M6A demethylase fat mass and obesity‐associated protein regulates cisplatin resistance of gastric cancer by modulating autophagy activation through ULK1

Abstract Drug resistance is an important factor for treatment failure of gastric cancer. N6‐methyladenosine (m6A) is the predominant mRNA internal modification in eukaryotes. The roles of m6A modification in drug resistance of gastric cancer remains unclear. In the present study, the m6A methylated RNA level was significantly decreased while the expression of m6A demethylase fat mass and obesity‐associated protein (FTO) was obviously elevated in cisplatin‐resistant (SGC‐7901/DDP) gastric cancer cells. Knockdown of FTO reversed cisplatin resistance of SGC‐7901/DDP cells both in vitro and in vivo, which was attributed to the inhibition of Unc‐51‐like kinase 1 (ULK1)‐mediated autophagy. Mechanistically, ULK1 expression was regulated in an FTO‐m6A‐dependent and YTHDF2‐mediated manner. Collectively, our findings indicate that the FTO/ULK1 axis exerts crucial roles in cisplatin resistance of gastric cancer.


| INTRODUC TI ON
Gastric cancer is one of the most common tumors that poses a potential threat to human health and quality of life. 1,2 At present, platinum-based chemotherapy is commonly used for patients with gastric cancer, which can improve prognosis and prolong the survival of patients. Unfortunately, poor response to platinum-based chemotherapy is often observed in gastric cancer patients due to the primary or obtained resistance, which becomes the major cause of treatment failure. 3,4 Over the past few decades, scientists have studied the mechanisms of cisplatin resistance. It is a complex cellular process that involves multiple pathways, including DNA repair, autophagy, and glutamine metabolism. [5][6][7] Consequently, the understanding of molecular mechanisms for platinum resistance and investigating potential therapeutic targets are urgent. Autophagy is a system of cellular degradation, which is a prevalent element of drug resistance to chemotherapy in tumor cells. 8,9 Chemotherapeutic drugs could induce autophagy, which avoids cancer cell apoptosis, thereby producing drug resistance. 10 Previous studies have shown that 5-fluorouracil induced autophagic death of gastric cancer cells and inhibited cell proliferation by upregulating Beclin1 expression by inhibiting microRNA-30. 11 It was found that autophagy, which was involved in the formation and maintenance of chemotherapy resistance, inhibited the killing effect of imatinib on tumor cells in BCR-ABL positive chronic myelogenous leukemia. [12][13][14] Some studies have found that paclitaxel increased the expression of BECN1 and decreased P62, inducing autophagy, in both cases. 15,16 Unc-51-like kinase 1 (ULK1) is a cytoplasmic kinase, which recruits the autophagy proteins to mediate the initiation of autophagy. 17 A selective inhibitor of ULK1 (SBI0206965) significantly inhibits the progress of autophagy in cisplatin-resistant non-small-cell lung cancer (NSCLC) cells. 18 In addition, knockdown of ULK1 in NSCLC cells makes cells more sensitive to cisplatin. 19 N 6 -methyladenosine (m 6 A), in which the sixth nitrogen (N) atom of adenine is methylated, is the most frequent modification on mRNAs in eukaryotes. Modification of m 6 A is regulated by methyltransferase complex ("writers"), demethylases ("erasers"), and RNAbinding proteins ("readers"). 20 The main discovered components of the methyltransferases include methyltransferase-like 3 (METTL3) and METTL14, while fat mass and obesity-associated protein (FTO) and α-ketoglutarate-dependent dioxygenase homolog 5 (ALKBH5) have been identified as demethylated enzymes. The main known "reader" proteins include YTH domain families (YTHDF1, YTHDF2, and YTHDC1), which mediate mRNA stabilization. 21,22 In the current study, we aimed to clarify the effects and mechanisms of m 6 A RNA modification on autophagy-related cisplatin resistance in gastric cancer cells.
Fat mass and obesity-associated protein, the first identified m 6 A demethylase, has important roles in various types of disease. [23][24][25] The present study found that FTO was obviously elevated in the cisplatin-resistant (SGC-7901/DDP) gastric cancer cells. Knockdown of FTO reversed cisplatin resistance both in vitro and in vivo, which was attributed to the inhibition of ULK1-mediated autophagy.
Furthermore, ULK1 expression was regulated in the FTO-m 6 A dependent and YTHDF2-mediated manner.

