NAT10 promotes gastric cancer progression by enhancing the N4-acetylcytidine modification of TNC mRNA

Abstract

Background

Gastric cancer (GC) is a very aggressive malignant tumor of the digestive system. Previous studies have shown that N-acetyltransferase 10 (NAT10) can regulate the N4-acetylcytidine (ac4C) modification of downstream mRNAs through certain pathways to promote the progression of various tumors. However, reports on the regulatory effects of NAT10 on GC are rare. This study aimed to explore the potential mechanism by which NAT10 regulated GC progression.

Methods

Clinical samples were used to study the correlation between NAT10 expression and poor prognosis in patients with GC by univariate analysis and multivariate analysis. In vitro and in vivo assays were performed to assess the effects of NAT10 and Tenascin C (TNC) on the malignant biological behaviors of GC cells. Acetylated RNA immunoprecipitation sequencing was conducted to explore the role of NAT10 in ac4C modification in GC. mRNA stability and translation efficiency assays were performed to investigate the effect of changes in NAT10 expression on its target TNC.

Results

Analysis of clinical samples revealed that NAT10 expression was abnormally elevated and positively correlated with TNC expression in GC, and increased NAT10 expression led to poor overall survival. In vitro and in vivo experiments revealed that high NAT10 expression promoted the invasive and proliferative capacity of GC cells. Rescue experiments suggested that TNC played an important role in the above process. Mechanistically, the acetylation-based RNA immunoprecipitation sequencing and acetylated RNA immunoprecipitation qPCR results indicated that NAT10 regulated the level of ac4C modification by binding to specific regions in TNC mRNA, increasing mRNA stability and translation, upregulating TNC expression, further activating the TNC/Akt/TGF-β1 positive feedback loop.

Conclusions

In summary, our results reveal that NAT10 plays a critical role in GC development by affecting TNC mRNA stability and translation efficiency, which ultimately activates the TNC/Akt/TGF-β1 positive feedback loop. This study is expected to provide a novel target and theoretical basis for improving the diagnosis and treatment of GC.

Background

Gastric cancer (GC) is a highly aggressive malignancy, which ranks fifth in both morbidity and mortality among malignancies worldwide [1]. Notably, despite great progress in currently available treatments, the prognosis of patients with advanced GC remains poor, and the 5-year survival rate is low [2]. Therefore, it is extremely important to explore the mechanisms underlying GC progression; such studies may provide new discoveries and potential targets for the diagnosis and treatment of GC.

Abnormal regulation of RNA epigenetic modifications plays a key role in affecting mRNA stability and translation efficiency in malignant tumors [3]. Among such modifications, the N4-acetylcytidine (ac4C) modification, which is a highly conserved RNA modification, is widely distributed in the human transcriptome [4], but most ac4C modifications are found in core coding sequences (CDSs) [5]. N-acetyltransferase 10 (NAT10), which is an ATP-dependent RNA acetyltransferase that mainly localizes to the nucleus, plays a key role in ac4C modification. NAT10 is involved in almost all ac4C modifications in eukaryotes [6]. Recent studies have shown that NAT10 is crucial for the occurrence and development of various malignant tumors, such as colon cancer, pancreatic cancer, breast cancer and bladder cancer [5, 7,8,9]. However, the functional role of NAT10, as well as the specific underlying mechanism in GC progression remain unclear. Tenascin C (TNC) is a multi-modular extracellular matrix protein that has four distinct domains [10]. TNC, which is a macromolecular glycoprotein, exhibits many significant features related to its structure, expression and function [11]. TNC has been shown to be highly expressed in the tumor tissues of a variety of malignant solid tumors, and it plays an important role in tumor growth, metastasis, angiogenesis and inhibition of immune surveillance [12, 13]. We are interested by the fact that relevant reports suggest that the activation of signaling pathways, such as the MAPK and Wnt pathways, can upregulate TNC expression. Interestingly, these two signaling pathways can also be activated by TNC [14]. This suggests that there is a positive feedback regulatory loop between TNC expression and signaling pathway activation in malignant tumors. Based on these findings, we believe that further exploration of whether NAT10 can regulate TNC expression and promote GC progression is valuable.

In the present study, we found that NAT10 overexpression could drive the malignant biological behaviors of GC cells by catalyzing the ac4C modification of TNC mRNA and then inducing the phosphorylation of Akt; these phenomena resulted in upregulated TGF-β1 expression, which in turn increased TNC expression. Therefore, the present study revealed a potential mechanism by which NAT10 promoted the malignant phenotype of GC via a TNC/Akt/TGF-β1 positive feedback loop.

Methods

Patients and tissue specimens

In this study, we used primary human GC tissues and paired adjacent normal tissues that were obtained from the Department of Gastrointestinal Surgery of the First Affiliated Hospital of Fujian Medical University (Fujian, China). A total of 158 GC patient specimens were obtained at the time of surgery for IHC analysis. Another 30 pairs were fetched for qRT‒PCR. Written informed consent was obtained from all the subjects. This study was approved by the Ethics Committee of the First Affiliated Hospital of Fujian Medical University.

