Tissue samples from HNSCC patients

All tissue specimens cited in this research were gathered from The First Affiliated Hospital of Sun Yat-sen University between 2016 and 2019. The collection comprised 100 HNSCC specimens and 90 paracancerous tissues located 2 cm away from HNSCC tissue. Patients who had undergone chemotherapy or radiotherapy prior to surgery were not included in the research. Before commencing the sample collection procedures, informed consent was meticulously obtained from each patient. Overall survival was computed based on the time span from the initial diagnosis until most recent follow-up in November 2023 or the time of decease. This research received authorization from the ethics board of the medical facility ([2016]074 and [2022]229).

Cell culture and transfection

HNSCC cell line HSC3 was sourced from the Japanese Collection of Research Bioresource cell bank, and HNSCC cell lines SCC9 and SCC25 were obtained from ATCC. Each cell line underwent authentication via short tandem repeat (STR) profiling.

FTO or ATIP1 short hairpin RNA (shRNA) and FTO cDNA (ID: NM_001080432) lentiviral vectors, and their respective negative control (NC) sequences were obtained from Jikai Biotechnology Co., Ltd. (Shanghai, China). MTUS1/ATIP1 cDNA (ID: NM_020749.4) and corresponding NC lentiviral vectors were obtained from iGene Co., Ltd. (Guangzhou, China). To determine the most effective shRNA sequences for knocking down the expression of human ATIP1 or FTO, three different shRNA sequences were evaluated. Following a thorough assessment, the most potent knockdown sequence for each gene was selected for subsequent studies. Lentiviruses were used to transduce cells at an MOI of 25. The Supplementary Table S1 contains the specific shRNA sequences for FTO or MTUS1/ATIP1.

Quantitative real-time PCR (qRT‒PCR)

TRIzol (ThermoFisher, 15596-018) was utilized for the isolation of RNA from tissues and cells associated with HNSCC. Following this, Takara RT kit (Takara, RR047A) was used for the reverse transcription of mRNA. This method enables the synthesis of complementary DNA from mRNA templates. The SYBR-green PCR kit (TOYOBO, QRT-201) was employed for RT-PCR analysis. The relative expression levels of all mRNAs (primers are detailed in Supplementary Table S2) were determined utilizing the ΔΔCt method.

Western blotting

Ice-cold RIPA buffer (CST, 9806 S) was used to harvested total cellular protein from living cells. Cellular proteins were loaded onto pre-configured SDS-PAGE gels of varying concentrations for separation, followed by transfer onto PVDF membranes. Following a blocking step at RT using 3% BSA (Aladdin, B265994), the membranes were kept overnight with 1st antibodies, and subsequently exposed to 2nd antibodies. Subsequently, the membranes were rinsed 3 × 5 min in TBST and identified through enhanced chemiluminescence (ECL; ThermoFisher, 34580), followed by visualization using the ProteinSimple imaging system (Bio-Techne, Minneapolis, MN). The primary antibodies used in this research include ATIP1 (N-terminal region) from Aviva Systems Biology (ARP44418_T100), MTUS1 from Cell signaling Technology (CST, 13436-1-AP), FTO from Proteintech (27226-1-AP), β-Actin from CST (4970), and GAPDH from CST (2118).

RNA m6A dot blot assay

RNA samples were initially subjected to a denaturation step at 95 °C for 3 min in RNA incubation buffer. The denatured samples were subsequently divided into two experimental groups, consisting of 400 ng and 200 ng, respectively, and then transferred onto Hybond-N + membranes (Beyotime, Shanghai, China). The membrane was UV cross-linked using a UV crosslinker (Spectrolinker XL-1500, Westbury, NY) at 254 nm. Following the blocking steps, the membranes were incubated with the m6A antibody (Abcam, ab284130) and subsequently treated with 2nd antibodies (CST, 7074 S) for a duration of 2 h. After rinsing the membranes in TBST, dot blot intensities were detected using ECL (ThermoFisher, 34580) and visualized with the ProteinSimple imaging system (Bio-Techne, Minneapolis, MN). Subsequent staining of the membrane was was carried using a pre-diluted methylene blue working solution (MCE, HY-B1359) to determine the RNA content.

MeRIP-seq and MeRIP-qPCR

The experiments were conducted following previously published protocols with some adjustments [23,24,25]. Total RNA samples from HNSCC tissues in our experiment were isolated and then subjected to fragmentation using ZnCl2. The RNA was subsequently treated treated with either m6A antibody (Abcam, ab284130) or 2nd antibody for a duration of 2 h. Following the addition of magnetic beads (MCE, HY-K0202), we put the mixture in four-degree refrigerator for another 2 h in order to facilitate the capture of immunoprecipitated fragments. Following the purification of the fragments, they were divided into two distinct experimental groups: one portion was designated for sequencing library construction, while the other was allocated for qRT-PCR analysis. Sequencing of the RNA samples was carried out utilizing a NextSeq 500 sequencer (Illumina, Hayward, CA). The obtained data was subjected to analysis as detailed in earlier protocols. The identification of methylated sites on the RNAs (peaks) was done through the utilization of MACS software.

