Culture of AML12 and HepG2 c ells and PA or Remodelin treatment

The AML12 and HepG2 cell were purchased from China National Collection of Authenticated Cell Cultures. AML12 cells were maintained in Gibco’s DMEM/F-12 (Shanghai, China) media enriched with 10% fetal bovine serum (FBS, Sigma, F8687, Australia), ITS Liquid Media Supplement (Sigma, I3146, MO, USA) and 40 ng/ml Dexamethasone (Sigma, D4902, Shanghai, China). In addition, AML12 cells were treated with 150 µmol/L PA (Sigma, P0500, Shanghai, China) or 20 µmol/L Remodelin (Beyotime, SD1168, Shanghai, China) for 48 h. HepG2 cells were cultured in Gibco’s DMEM (Shanghai, China) media containing 10% FBS.

Real-time quantitative reverse transcription PCR (RT-qPCR)

RNAs were isolated from AML12 cells or liver tissues using Triquick reagent (Solarbio, R1100, Beijing, China), and subsequently were reverse transcribed to cDNA (Solarbio, RP1100, Beijing, China). qPCR with 2×SYBR Green PCR Mastermix (Solarbio, SR1110, Beijing, China) was used to detect relative mRNA expression, normalized to Gapdh expression. Primer sequences are provided in Supplementary Table 1.

Oil red O staining and FFA quantification

After AML12 cells were incubated with 150 µmol/L PA for 48 h, the cells were first washed with PBS and fixed with 4% paraformaldehyde, followed by multiple washes with water. Next, the cells were then washed with water for 3 times, incubated with diluted Oil red O (Beyotime, C0157S, Shanghai, China) for 10 min, and were washed with water again. The photographs were taken with microscope (Leica DM IL LED Fluo, Wetzlar, Germany). The liver tissue slices were added to diluted Oil red O, and incubated for 20 min, decolorized with 70% alcohol, and then washed with water again. Next, the liver slices were further stained using hematoxylin. The cellular FFA content were detected with FFA content assay kit (Solarbio, BC0590, Beijing, China).

Immunoblotting

The AML12 cells or liver tissue protein samples were extracted with ProteoPrep® total protein isolation kit (Sigma, MO, USA), and the content was detected using the BCA quantitative analysis kit (Sigma, Shanghai, China). Subsequently, approximate 100 µg protein was separated by SDS-PAGE (70 V, 30 min then 120 V, 90 min) and transferred (300 mA, 60 min) and transferred to PVDF membranes (Millipore, MA, USA). The membrane was then blocked with 5% non-fat milk for 2 h, and incubated in diluted primary antibody including NAT10 (13365-1-AP), SREBP-1 (14088-1-AP), ACC1 (21923-1-AP), FASN (10624-2-AP) and β-actin (66009-1-Ig) [all provided by Proteintech, Wuhan, China], SCAP (Abcam, ab308060, MA, USA) and Cd36 (Abcam, ab255331, MA, USA). The next day, the membranes were incubated with HRP-labeled secondary antibodies for 1 h, and after washed for 3 times, chemiluminescence reaction was performed by using BeyoECL plus kit (Beyotime, P0018S, Beijing, China). Images were captured using the GelDoc XR Biorad system (Bio-Rad, CA, USA) for further analysis.

mRNA ac4C dot blot

The mRNA was isolated from about 10 µg of total RNA using Dynabeads™ mRNA purification kit (Invitrogen, CA, USA), and mRNA was denatured at 95 °C for 5 min, and was placed at ice immediately for 1 min. Next, the mRNA was added to PVDF membranes (Millipore, MA, USA), the membrane was cross-linked for 3 times at ultraviolet 254 nm, and subsequently were blocked with 5% non-fat milk containing 0.1% PBST. The membrane was incubated with anti-ac4C antibody (Abcam, ab252215, MA, USA) overnight at 4 °C. The membrane was washed for 3 times and incubated using an anti-rabbit IgG-HRP (Sigma, MO, USA) for 1.5 h at room temperature, and then washed for 3 times. The membrane was incubated with ECL (Beyotime, P0018S, Shanghai, China) and 0.02% methylene blue (Solarbio, M8030, Beijing, China) was used to stain membrane. After that, membrane was washed with RNase-free water (Solarbio, R1600, Beijing, China) for 1.5 h, and the membrane was photographed with ChemiDoc imaging system (Bio-Rad, CA, USA).

Hepatocyte-targeted mouse adeno-associated virus (AAV) shRNA-NAT10 plasmid construction

The overexpressed NAT10 (OE-NAT10, NM_153126. 4), OE-Srebf1, OE-Scap and overexpressed negative control plasmid (OE-NC), siRNA NAT10 (siRNA-NAT10) and the siRNA negative control plasmid (siRNA-NC) was constructed by Shanghai Sanggong Biotechnology Co., Ltd, China. Hepatocyte-targeted AAV-shRNA-NAT10 and AAV-shRNA-NC plasmids were designed and constructed by Shanghai Gene Chemistry Co., Ltd, China. The NAT10 target sequence was 5’-GCTGGATTTGTTCCTGTCTAT-3’, and element sequences was pAAV-ApoE/hAATp-EGFP-MIR155(MCS)-SV40 PolyA. The cells were seeded in six-well palate and when cells growth to 80–90% confluence and then were transfected by 50nM siRNA-NAT10 and siRNA-NC.