Bioinformatics analysis

Bioinformatics data of NAT10 expression levels in GC and normal tissues from The Cancer Genome Atlas (TCGA) were collected from and analyzed with the UALCAN database (https://ualcan.path.uab.edu/index.html).

Quantitative real‑time polymerase chain reaction (qRT‒PCR)

TRIzol reagent (Takara, Shiga, Japan) was used to isolate total RNA from cells and tissues according to the manufacturer’s instructions. A NanoDrop 2000 (Thermo Fisher Scientific, Waltham, MA, USA) was used to measure RNA concentrations. One microgram of RNA per sample was reverse transcribed using HyperScript III RT SuperMix for qPCR with gDNA Remover (EnzyArtisan Biotech, Shanghai, China). qRT‒PCR was performed using 2 × S6 Universal SYBR qPCR Mix (EnzyArtisan Biotech). Primer sequences used in qRT‒PCR were listed in Supplementary Table 1. Relative target gene expression was normalized to GAPDH expression. Each sample was analyzed in triplicate, and the relative mRNA levels were calculated using the 2^−ΔΔCt method.

Western blotting analysis

Tissues and cells were collected and lysed with RIPA lysis buffer (NCM Biotechnology, Jiangsu, China) supplemented with protease and phosphatase inhibitors (NCM Biotechnology) for 30 min at 4 °C. A BCA protein assay kit (Beyotime Biotechnology, Shanghai, China) was used to determine the protein concentrations. Equal amounts of proteins were separated by SDS‒PAGE and transferred to polyvinylidene fluoride (PVDF) membranes (Millipore, Billerica, MA, USA). The membranes were incubated with primary antibodies at 4 °C overnight after blocking with 5% skim milk for one hour at room temperature, and then, the membranes were incubated with a horseradish peroxidase (HRP)-conjugated anti-mouse or anti-rabbit secondary antibody (Cell Signaling Technology, Beverly, MA, USA). An enhanced chemiluminescence reagent (Millipore) was used to detect the protein bands. The primary antibodies used in Western blot were shown in Supplementary Table 2. All the blots were processed in parallel.

Immunohistochemical staining

For epitope retrieval, tissue sections were dewaxed, rehydrated, and heated with citrate buffer (0.01 M, pH 6.0). The tissue sections were incubated with anti-NAT10 (1:500; Abcam), anti-p-Akt (1:100; Santa Cruz), anti-TGF-β1 (1:100; Santa Cruz), and anti-TNC (1:400; Abcam) antibodies at 4 °C overnight after exposure to 3% hydrogen peroxide for 10 min to block endogenous peroxidase activity. Then, a GTVisionTM III Detection System/Mo & Rb (GeneTech, Shanghai, China) was used to detect the primary antibody after incubation for 30 min at room temperature. After washing, the slides were counterstained with hematoxylin and mounted with coverslips. The immunohistochemical score was calculated based on the staining intensity and the percentage of positive cells. In details, the proportion of positively stained cells was categorized into five levels (percentage scores): less than 10% (0), 10–25% (1), 25–50% (2), 50–75% (3), and over 75% (4). The depth of staining was graded into four levels (intensity scores): absence of staining (0), lightly brown (1), brown (2), and deeply brown (3). A positive NAT10 stain was calculated using the formula: (percentage score) × (intensity score) = total score. A total score ranging from 0 to 3 was considered negative expression; 4 to 6 was considered low expression; and 8 to 12 was considered high expression. Two pathologists independently evaluated all samples to minimize bias.

Immunofluorescence (IF) staining

Cells were seeded in confocal dishes (Thermo Fisher Scientific) for 24 h. Afterward, the cells were fixed with 4% paraformaldehyde and permeabilized with 0.1% Triton X-100 at room temperature. After blocking, specific primary antibodies were incubated with the cells at 4 °C at a suitable concentration overnight. Then, the cells were incubated with species-appropriate mouse or rabbit secondary antibodies conjugated with FITC dyes (Abcam) and DAPI (Dojindo, Kumamoto, Japan) at room temperature for one hour. Images were acquired with a Leica microscope (Wetzlar, Germany). The primary antibodies that were used were as follows: anti-NAT10 (1:2000; Abcam), anti-TNC (1:200; Abcam), anti-p-Akt (1:100; Santa Cruz), anti-Akt (1:100; Santa Cruz) and anti-TGF-β1 (1:100; Santa Cruz).

Cell culture, transfection, and generation of stable cell lines

The normal gastric epithelial cell line GES-1 and human GC cell lines (AGS, HGC-27, MKN28, MKN45, and NCI-N87) were preserved by the Department of Gastrointestinal Surgery of the First Affiliated Hospital of Fujian Medical University. The cell lines were cultured in RPMI-1640 medium supplemented with 10% FBS (Thermo Fisher Scientific, Waltham, MA, USA) and 1% penicillin–streptomycin (NCM Biotechnology) in a humidified atmosphere containing 5% CO2 at 37 °C.