| Cell culture
The human gastric epithelial cell line GES-1 and the human gastric cancer cell line SGC-7901 were obtained from the Health Science Research Resources Bank and used in our previous study. 26

| Cell viability assay
Cell viability assays were performed using the CCK-8 method. Cells at log-phase growth were seeded into 96-well plates at 5000 cells/ well. After treatment, 10 μl CCK-8 solution (Beyotime) was added into each well, followed by an additional incubation for 2 h at 37°C.
Finally, the 450 nm absorbance (OD) was determined using a microplate reader (Bio-Tek). The percentage cell viability was calculated.

| m 6 A RNA methylation quantification
Total RNA was extracted from GES-1, SGC-7901, and SGC-7901/DDP cells using TRIzol reagent. The m 6 A content in the total RNA was detected using the m 6 A RNA Methylation Assay Kit (Abcam). The percentage of m 6 A in total RNA was carried out using the following formula:

| Western blot analysis
Cells or tumor tissues were lysed in RIPA lysis buffer with protease and phosphatase inhibitors (Beyotime). Thirty micrograms of total protein was subjected to SDS-PAGE gels and subsequently transferred to PVDF membranes. The membranes were blocked and then incubated with primary Abs overnight at 4°C. After washing three times, the membranes were incubated with the secondary Abs for 2 h at room temperature. Finally, protein bands were visualized by a chemilumi-

| Transmission electron microscopy
Autophagosomes refer to endogenous substances, including damaged organelles or excess glycogen stored in cells due to physiological or pathological reasons, which can be formed by wrapping

| Immunofluorescence analysis
Cells were fixed with 4% paraformaldehyde and then permeabilized with Triton X-100 (Beyotime) at room temperature. Cells were incubated with anti-LC3B Abs (Abcam) at 4°C overnight after blocking with the Immunol staining blocking buffer (Beyotime) for 60 min.
After washing three times with TBST solution, cells were incubated with Cy3 goat anti-rabbit IgG (Abclonal) at room temperature for 1 h.
Cell nuclei were stained with DAPI (Beyotime) for 10 min at room temperature. Cells were observed under fluorescence microscope and images were collected after adding antifluorescence attenuation sealer.

| Quantitative real-time PCR
TRIzol reagent (Invitrogen) was used to extract total RNA from cells, and a reverse transcription kit (Tiangen) was used for reverse transcription. cDNA was generated according to the manufacturer's instructions. In addition, quantitative real-time PCR (qPCR) was carried out to analyze gene expression using the SYBR Green PCR Kit

| N 6 -methyladenosine RIP qPCR assay
This procedure was undertaken with the m 6 A RIP (MeRIP) kit (Bersinbio) according to the manufacturer's instructions. Briefly, total RNA was extracted from 2 × 10 7 SGC-7901/DDP cells using TRIzol reagent, and RNA was further fragmentated using ultrasound.

| RNA immunoprecipitation PCR
RNA immunoprecipitation was undertaken with the RIP kit (Bersinbio) according to the manufacturer's instructions. Briefly, total RNA was extracted from 2 × 10 7 cells using TRIzol reagent.
RNA samples were divided into three: the IP group, IgG group and Input group. The Input samples were stored at −80°C. Anti-FTO, -YTHDF2, or -IgG Abs were conjugated to protein A/G magnetic beads in IP buffer for 16 h at 4°C in a vertical mixer. The RNA was eluted from the beads with proteinase K at 55°C for 45 min. Finally, the precipitated RNA and input RNA were detected by qRT-PCR.

| mRNA stability analysis
Cells were transfected with either the NC siRNA or YTHDF2 siRNA for 24 h and then treated with 5 μg/ml actinomycin D (MCE) to inhibit mRNA transcription. The cells were collected and total RNA was extracted by TRIzol reagent after treatment for the indicated times. The levels of mRNA were detected by qRT-PCR.

