Quetiapine inhibits the oxidative phosphorylation in head and neck squamous cell carcinoma through suppressing NAT10-mediated ac4C modification

Abstract

Background

N4-acetylcytidine (ac4C) is an RNA epigenetic modification, newly discovered to be catalyzed by the enzyme N-acetyltransferase 10 (NAT10). This study aimed to elucidate the functional role and regulatory mechanism of NAT10-mediated ac4C RNA modification in head and neck squamous cell carcinoma (HNSCC), and to explore its potential as a druggable target.

Methods

Bioinformatics analysis of The Cancer Genome Atlas (TCGA) data was performed to assess NAT10 expression and its clinical correlation in HNSCC. Gene set enrichment analysis (GSEA) was conducted on differentially expressed genes from NAT10-high versus NAT10-low patients. The L1000 FireWorks Display (L1000FWD) platform was utilized to predict potential NAT10-targeting drugs. The anti-tumor effects and mechanisms of the top candidate, quetiapine, were validated through in vitro experiments, including binding assays, functional phenotyping (proliferation, migration, and apoptosis), and assessment of mitochondrial function via oxygen consumption rate (OCR) measurements.

Results

NAT10 was significantly upregulated in HNSCC, and its high expression was correlated with advanced tumor stage, higher grade, poor overall survival, and specific immune cell infiltration patterns. GSEA revealed a strong association between NAT10 and the oxidative phosphorylation (OXPHOS) pathway. Quetiapine was identified as a top candidate targeting the NAT10-associated signature. In vitro experiments confirmed that quetiapine directly bound to NAT10, inhibited its expression, and reduced global ac4C levels. Quetiapine treatment potently suppressed HNSCC cell proliferation and migration, while promoting apoptosis. Mechanistically, quetiapine-mediated NAT10 inhibition downregulated key OXPHOS components and substantially decreased cellular OCR, indicating impaired mitochondrial respiration.

Conclusions

NAT10 functions as a critical oncoprotein in HNSCC, potentially by enhancing OXPHOS-driven energy metabolism. The repurposed drug quetiapine suppresses tumor growth by targeting the NAT10/ac4C axis and disrupting mitochondrial respiratory function, positioning it as a promising therapeutic agent for HNSCC.

Highlight box.

Key findings

• N-acetyltransferase 10 (NAT10) is an oncogenic driver and prognostic marker in head and neck squamous cell carcinoma (HNSCC).

• NAT10 mediated N4-acetylcytidine (ac4C) modification promotes tumor progression via oxidative phosphorylation (OXPHOS).

• The antipsychotic drug quetiapine is a novel NAT10 inhibitor with anti-tumor efficacy.

What is known and what is new?

• NAT10 catalyzes the ac4C modification and is implicated in cancer and HNSCC is an aggressive cancer with a need for new therapeutic targets.

• This study newly establishes NAT10 as a critical oncoprotein specifically in HNSCC, showing its significant upregulation and correlation with advanced stage, grade, and poor survival. It reveals a novel functional mechanism, connecting the NAT10/ac4C axis to the OXPHOS pathway in HNSCC, suggesting it fuels tumors by boosting mitochondrial energy production.

What is the implication, and what should change now?

• This positions the NAT10/ac4C axis as a novel therapeutic target and quetiapine as a promising repurposed drug. Current efforts should now pivot to include RNA-modification targeting in HNSCC strategy and prioritize the clinical translation of quetiapine based on this mechanism-driven evidence.

Introduction

Head and neck squamous cell carcinoma (HNSCC) is the sixth most prevalent malignancy worldwide and represents a major global health burden (1). The latest GLOBOCAN 2022 data reporting over 700,000 new diagnoses and more than 300,000 deaths annually for HNSCC further underscores its severe clinical burden (2). The traditional treatment for HNSCC includes surgery and chemoradiotherapy. Despite the advances in treatment strategy (3), with a rate exceeding 50%, patients suffer from recurrence or metastasis (4). First-line treatment for these patients typically combines platinum-based chemotherapy with epidermal growth factor receptor (EGFR) monoclonal antibody cetuximab, but its utility is limited by the acquired drug resistance (5). Immune checkpoint inhibitors, approved by the U.S. Food and Drug Administration (FDA), improve the survival rate of recurrent/metastatic HNSCC patients effectively. However, their benefit is restricted to programmed death-ligand 1 (PD-L1)-expressing patients. Given the limitations of current therapeutic methods, the median overall survival for patients with recurrent/metastatic HNSCC remains as low as 11.6 months (6). Therefore, uncovering novel diagnostic and prognostic biomarkers for early diagnosis, complemented by the development of more effective and better-tolerated therapies, is crucial to ultimately enhance the overall survival of HNSCC patients.

N4-acetylcytidine (ac4C) RNA modification is a highly conserved RNA modification (7) that plays an important role in multiple cell processes, including RNA processing, decay, and translation (8). N-acetyltransferase 10 (NAT10) is the only identified enzyme responsible for ac4C RNA modification (9). Through facilitating the formation of ac4C, NAT10 enhances both the stability and translation efficiency of target transcripts, thereby exerting post-transcriptional control over gene expression (10). Recently, NAT10 has been widely observed in tumor cells, where it regulates the proliferation and apoptosis and is closely related with increased risk for tumor metastasis and recurrence, chemotherapy resistance and radioresistance. Meanwhile, increasing evidence has indicated that NAT10 mediated ac4C modification promotes the expression of pro-tumorigenic messenger RNA (mRNA) and consequently causes the progression of various tumor including HNSCC (11-14). For example, NAT10 mediated ac4C modification on B-cell CLL/lymphoma 9 like protein (BCL9L), SRY-box transcription factor 4 (SOX4), and AKT serine/threonine kinase 1 (AKT1) mRNA improves the stability and translation efficiency of these RNAs, ultimately playing an oncogenic role in bladder cancer (15). NAT10 has also been determined to facilitate the metastasis of gastric cancer through enhancing the process of epithelial-mesenchymal transition (EMT) via the ac4C modification of COL5A1 mRNA (12). In HNSCC, previous study showed that NAT10 is overexpressed and related with the poor prognosis (16). Ac4C modification regulated via NAT10 could promote HNSCC metastasis and remodels tumor microenvironment through mitogen activated protein kinase/extracellular signal regulated kinase (MAPK/ERK) signaling pathway (11). Accordingly, those compelling evidence has positioned NAT10 among one of the most promising HNSCC prognostic biomarkers. Meanwhile, targeting NAT10 represents a promising therapeutic strategy for HNSCC. However, the precise nature of the mRNA substrates and biological role of ac4C modification regulated via NAT10 in HNSCC remain an open area of further research and deserve further research.

