Omeprazole improves chemosensitivity of gastric cancer cells by m⁶A demethylase FTO-mediated activation of mTORC1 and DDIT3 up-regulation

Gastric cancer (GC) is one of the most common cancers all over the word that is currently the second-leading cause of cancer related death among both men and women in China [1]. In recent years, significant improvements including surgery, radiation therapy, chemotherapy, targeted therapy or immunotherapy alone or in combination have been made in the treatment measures for gastric cancer [2]. However, the curative effect for patients with advanced gastric cancer is still unsatisfactory. Hence, a better understanding of the factors and potential molecular mechanism involved in improving the response efficiency of GC cells to therapeutic methods has very important clinical significance.

Proton pump inhibitors (PPIs), such as omeprazole, esomeprazole and pantoprazole (PPZ), have been widely used for the treatment of acid-related diseases, such as gastro-esophageal reflux disease, peptic ulcer disease, and the eradication of Helicobacter pylori infection [3]. Previously, the conclusion that PPIs could function as a promising treatment strategy for gastric cancer due to their significantly sensitize gastric cancer cells to antitumor drugs has been made by our research group and others [4–6]. It was found that PPIs directly inhibited cancer cell proliferation, drug resistance-induced metastasis and regulate stemness of GC cancer cells [5,7,8].

N6-methyladenosine (m⁶A) is reported to be the most common modification among RNA modifications. In recent years, with the improvement of experimental techniques, m⁶A has been found to be involved in the control of various cell functions. M⁶A exerts its regulatory function mainly through three kinds of proteins: RNA methylates (writers) and demethylates (erasers) and a third group of ´readers’ bind to m6A sites and determine the fate of the modified mRNA [9]. However, up to now, little has been reported on whether m⁶A modification participated in the antineoplastic drugs sensitivity of cancer cells.

Fat mass and obesity associated (FTO) is one of the first identified m⁶A eraser proteins that has an important role in promoting the occurrence of GC, and it may be a vital molecular marker in the diagnosis and prognosis of GC patients. Down-regulation of FTO significantly inhibited the proliferation, migration and invasion of GC cell lines [10,11].

In this research, for the first time, we found that omeprazole pretreatment could enhance the inhibitory effect of 5-Fu, DDP and TAX on gastric cancer cells. Omeprazole pretreatment promoted the total m⁶A level of gastric cancer cells due to the FTO inhibition induced by omeprazole. FTO inhibition induced by omeprazole enhanced the activation of mTORC1 signal pathway that inhibited the prosurvival autophagy so as to improve the antitumor efficiency of chemotherapeutic drugs on GC cells. Meanwhile, omeprazole induced FTO inhibition could increase the transcript level of DDIT3, which is an apoptosis-related tumor suppressor gene downstream of mTORC1 signaling through an m⁶A-dependent mechanism.

The human GC cell lines, AGS and HGC27, were purchased from the Hunan Yearth Biotechnology Co., Ltd. (Changsha, China). Both cell lines were cultured in RPMI-1640 medium supplemented with 10% fetal bovine serum (FBS; Gibco, Canada) and antibiotics (1% penicillin/ and100 μg/ml streptomycin sulfates, Selleck, China) in a humidified atmosphere of 95% air and 5% CO₂ at 37°C. Proton pump inhibitor omeprazole, antitumor drugs 5-Fu, DDP, TAX and autophagy pharmacological agents 3-MA and Rapamycin were all obtained from Selleckchem (Shanghai, China).

Cells were pretreated with 40 μg/ul omeprazole overnight, then, about 5000 cells were seeded in 96-well plates with 100 μl of medium with1μM 5-Fu, 10 mg/l DDP, 100 nM TAX for another 24 h. The relative cell survival rate was assessed using MTT assay according to manufacturer’s instructions.

Annexin V-FITC/propidium iodide (PI) staining assay was used to detect apoptosis of GC cells according to manufacturer’s instructions (Kengen biotech, China). Finally, apoptosis cells were assessed by gating PI and Annexin V-positive cells on a fluorescence-activated cell-sorting (FACS) flow cytometer (BD Pharmingen U.S.A.).

The coefficient of drug interaction (CDI) was calculated according to previous studies [12,13], and CDI value <1, =1 or >1 indicates that the drugs are synergistic, additive or antagonistic, respectively.