The TNC overexpression plasmid and TNC siRNA were synthesized by Genomeditech (Shanghai, China). The sequence of siRNA used in transient transfection were listed in Supplementary Table 3. For transient transfection, proliferating cells were incubated in serum-free medium containing plasmid or siRNA and Lipofectamine 2000 (Thermo Fisher Scientific) in 6-well cell culture plates for 6 h. According to the product manual, taking a 6-well plate as an example, the transfection dose for the TNC overexpression plasmid or siTNC was 2 µg. Then, the cells were incubated in complete medium for 48 h for further experiments.

The NAT10 lentiviral vector and NAT10 shRNA lentiviral vector were obtained from Genomeditech. Lentiviruses were loaded onto GC cells, and then, the stably transfected cells were selected with blasticidin (InvivoGen, San Diego, CA, USA). The cell lines were divided into the following groups: mock group, control cells that were not transfected; Ctrl group, cells that were transfected with a lentiviral vector carrying the control fragment or the control shRNA; MKN28-NAT10-OV group, cells that were transfected with the lentivirus carrying the NAT10 fragment; and MKN45-sh-NAT10 group, cells that were transfected with NAT10 lenti-shRNA.

Cell growth assay

Cell Counting Kit-8 (CCK-8) assays were conducted to evaluate cellular proliferation in vitro. Briefly, MNK28 and MNK45 cells (5000 cells/well) were seeded in triplicate in 96-well cell culture plates and incubated for the indicated durations (0, 24, 48, or 72 h). Cell growth curves were generated by the CCK-8 assay (Dojindo), and the absorbance at 450 nm was measured on a Gen5 microplate reader (BioTek, Winooski, VT, USA). A Cell-Light EdU Apollo488 In Vitro Kit (RiboBio, Guangdong, China) was used to measure the DNA synthesis rate. Cells (1 × 104/well) were seeded in 96-well plates, cultured for 24 h, incubated with 50 µmol/L EdU solution for 2 h and fixed with 4% paraformaldehyde. Subsequently, the cells were stained with Apollo488 and Hoechst 33,342 solution. Cells were calculated by Image J. The ratio of EdU-positive cells to Hoechst-positive cells was determined to evaluate cell proliferation. These experiments were performed independently in triplicate.

Flow cytometry analysis of cell apoptosis

An Annexin V-FITC/PI Apoptosis Kit (MultiSciences, Zhejiang, China) was used to analyze cell apoptosis. Cells were collected and resuspended in 500 µL of 1 × binding buffer and then stained with 5 µL of Annexin V-FITC and 10 µL of PI for 5 min in the dark. Apoptotic cells were analyzed by flow cytometry (BD Biosciences, San Jose, CA, USA). The experiment was performed independently in triplicate.

Cell migration and invasion assay

A Transwell 24-well Boyden chamber (Corning, New York, NY, USA) with a polycarbonate membrane with 8.0-µm pores was used without (to assess migration) or with (to assess invasion) Matrigel (Corning). A total of 2.0 × 104 cells in 200 µL of serum-free medium were seeded in the upper chambers, while the lower chambers were filled with 600 µL of medium supplemented with 10% FBS as a chemoattractant. After 48 h, the cells that had migrated or invaded the membranes of the lower chambers were fixed with methyl alcohol and then stained with 0.1% crystal violet solution. These experiments were performed independently in triplicate.

Nude mouse subcutaneous xenograft models

All animal experiments were approved by the Animal Care Committee of the First Affiliated Hospital of Fujian Medical University and performed according to the animal welfare guidelines of the committee. This study was conducted in compliance with the Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines. Four-week-old male BALB/C nude mice were used to establish a subcutaneous xenograft model. After the mice were randomly divided into two groups (n = 5), 5 × 106 cells were injected into the subcutaneous tissues of the nude mice. The basic conditions of the nude mice, such as their food consumption and growth, were observed throughout the experiment. Three weeks later, all the mice were sacrificed by 2% pentobarbital sodium (200 mg/kg, intraperitoneal injection), and the tumors were excised. Tumor volume was calculated using the following formula: volume = width2 × length × 0.5. All the animal experiments were approved by the Animal Care Committee of the First Affiliated Hospital of Fujian Medical University.

Ac4C dot blotting

Total RNA was collected from GC tissues and paired adjacent normal tissues, denatured at 75 °C for five minutes, immediately cooled for one minute, and then added to Amersham Hybond-N+ membranes (Solarbio, Beijing, China). The membranes were cross-linked, blocked, and then incubated with an anti-ac4C antibody (1:2000, Abcam) at 4 °C overnight. After the membranes were washed and probed with an HRP-conjugated secondary anti-rabbit IgG at room temperature for 1 h, they were washed three times, and visualized with a chemiluminescent HRP substrate (Millipore). After exposure, the membranes were stained with 0.02% methylene blue staining buffer. Then, the input RNA was scanned to determine its total content.

