pathway by CDKN2A-DT, thereby promoting cell cycle progression through enhanced protein synthesis (22-24). Regarding adhesion and migration, elevated phosphorylation levels of Paxillin are directly correlated with enhanced cellular adhesion (25), while Zhang et al. (26) indicate that in gastric cancer (GC) cells, inhibiting the phosphorylation of Paxillin can suppress the metastasis of tumor cells. This suggests that an increase in the phosphorylation level of Paxillin may have the potential to promote tumor cell metastasis. Notably, P85α overexpression may indirectly influence integrin trafficking via modulation of Rab5 activity (27), which contributes to the stabilization of the Paxillin signaling complex (28). This promotes focal adhesion maturation and establishes a positive feedback loop between adhesion and migration. Furthermore, upregulation of MMP-2 facilitates physical space creation for cell migration through degradation of the extracellular matrix (29). Ultimately, during invasion, these alterations operate synergistically: p-S6-driven proliferation provides the quantitative foundation for OSCC invasion, while MMP-2-mediated matrix degradation combines with p-Paxillin-enhanced motility to cooperatively potentiate cellular migration and invasion. Additionally, p-AKT may sustain long-term MMP-2 overexpression via transcriptional regulation (30), ultimately endowing OSCC cells with heightened invasive potential. In summary, these changes collectively form a coordinated network encompassing “signal amplification, adhesion remodeling, and matrix degradation”, systematically promoting the aggressive phenotype of OSCC.

While the specific mechanism underlying CDKN2A-DT-mediated regulation of PI3K expression remains uninvestigated, the primary pathways by which lncRNAs modulate PI3K p85α are summarized below, offering a framework to guide future studies. Firstly, as for the regulation at the transcriptional level, as Wan et al. (31) have indicated, lncRNA Non-coding RNA Activated by DNA Damage (NORAD) acts as a molecular sponge for miR-202-5p, representing a key mechanism in activating the PI3K/AKT/mTOR pathway. This mechanistic model provides a valuable reference for further investigations into CDKN2A-DT to identify potential regulatory targets. Secondly, certain lncRNAs can directly bind to the promoter regions of PI3K or AKT and modulate their transcriptional activity. Such interactions may either promote or suppress gene transcription, thereby altering the expression level of p85α and the overall activity of the PI3K signaling pathway (32). Furthermore, lncRNAs may regulate the production of p85α and its isoforms (e.g., p55α) by modulating alternative splicing of the PIK3R1 gene. As PIK3R1 generates multiple functionally distinct isoforms via alternative splicing, lncRNAs can influence this process by interacting with splicing factors or modifying chromatin status, thereby controlling the expression ratio of p85α (33,34). Finally, p85α not only forms the PI3K complex with the catalytic subunit p110 in the cytoplasm, but also translocates into the nucleus independently of p110 to participate in transcriptional regulation (35,36). Although it remains unclear whether lncRNAs directly regulate the nuclear translocation of p85α, emerging evidence has demonstrated that lncRNAs can modulate protein subcellular localization (34,37), suggesting that this regulatory axis warrants further investigation.

Implications and actions needed

This study establishes CDKN2A-DT as an independent prognostic biomarker for OSCC, with its prognostic value being independent of conventional TNM staging. It may complement the existing risk stratification system to accurately identify high-risk patients and guide individualized follow-up intensity and treatment strategies. Patients with high CDKN2A-DT expression are associated with poorer survival outcomes and may benefit from intensified surveillance and targeted interventions. Mechanistically, CDKN2A-DT activates oncogenic signaling by upregulating PI3K p85α. Future research should elucidate the molecular mechanisms underlying its regulation of p85α, its impact on the PI3K catalytic subunit p110, and its interactions with other signaling pathways. In terms of clinical translation, the prognostic efficacy of CDKN2A-DT should be validated in multi-center cohorts. Targeted therapeutics such as CDKN2A-DT siRNA/antisense oligonucleotides should be developed, and their synergistic effects with PI3K/AKT/mTOR inhibitors evaluated. Additionally, a detection method for formalin-fixed paraffin-embedded tissues should be established, and prospective clinical trials initiated. Basic research should involve RNA immunoprecipitation to identify interacting molecules, the construction of patient-derived xenograft/organoid models, and single-cell sequencing to dissect its role in the tumor microenvironment. In summary, CDKN2A-DT represents a promising therapeutic target and prognostic indicator for OSCC, warranting accelerated clinical validation and mechanistic investigations to enable precise clinical application.