Metabolic reprogramming is a fundamental adaptation in cancer cells that supports their proliferative demands and survival under stress. Cancer cells predominantly utilize glycolysis for rapid adenosine triphosphate (ATP) production, a phenomenon termed the Warburg effect, which has been a fundamental concept in cancer metabolism for nearly a century (17,18). Recent years have witnessed a paradigm shift in cancer metabolism research. Growing evidence demonstrates the coexistence of high glycolytic rates with functional oxidative phosphorylation (OXPHOS) in cancer cells (19). OXPHOS is the primary mitochondrial pathway for ATP production through nutrient oxidation, and frequently undergoes dysregulation in cancers, influencing their growth, invasiveness, and therapeutic resistance (20,21). Moreover, it could modulate the tumor microenvironment to promote immune escape by impairing immune cell function (22). Thus, OXPHOS emerges as a functionally significant and compelling target for novel cancer therapies (23). In HNSCC specifically, residual OXPHOS activity in tumor-initiating cells has been mechanistically linked to radioresistance and recurrence (24). Interestingly, preliminary evidence suggests that NAT10 may interface with metabolic pathway, especially in glycolysis. NAT10 facilitates the ac4C modification of forkhead box protein p1 (FOXP1) mRNA and induces the expression of glucose transporter type 4 (GLUT4) and ketohexokinase (KHK), eventually resulted in increased glycolysis and a continuous increase in lactic acid secretion by cervical cancer (CCa) cells (14). NAT10 mediated ac4C acetylation is involved in regulation of glycolysis and promotes osteosarcoma (25). However, whether NAT10-mediated ac4c modification involved in the OXPHOS is still unknown. Given NAT10’s established prognostic value in HNSCC and emerging links to cancer metabolism, we hypothesized that NAT10 could drive HNSCC progression through modulation of OXPHOS activity.

Accordingly, in the present work, we set out to elucidate the role and the underlying mechanism by which the NAT10/ac4C axis accelerates the progression of HNSCC. Subsequently, we screened for existing drugs capable of targeting NAT10, which may provide a potential therapeutic strategy for HNSCC. We present this article in accordance with the MDAR reporting checklist (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1-2683/rc).

Methods

Bioinformatics analysis

RNA sequencing (RNA-seq) data of 33 types of tumors were downloaded from The Cancer Genome Atlas (TCGA) database (https://portal.gdc.cancer.gov/) (dbGaP Study Accession: phs000178). The representative immunohistochemistry (IHC) data of NAT10 are from the human protein atlas (HPA) (https://www.proteinatlas.org/) (The patients id are 4012 and 2608, respectively). The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

For the analysis of immune cell infiltration, we first retrieved 24 markers of immune cells in the literature (26) and extracted the corresponding markers and the expression levels of NAT10 from the RNA-seq of 33 tumor patients in the TCGA database. Subsequently, based on the expression levels of markers in immune cells, the infiltration levels of immune cells in each sample were calculated through the ssGSEA algorithm provided in the “GSVA” (DOI: 10.18129/B9.bioc.GSVA) package of R software. Finally, Pearson correlation analysis was conducted to analyze the correlation between the expression levels of NAT10 and 24 types of immune cells in each sample and the heat map was drawn.

RNA-seq data and clinical information of HNSCC patients were downloaded from TCGA database, too (https://portal.gdc.cancer.gov/) (dbGaP Study Accession: phs000178). Delete patients without complete clinical information and integrate the patients’ clinical information with the expression level of NAT10. For the relationship between NAT10 expression and tumor stages, the clinical correlation between NAT10 and HNSCC patients, the efficacy of diagnosing HNSCC, and the overall survival rate were analyzed respectively using the “stats” (DOI: 10.18129/B9.bioc.stats), “rms” (DOI: 10.18129/B9.bioc.rms), “ROC” (DOI: 10.18129/B9.bioc.ROC), and “survival” (DOI: 10.18129/B9.bioc.survival) packages of R software.

To explore the underlying molecular mechanisms by which NAT10 modulates HNSCC progression, the expression data of NAT10 were extracted from RNA-seq of HNSCC patients in the TCGA database. Subsequently, patients were stratified into low- and high-expression groups based on the median expression level of NAT10. The “DESeq2” (DOI: 10.18129/B9.bioc.DESeq2) package was used to screen differentially expressed genes (DEGs) in the two groups. The differential expression genes were used to perform Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways, Gene Ontology (GO) analysis by Database for Annotation Visualization and Integrated Discovery (DAVID; https://davidbioinformatics.nih.gov/).

Prediction of potential small molecule drugs

DEGs were divided into up- and downregulated groups. Both groups of genes were uploaded to the L1000 FireWorks Display (L1000FWD) platform (https://maayanlab.cloud/L1000FWD/) to obtain ranked prediction results (27). The chemical structures of the drugs were visualized via the PubChem (https://www.ncbi.nlm.nih.gov/).

Molecular docking

The structure of quetiapine was retrieved from PubChem (https://pubchem.ncbi.nlm.nih.gov/). Target proteins structures were acquired from Protein Data Bank (http://www.rcsb.org/). The water molecules and small molecular ligands of the protein was removed via PyMOL software (Version 3.1). The protein molecules were subsequently hydrogenated and converted into Protein Data Bank, Partial Charge, Atom Type format (PDBQT) using AutoDockTools (Version 4.2.6), with their active sites identified. Molecular docking and binding energy calculations were performed using AutoDock Vina (Version 1.1.2). The most favorable binding pose was visualized with PyMOL (Version 2.6). A lower binding energy signifies a stronger interaction between the molecule and the target protein. A binding energy ≤−5.0 kcal/mol suggests potential binding activity, while a value ≤−7.0 kcal/mol indicates a robust and excellent binding affinity.

