miR-498 inhibits cell proliferation and metastasis by targeting FOXK1 in gastric cancer

Objective: MiR-498 has emerged as a potential molecular target for several cancer. In this study, we aimed to investigate the important function and mechanisms of miR-498 in gastric cancer. Methods: To detect the important roles of miR-498 in gastric cancer, we rst measured its expression by RT-qPCR in gastric cancer cell lines. The impact of the miR-498 on gastric cancer cell proliferation were detected by CCK-8 and colony formation assays. The effect of the miR-498 on the cell apoptosis and cell cycle were detected by ow cytometry. We also used the scratch and transwell chamber assays to measure the cell migration and invasion. The expression levels of related proteins were assessed by western blot. The bioinformatics analysis was used to explore the target gene of miR-498. RT-qPCR and western blot assays were used to detect the expression levels of FOXK1 in response to miR-498 overexpression. In order to prove the role of FOXK1 in mediating the effect of miR-498 on the gastric cancer, CCK-8, colony formation, transwell chamber and ow cytometry assays were used for the further investigations. Results: The expression level of miR-498 is downregulated in gastric cancer cell lines. Overexpression of miR-498 inhibited proliferation and migration/invasion, while promoted the apoptosis of gastric cancer cells. Bioinformatics analysis indicated that miR-498 targeted on FOXK1 to inhibit the gastric cancer in vitro. Conclusion: MiR-498/FOXK1 axis may be a potential therapeutic target for the treatment of gastric cancers. identied a novel target, FOXK1, mediated the inhibitory effect of miR-498 on the gastric cancer. ndings highlight the potential therapeutic target in gastric cancer and are worthy for future investigation.


Introduction
Gastric cancer (GC) is the fth most common malignancy with high morbidity and mortality in the world, while it is the third cause of cancer death after lung cancer and liver cancer. It was estimated in 2018, there were about over 1 million patents diagnosed with gastric cancer and the cases of death was over 780,000 [1] . With medical advanced, though the 5-year survival of gastric cancer patients is less than 30% [2] .The main pathogenesis of gastric cancer is the infection of helicobacter pylori, the genetic susceptibility of host, the environmental factors, including the diet, smoking, and use of alcohol [3][4] . The tumorigenesis of gastric cancer is an extremely complicated process, which contains genetic and epigenetic changes of proto-oncogenes and cancer suppressors. Of note, dysregulated microRNAs (miRNAs) played an important role in gastric cancer [4] . For example, miRNA-21 may promote the growth of gastric cancer cells by adjusting and controlling PTEN/Akt signal passage mediated PEG2 [5] . MiR-4317 represses the proliferation of gastric cancer cell by targeting and suppressing ZNF322 [6] . MiRNA-194 is oncogenic and promotes GC cell proliferation and migration by activating Wnt/βcatenin signaling via suppression of SUFU [7] .
MiRNAs are noncoding RNA molecules consist of 22 nucleotides, which could bind to a complimentary mRNA sequence, and thus resulting in posttranslation repression or degradation and silencing. The miRNAs mediate the cell survival, proliferation, apoptosis, tumor growth, and metastasis by acting as a regulators of genes expression [8] . It has been widely reported that miR-498 were involved in the regulation of various cancers and were considered to be the prognostic biomarker and therapeutic target for prostate cancer [9] , triple negative breast cancer [10] , esophageal squamous cell carcinoma [11] , liver cancer [12] and non-small cell lung cancer [13] . A previous study has found the overexpression miR-498 inhibits gastric cancer cell proliferation, migration and invasion and the long non-coding RNA UFC1 promotes gastric cancer progression by regulating miR-498/Lin28b [14] .
The lentiviral vector expressing FOXK1 was transfected into 293T packaging cells to obtain high levels of lentiviral particles in the culture supernatant.
The AGS and SGC-7901 cells were cultured in 25-cm 2 dishes. Transfection was performed by adding polybrene (8 µg/ml) and 20 µl each viral dilution to the cells, thoroughly and gently mixing the solutions, and incubating the cells in 5% CO 2 at 37°C. After 18 h, the viral particles remaining in the supernatant were removed and the medium was replaced with fresh medium supplemented with 10% FBS (Thermo Fisher Scienti c, USA.). The cells were incubated in 5% CO 2 at 37°C for an additional 72 h.