Conclusions

In summary, this study provides novel insights into the oncogenic role of lncRNA CDKN2A-DT in OSCC, demonstrating its capacity to drive malignant progression through activation of the PI3K/AKT/mTOR signaling axis. While our in vitro and in vivo data robustly support this mechanism, validation of the CDKN2A-DT/PI3K/AKT/mTOR axis in larger clinical cohorts using IHC and RNA in situ hybridization (RNA-ISH) is warranted. These findings identify CDKN2A-DT as a promising candidate for further translational research in OSCC.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the MDAR and ARRIVE reporting checklists. Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1-2790/rc

Data Sharing Statement: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1-2790/dss

Peer Review File: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1-2790/prf

Funding: This study was supported by the National Natural Science Foundation of China (No. 82360187), Joint Project on Regional High-Incidence Diseases Research of Guangxi Natural Science Foundation (No. 2023GXNSFAA026326), and National Natural Science Foundation of China (No. 82460462).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1-2790/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The clinical study was approved by the Ethics Committee of the Stomatological Hospital Affiliated to Guangxi Medical University (Approval No. 2022079), and informed consent was obtained from all individual participants. All animal experiments were performed under a project license (No. 202305001) granted by Animal Ethics Committee of Guangxi Medical University, in compliance with institutional guidelines for the care and use of animals.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Nagdeve SN, Suganthan B, Ramasamy RP. An electrochemical biosensor for the detection of microRNA-31 as a potential oral cancer biomarker. J Biol Eng 2025;19:24. [Crossref] [PubMed]
2. Wang C, Sarubo M, Chen S, et al. Partial-EMT in oral squamous cell carcinoma: molecular circuitry and clinical translation. Int J Oral Sci 2026;18:15. [Crossref] [PubMed]
3. Pandey M, Choudhury H, Ying JNS, et al. Mucoadhesive Nanocarriers as a Promising Strategy to Enhance Intracellular Delivery against Oral Cavity Carcinoma. Pharmaceutics 2022;14:795. [Crossref] [PubMed]
4. Alsahafi E, Begg K, Amelio I, et al. Clinical update on head and neck cancer: molecular biology and ongoing challenges. Cell Death Dis 2019;10:540. [Crossref] [PubMed]
5. Yuan Y, Tang Y, Fang Z, et al. Long Non-Coding RNAs: Key Regulators of Tumor Epithelial/Mesenchymal Plasticity and Cancer Stemness. Cells 2025;14:227. [Crossref] [PubMed]
6. Lu N, Jiang Q, Xu T, et al. LncOCMRL1 promotes oral squamous cell carcinoma growth and metastasis via the RRM2/EMT pathway. J Exp Clin Cancer Res 2024;43:267. [Crossref] [PubMed]
7. Humphray SJ, Oliver K, Hunt AR, et al. DNA sequence and analysis of human chromosome 9. Nature 2004;429:369-74. [Crossref] [PubMed]
8. Zhao Q, Dong D, Chu H, et al. lncRNA CDKN2A-AS1 facilitates tumorigenesis and progression of epithelial ovarian cancer via modulating the SOSTDC1-mediated BMP-SMAD signaling pathway. Cell Cycle 2021;20:1147-62. [Crossref] [PubMed]
9. Wang GD, Huang JT, Liu Y, et al. Deletion of C9orf53 promotes the growth of head and neck squamous cell carcinoma and is associated with poor prognosis of patients with head and neck squamous cell carcinoma. Oncol Lett 2019;17:1223-8. [PubMed]
10. Liu B, Chen Y, Yang J. LncRNAs are altered in lung squamous cell carcinoma and lung adenocarcinoma. Oncotarget 2017;8:24275-91. [Crossref] [PubMed]
11. Benichou E, Seffou B, Topçu S, et al. The transcription factor ChREBP Orchestrates liver carcinogenesis by coordinating the PI3K/AKT signaling and cancer metabolism. Nat Commun 2024;15:1879. [Crossref] [PubMed]
12. Rajendran P, Sekar R, Dhayasankar PS, et al. PI3K/AKT Signaling Pathway Mediated Autophagy in Oral Carcinoma - A Comprehensive Review. Int J Med Sci 2024;21:1165-75. [Crossref] [PubMed]
13. Jiang Q, Xiao J, Hsieh YC, et al. The Role of the PI3K/Akt/mTOR Axis in Head and Neck Squamous Cell Carcinoma. Biomedicines 2024;12:1610. [Crossref] [PubMed]
14. Harsha C, Banik K, Ang HL, et al. Targeting AKT/mTOR in Oral Cancer: Mechanisms and Advances in Clinical Trials. Int J Mol Sci 2020;21:3285. [Crossref] [PubMed]