Surface plasmon resonance (SPR)

SPR binding assays were conducted using a Biacore 8K (GE Healthcare, Boston, USA). Phosphate Buffered Saline with Tween-20 (PBST) buffer was employed as the running buffer. The protein sample was dissolved in coupling buffer and immobilized on a CM5 chip, which had been pre-equilibrated with PBST; 1 µM quetiapine was serially diluted and injected at a flow rate of 30 µL/min, with both the association and dissociation phases lasting 90 seconds. Data were collected using the Biacore 8K evaluation software.

Cell culture and drug treatment

The human oral keratinocytes (HOK) and three HNSCC cell lines, namely WSU-HN6, SCC-15 and CAL-27, were acquired from the American Type Culture Collection. Both cell lines were cultivated in Dulbecco’s modified Eagle medium (DMEM) (Gibco, Grand Island, USA), supplemented with 10% fetal bovine serum (FBS) (Gibco), 1% penicillin/streptomycin (Solarbio, Beijing, China), and maintained under conditions of 5% CO₂ and a temperature of 37 ℃.

A stock solution of quetiapine (Selleck, Houston, USA) was prepared by dissolving the compound in dimethyl sulfoxide (DMSO) and stored at −20 ℃. For the quetiapine treatment, cells were plated in the corresponding plates according to the experimental requirements and cultured for 24 hours to allow adherence before drug exposure. After incubation, the culture medium was replaced with fresh medium containing the drug. To control for potential solvent effects, the control group was treated with an equivalent final concentration of DMSO (0.2% v/v) for every experimental condition.

Western blotting analysis

Total proteins were extracted from SCC-15 cells by RIPA lysis buffer (Solarbio) supplemented with a 100 X protease inhibitor cocktail (Beyotime, Shanghai, China). The protein concentration was then determined using a BCA Protein Assay Kit (Beyotime) following the manufacturer’s protocol. An equal quantity of proteins (30 µg per lane) was separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gels ranging from 4–20% and then transferred onto polyvinylidene difluoride (PVDF) membranes. These membranes were blocked with 5% bovine serum albumin (BSA) for 2 hours at room temperature, followed by overnight incubation with primary antibodies at 4 ℃. The primary antibodies used were anti-NAT10 (1:1,000, Proteintech, Wuhan, China), anti-OXPHOS (1:1,000, Abcam, Cambridge, USA), and anti-GAPDH (1:10,000; Proteintech). The membranes were then incubated with horseradish peroxidase (HRP)-conjugated goat anti-rabbit immunoglobulin G (IgG) (1:10,000, Proteintech) or HRP-conjugated goat anti-mouse IgG (1:10,000, Proteintech) for 2 hours at room temperature. The target proteins were finally detected using a SuperEnhanced chemiluminescence kit (Beyotime).

RNA isolation and real-time polymerase chain reaction (PCR) analysis

Total RNA from SCC-15 cells was obtained using Trizol (Invitrogen, Carlsbad, USA). An equivalent amount of total 1 µg RNA from each group was subsequently transcribed into cDNA employing the Reverse Transcription Kit (TaKaRa, Tokyo, Japan). A real-time PCR reaction mixture was prepared by combining 1 µL of cDNA, 1 µL of forward and reverse primers (each at 10 µM final concentration, Tsingke Biotechnology Co., Ltd., Beijing, China), 5 µL of universal SYBR Green Fast qPCR Mix (ABclonal, Wuhan, China), and 3 µL of RNase-free water. This mixture was then subjected to real-time PCR analysis using the ABI 7500 Sequence Detection System (Invitrogen). The real-time PCR program is shown in Table 1. GAPDH was utilized as the internal reference gene. The comparative CT method (ΔΔCT) was applied to analyze all real-time PCR data. The sequences of the primers used are provided in Table 2.

Table 1. The real-time PCR program.

Step	Temperature (℃)	Time	Cycle
Initial denaturation	95	2 min	–
Denaturation	95	15 s	40
Annealing/extension	60	30 s
Melting curve analysis	95	15 s	–
	60	1 min	–
	95	Continuous acquisition	–

PCR, polymerase chain reaction.

Table 2. Primers for real-time PCR.

Gene	Primer sequence
NAT10	F: GGGATTGGCCTGCAGCATA
NAT10	R: GGCTCCATGACCACATCCTT
SDHD	F: CTTCGAACTCCAGTGGTCAGA
SDHD	R: CAACTTGTCCAAGGCCCAATG
UQCRH	F: GGACTGGAGGACGAGCAAAA
UQCRH	R: GCACTGAAAGCCTCAGTCCT
PRMT1	F: ATCGGGATCGAGTGTTCCAG
PRMT1	R: TAGATGTCCACCTCCAGCCAC
UQCR10	F: GGGGTGACAGTGGAGTAGAGA
UQCR10	R: GCGAATGGTGGTCCATGAAG
NDUFS6	F: GTTCAGACAGCACCACCACT
NDUFS6	R: CACCAGGAATACCCTTCGCA
NDUFB2	F: GTATGTCCGCTCTGACTCGG
NDUFB2	R: AACTCGCTCTGGAACACCTG
NDUFA11	F: CTGCCTCGAAGTCCTCTCTG
NDUFA11	R: GTGTGTCCGCCCTTTCTCAG
ATP5MC1	F: AGGGCTAAAGCTGGGAGACT
ATP5MC1	R: CCACCTGGAGTGGGAAGTTG
UQCRC1	F: TGTCTCGTGCAGACTTGACC
UQCRC1	R: GGCGAGGTCTAACAGTTGCT

F, forward; PCR, polymerase chain reaction; R, reverse.