RNA extraction and RT-qPCR
Total RNAs were extracted from the treated cell lines GES-1, AGS, MGC-803, SGC-7901 and SNU-1 with TRIZOL reagent (Thermo Fisher Scienti c, USA), and the cDNA were reverse-transcribed by Revert Aid First Strand cDNA Synthesis kit (Thermo Fisher Scienti c, USA). The reverse transcription reactions were: 37℃ for 15 min, 85℃ for 5 sec. RT-qPCR was conducted by SYBR-Green PCR Master Mix kit (Takara, Tokyo, Japan) and 7900 HT Fast system (Applied Biosystems, California, USA) with the following protocol: 94°C, 5 min; 40 cycles of ampli cation (94°C, 30 sec and 62°C, 40 sec); and then 72°C, 10 min. U6 were used as the internal control.
Cell counting kit-8 (CCK-8) assay After transfection with NC mimic and miR-498 mimic, cells (2×10 3 cells/well) were seeded into 96-well plates and incubated in the humidi ed incubator at 37℃and 5% CO 2 for 0 h, 24 h, 48 h, and 72 h after inoculation. At each time point, 10 μl of the CCK-8 solution (Syngene, Nanjing, China) was added into each well, followed by incubation at 37℃ for another 2 h. The optical density value was detected at a 450 nm wavelength on a microplate reader (Thermo Fisher Scienti c, USA)..

Colony formation assay
After transfection with miR-498 mimic, the cells were trypsinized, counted, seeded for colony formation assay in 6-well plates at the density of 100 cells per well. During colony growth, the culture medium was replaced every 3 days. The colony was counted only if contained more than 50 cells, and the number of colonies was counted the 6 th day after seeding. Colony formation rate was calculated with equation; colony formation rate = number of colonies/ number of seeded cells ×100%. Each treatment was carried out in triplicate.
Flow cytometry analysis for apoptosis AGS and SGC-7901 cell lines were transfected with either NC mimic or miR-498 mimic for 48 h. After that, the cells were washed, digested and transferred into centrifuge tubes and centrifuged at 1000g for 5 min for collection. 100 μl buffer solution were added to resuspend the cells to make sure the cell concentration was 1x10 6 . 5 μl Annexin V-FITC and 10 μl propidium iodide (PI) dye were further added into each tube for 10 min. Finally, the cell apoptosis was detected within 1h by Flow cytometer. The green uorescence of FITC was detected in FL1 channel, while red uorescence of PI was in FL2 or FL3 channel and then draw the two colors dot plot, the abscissa is FITC and the ordinate is PI.
Flow cytometry analysis for cell cycle AGS and SGC-7901 cell lines were harvested, washed with 10 ml PBS with that centrifuged and resuspended to make sure the cell density at 1 × 10 6 . The cells were xed by 4.5 ml pre-chilled 70% ethanol (−20°C) and incubated at least 4 h at −20°C. After that, the cells were centrifuged for 3 min at 300 g and washed with 5 ml FACS buffer. Cells were stained with 0.5 ml Hoechst 33342 and Pyronin Y staining solution and were incubated at room temperature for 20 min and analyzed by ow cytometry. The ow cytometer was set up and adjusted with UV (355 nm) and blue (488 nm) lasers as well as proper lter sets. Doublets were excluded by creating a combination of same-channel bivariate plots utilizing Area vs. Height or Area vs. Width. Acquire the uorescence and analyze cell cycle stages of each sample.
Western blot analysis AGS and SGC-7901 cell lines transfected with NC mimic and miR-498 mimic were collected and then cracked in RIPA lysis buffer plus PMSF in low temperature. After extracted, the total protein concentration was detected by BCA assay kit (Santa Cruz, USA). Prepared protein samples were loaded in SDS-PAGE and then transferred into 0.22μm PVDF membranes and incubated with prepared antibodies, the primary antibody incubated overnight, and the secondary antibody for 2 h. Finally, enhanced chemiluminescence (Thermo Fisher Scienti c, USA) visualized this membrane. The antibody β-actin was purchased from CST (Beverly, USA). Antibodies against the Cyclin D1, CDK2, P21, Cox-2, MMP-2, MMP-9 and FOXK1 were purchased from Abcam (Shanghai, China).