Gastric cancer cells that have undergone various treatments after infected with AAV- mRFP-GFP-LC3 (Hanbio, Wuhan, China) were fixed with 4% paraformaldehyde and then mounted onto microscope slides with FluorSave™ Reagent (Calbiochem, U.S.A.). Localization of RFP-LC3 or GFP-LC3 was evaluated and observed using a fluorescent inverted microscope (motic).

In order to construct FTO loss-of-function cell models, three siRNA sequences targeting FTO were designed and synthesized in genepharma biotech (Suzhou, China), and the one with best interference efficiency in GC cells was used for subsequent experiments. The sequence information of FTO-siRNA was: 5′-GCACAAGCATGGCTGCTTA-3′. To obtain FTO gain-of-function cell models, FTO overexpressed lentivirus particles was purchased from genepharma biotech (Suzhou, China). Relative transfection and infection experiment was performed according to manufacturer’s instructions.

Total RNA from cell lines of different treatment were extracted using TRIzol reagent (Invitrogen, CA) and treated with deoxyribonuclease I (Sigma, Shanghai, China). After RNA quality inspection, the commercial m⁶A RNA methylation quantification kit (ab185912; Abcam, U.S.A.) was used to detect the total m⁶A level. Briefly, 250 ng RNAs were seeded in each well, followed by adding capture antibody solution and detection antibody solution according to the manufacturer’s instruction. The m⁶A levels were then measured colorimetrically by reading the absorbance of each well at a wavelength of 450 nm.

Total RNAs were extracted from different group of cells using the TRIzol Plus RNA Purification Kit (Invitrogen, Carlbad, CA, U.S.A.). Quantitative real-time PCR was performed to detect the relative mRNA levels of genes of interest using QPCR ThermoScript One-Step System (Invitrogen, Carlbad, CA, U.S.A.) in the LC480 Real-Time PCR Detection System (Roche) according to the manufacturer’s instructions. All tests were carried out with three independent experiments. The primer sequences are listed in Table 1. The quantification analysis for relative mRNA levels of target genes was performed using the relative quantification comparative CT method.

Western blot was performed to detect the target protein expression of interest according to the methods described in the literature [14]. The primary antibodies we used were anti-p62 (18420-1-AP, Proteintech, China), anti-LC3 (18725-1-AP, Proteintech, China), anti-Bax (50599-2-lg, Proteintech, China), anti-caspase-3 (12789-1-AP, Proteintech, China), anti-S6 (#2217, CST, U.S.A.), anti-phospho-S6 (Ser235/236) (#4856, CST, U.S.A.), 4EBP1 (#9452, CST, U.S.A.), phospho-4EBP1(T37/46) (#2855, CST, U.S.A.).

Real-time quantitative PCR (qPCR) was performed to assess the relative abundance of the selected mRNA in m⁶A antibody IP samples and input samples between the Lv-FTO, siRNA-FTO cell lines and NC cell lines. The gene-specific m⁶A qPCR was performed to determine the m⁶A abundance on the transcripts according to the method and steps reported in the previous literature [15].

All experiments were repeated for three times and experiments were performed in triplicate. Statistical analyses were performed using GraphPad Prism 7.00 (GraphPad Software, La Jolla, CA, U.S.A.). Student’s t tests were used for comparison between the treatment and control groups. One-way ANOVA with Tukey’s post test for multiple comparisons was used when comparing more than two groups. For all analyses, a P value of ≤ 0.05 was considered significant.

Gastric adenocarcinoma cell lines AGS and HGC-27 were treated with 5-Fu, DDP and TAX for 24 h with or without omeprazole pretreatment. The results showed that three antitumor drugs can significantly reduce the relative survival rate of AGS and HGC-27 cells. At the same time, the relative survival and apoptosis rate of cells pretreated with omeprazole was significantly lower and higher respectively than that of cells untreated with omeprazole. This result suggested that, to some extent, omeprazole preconditioning could improve the response of gastric cancer cells to antitumor drugs (Table 1; Figure 1A,B). In recent years, more and more literatures have confirmed that antitumor drugs induced prosurvival autophagy of tumor cells that is a key reason for reducing chemosensitivity. Our research also confirmed the conclusion using Western blot detection of autophagic substrate P62 and classic autophagic marker LC3. Antitumor drugs reduced the expression of P62 and increased the ratio of LC3II/I (Figure 1C).