Ac4C quantification assay

The total ac4C levels in SCC-15 cells were quantified by dot blot assay. Briefly, total RNA was harvested using TRIzol reagent; 1 µg RNA were spotted onto a nylon membrane, air-dried, and immobilized via UV cross-linking. The membrane was then blocked with 5% BSA for 1 hour at room temperature, followed by overnight incubation at 4 ℃ with ac4C primary antibody. After three washes, the membrane was probed with an HRP-conjugated secondary antibody for 1 hour at room temperature. Signals were detected using an enhanced chemiluminescence kit and captured with a chemiluminescence imaging system. All experiments were performed in triplicate to ensure reproducibility. Finally, Image J software (version 1.54g) was used to analyze the expression level of ac4C.

Cell Counting Kit-8 (CCK-8) assay

Cell viability was assessed utilizing CCK-8 kit (Beyotime), following the manufacturer’s instruction. SCC-15 cells were plated into 96-well plates with a density of 2,000 cells per well. After the corresponding cultivation period (0, 24, 48, or 72 hours), the supernatant was substituted with DMEM mixed with CCK-8 solution in a ratio of 10:1, and then incubated at 37 ℃ for 2 hours. The optical density (OD) value at 450 nm was determined using an automatic microplate reader (BioTek ELX808, Winooski, USA). For each biological replicate, we performed quintuplicate technical replicates (5 wells per condition). The viability for each biological replicate was indeed calculated from the average OD value of its five technical wells, as per the formula provided.

The formula for calculating cell viability is as follows: cell viability = [(OD of experimental wells − OD of blank wells)/(OD of control wells − OD of blank wells)] * 100%.

Colony formation assay

SCC-15 cells were seeded into 6-well plates at a density of 400 cells per well and incubated for 7 days. The culture medium was replaced every 48 hours. On the seventh day, the cell clones were rinsed with PBS, fixed using 4% formaldehyde for 15 min, and subsequently stained with 0.1% crystal violet solution for 20 min. Finally, the number of colonies was counted and analyzed using the Image J software (version 1.54g).

Flow cytometric analysis

SCC-15 cells were plated into 6-well plates with a concentration of 3×10⁵ cells per well and incubated for 48 hours. Cells were harvesting, washing twice with cold PBS. Then cells were resuspending in 1× binding buffer, stained with 5 µL of Annexin V-fluorescein isothiocyanate (FITC) and 10 µL of propidium iodide (PI) and incubated in the dark at room temperature for 15 min employing the Annexin V-FITC Apoptosis Detection Kit (Solarbio). Finally, flow cytometry (BD FACS Calibur, San Jose, USA) was used to detect the percentage of apoptotic cells.

Migration assay

Transwell migration and wound healing assays were conducted to assess the ability of cell migration. In the transwell migration assay, cells were seeded onto the upper chamber (pore size, 8 µm; Corning, New York, USA) contained serum-free DMEM medium with a concentration of 8×10⁴ cells per well. The bottom chamber contained 700 µL of DMEM medium supplemented with 20% FBS. Following an incubation of 19 hours, the cells adhering to the underside of the insert were rinsed with PBS, fixed in 95% ethanol for 30 min, and subsequently stained with 0.1% crystal violet for 20 min. A minimum of five randomly selected fields were photographed using a microscope and analyzed using ImageJ software.

For the wound healing assay, 2×10⁴ cells were plated in 6-well plates. Once the cell density reaches 100%, draw a straight line vertically with a sterile pipette tip in the cell culture dish and take a photo under a microscope. After 24 hours of cultivation, the Olympus BX60 microscope (Olympus, Japan) was used to take another photo. Image J software was employed to calculate the area of cell scratch healing.

Seahorse analysis

A Seahorse XF Cell MitoStress Test Kit, along with a Bioscience XF96 Extracellular Flux Analyzer, was employed to quantify the oxygen consumption rate (OCR). SCC-15 cells were pre-treated with quetiapine or DMSO (control) for 48 hours in standard culture plates. Then, 2×10⁴ SCC-15 cells were trypsinized and plated in 96-well Seahorse plates. The cells were then incubated overnight in the continued presence of their respective pretreatment concentrations of quetiapine or DMSO to ensure stable attachment and sustained drug exposure. Following washing with Seahorse buffer, oligomycin, carbonyl cyanide-4-(trifluoromethoxy)phenylhydrazone (FCCP), and rotenone were automatically dispensed to determine the OCR. The OCR values were normalized to the cell count of SCC-15 cells. Briefly, following the Seahorse assay, the cell number in each well was quantified using the Hoechst 33342 staining (Beyotime). These counts were then set as the Normalization Values in the Agilent Wave software with the Normalization Unit as “per 103 cells” and the Normalization Scale Factor as 1. The results are depicted as the mean ± standard deviation (SD).

Statistical analysis

The statistical analysis was conducted using GraphPad Prism (version 8.4.3; San Diego, CA, USA) and R software (version 4.5.2; Boston, MA, USA). To validate the data, all experiments were conducted in triplicate. The error bars were represented as the mean ± SD (n=3). Between-group normally-distributed data were examined through Student’s t-tests, whereas multi-group analyses employed one-way analysis of variance (ANOVA). Non-normally-distributed data were analyzed with Wilcoxon rank-sum test. P<0.05 was deemed statistically significant.

Results

The mRNA expression of NAT10 in pan cancers

To investigate the role of NAT10 in the advancement of malignant tumors including HNSCC, we initiated an analysis of NAT10 mRNA expression across 33 human cancer types (RNA-seq data sourced from TCGA), utilizing their corresponding normal tissues as controls (Figure 1A). Among the 33 types cancer, excluding nine cases only tumor tissues were available, NAT10 was upregulated in 13 cancers. To further solidify our analysis concerning the differential NAT10 expression across multiple tumors from TCGA, we conducted a comparison of NAT10 expression levels between these human tumors and their paired normal tissues (Figure 1B), and similar results across tumors, indicating the promotion role of NAT10 in tumor progression.

Figure 1.