Scratch assay
Making 3 horizontal lines on the back of each hole of the 6-well plates and then culturing the AGS and SGC-7901 cells in the plates for a night after transfected with NC mimic and miR-498 mimic. When cells were grown to 90% con uence, creating a single wound in each hole perpendicular to the 3 horizontal lines and then washing the cells and culturing them in 37 ℃ 5% CO 2 incubator with medium without FBS. The images were digitally photographed immediately after incubating for 0 h, 6 h, 12 h, 24 h. The original opening distances of the wound were set as 100%. The opening distances after 24 h were measured from three areas randomly selected per well, and the distances in three wells of each group were quanti ed and normalized by the original opening distance. The experiment was performed three times in triplicate, and the percentage of the migration rate was calculated by measuring the length of cell migration and expressed as a percentage compared to the control group. Migration rates = (treatment group cell migration distance/control group migration distance) × 100%.

Transwell chamber assay
Transwell chamber assay was used to detect the capacity for cell migration and invasion (8 μm diameters; Shanghai, China). AGS and SGC-7901 cells were collected and washed with PBS after transfected with NC mimic and miR-498 mimic for 48 h. Then resuspended the cells in DMEM without FBS. In total, 100 μl of suspension containing 5×10 4 cells was added into the upper chambers. The transwell chambers were placed into a 24-well plate that had already been covered with 500 μl of DMEM containing 10% FBS. 24-h later, non-migratory cells were gently removed, and the migratory cells were xed with 100% methanol, stained with 0.5% crystal violet, washed with PBS, and imaged using an inverted microscope (Olympus, Tokyo, Japan). The migratory and invasive abilities were measured by counting respectively the migratory and invading cells.

Luciferase reporter assay
The 293T cell was transfected with NC mimic and miR-498 mimic to be assessed. After culturing the 293T for 48 h we rstly removed the growth medium and rinsed the cells with PBS and then added the cell culture lysis reagent to cover the cells (250 µl for a 60 mm dish), with that incubated the 293T for a short period. Finally, we extracted the cells with luciferase assay reagent (Promega, Shanghai, China) and measured the activity of promoters. The luciferase reporter was constructed by cloning 3'UTR region of the FOXK1 WT, FOXK1 Mut (mutant of functional miR-498 binding domain), whose sequences that bind to miR-498 were partly mutated in order to identify the binding speci city.

Statistical analyses
All experiments were repeated three times and the data are presented as the mean ± standard deviation (SD) using SPSS 18.0 (SPSS, Inc., USA). An unpaired t-test was performed to compare the difference of the two groups. At the meanwhile, a one-way ANOVA followed by Bonferroni multiple comparison test was used to detect differences between the two or more groups. P < 0.05 was considered as statistically signi cant.