Immunohistochemistry (IHC) staining

Tissue sections from collected specimens were deparaffinized and rehydrated to restore their original moisture levels, followed by endogenous peroxidase quenching and antigen extraction. After the blocking step, the slides were incubated with ATIP1 (N-terminal region) (Aviva Systems Biology, ARP44418_T100) antibody or FTO (Proteintech, 27226-1-AP) antibody, and subsequently treated with a 2nd antibody. After being rinsed in PBS, sections were dyed using a DAB kit (Proteintech, PR30010) and hematoxylin. Each sample was evaluated using the scoring system describedby Pirker et al. [26]. The Immunohistochemistry score is a continuous scale (0-300), determined by multiplying the staining intensity score (0–3) with the proportion of cells corresponding to each level.

Protein and RNA stability assays

These experiments were carried out on HNSCC cells to examine how FTO influences the stability of MTUS1/ATIP1 mRNA and its corresponding protein. In the protein decay experiment, the indicated cells were exposed to cycloheximide (CHX; APExBIO, A8244-1000) at concentrations of 100 mg/mL for 0, 2, 4, and 8 h. Total cellular protein was collected, and the relative abundance of MTUS1/ATIP1 protein (compared to 0 h) was measured through western blotting. In the RNA decay experiment, the indicated cells were exposed to ActD at concentrations of 5 mg/mL for 0, 3, and 6 h. Subsequently, the relative level of MTUS1/ATIP1 mRNA was measured via qRT-PCR.

Cell proliferation assay

The impact of FTO and MTUS1/ATIP1 on the proliferation of HNSCC cells was investigated using Cell Counting Kit-8 (CCK-8; Dojindo, Kumamoto, Japan) assays. The HNSCC cell solution was initially diluted and then seeded into a 96-well plate (3000 cells/well). The cells were allowed to incubate for 24, 48, 72, or 96 h. After a 2-hour incubation period with the CCK-8 solution, the absorbance was measured at 450 nm using a microplate reader (Bio-Tek, Winooski, VT).

Plate colony formation assay

Cells were seeded in 6-well plates at a concentration of 180 cells/mL and cultured until visible colonies formed (approximately 2 weeks). The colonies were fixed using 4% paraformaldehyde and subsequently stained with crystal violet, followed by examination and photography.

Wound healing assay

The HNSCC cell solution was initially diluted and then seeded in a 6-well plate until they reached 100% confluence. Subsequently, a fixed-width band was created by scratching using a sterile 200µL tip, followed by incubation in medium containing 2% FBS for 24 h. Wound closure of all cells in the 6-well plate was examined using the Zeiss Z1 microscope to track the migrating front.

Ribosome-nascent chain complex (RNC)-qRT-PCR assay

The ribosome-Nascent Chain Complex was extracted following established protocols with minor modifications [27]. Cycloheximide (100 mg/mL) was initially administered to HNSCC cells, followed by the addition of a ribosome buffer containing 1% Triton X-100. After a 30-minute ice bath, the lysates were centrifuged to eliminate cell debris and nuclei for a duration of ten minutes at 16,200×g and a temperature of 4 degrees Celsius.

The supernatants were subsequently gathered and placed atop 20 mL of a buffer containing 30% sucrose.

We then utilized ultracentrifugation to gather RNC under conditions of 185,000×g and 4 degrees Celsius for a duration of five hours. Finally, RNA from the polysome fraction was analyzed using qRT-PCR as described in the aforementioned methods.

Animals study

BALB/c nude mice were obtained and housed in a controlled specific pathogen-free (SPF) environment at the Animal Experiment Center affiliated with the First Affiliated Hospital of Sun Yat-sen University (Guangzhou, China). The animal experiments conducted during this research received ethical approval from the Ethics Committee of the First Affiliated Hospital of Sun Yat-sen University ([2022]229).

All models were created based on previously published methods, incorporating specific modifications [23, 28, 29]. In the subcutaneous tumorigenesis model, the indicated HNSCC SCC9 cells were injected into the right armpits of six-week-old female mice at a concentration of 1 × 106 cells in 100µL of vehicle composed of DMEM/Matrigel matrix (1:1). The sizes of the xenografts were measured, computed and recorded every three days: tumor volume (mm3) = ½×length×width2. All mice in the subcutaneous injection experimental group were euthanized were euthanized 28 days post-injection. In the in situ tongue cancer model, the indicated HNSCC SCC9 cells were injected into the right lateral tongue of six-week-old female mice at a concentration of 1 × 105 cells in 20 µL of vehicle. The xenografts were measured, computed and recorded every three days. At the 21-day observation, we found the mice weight in tongue cancer groups significantly decreased with tumor growth. Consequently, we terminated the experiment for the tongue cancer groups at that time. Following euthanasia, the tumors were dissected, weighed, measured, and subsequently embedded, sectioned, and stained. For the euthanasia procedure, the mice were placed in the euthanasia box, and carbon dioxide was gradually introduced over a period of 15 min. After euthanization by carbon dioxide asphyxiation, each mouse was individually checked, followed by cervical dislocation. The euthanasia methods employed for all mice were approved by The Ethics Committee of the First Affiliated Hospital of Sun Yat-Sen University and are in accordance with the AVMA Euthanasia Guidelines of Animals (2013 edition).

Statistical analysis

Statistical analysis of the data in all experiments was carried out utilizing GraphPad Prism v10.1 along with SPSS v27.0. The relationships between FTO and ATIP1 were evaluated through IHC scores and analyzed using Pearson’s correlation coefficient. For survival analysis, a Kaplan-Meier curve was generated and accompanied by a log-rank test. Unpaired two-tailed Student’s t-tests were used to assess differences between two groups. In order to evaluate variations involving multiple groups, one-way ANOVA was used. The results are presented as mean ± SD or SEM. P < 0.05 was deemed statistically significant.