The expression of NAT10 in various tumors. A comparison of NAT10 mRNA expression in diverse human cancers versus unpaired (A) or paired (B) normal tissues, utilizing data sourced from the TCGA database. The red boxes represent the mRNA expression of NAT10 in HNSCC versus unpaired or paired normal tissues, respectively. Data are presented as mean ± SD. ns, no significance; *, P<0.05; **, P<0.01; ***, P<0.001 by unpaired t-test (A) and paired t-test (B). ACC (normal =0, tumor =79), BLCA (normal =19, tumor =412), BRCA (normal =113, tumor =1113), CESC (normal =3, tumor =306), CHOL (normal =9, tumor =35), COAD (normal =41, tumor =480), DLBC (normal =0, tumor =48), ESCA (normal =11, tumor =163), GBM (normal =5, tumor =169), HNSC (normal =44, tumor =504), KICH (normal =25, tumor =65), KIRC (normal =72, tumor =541), KIRP (normal =32, tumor =291), LAML (normal =0, tumor =150), LGG (normal =0, tumor =532), LIHC (normal =50, tumor =374), LUAD (normal =59, tumor =539), LUSC (normal =49, tumor =502), MESO (normal =0, tumor =87), OV (normal =0, tumor =381), PAAD (normal =4, tumor =179), PCPG (normal =3, tumor =184), PRAD (normal=52, tumor=501), READ (normal =10, tumor =167), SARC (normal =2, tumor =263), SKCM (normal =1, tumor =472), STAD (normal =32, tumor =375), TGCT (normal =0, tumor =156), THCA (normal =59, tumor =512), THYM (normal =2, tumor =120), UCEC (normal =35, tumor =554), UCS (normal =0, tumor =57), UVM (normal =0, tumor =80). (A) Except for ACC, DLBC, LAML, LGG, MESO, OV, SARC, SKCM, TGCT, THYM, UCS, and UVM, the P values of the other items from left to right were <0.001, 2.97e−27, 0.36, 2.82e−09, 3.09e−22, <0.001, 0.08, 5e−11, 0.69, 4.7e−11, 0.03, 1.76e−23, 7.88e−17, 5.87e−20, 0.54, 0.15, 0.17, 1.05e−06, 1.04e−16, 0.01, 0.17, respectively. (B) Except for SARC, SKCM, and THYM, the P values of the other items from left to right were 7.63e−06, 8.17e−11, >0.99, 0.008, 9.09e−13, 0.008, 2.59e−09, 0.89, 5.13e−10, <0.001, 2.58e−09, 2.27e−08, 1.27e−10, 0.62, 0.75, 0.01, 0.01, 4.92e−07, 0.01, and 0.16, respectively. ACC, adrenocortical carcinoma; BLCA, bladder urothelial carcinoma; BRCA, breast invasive carcinoma; CESC, cervical squamous cell carcinoma; CHOL, cholangiocarcinoma; COAD, colon adenocarcinoma; DLBC, lymphoid neoplasm diffuse large b-cell lymphoma; ESCA, esophageal squamous cell carcinoma; GBM, glioblastoma multiforme; HNSC, head and neck squamous cell carcinoma; KICH, kidney chromophobe; KIRC, kidney renal clear cell carcinoma; KIRP, kidney renal papillary cell carcinoma; LAML, acute myeloid leukemia; LGG, brain lower grade glioma; LIHC, liver hepatocellular carcinoma; LUAD, lung adenocarcinoma; LUSC, lung squamous cell carcinoma; MESO, mesothelioma; mRNA, messenger RNA; OV, ovarian serous cystadenocarcinoma; PAAD, pancreatic adenocarcinoma; PCPG, pheochromocytoma and paraganglioma; PRAD, prostate adenocarcinoma; READ, rectum adenocarcinoma; SARC, sarcoma; SD, standard deviation; SKCM, skin cutaneous melanoma; STAD, stomach adenocarcinoma; TCGA, The Cancer Genome Atlas; TGCT, testicular germ cell tumors; THCA, thyroid carcinoma; THYM, thymoma; TPM, transcripts per million; UCEC, uterine corpus endometrial carcinoma; UCS, uterine carcinosarcoma; UVM, uveal melanoma.

The relationship between NAT10 and tumor immune cell infiltration

Given the crucial role of immune cells infiltrating in tumor progression, we analyzed the relationship between NAT10 and immune cell infiltration across various cancer types. In HNSCC, NAT10 expression was positive correlated with the infiltration levels of T helper cells, central memory T cell (Tcm), Th2 cells, and negatively correlated with the infiltration levels of B cells, CD8 T cells, cytotoxic cells, dendritic cell (DC), immature dendritic cell (iDC), mast cells, natural killer (NK) CD56dim cells, plasmacytoid dendritic cell (pDC) cells, T cells, T follicular helper (TFH), Th17 cells, and regulatory T cell (Treg) (Figure 2). Meanwhile, significant correlations between NAT10 expression and immune infiltration landscapes were also observed in other cancer types. Accordingly, our findings revealed that the expression of NAT10 was generally correlated with multiple immune cells.

Figure 2.

The relationship between NAT10 and the infiltration level of immune cells. *, P<0.05 by Pearson correlation analysis. The red box represent the correlation between NAT10 and immune cells in HNSCC. For HNSC, the P values from top to bottom were 0.549, <0.001, 3.86e−05, 9.81e−07, 1.99e−05, 0.37, 3.71e−06, 0.60, 0.002, 0.052, 0.31, 1.84e−05, 0.89, 2.82e−06, 2.11e−06, 0.03, 0.004, 0.26, 4.72e−05, 0.15, 0.11, 4.81e−08, 2.08e−09, and 0.04 , respectively. aDC, activated dendritic cell; DC, dendritic cell; HNSC, head and neck squamous cell carcinoma; iDC, immature dendritic cell; NK, natural killer; pDC, plasmacytoid dendritic cell; Tcm, central memory T cell; Tem, effector memory T cell; TFH, T follicular helper; Tgd, γδ T cell; Treg, regulatory T cell.