In order to explore whether omeprazole was associated with autophagy regulation in the antitumor synergism of gastric cancer cells, Western blot and LC3 puncta detections were performed in cells pretreated with or without omeprazole before antitumor drug treatment. When cells were pretreated with omeprazole, the expression of autophagic substrate P62 was relatively higher, while the ratio of LC3II/I was relatively lower compared with cells without omeprazole pretreatment (Figure 1D). Besides, in cells with omeprazole pretreatment, the number of LC3 puncta reduced obviously, and the fluorescence intensity was also weaker when compared with cells without omeprazole pretreatment (Figure 1E,F).

m⁶A is one of RNA modifications which is involved in the control of various cell functions, such as tumor progression [16,17] and stem cell development [18]. However, little research has been done on whether m⁶A modification is involved in the regulation of chemosensitivity. Based on this idea, we tried to find out whether omeprazole preconditioning could change the total m⁶A level of gastric cancer cells first. What is amazing is that omeprazole treatment caused a significant concentration-dependent increase in the total m⁶A level of both AGS and HGC-27 cells (Figure 2A).

Then, the relative mRNA levels of classic m⁶A regulators were detected using qRT-PCR. The results showed that omeprazole treatment reduced the expression of m⁶A eraser FTO significantly, which was consistent with the promotion of total m⁶A level in both AGS and HGC-27 cells (Figure 2B). Western blot was performed to further confirm the expression of FTO, and a consistent conclusion was reached (Figure 2C). FTO expression was reported to have important roles in promoting gastric cancer occurrence and could be an vital molecular marker in the diagnosis and prognosis of GC patients [11]. We evaluated the expression and significance of FTO in overall survival (OS) and disease-free survival (DFS) in GC through GEPIA database (http://gepia.cancer-pku.cn). It was found that FTO expression is significantly elevated in GC tissues (T) compared with noncancerous tissues (N). Meanwhile, stronger FTO protein expression was associated with lower disease-free survival rate in GC patients (Figure 2D,E). In order to further analyze the function of FTO, the knockdown and overexpression models of FTO were constructed and validated in both GC cells (Figure 2F).

To further confirm that FTO is one of the targets of omeprazole on GC cells, gain-of-function cell models were established in both AGS and HGC-27 cells. It was found that, when FTO was overexpressed, the enhanced chemosensitivity of GC cells induced by omeprazole pretreatment significantly reduced and the relative survival rate of cells increased significantly accompanied by decreased cell apoptosis (Figure 3A,B). Because the present study has clarified that omeprazole pretreatment inhibited chemotherapy-induced prosurvival autophagy and improve the chemosensitivity of GC cells. So does FTO, the target of omeprazole, have the ability to regulate autophagy and apoptosis? First, Western blot was performed to detect the classic autophagy marker proteins in FTO gain-of-function cell models treated with antitumor drugs with omeprazole preconditioning. The results showed that when FTO was enhanced, autophagic substrate P62 decreased and the ratio of LC3II/I up-regulated obviously compared with negative control cells treated with antitumor drugs with omeprazole preconditioning (Figure 3C). Meanwhile, the number of LC3 puncta in FTO overexpressed GC cells increased obviously when compared with negative control cells without FTO up-regulation (Figure 3D). Second, Western blot was also performed to detect the classic apoptosis-related proteins in FTO gain-of-function cell models treated with antitumor drugs with omeprazole preconditioning. The results showed that when FTO was upregulated, apoptotic executive protein cleaved caspase-3 and proapoptosis protein Bax were both decreased obviously (Figure 3C). When we tried to block the autophagy of FTO-overexpressed cells again with autophagy inhibitor 3-MA, the synergistic inhibitory effect of omeprazole and antineoplastic drugs appeared again (Figure 3E). The above results suggested that FTO could restore the prosurvival autophagy inhibition induced by omeprazole pretreatment and reduce the sensitivity of GC cells to antineoplastic drugs.

mTORC1 plays a role in switching the prosurvival autophagy [19]. Our results found that omeprazole treatment could enhance the phosphorylation S6 (Ser235/236) and 4EBP1 (Ser37/46), both of which are downstream targets of mTORC1 [20], suggested that omeprazole treatment activated mTORC1 to some extent (Figure 4A). Besides, when FTO was silenced, the phosphorylation S6 (Ser235/236) and 4EBP1 (Ser37/46) promoted. In the other hand, the phosphorylation S6 (Ser235/236) and 4EBP1 (Ser37/46) attenuate when FTO was overexpressed (Figure 4B). These results demonstrated that FTO/mTORC1 signal axis played roles in omeprazole induced autophagy mediation.