Cell transfection

When AML12 cells were grown in approximately 80% confluency, the cells were transfected with OE-NAT10, OE-Srebf1, OE-Scap, OE-NC, siRNA-NAT10 and siRNA-NC with Lipo 5000 (Solarbio, L3200, Beijing, China) according to operation manual. After 48 h, the cells were treated with or without 150 µmol/L PA.

ac4C-specific RNA immunoprecipitation (RIP) PCR (acRIP-PCR) and NAT10-RIP-PCR

After AML12 cells transfected with siRNA-NAT10 or siRNA-NC for 48 h, the total RNA was isolated and mRNA was purified using a mRNA purification kit (Invitrogen, CA, USA). The acRIP (Millipore, 17–700, MA, USA) was performed. Briefly, the mRNA was incubated with anti-ac4C antibody (Abcam, ab252215, MA, USA) and anti-IgG antibody (Abcam, ab172730, MA, USA) using Protein G magnetic beads. The ac4C modified mRNA was collected and further were amplified using RT-qPCR. NAT10-RIP-PCR method was performed according to acRIP-PCR. The Srebf1 and Scap primers were designed according to ac4C modification sites predicted by PACES software (http://www.rnanut.net/paces/). The primer sequences are listed in Supplementary Table 1.

RNA decay assay

After AML12 cells were transfected with siRNA-NAT10 and siRNA-NC for 48 h, the two group of cells were harvested after 5 µg/mL Actinomycin D (Selleck, S8964, TX, USA) treatment for 0, 2, 4, 6 h. The total RNA was extracted and reverse transcribed, and then the Srebf1 and Scap expression were amplified using qPCR.

Dual-luciferase reporter assay

According to Srebf1 and Scap mRNA ac4C motif site predicted by PACES software, the pGL3-basic-Srebf1-wild type (Srebf1-WT), pGL3-basic-Scap-wild type (Scap-WT), pGL3-basic-Srebf1-mutation (Srebf1-MUT, mutation sequence: CTGCTGGCAAGTCTC) and pGL3-basic-Scap-mutation (Scap-MUT, mutation sequence: CGACCGAATACCCGA) plasmid were constructed. Approximately 1 µg plasmid was transfected into HEK293T and evaluation of firefly luciferase and renilla luciferase was carried out using a dual-luciferase reporter assay kit (Solarbio, D0011, Beijing, China).

In vivo study

To explore NAT10 and mRNA ac4C expression in liver tissue following a high-fat diet in mice, 6-weeks-old male mice were fed by chow diet (‌D12450J)‌ (n = 5) and high-fat diet (D12492), all feed were provided by Research Diets, NJ, USA (n = 5) for 12 weeks, the liver tissue were collected at 12th week. In addition, for AAV mediated NAT10 hepatocyte-targeted silencing experiment, 6-weeks-old male C57BL/6J mice were injected with hepatocyte-targeted AAV-shRNA-NAT10 (4 × 1011v.g, n = 5) or AAV-shRNA-NC (4 × 1011v.g, n = 5) through tail vein injection, and then subsequently were fed with a high-fat diet for 12 weeks. Additionally, two group of mice were injected again using same dose at 4th week. Furthermore, for Remodelin gavage experiment, male 6-weeks-old C57BL/6J mice were gavage with DMSO (n = 5) and Remodelin (100 mg/kg/day, n = 5), respectively, and two groups also were fed with high-fat diet for 12 weeks. All animal procedures were conducted in accordance with ethical guidelines approved by the Animal Ethics Committee of The Third Xiangya Hospital of Central South University, Changsha, China, (Approval NO: 2023-S027).

Hematoxylin and eosin (H&E) staining

The liver tissues were fixed with 4% paraformaldehyde for overnight at 4 °C, and the liver tissue were paraffin imbedded and cut into 5µM sections. After deparaffinized and rehydrated, the specimens were stained with H&E staining kit (Solarbio, G1120, Beijing, China). Microscopic images were captured using a Leica DM IL LED Fluo microscope (Wetzlar, Germany).

Biochemical analysis

The alanine aminotransferase (ALT) and aspartate aminotransferase (AST) activity were assayed by corresponding assay kits (BC1555 and BC1565, Solarbio, Beijing, China) along with the triglyceride (TG) and cholesterol (TC) assay kits (BC0625 and BC1985, Solarbio, Beijing, China). 100 mg of liver tissue were added 1mL of a mixture containing n-heptane and isopropyl alcohol at a volumetric ratio of 1:1, and were centrifuged at 10,000 rmp at 4℃. The supernatant was used to detect liver TG content.

Glucose tolerance test (GTT) and insulin tolerance test (ITT)

For AAV mediated NAT10 hepatocytes targeted silencing experiment and Remodelin gavage experiment, after 12 weeks, mice were subjected to a fasting period of 12 h, the mice were then intraperitoneally injected with 1.5 g/kg of glucose and their blood glucose levels were monitored at various time points − 0, 30, 60, 90, and 120 min post-injection. In addition, for ITT, after mice were fasted for 12 h, they were injected with 1 IU/kg of insulin (Novo Nordisk, Copenhagen, Denmark), the blood glucose levels were measured at the same intervals as the GTT experiment. Area under the curve (AUC) for GTT and ITT were calculated by Graphpad Prism 10 software.

Statistics analysis

The data analysis was conducted using GraphPad Prism 10 software. A Student’s t-test was used to compare data from two groups, while one-way analysis of variance (ANOVA) was used to analyze and compare multiple groups. The significance level was set at P < 0.05 to determine statistically significant differences among the groups.