The clinical significance of NAT10 in HNSCC

To obtain the association between NAT10 and HNSCC patients’ clinicopathological features, we downloaded the IHC image of NAT10 in HNSCC tissues form the HPA (https://www.proteinatlas.org/). As the representative images shown in Figure 3A, NAT10 expression was markedly upregulated in tumor tissues compared to normal tissues. Meanwhile, NAT10 was elevated in HNSCC patients with higher stage and grades (Figure 3B). Furthermore, integrated the expression level of NAT10 and the survival information of HNSCC patients (delete one case of HNSCC patient with no survival information), and drew the Kaplan-Meier curve by dividing them into high- and low-expression groups (n=252 and n=251, respectively) according to the expression level of NAT10 (median value: 3.91365537748694). The results showed that high NAT10 expression was associated with reduced overall survival in HNSCC (Figure 3C). Receiver operating characteristic (ROC) curves for predicting survival probabilities based on NAT10 expression levels showed good discriminatory ability (area under the curve was 0.912), indicating that NAT10 expression can serve as a useful prognostic marker for HNSCC patients (Figure 3D). Univariate analysis revealed that advanced tumor (T) stage, node (N) stage and high NAT10 expression were significantly associated with poorer survival. Importantly, in the multivariate Cox regression model adjusting for these clinical features, high NAT10 expression remained an independent predictor of adverse prognosis (Figure 3E). Collectively, these results indicated that high NAT10 expression association with unfavorable prognosis in patients with HNSCC.

Figure 3.

NAT10 was associated with poor prognosis in patients with HNSCC. (A) Representative images of IHC staining for NAT10 in tumor tissues compared to normal tissues (http://www.rcsb.org/, https://www.proteinatlas.org/ENSG00000135372-NAT10/tissue/oral+mucosa; https://www.proteinatlas.org/ENSG00000135372-NAT10/cancer/head+and+neck+cancer). (B) The relationship between NAT10 expression and T stage and grade; n=447 and n=483, respectively. P=0.02 (left) and P=0.01 (right). (C) ROC curves and AUC values; n=503. (D) Kaplan-Meier curve for cumulative survival time. HNSCC patients were separated by the median expression of NAT10; low =251, high =252. (E) Univariate and multivariate Cox regression analysis; n=503. *, P<0.05 by log-rank test. AUC, area under the curve; CI, confidence interval; FPR, false positive rate; G, grade; HNSCC, head and neck squamous cell carcinoma; HR, hazard ratio; IHC, immunohistochemistry; N, node; ROC, receiver operating characteristic; T, tumor; TPM, transcripts per million; TPR, true positive rate.

Functional enrichment analysis of NAT10 and identification of potential targeted drugs

Patients were stratified into low-expression (n=252) and high-expression groups (n=252) based on the median expression level of NAT10 (median value: 3.91365537748694). A total of 664 DEGs were screened based on the criteria of |log fold change (FC) >1| and P.adj<0.05, consisting of 258 upregulated and 406 downregulated (Figure 4A). To further explore the biological functions of NAT10, we analyzed 664 DEGs related to NAT10. The GSEA results showed that the DEGs enriched the OXPHOS signaling pathway (Figure 4B). The most enriched GO terms including GTP binding, mitochondrial matrix and mitotic nuclear division (Figure 4C). Accordingly, these results indicates that NAT10 is correlated with the process of OXPHOS.

Figure 4.

Enrichment analysis revealed that NAT10 could regulate the process of OXPHOS. (A) Volcano plot depicting DEGs. (B) GSEA enrichment map of DEGs. (C) GO enrichment map of DEGs. BP, biological process; CC, cellular component; CXCR, C-X-C chemokine receptor; DEG, differentially expressed gene; FDR, false discovery rate; GO, Gene Ontology; GSEA, gene set enrichment analysis; GTP, guanosine triphosphate; NES, normalized enrichment score; KEGG, Kyoto Encyclopedia of Genes and Genomes; MF, molecular function; OXPHOS, oxidative phosphorylation; sig, significance.

Quetiapine treatment regulated NAT10-mediated ac4C modification in HNSCC

Subsequently, the L1000FWD platform was utilized to predict potential drugs targeting these 664 DEGs. Among the potential drugs, quetiapine exhibited the highest binding score (Figure 5A). The structure of quetiapine was shown in Figure 5B. To validate the binding between NAT10 and quetiapine, Molecular docking analysis was performed and the result showed that NAT10 and quetiapine could be bonded by hydrogen and hydrophobic bonds (Figure 5C). To further validate this binding, SPR were performed, which confirmed that quetiapine could effectively bind with NAT10 (Figure 5D). Additionally, we analyzed the mRNA and protein expression levels of NAT10 in HNSCC cell line. Compared with HOK, the expression of NAT10 in HNSCC cells is significantly increased (Figure 5E,5F). At the same time, the expression level of NAT10 in SCC-15 cells is the highest among the three HNSCC cell lines (Figure 5E,5F). Given that the SCC-15 cells exhibited the highest basal expression of NAT10 among the tested HNSCC cell lines, it was selected as the primary model for subsequent in vitro functional experiments. Then, to determine the regulatory role of quetiapine in the NAT10/ac4C axis, we evaluated NAT10 expression and ac4C levels. The appropriate working concentrations of quetiapine in this study were selected based on preliminary dose-response experiments. The half-maximal inhibitory concentration (IC₅₀) according to the CCK-8 analysis was calculated to be 32.56 µM (Figure S1). Based on this result, a concentration of 20 µM, which represents effective dose, was selected for all further experimental investigations. Compared to the control group, quetiapine treatment markedly downregulated NAT10 mRNA and protein (Figure 5G,5H). Consequently, the global ac4C abundance in HNSCC cells were also decreased (Figure 5I). These findings collectively indicate that quetiapine suppressed NAT10-mediated ac4C modification in HNSCC.

Figure 5.