It was reported that when FTO was inhibited by R-2HG treatment, the transcripts with increased m⁶A levels (m⁶A-hyper) were also significantly enriched for target genes of mTORC1 signaling pathway [21]. This aroused our strong interest. Through the comparison and analysis of GSE data sets (GSEA 87190, GSEA 87187 and GSEA 87189), 3 mTORC1 pathway target genes were identified that may be regulated by FTO in m⁶A-dependent manner (Figure 5A). Besides, based on m⁶A-seq data GSEA87190, when FTO was inhibited by R-2HG, the m⁶A abundance of DDIT3 mRNA increased, especially in the CDS and 3′-UTR regions (Figure 5B). Subsequently, qRT-PCR and Western blot were performed in omeprzole pretreated GC cells. It was confirmed in AGS and HGC-27 cells. DDIT3, an apoptosis-related tumor suppressor gene downstream of mTORC1 signaling, showed significant enhancement when compared with cells without ome-treatment (Figure 5C). To explore whether ome-induced DDIT3 up-regulation is mediated through m⁶A-dependent manner in AGS and HGC-27 cells studied in this research, gene-specific m⁶A qPCR assays were performed to evaluate relative m⁶A level of DDIT3 in omeprazole pretreated GC cells. The results showed that omeprazole treatment significantly promoted the relative m⁶A level of DDIT3 in both 3′UTR and CDS regions in GC cells compared with PBS-treated cells. However, the m⁶A level of DDIT3 did not increase significantly when FTO overexpressed GC cells (Lv-FTO) were pretreated with omeprazole (Figure 5D). In order to further certificate that FTO negatively regulates DDIT3 through m⁶A-dependent manner, relative m⁶A levels of DDIT3 in different FTO functional GC cells were assessed. The results demonstrated that FTO silence could enhance m⁶A levels of DDIT3 in both CDS and 3′UTR region. Conversely, FTO overexpression obviously attenuated m⁶A levels of DDIT3 in both CDS and 3′UTR region (Figure 5E).

DDIT3 was an apoptosis-related tumor suppressor gene downstream of mTORC1 signaling. Omeprazole/FTO/m⁶A modification axis up-regulated DDIT3 expression. Thus, is this related to the synergistic effect of omeprazole and antineoplastic drugs to GC cells? To clarify the problem, the apoptosis of FTO overexpressed GC cells harboring DDIT3 CDS containing wild-type or mutant m⁶A sites (m⁶A was replaced with T) was measured. The results showed that when GC cells were treated with omeprazole in combination with antineoplastic drugs, the attenuated cell apoptosis resulted from FTO up-regulation was significantly restored due to the putative m⁶A sites were mutated in DDIT3 CDS region (Figure 5F). Correspondingly, the expression of apoptotic marker proteins c-caspase 3 and Bax was also enforced when compared with wild type group cells (Figure 5G).

Several researches demonstrated that PPIs was promising antitumor and chemosensitizing efficacy that could restore chemosensitivity in drug-resistant cancer cells due to tumor acidity was considered as key determinant of drug-resistance and tumor progression [22]. A few studies suggested that PPI combined with chemotherapeutic drugs can synergistically inhibit the production of endometrial cancer and melanoma [23,24]. To our knowledge, whether drug combination has synergistic effect should be judged by the coefficient of drug interaction (CDI) value. CDI value <1, =1 or >1 indicates that the drugs are synergistic, additive or antagonistic, respectively. Based on this, in the present study, it was reconfirmed that omeprazole preconditioning could indeed produce synergistic inhibitory effect on gastric cancer cells with antitumor drugs on the basis of CDI value.