Liquid chromatography-tandem mass spectrometry (LC‒MS/MS)

LC–MS/MS was conducted by MetWare Biotechnology (Wuhan, China) to quantify the corresponding ac4C levels. Briefly, buffer, S1 nuclease (Takara), alkaline phosphatase (Takara) and phosphodiesterase I (Sigma‒Aldrich, St. Louis, MO, USA) were added to one microgram of RNA, and the mixture was subsequently incubated at 37 °C. After the RNA was completely digested into nucleosides, the mixture was extracted with chloroform. The resulting aqueous layer was collected for analysis via LC‒MS/MS with the AB Sciex QTRAP 6500 LC‒MS/MS platform. The electrospray ionization (ESI) source operation parameters were as follows: ion source, ESI + ; source temperature, 550 °C; ion spray voltage (IS), 5500 V; and curtain gas (CUR), 35 psi. RNA modifications were analyzed using scheduled multiple reaction monitoring (MRM). Mass spectrometer parameters, including the declustering potential (DP) and collision energy (CE), for individual MRM transitions were determined with further DP and CE optimization. A specific set of MRM transitions was monitored for each period according to the metabolites that were eluted within this period.

Acetylated RNA immunoprecipitation (acRIP)

To measure the level of ac4C modification of a given gene, acRIP was performed. In brief, after mixing an anti-ac4C antibody (Abcam) with the beads overnight, the complex was incubated with the RNA samples. Then, the RNA was eluted from the beads. qRT‒PCR was performed as previously described. The acRIP-seq analysis of GC tissues and paired adjacent normal tissues was performed by NEWCORE BIOTECH (Shanghai, China).

RNA stability assay

Cells were treated with 200 μM actinomycin D (MedChemExpress, Monmouth Junction, NJ, USA) for 0, 3, 6, 9, or 12 h in 6-well plates. The half-life of TNC mRNA was predicted by analyzing the total RNA that was collected with the method described above for qRT‒PCR analysis. The half-life (t_1/2) of mRNA was measured from the initial and final amounts of RNA:

$$K = \frac{{\ln C_{0} - \ln C_{12} }}{t12 - t0};t_{{{1}/{2}}} = 0.{693}/K.$$

Translation efficiency analysis

Cells were incubated with 1 μg/mL puromycin (InvivoGen) for 0, 15, 30, 60, or 120 min. Afterward, the efficiency of TNC translation was examined by Western blotting analysis.

Statistical analysis

SPSS 22.0 statistical software (SPSS, Chicago, IL, USA) was used to analyze the data. Pearson’s correlation was used to predict the relationships between variables. Student’s test and one-way analysis of variance test were used to compare data between the different groups. In addition, overall survival (OS) was estimated by Kaplan–Meier analysis. The log-rank test was employed to assess differences among the survival curves. The chi-square test was used to determine associations between the clinicopathological characteristics of GC patients and the expression of NAT10. The Cox proportional hazards model was used to adjust for possible confounding variables, investigate independent prognostic factors and multivariate analysis. For all the analyses, p < 0.05 was considered to indicate statistical significance.

Results

Analysis of NAT10 expression and its prognostic value in GC

To compare the mRNA expression level of NAT10 in GC tissues and adjacent normal tissues, we collected 30 pairs of clinical samples for qRT‒PCR analysis. The results indicated a significantly higher level of NAT10 mRNA expression in GC tissues than in paired adjacent normal tissues (27/30, Fig. 1a). Consistently, this result was further supported by Western blotting and immunohistochemical analysis (Fig. 1c- e). Additionally, analysis using the TCGA database on the UALCAN platform demonstrated increased NAT10 expression in GC tissues compared with normal tissues (Fig. 1f). Then, Kaplan‒Meier survival curves revealed significantly lower survival of patients with higher NAT10 expression than of those with lower NAT10 expression (Fig. 1g). Subsequently, analysis of the relationship between the expression of NAT10 and the clinicopathological features of patients with GC showed that the expression of NAT10 was correlated with the T stage and UICC stage (p < 0.05) (Table 1). Univariate analysis indicated that T stage, N stage, M stage, UICC stage, differentiation degree and NAT10 expression were prognostic factors for OS of patients with GC. Multivariate analysis revealed that NAT10 was an independent risk factor for OS of GC patients (Table 2).

Fig. 1

NAT10 expression was upregulated and positively correlated with TNC expression in GC tissues. a The mRNA expression of NAT10 and TNC in GC tissues and paired adjacent normal tissues was measured by qRT‒PCR, which showed that higher NAT10 and TNC mRNA expression in GC tissues compared with paired normal tissues. b The relationship between NAT10 and TNC mRNA expression was analyzed by Pearson’s correlation based on qRT‒PCR results, a positive correlation was verified. c The protein expression of NAT10 and TNC in representative paired samples of GC tissues and adjacent normal tissues was measured by Western blotting analysis, which showed that higher NAT10 and TNC protein expression in GC tissues compared with paired normal tissues. d Representative images of immunohistochemical staining for NAT10 obtained from GC tissue microarray with varying degrees of differentiation. Original magnification, 40× (200× for insert images). e Comparison of immunohistochemical scores between GC tissues and normal tissues. f The expression of NAT10 in GC tissues and normal tissues from the UALCAN platform was analyzed. g OS of GC patients with different NAT10 expression levels were analyzed by Kaplan‒Meier analysis, higher NAT10 expression led to lower survival