Quetiapine could regulate NAT10-mediated ac4C modification in HNSCC. (A) The mainly potential drugs identified by L1000FWD database. (B) The chemical structure of quetiapine. (C) Molecular docking of NAT10 and quetiapine. (D) The binding sensorgram of the interactions between NAT10 and quetiapine. (E) The protein expression of NAT10 was detected through western blotting (left) and quantitatively analyzed (right); n=3. From left to right, P=0.001, <0.001, and 0.002, respectively. (F) The mRNA expression of NAT10 in HNSCC cells; n=3. From left to right, P=0.03, 0.002, and 0.08, respectively. (G) The protein expression of NAT10 after SCC-15 cells were treated with quetiapine was detected through western blotting (left) and quantitatively analyzed (right); n=3. P<0.001. (H) The mRNA expression of NAT10 after SCC-15 cells were treated with quetiapine; n=3. P<0.001. (I) Dot blot assay was conducted to assess the ac4C level. P<0.001. Data are presented as mean ± SD. ns, no significance; *, P<0.05; **, P<0.01; ***, P<0.001 by unpaired t-test. EGFR, epidermal growth factor receptor; HDAC, histone deacetylase; HNSCC, head and neck squamous cell carcinoma; MEK, methyl ethyl ketone; MOA, mechanism of action; mRNA, messenger RNA; NF-κB, nuclear factor kappa-B; PARP, poly ADP-ribose polymerase; PLK, polo-like kinase; RAF, Raf kinase; RU, response unit; SD, standard deviation.

Inhibition of NAT10 via quetiapine diminished cell proliferation, migration, and facilitated cell apoptosis

Then we treated SCC-15 cells with quetiapine and conducted a comprehensive evaluation of the effect of NAT10 in HNSCC progression in vitro. After 24, 48, and 72 hours of quetiapine treatment, the cell viability rates decreased by 31.55%, 34.6% and 62.37%, respectively, suggesting that quetiapine inhibited cell proliferation (Figure 6A). Comparable results were obtained from the colony formation assay, where the number of colonies decreased by about 60% after the treatment of quetiapine (Figure 6B,6C). These findings thus suggested that quetiapine has an inhibitory effect on proliferation in HNSCC cells. To further investigate the role of quetiapine in HNSCC cells, we examined cell apoptosis following the treatment of quetiapine. Compared to the control group, cells treated with quetiapine expression exhibited a notably increased apoptosis rate by approximately 10.28% (Figure 6D,6E, Figure S2). Results from the transwell migration assay revealed that the migration number of SCC-15 cells was significantly reduced after quetiapine treatment, by approximately 69.43% (Figure 6F,6G). Similarly, the wound healing capacity of SCC-15 cells was also significantly impaired. The relative wound healing percentage after 24 hours decreased by 20.66% upon quetiapine treatment (Figure 6H,6I). In summary, these data implied that quetiapine treatment effectively inhibited the malignant phenotypes in HNSCC cells in vitro.

Figure 6.

Quetiapine treatment inhibited HNSCC malignant phenotypes in vitro. (A) CCK-8 assay showed that quetiapine treatment inhibited the proliferation of SCC-15 cells; n=5. P<0.001. (B) Representative images of SCC-15 clones treated with quetiapine using crystal violet solution staining. (C) The quantitation of colony number for SCC-15 cells treated with quetiapine; n=3. P=0.001. (D,E) Apoptosis assay results of SCC-15 cells treated with quetiapine; n=3. P=0.02. (F) Representative imaged of SCC-15 transwell migration treated with quetiapine using crystal violet solution staining. (G) Quantified of transwell migration in SCC-15 cells treated with quetiapine; n=3. P=0.003. (H,I) Wound healing assays showed the treatment of quetiapine inhibited migration of SCC-15 cells. The migration distance was quantified at 24 hours after scratching; n=3. P<0.001. *, P<0.05; **, P<0.01; ***, P<0.001 by unpaired t-test. CCK-8, Cell Counting Kit-8; FITC, fluorescein isothiocyanate; HNSCC, head and neck squamous cell carcinoma; PI, propidium iodide.

Inhibition of NAT10 via quetiapine repressed the process of OXPHOS

With an aim to identify whether the inhibited of NAT10 via quetiapine could regulate the process of OXPHOS, we first detected the mRNA expression of the key genes involved in the process of OXPHOS. As results showed in the Figure 7A, the mRNA expression levels of OXPHOS were decreased after quetiapine treated. Moreover, the protein expression level of OXPHOS were decreased after quetiapine treated (Figure 7B). Furthermore, we conducted the seahorse analysis to detect OCR, which reflects mitochondrial respiration. Seahorse analysis also indicated a decreased OCR in quetiapine treated cells (Figure 7C,7D). Our study reveals the mechanism by which quetiapine inhibits the progression of HNSCC through OXPHOS.

Figure 7.

Quetiapine treatment inhibited the process of OXPHOS. (A) The mRNA expression of OXPHOS related genes after SCC-15 cells were treated with quetiapine; n=3. From left to right, P values <0.001, <0.001, <0.001, 0.02, <0.001, 0.003, 0.004, <0.001, and <0.001, respectively. (B) The protein expression of OXPHOS related proteins after SCC-15 cells were treated with quetiapine were detected through western blotting (left) and quantitatively analyzed (right); n=3. From left to right, P=0.002, <0.001, <0.001, 0.01, and 0.003, respectively. (C) OCR, which reflects mitochondrial respiration, was decreased in quetiapine treated SCC-15 cells. (D) Statistical chart of OCR various parameters about SCC-15 cells after treated with quetiapine; n=3. From left to right, P values <0.001, 0.002, 0.64, <0.001, and 0.07, respectively. ns, no significance; *, P<0.05, **, P<0.01; ***, P<0.005 by unpaired t-test. mRNA, messenger RNA; FCCP, carbonyl cyanide 4-(trifluoromethoxy)phenylhydrazone; OCR, oxygen consumption rate; OXPHOS, oxidative phosphorylation.