| Hematoxylin−eosin and immunohistochemistry staining
Tumor tissues of the mice were fixed in 4% paraformaldehyde and then embedded in paraffin. After embedding and sectioning, the tumor tissues were stained with H&E. Immunohistochemistry staining was carried out using an anti-Ki-67 mAb (Cell Signaling Technology). The protein expression levels of Ki-67 in tumor tissues were observed under microscopy.

| Statistical analysis
The data are presented as the mean ± SD. The statistical analyses were undertaken using a two-tailed Student's t-test or ANOVA, which was used to analyze the differences when there were more than two groups. The statistical analyses were carried out using SPSS 19.0 software. P values less than 0.05 were considered to be statistically significant.  Figure 1K). We also observed that FTO overexpression significantly reduced the sensitivity of SGC-7901 cells to cisplatin ( Figure 1L).

| Fat mass and obesity-associated protein is
These results indicate that m 6 A demethylase enzyme FTO mediates cisplatin resistance of gastric cancer cells.

| Fat mass and obesity-associated protein promotes cisplatin resistance by facilitating autophagy in gastric cancer cells
Autophagy is a common cellular process in eukaryotic cells, which is largely involved in the cisplatin resistance of tumors. 6,27,28 During autophagy, LC3 I is converted to LC3 II through lipidation by a ubiquitin-like system, and P62 can be degraded by the autophagosome. 29,30 Consequently, degradation of P62 and ac-  was downregulated, indicating that ULK1 activity decreased 31,32 ( Figure 3E,F). Moreover, knockdown of ULK1 obviously promoted the inhibitory effects of cisplatin on growth in SGC-7901/DDP cells ( Figure 3G). These results indicate that ULK1 is functionally important for autophagy and cisplatin resistance in gastric cancer cells.

| Fat mass and obesity-associated protein regulates autophagy and cisplatin resistance through targeting ULK1 in an m 6 A-dependent manner
To confirm whether FTO influenced autophagy and cisplatin resistance through targeting ULK1, we undertook rescue experiments and observed that ULK1 inhibition reversed the upregulated LC3B II, P-ATG13 and increased the downregulated P62 in SGC-7901/DDP cells with FTO overexpression ( Figure 4A,B). In addition, immunofluorescence assays showed that the elevated LC3B puncta induced by FTO overexpression was compromised in SGC-7901/DDP cells when treated with siULK1 ( Figure 4C). Overexpression of FTO significantly attenuated the inhibitory effects of cisplatin on growth in SGC-7901/DDP cells. Transfection with siULK1 obviously reversed these effects ( Figure 4D). To further explore the potential underlying mechanisms of m 6 A in autophagy regulation, MeRIP-qPCR and RIP-qPCR assays were carried out. As shown in Figure 4E, silencing of FTO significantly increased the m 6 A levels on mRNA of ULK1 at three sites. Furthermore, RIP-qPCR analysis revealed that ULK1 was a direct target of FTO ( Figure 4F). These results indicate that FTO regulates the expression of ULK1 in an m 6 A-dependent manner.

| Fat mass and obesity-associated protein modulates ULK1 expression through YTHDF2
It is well known that m 6 A methylation mediates targeted mRNAs by specific RNA-binding proteins. YTHDF2, a major m 6 A reader, is reported to selectively bind and destabilize m 6 A-modified mRNAs. 23,33 To investigate whether the expression of ULK1 was affected by  Figure 5G). Together, our data suggest that FTO regulates ULK1 in a YTHDF2-dependent manner.

| Knockdown of FTO improves the sensitivity of cisplatin-resistant gastric cancer cells to cisplatin in vivo
We In the present study, we first revealed the relationship between Previous studies have indicated that FTO positively regulates autophagy in cervical cancer HeLa cells 24 and mouse pre-adipocyte 3T3-L1 cells. 23 Autophagy functions as a protective factor in resistance of cancer cells exposed to anticancer drugs. 28

CO N FLI C T O F I NTE R E S T
The authors declare no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The datasets generated for this study are available on request to the corresponding author.

E TH I C S S TATEM ENT
Approval of the research protocol by an institutional review board: N/A.