Table 1 Relationship between NAT10 expression levels and clinicopathologic characteristics of GC patients (n = 158)

Table 2 Univariate and multifactorial analysis of OS

NAT10 promoted the malignant biological behaviors of GC cells in vitro

qRT‒PCR and Western blotting analysis were performed to measure the expression of NAT10 in various GC cell lines and the normal gastric epithelial cell line GES-1 (Fig. 2a, b). To explore the potential function of NAT10 in GC cells, we selected MKN28 and MKN45 to establish stable NAT10-overexpressing and NAT10-knockdown cell lines (Fig. 2c). Subsequently, CCK-8 assay results suggested that increased expression of NAT10 promoted GC cell growth, while NAT10 knockdown decreased GC cell proliferation (Fig. 2d). The results were verified by EdU assay (Fig. 2e). Then, flow cytometry analysis suggested that increased NAT10 expression was associated with decreased percentages of apoptotic cells (Fig. 2f). Transwell assays demonstrated that NAT10 overexpression promoted the migration and invasion of GC cells, while NAT10 knockdown had the opposite effect (Fig. 2g).

Fig. 2

The effects of NAT10 on malignant phenotypes in GC cells. a The mRNA expression of NAT10 in GC cells and GES-1 cells was measured by qRT‒PCR. b The protein expression of NAT10 in GC cells and GES-1 cells was measured by Western blotting analysis, MKN28 has the lowest NAT10 expression compared with other GC cells and MKN45 has the highest NAT10 expression compared with other GC cells. c MKN28 and MKN45 were selected to construct stable NAT10-overexpressing and NAT10-knockdown cell lines, the protein expression of NAT10 in NAT10-overexpressing or NAT10-knockdown stable cell lines was verified by Western blotting analysis. d CCK-8 assay showed a promotion or inhibition of cell proliferation in NAT10-overexpressing or NAT10-konckdown GC cells, which could be partly reversed by transfection with siTNC or a TNC overexpression plasmid. e EdU staining assay showed a promotion or inhibition of DNA synthesis in NAT10-overexpressing or NAT10-konckdown GC cells, which could be partly reversed by transfection with siTNC or a TNC overexpression plasmid. f Flow cytometry analysis showed a lower or higher apoptosis ratio in NAT10-overexpressing or NAT10-konckdown GC cells, which could be partly reversed by transfection with siTNC or a TNC overexpression plasmid. g Transwell assay showed a promotion or suppression of cell migration and invasion in NAT10-overexpressing or NAT10-konckdown GC cells, which could be partly reversed by transfection with siTNC or a TNC overexpression plasmid. (*p < 0.05, **p < 0.01, ***p < 0.001)

NAT10 promoted GC cell proliferation in vivo

To elucidate the role of NAT10 in promoting GC cell proliferation within a biological system, we established a subcutaneous xenograft model in nude mice. The results indicated that the size of the xenograft tumors in the MKN45-shNAT10 group was significantly reduced compared with that in the MKN45-Mock group (Fig. 3a). Immunohistochemical analysis and IHC score revealed that the NAT10 staining intensity in the MKN45-shNAT10 group was notably lower than that in the MKN45-Mock group (Fig. 3b, c). These observations suggested that NAT10 exerted a stimulatory effect on the proliferation of GC cells in vitro and in vivo.

Fig. 3

Knockdown of NAT10 inhibited the tumorigenicity of GC cells in nude mice. a Xenografts of subcutaneous tumors in nude mice, NAT10 knockdown reduced the size of the xenograft tumors. b Representative immunohistochemical staining for NAT10, TNC, p-Akt and TGF-β1 in subcutaneous tumors, NAT10 knockdown led to lower expression of above markers. Original magnification: 100×. c Comparison of immunohistochemical scores in subcutaneous tumors generated by MKN45-Ctrl cells and MKN45-sh-NAT10 cells. (**p < 0.01, ***p < 0.001)

ac4C modification of TNC was affected by NAT10

To further explore the underlying mechanism by which NAT10 promoted GC process, the overall level of ac4C modification was analyzed in GC tissues and paired adjacent tissues using ac4C dot blotting and LC‒MS according to previous research [15]. The experimental results revealed a significant increase in the overall level of ac4C modification in GC tissues compared with adjacent tissues (Fig. 4a, b). Then, three pairs of GC tissues and paired adjacent tissues were subjected to acRIP-seq analysis. The results suggested that the ac4C modification peak in GC tissues and paired adjacent tissues was mainly distributed in the CDS region (45.3% vs. 45.8%; Fig. 4c, d). Consistent with previous studies, "CXXCXXCXX", which is the most prevalent motif for ac4C modification [16], was enriched among the peak sequences and was observed in both GC tissues and adjacent normal tissues (Fig. 4e). Volcano plot analysis demonstrated alterations in the ac4c modification levels of different genes between GC tissue samples and matched adjacent tissue samples (Fig. 4f). Subsequently, through the combined analysis of acRIP-seq results and input transcript sequencing results, we found that the PI3K-Akt, MAPK, TGF-β and other signaling pathways, which are essential for tumor development, were enriched (Fig. 4g, h). A total of 11 genes had increased levels of ac4C modification and were upregulated in GC tissues according to acRIP-seq combined with input transcript sequencing (Fig. 4i). Here, we focused on TNC as a crucial oncogene. Previous studies have reported that TNC expression is increased in tumor tissues [11]. As shown in Fig. 4j, the level of ac4C modification of TNC was higher in GC tissues than in normal tissues. Moreover, qRT‒PCR revealed that the expression of TNC was increased in 83.33% (25/30) of the GC tissues compared with the adjacent normal tissues (Fig. 1a), which was confirmed by Western blot (Fig. 1c). Additionally, Pearson’s correlation analysis revealed a significant positive correlation between the expression levels of NAT10 and TNC in GC tissues (r = 0.626, p < 0.001; Fig. 1b). In summary, TNC might be an important downstream target that is involved in the promotion and maintenance of malignant biological behaviors initiated by NAT10 in GC.