Discussion

Tumor cells can maximize the energy supply for their life activities such as proliferation, invasion, and metastasis by regulating their metabolic energy supply. As early as the last century, Warburg pioneered the idea that even in oxygen rich environments, tumor cells still tend to use glycolysis for glucose metabolism to produce ATP, rather than OXPHOS within mitochondria to achieve complete glucose oxidation (28). Warburg initially hypothesized that tumor cells possess developmental defects in mitochondria, resulting in impaired aerobic respiration and consequent reliance on glycolytic metabolism. Furthermore, earlier study generally posited that the upregulation of glycolysis in cancer cells was accompanied by a concomitant suppression of OXPHOS (29). However, recently, studies related to tumor metabolism have shown that there is not a sufficient and necessary relationship between the use of aerobic glycolysis by tumor cells and mitochondrial damage (30,31). Many tumors that utilize aerobic glycolysis also have complete mitochondrial respiration (32,33). Although malignancies such as HNSCC, leukemia, lymphoma, pancreatic ductal adenocarcinoma, and endometrial carcinoma exhibit enhanced glycolysis, they often maintain functional OXPHOS (19). Beyond supplying sufficient energy for tumor survival, OXPHOS also regulates critical processes including cancer cell proliferation, invasion, and metastasis. In HNSCC, it is reported that OXPHOS is enriched in recurrent human papillomavirus (HPV)-associated HNSCC, which may contribute to treatment failure (24), meanwhile, enhanced OXPHOS process regulating could promote the tumor proliferation (34). Moreover, it can suppress immune cell function within the tumor microenvironment, thereby promoting immune escape. Consequently, targeting OXPHOS represents a functionally significant and promising therapeutic strategy in oncology (35,36). Here, our experimental results established that the suppression of OXPHOS via the pharmacological inhibition of NAT10 as a key mechanism underlying the therapeutic efficacy in HNSCC.

RNA modifications, which are crucial for mRNA stability, transcription, and translation, have emerged as key players in the pathogenesis of various human diseases, including tumor. Among these, ac4C modification has gained significant attention for its essential role in tumor progression (37). Served as the only identified enzyme responsible for ac4C modification, accumulating evidence underscores the pivotal contribution of NAT10-mediated ac4C modification to tumor progression (38). Studies utilizing NAT10-deficient mouse models of bladder cancer have demonstrated an ac4C-dependent reduction in tumor burden. Similarly, NAT10-driven ac4C modification of FSP1 mRNA has been shown to facilitate colon cancer progression and metastasis (13). However, the functional significance of NAT10-mediated ac4C acetylation in HNSCC remains to be elucidated. In this study, we delineated that high NAT10 expression was associated with the poor survival. Furthermore, we established NAT10 as a central node by identifying DEGs between NAT10-high and NAT10-low expressing HNSCC patient tissues. Using the L1000FWD platform, we predicted potential therapeutic drugs targeting these DEGs and subsequently validated through experiments that the identified drug quetiapine not only could bind to NAT10 but also suppress malignant phenotypes in HNSCC.

Quetiapine, an atypical antipsychotic, has demonstrated anti-proliferative effects in some cancers (39). In hepatocellular carcinoma, quetiapine exerts its anti-tumor effect by suppressed the phosphorylation protein expression of ERK and NF-κB (40). Given that both pathways play pivotal pro-tumorigenic roles in HNSCC (41,42), this shared mechanism highlights the potential of quetiapine as a promising candidate for therapeutic. Despite this compelling rationale, its role and underlying mechanism on HNSCC progression is unknown. In this study, based on integrated bioinformatics analysis and experimental validation, we found that quetiapine could suppress malignant phenotypes of HNSCC by directly binding to NAT10 and consequently inhibiting the OXPHOS process. This finding elucidates a novel molecular mechanism underlying quetiapine’s anti-tumor activity, suggesting it as a promising treatment strategy for HNSCC treatment. Meanwhile, we proposed the translational potential of quetiapine, an FDA-approved drug with a known safety profile, which could significantly accelerate its path to clinical application and offer a new treatment option for HNSCC patients.

This study identifies potential therapeutic implications. Firstly, by the analysis of HNSCC patient samples and experimental validation, we observed a significant correlation between NAT10 expression levels and tumor grade, with higher expression levels associated with more aggressive tumors. Furthermore, patients with high NAT10 expression had poorer prognosis, suggesting that NAT10 may serve as a prognostic biomarker for HNSCC. These findings are consistent with previous studies reporting the association between NAT10 expression and cancer progression and prognosis in various other cancer types (16). Secondly, given the critical role of OXPHOS in cancer cell metabolism, targeting this process has emerged as a promising strategy for cancer treatment. Our results suggest that inhibiting NAT10 via quetiapine could disrupt the enhanced OXPHOS activity in HNSCC cells, leading to impaired cell growth and proliferation. Thirdly, our bioinformatics analysis revealed that NAT10 expression is correlated with the infiltration levels of multiple immune cell types in HNSCC. Interestingly, the alteration of OXPHOS could damage the function of immune cells infiltration in the tumor microenvironment, leading to immune evasion (43). Thus, we reasonably assumed that quetiapine could overcome immune evasion to achieve a better therapeutic effect on HNSCC, which warrants further investigation.

However, several questions remain to be addressed in future studies. For instance, the mechanism by which NAT10 regulates OXPHOS is multifaceted. It is reported that the enhanced OXPHOS process regulated by integrin subunit beta 2 (ITGB2)/phosphatidylinositol 3-kinase (PI3K)/AKT/mammalian target of rapamycin (mTOR) signaling pathway could promote the tumor proliferation. Previous studies reported that NAT10 could regulate PI3K/AKT signaling pathway to promote the migration and invasion of hepatocellular carcinoma (44). This finding provides a plausible explanation that NAT10 may directly interact with PI3K/AKT signaling pathway to regulate the process OXPHOS. Nonetheless, the precise molecular mechanism by which NAT10 regulates OXPHOS needs further elucidation. Additionally, it would be interesting to investigate the potential cross-talk between NAT10-regulated OXPHOS and other metabolic pathways, such as glycolysis and fatty acid synthesis, which are also known to be altered in cancer cells. Furthermore, the identification of additional biomarkers that could predict the response to NAT10 inhibition would be valuable for patient selection and personalized treatment strategies.

Conclusions

To summarize, our work identifies quetiapine as a molecular inhibitor of the NAT10-ac4C regulatory axis in HNSCC. By targeting this axis, quetiapine effectively disrupts OXPHOS and impedes tumor growth. This study provides a novel metabolic mechanism for drug repurposing and highlights a viable strategy for targeting RNA modification in cancer therapy.

Supplementary

The article’s supplementary files as

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.

Data Sharing Statement

Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1-2683/dss

tcr-15-03-175-dss.pdf ^{(101.9KB, pdf)}

DOI: 10.21037/tcr-2025-1-2683