The potential functional role of m⁶A-related genes in gastric cancer (GC) prognosis has not been systematically researched. At the same time, little research has been done on whether m⁶A modification is involved in the regulation of chemosensitivity. FTO was reported to enhance the chemoresistance of cervical squamous cell by reducing m⁶A levels of β-catenin mRNA [25]. Based on this, we hypothesize whether the increased chemosensitivity of gastric cancer cells induced by omeprazole preconditioning is related to m⁶A modification.

The results of colorimetric assay showed that omeprazole treatment could significantly promote the overall m⁶A level of GC cells which implied that omeprazole pretreatment leads to the desregulation of some m⁶A related genes. So, we detected mRNA levels of a series of m⁶A classical related factors by qRT-PCR. It was found that after omeprazole treatment, m⁶A eraser FTO decreased significantly, which was in line with the result that the overall proportion of m⁶A increased after omeprazole treatment. However, when our team attempted to reconfirm the result that the overall level of m⁶A alteration from the blood samples collected clinically, the low quality of RNA from blood sources led to the failure of the validation.

Up to now, little research has been done on the involvement of m⁶A-related factors in autophagy regulation. However, it is worth mentioning that research has identified that m⁶A eraser FTO as an oncogenic role via autophagy. In addition, FTO was considered or associated with autophagy regulation [26,27]. Our results suggested that FTO promoted survival-promoting autophagy of gastric cancer cells, at least in the context of the present study.

R-2HG was a FTO inhibitor, after R-2HG treatment, the transcripts with increased m⁶A levels (m⁶A-hyper) were also significantly enriched for target genes in mTORC1 signaling pathways [21]. The mTORC1 signal that was reported to coordinate the autophagy and apoptosis was first thought of by us [19]. This result demonstrated that FTO could negatively regulate mTORC1 activation that could in turn inhibit autophagy. Further, based on the bioinformatics analysis of data, we identified three mTORC1 pathway target genes that may be regulated by FTO in m⁶A-dependent manner. Among them, DDIT3 which is an apoptosis-related tumor suppressor gene was found to be up-regulated with the down-regulation of FTO. Based on these, it was speculated omeprazole induced FTO inhibition could increase the transcript level of DDIT3, which is an apoptosis-related tumor suppressor gene downstream of mTORC1 signaling through an m⁶A-dependent mechanism. Fortunately, we have also substantiated this scientific hypothesis through experiments.

In our study, the molecular mechanism of the effect of omeprazole on GC cells through regulating FTO / mTORC1 signal axis includes two aspects: first, FTO inhibition induced by omeprazole enhanced the activation of mTORC1 signal pathway that inhibited the prosurvival autophagy so as to improve the cell apoptosis on GC cells induced by chemotherapeutic drugs. In this part, we basically confirmed the omeprazole mainly to inhibit prosurvival autophagy to promote apoptosis by setting autophagy inhibitor 3-MA group (we only measured the survival capacity to laterally reflect the cell apoptosis, but did not directly detect the cell apoptosis, this was a deficiency).

Second, It was found that omeprazole-induced FTO silence could activate mTORC1, then up-regulated DDIT3 through an m6A-dependent mechanism. DDIT3 is an apoptosis-related tumor suppressor gene which directly promote the apoptosis of tumor cells. However, whether DDIT3 could promote the apoptosis of GC cells by regulating autophagy deserves further study.

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

The authors declare that there are no competing interests associated with the manuscript.

The project is funded by The Science and Technology Program of Xiamen city [grant number 3502Z20184049]; The Science and Technology Program of Haicang District of Xiamen city [grant number 350205Z20174005)]; The Tumor Treatment Center Special Fund and the Medical Science Innovation Fund of the Nanjing Military Region of the People’s Liberation Army [grant number 14ZD33]; Scientific research project of Xiamen science and Technology Bureau [grant number 3502Z20199106].

Doctor Shuitu Feng provided the experimental design and the major contributor in writing the manuscript. Dr Guoqin Qiu, Dr Lihong Yang and Lihua Feng are the main experimenter performing all the experiments. Dr Xin Fan, Fang Ren and Kaida Huang are mainly responsible for acquisition and analysis of data. Professor Yide Chen revised the manuscript critically for important intellectual content.

CDI

coefficient of drug interaction

DFS

disease-free survival

GC

gastric cancer

OS

overall survival

PPI

proton pump inhibitor

PPZ

pantoprazole