Fig. 4

acRIP-seq revealed the role of NAT10 in catalyzing the ac4C modification of TNC in GC. a, b The total ac4C levels in paired samples of GC tissues and paired adjacent normal tissues were determined by dot blotting and LC‒MS/MS, NAT10 overexpression increased the overall level of ac4C modification in GC tissues compared with adjacent tissues. c, d The distribution of ac4C-containing peaks across mRNA in GC tissues and paired adjacent normal tissues, the CDS region occupied the largest proportion. e Consensus motifs of GC tissues and paired adjacent normal tissues. f Volcano plots of mRNA in paired samples segregated by ac4C modification levels. g Representative gene ontology terms showing biological processes, cellular components, molecular functions and KEGG pathways of genes that were significantly enriched by acRIP-Seq and input transcript. h Joint analysis of KEGG pathways that were enriched by acRIP-Seq and input transcript, including the PI3K-Akt, MAPK, TGF-β and other signaling pathways. i Four-quadrant diagram showing related genes with different mRNA expression and ac4C modification levels in paired samples analyzed by acRIP- seq and input transcript, TNC had increased levels of ac4C modification and upregulated expression. j ac4C modification distributions across TNC analyzed by acRIP-seq. (***p < 0.001)

ac4C modification affected mRNA stability and translation efficiency of TNC in GC cells

To further validate our acRIP-seq results, we carried out acRIP-qPCR experiments, and we found that the level of ac4C modification of TNC was significantly increased in NAT10-overexpressing GC cells (Fig. 5a). Afterward, RNA stability assay demonstrated that overexpression of NAT10 significantly prolonged the half-life of TNC mRNA, increasing its stability (Fig. 5b). Furthermore, a puromycin incorporation assay suggested that knockdown of NAT10 significantly inhibited the translation efficiency of TNC (Fig. 5c).

Fig. 5

NAT10 increased the mRNA stability and translation efficiency of TNC by catalyzing ac4C modification. a Changes in the levels of TNC ac4C modification of NAT10-overexpressing or control GC cells were verified by acRIP-qPCR, NAT10 overexpression had a higher level of ac4C modification of TNC. b The mRNA expression levels of TNC after treatment with actinomycin D (0, 3, 6, 9 and 12 h) were quantified to analyze TNC mRNA stability, the mRNA stability of TNC was reduced by NAT10 knockdown. c Representative images of puromycin incorporation assay to determine translation efficiency by Western blotting analysis, the translation of TNC was decreased by NAT10 knockdown. (*p < 0.05)

TNC significantly enhanced the growth, migration and invasion of GC cells

To explore the effect of TNC on the biological characteristics of GC cells, we constructed a TNC overexpression plasmid and TNC-specific siRNA to increase and inhibit the expression of TNC in GC cells, respectively (Fig. 6a). The CCK-8 assay results showed that TNC upregulation or downregulation promoted or suppressed the proliferation of GC cells, respectively (Fig. 6b). Transwell assays showed that migration and invasion were enhanced by increased TNC expression, while downregulation of TNC expression by siRNA had the opposite effect (Fig. 6c).

Fig. 6

TNC promoted GC cell proliferation, migration and invasion. a MKN28 cells were transfected with TNC overexpression plasmid, and MKN45 cells were transfected with siTNC-1 and siTNC-2. b Cell proliferation ability was detected by CCK-8 assay, which showed a promotion or inhibition of cell proliferation after TNC upregulation or downregulation. c Migration and invasion abilities were analyzed by Transwell assay, which showed a promotion or inhibition of cell migration and invasion after TNC upregulation or downregulation. (***p < 0.001)

siTNC partially reversed NAT10-induced malignant phenotypes

Rescue experiments were performed to further investigate the relationship between NAT10 and TNC. The results of CCK-8, EdU, flow cytometry, Transwell invasion and migration assays showed that the increased proliferation, decreased apoptosis, and increased migration and invasion of NAT10-overexpressing GC cells were partially impaired after transfection with siTNC. After transfection with the TNC overexpression plasmid, the decreased proliferation, increased apoptosis, and decreased migration and invasion in NAT10-knockdown GC cells were partially reversed (Fig. 2d–g).