15. Papadakos S, Issa H, Alamri A, et al. Rapamycin as a Potential Alternative Drug for Squamous Cell Gingiva Carcinoma (Ca9-22): A Focus on Cell Cycle, Apoptosis and Autophagy Genetic Profile. Pharmaceuticals (Basel) 2024;17:131. [Crossref] [PubMed]
16. Liu W, Zhan J, Zhong R, et al. Upregulation of Long Noncoding RNA_GAS5 Suppresses Cell Proliferation and Metastasis in Laryngeal Cancer via Regulating PI3K/AKT/mTOR Signaling Pathway. Technol Cancer Res Treat 2021;20:1533033821990074. [Crossref] [PubMed]
17. Shan F, Liang L, Feng C, et al. LAMC2 regulates proliferation, migration, and invasion mediated by the Pl3K/AKT/mTOR pathway in oral. Oncol Res 2023;31:481-93. [Crossref] [PubMed]
18. Aljedani SS, Sandholu A, Aldehaiman A, et al. SH3 domains selectively activate the PI3 kinase through non-conventional tertiary contacts. Commun Biol 2026; Epub ahead of print. [Crossref] [PubMed]
19. Thapa N, Chen M, Cryns VL, et al. A p85 isoform switch enhances PI3K activation on endosomes by a MAP4- and PI3P-dependent mechanism. Cell Rep 2024;43:114119. [Crossref] [PubMed]
20. Cheung LW, Walkiewicz KW, Besong TM, et al. Regulation of the PI3K pathway through a p85α monomer-homodimer equilibrium. Elife 2015;4:e06866. [Crossref] [PubMed]
21. Carracedo A, Pandolfi PP. The PTEN-PI3K pathway: of feedbacks and cross-talks. Oncogene 2008;27:5527-41. [Crossref] [PubMed]
22. Wu X, Xie W, Xie W, et al. Beyond controlling cell size: functional analyses of S6K in tumorigenesis. Cell Death Dis 2022;13:646. [Crossref] [PubMed]
23. Ma S, Dong Z, Huang Y, et al. eIF3a regulation of mTOR signaling and translational control via HuR in cellular response to DNA damage. Oncogene 2022;41:2431-43. [Crossref] [PubMed]
24. Jiao D, Sun H, Zhao X, et al. mTORC1/S6K1 signaling promotes sustained oncogenic translation through modulating CRL3(IBTK)-mediated ubiquitination of eIF4A1 in cancer cells. Elife 2024;12:RP92236. [Crossref] [PubMed]
25. Gurgul I, Janczy-Cempa E, Mazuryk O, et al. Inhibition of Metastasis by Polypyridyl Ru(II) Complexes through Modification of Cancer Cell Adhesion - In Vitro Functional and Molecular Studies. J Med Chem 2022;65:10459-70. [Crossref] [PubMed]
26. Zhang L, Yao L, Zhou W, et al. miR-497 defect contributes to gastric cancer tumorigenesis and progression via regulating CDC42/ITGB1/FAK/PXN/AKT signaling. Mol Ther Nucleic Acids 2021;25:567-77. [Crossref] [PubMed]
27. Wang X, Zheng S, Yang F, et al. lncRNA HITT inhibits metastasis by attenuating Rab5-mediated endocytosis in lung adenocarcinoma. Mol Ther 2022;30:1071-88. [Crossref] [PubMed]
28. Ripamonti M, Liaudet N, Azizi L, et al. Structural and functional analysis of LIM domain-dependent recruitment of paxillin to αvβ3 integrin-positive focal adhesions. Commun Biol 2021;4:380. [Crossref] [PubMed]
29. Liu J, Chen T, Li S, et al. Targeting matrix metalloproteinases by E3 ubiquitin ligases as a way to regulate the tumor microenvironment for cancer therapy. Semin Cancer Biol 2022;86:259-68. [Crossref] [PubMed]
30. Liu F, Liu T, Li H. Aloperine inhibits the progression of non-small-cell lung cancer through the PI3K/Akt signaling pathway. Cancer Cell Int 2021;21:662. [Crossref] [PubMed]
31. Wan Y, Yao Z, Chen W, et al. The lncRNA NORAD/miR-520a-3p Facilitates Malignancy in Non-Small Cell Lung Cancer via PI3k/Akt/mTOR Signaling Pathway. Onco Targets Ther 2020;13:1533-44. [Crossref] [PubMed]
32. Hashemi M, Mohandesi Khosroshahi E, Asadi S, et al. Emerging roles of non-coding RNAs in modulating the PI3K/Akt pathway in cancer. Noncoding RNA Res 2025;10:1-15. [Crossref] [PubMed]
33. Gupta I, Gaykalova DA. Unveiling the role of PIK3R1 in cancer: A comprehensive review of regulatory signaling and therapeutic implications. Semin Cancer Biol 2024;106-107:58-86. [Crossref] [PubMed]
34. Yang M, Lu H, Liu J, et al. lncRNAfunc: a knowledgebase of lncRNA function in human cancer. Nucleic Acids Res 2022;50:D1295-306. [Crossref] [PubMed]
35. Wang P, Mak VC, Rao L, et al. p85β acts as a transcription cofactor and cooperates with BCLAF1 in the nucleus. Nat Commun 2025;16:2042. [Crossref] [PubMed]
36. Zhang Y, He B, Zhang D, et al. FAK-mediated phosphorylation at Y464 regulates p85β nuclear translocation to promote tumorigenesis of ccRCC by repressing RB1 expression. Cell Rep 2023;42:112188. [Crossref] [PubMed]
37. Ge X, Shen Z, Yin Y. Comprehensive review of LncRNA-mediated therapeutic resistance in non-small cell lung cancer. Cancer Cell Int 2024;24:369. [Crossref] [PubMed]