NAT10 promoted the progression of GC by regulating TNC to activate the TNC/Akt/TGF-β1 positive feedback loop

Western blotting analysis indicated that the protein expression of the apoptosis-related protein Bcl2 was increased after NAT10 overexpression, while the protein expression levels of Bax, Cleaved Caspase-9, Cleaved Caspase-3 were decreased; the opposite results were observed after NAT10 knockdown (Fig. 7a). Moreover, we measured the expression of epithelial mesenchymal transition (EMT)-related proteins. As shown in Fig. 7b, NAT10 overexpression increased the protein expression of N-cadherin and Vimentin and decreased the protein expression of E-cadherin. The opposite effects were observed after NAT10 knockdown. These results further indicated that NAT10 inhibited GC cell apoptosis and promoted EMT in GC.

Fig. 7

NAT10 activated the TNC/Akt/TGF-β1positive feedback loop to drive the malignant phenotypes of GC cells. a The protein expression of apoptosis-related protein markers in stable NAT10-overexpressing or NAT10-knockdown cell lines was measured by Western blotting analysis. NAT10 overexpression resulted in increased expression of Bcl2 and decreased expression of Bax, Cleaved Caspase-9 and Cleaved Caspase-3, which could be partly reversed by siTNC, while NAT10 knockdown caused the opposite alteration. b The protein expression of EMT-related protein markers in stable NAT10-overexpressing or NAT10-knockdown cell lines was measured by Western blotting analysis. NAT10 overexpression resulted in increased expression of N-cadherin and Vimentin and decreased expression of E-cadherin, which could be partly reversed by siTNC, while NAT10 knockdown caused the opposite alteration. c Representative images of immunofluorescence staining for the indicated markers. NAT10 overexpression led to increased expression of TNC, p-Akt and TGF-β1, while NAT10 knockdown caused the opposite alteration. Original magnification: 400×. d Stable NAT10-overexpressing cells were treated with siTNC or LY294002, and stable NAT10-knockdown cells were treated with TNC overexpression plasmid, followed by Western blotting analysis of the expression of the indicated markers. siTNC or LY294002 partly reversed the upregulated expression of TNC, p-Akt and TGF-β1 caused by NAT10 overexpression, while TNC overexpression plasmid partly reversed the downregulated expression of TNC, p-Akt and TGF-β1 caused by NAT10 knockdown. e MKN28 cells were treated with TGF-β1, and MKN45 cells were treated with P144, followed by Western blotting analysis of the expression of the indicated markers. TGF-β1 could upregulate the expression of TNC and p-Akt, while P144 caused the opposite alteration

The aforementioned immunohistochemical staining of subcutaneous nodules from nude mice showed that the TNC, p-Akt and TGF-β1 levels were lower in the MKN45-sh-NAT10 group than in the control group (Fig. 3b). Subsequently, immunofluorescence staining showed that NAT10 upregulation led to an increase in TNC and TGF-β1 expression and notably increased Akt phosphorylation, whereas NAT10 downregulation had the opposite effects (Fig. 7c). These results were confirmed by Western blotting analysis. Moreover, we found that these NAT10 overexpression-induced changes were partially reversed by siTNC or LY294002 (an Akt inhibitor); while in NAT10-knockdown cells that were transfected with the TNC overexpression plasmid, opposite changes were observed (Fig. 7d). Further experiments revealed that when MKN28 cells were treated with exogenous TGF-β1, the expression of TNC was increased, and Akt phosphorylation was increased. Conversely, the treatment of MKN45 cells with P144 (a TGF-β1 inhibitor) decreased TNC expression and reduced Akt phosphorylation (Fig. 7e).

Taken together, these results suggested that NAT10 might stimulate a positive feedback loop that involved TNC, Akt, and TGF-β1 by modulating TNC expression in GC cells, which in turn could facilitate the progression of GC.

Discussion

Ac4C, which is an evolutionarily conserved posttranscriptional modification that predominantly occurs in tRNA and rRNA [17], has been implicated in a variety of pathological conditions [4, 18,19,20]. NAT10 is an enzyme that is pivotal for the ac4C modification of mRNA, and it or its homologs can catalyze ac4C production [21]. The association of NAT10 with neoplastic progression is particularly noteworthy. It has been reported that NAT10-mediated ac4C modification enhances the mRNA stability and translation efficiency of KIF23 mRNA, which activates the Wnt/β-catenin signaling pathway to promote colorectal cancer progression [16]. As for non-digestive tract tumors, provious study has reported that the ac4C modification mediated by NAT10 significantly increases the stability and translation efficiency of HNRNPUL1 mRNA, resulting in elevated HNRNPUL1 to promote the proliferation, migration, and invasion of cervical cancer cells [22]. The above-mentioned research indicates that NAT10 promotes tumor progression by enhancing the mRNA stability and translation efficiency of downstream target genes.

In the current study, we conducted relevant experiments to investigate the relationship between NAT10 expression and GC progression and found that the expression of NAT10 was greater in GC tissues than in paired adjacent normal tissues. NAT10 expression was significantly correlated with adverse GC prognosis. In vitro and in vivo assays confirmed that increased NAT10 expression was associated with the malignant biological behaviors of GC cells.

On the basis of these findings, our current investigation explored the underlying mechanism by which NAT10 influenced GC progression. By employing dot blotting and LC‒MS/MS, we further confirmed that ac4C levels were increased in GC tissues compared with paired adjacent normal tissues. Hence, we conducted acRIP-seq and input sequencing and revealed that ac4C modifications were predominantly enriched within CDS regions, with the "CXXCXXCXX" motif emerging as the most commonly modified motif. KEGG pathway enrichment analysis highlighted the enrichment of signaling pathways that were pivotal to oncogenesis and tumor progression, such as the PI3K-Akt, MAPK, and TGF-β pathways. Most importantly, we focused on TNC, whose ac4C modification level was increased and whose expression was upregulated. Previous studies have indicated that increased TNC expression can promote the proliferative and metastatic capabilities of various cancers [23,24,25], which was confirmed by our subsequent experimental findings. Meanwhile, TNC is closely related to the activation of various tumor-associated signaling pathways, indicating that TNC plays an important role in the occurrence and development of tumors [26]. These findings, along with our experimental results, indicated that TNC could promote the development of GC.

Drawing upon these findings, we hypothesized that TNC might serve as a critical downstream effector that was involved in the exacerbation and maintenance of the malignant phenotypes of GC cells in the context of NAT10 overexpression.

To test this hypothesis, we initially compared TNC levels between GC and normal tissues, and the results showed that TNC expression was increased in GC tissues. Pearson’s correlation analysis revealed a significant positive correlation between the expression levels of NAT10 and TNC in GC tissues. Afterward, acRIP-qPCR confirmed that NAT10 overexpression increased the level of ac4C modification of TNC in GC cells. RNA stability assays and translation efficiency analyses revealed increased RNA stability and translational efficiency after NAT10 overexpression. Western blotting analysis and qRT‒PCR further revealed that TNC expression was markedly upregulated upon NAT10 overexpression and downregulated upon NAT10 knockdown.

Integrating these results, we proposed that NAT10 modulated the expression of TNC, thereby affecting the activity of downstream signaling pathways. Consistent with our acRIP-Seq analysis, previous studies have shown that TNC can activate the PI3K-Akt signaling pathway via Akt phosphorylation to promote angiogenesis and migration in glioma [27]. Moreover, the PI3K-Akt pathway is the upstream activator of the nuclear transcription factor-activating protein complex AP-1 (c-Fos/c-Jun), and AP-1 can transcriptionally increase TGF-β1 expression by binding to the TGF-β1 promoter region [28]. It has also been reported that TGF-β1 can promote TNC expression through a transcriptionally active complex formed by Smad3, Sp1 and Ets1 [29]. These studies indicated that there might be a TNC/Akt/TGF-β1 positive feedback loop in the TNC regulatory network. Subsequent experiments revealed that the overexpression of NAT10 led to the upregulation of TNC and TGF-β1 expression as well as an increase in Akt phosphorylation, while the knockdown of NAT10 exerted the opposite effect. Moreover, treatment of MKN28 cells with TGF-β1 increased TNC expression and Akt phosphorylation, while treatment of MKN45 cells with a TGF-β1 inhibitor decreased TNC expression and Akt phosphorylation; these results suggested that TGF-β1 could regulate TNC expression and Akt phosphorylation. Collectively, these findings suggest the presence of a positive feedback loop involving TNC/Akt/TGF-β1. Furthermore, rescue experiments demonstrated that the positive or negative regulation of GC cells induced by NAT10 overexpression or knockdown could be partly reversed by siTNC or TNC overexpression plasmids. siTNC or TNC overexpression plasmids partially counteracted the changes in TGF-β1 expression and Akt phosphorylation that were induced by NAT10 overexpression or knockdown. Therefore, we hypothesized that NAT10 activated the TNC/Akt/TGF-β1 positive feedback loop, promoting the malignant phenotypes of GC cells.

Collectively, our data suggested that the aberrant overexpression of NAT10 in GC increased the ac4C modification of TNC, increased the RNA stability and translational efficiency of TNC, and increased the expression levels of TNC, thereby activating the TNC/Akt/TGF-β1 positive feedback loop to drive GC progression (Fig. 8).

Fig. 8

A possible mechanism was proposed. Abnormally high expression of NAT10 in GC, which catalyzed the ac4C modification of TNC, could activate the TNC/Akt/TGF-β1 positive feedback loop to promote the progression of GC

It’s essential to recognize certain limitations in our research. Our experiments are currently limited to the cellular and animal levels, with only a small sample of clinical validation conducted. In the future, we will expand the sample size to provide more valuable theoretical foundations for the diagnosis and treatment of GC. Meanwhile, the upstream regulatory network of NAT10 is also one of the targets for our future research endeavors.

Conclusions

Overall, this study preliminarily revealed that NAT10 promoted GC progression by increasing the ac4C modification of TNC and thereby activating the TNC/Akt/TGF-β1 positive feedback loop. Our findings suggested that NAT10 and TNC might be prospective targets for novel strategies for treating GC.