The expression level of miR-498 is downregulated in gastric cancer cell lines.
To detect the potential function of miR-498 in gastric cancer, we rst measured its expression in gastric cancer cell lines. RT-qPCR analysis indicated that the expression level of miR-498 was decreased in gastric cancer cell lines, including AGS, MGS-803, SGC-7901, SNU-1, when compared with the normal gastric cell GES-1 (Fig. 1A). Among these gastric cancer cells, the expression of miR-498 was lower expressed in the AGS and SGC-7901, thus we selected these cells for further investigation. To further identify the role of miR-498 in the gastric cancer progression, we overexpressed the miR-498 mimic into AGS and SGC-7901. As shown in Fig. 1B, RT-qPCR analysis revealed the miR-498 mimic upregulated in AGS and SGC-7901 cells.
2. Overexpression of miR-498 inhibits the proliferation and promotes the apoptosis of gastric cancer cells.
As shown in Fig. 2AB, the CCK-8 and colony formation assays revealed that miR-498 mimic inhibited the proliferation of AGS and SGC-7901. In order to explore the function of miR-498 on apoptosis, the ow cytometry assays were operated. After transfection with NC mimic and miR-498 mimic in AGS and SGC-7901, the miR-498 mimic increased the percentage of apoptosis of AGS and SGC-7901 cells (Fig. 2C). Meanwhile, the cell cycle was analyzed by ow cytometry. We found that the miR-498 mimic upregulated the numbers of the AGS and SGC-7901 cells in G0/G1 phase, while decreased those cell numbers in phase S and G2/M (Fig. 2D). At the molecular level, western blot assay were performed to examine the effect of miR-498 on cell cyclerelated proteins, including Cyclin D1, CDK2 and P21. Compared to NC mimic, miR-498 mimic inhibits the protein expression levels of Cyclin D1 and CDK2, whereas promotes the protein expression levels of P21 in AGS and SGC-7901 cells (Fig. 2E).

The overexpression level of miR-498 inhibits gastric cancer cells migration and invasion.
After transfection with NC mimic and miR-498 mimic, the migration and invasion of gastric cancer cells were demonstrated. The scratch assays indicated miR-498 signi cantly inhibited the migration of AGS and SGC-7901 cells (Fig. 3A). The transwell chamber assays showed miR-498 overexpression inhibited the migration and invasion of AGS and SGC-7901 cells (Fig. 3B). Western blot is also used to detect the expression level of migration and invasion-related proteins, such as Cox-2, which belongs to the cyclooxygenase (COX) enzymes family and has a crucial role in angiogenesis, invasion and immune suppression [18] , as well as the MMP-2 and MMP-9, which are the members of matrix metalloproteinases (MMPs), which are zinc-dependent endopeptidases involved in the invasion or migration [19] . As shown in Fig. 3C, miR-498 mimic inhibited the expression of Cox-2, MMP-2 and MMP-9 both in AGS and SGC-7901 cells.

miR-498 targets on FOXK1.
MicroRNA targets are recognized through pairing between the miRNA seed region and complementary sites within target mRNAs [20] . By using the StarBase, the bioinformatics analysis indicated that FOXK1 possessed the potential binding sites of miR-498 (Fig. 4A). Next, the luciferase reporter assay is carried out to investigate the effect of miR-498 on FOXK1 expression. The results indicated miR-498 mimic suppressed the luciferase activity of FOXK1 WT, whereas exhibited modest impact on the FOXK1 MUT in 293T cell lines (Fig. 4B). Coincidence with these results, RT-qPCR and western blot analyses indicated that miR-498 mimic signi cantly inhibited the expression of FOXK1 in AGS and SGC-7901 cell lines ( Fig. 4C and D).

FOXK1 mediates the effect of miR-498 on the gastric cancer progression
After infection with the lentivirus expressing FOXK1, the RT-qPCR revealed the expression levels of FOXK1 were increased AGS and SGC-7901 cells (Fig.  5A). CCK-8 and Colony formation assays indicated that miR-498 mimic inhibited the proliferation of AGS and SGC-7901 cell lines, while FOXK1 partially blocked such inhibition ( Fig. 5B and C). In addition, the apoptosis was detected by ow cytometry analysis. The results showed that miR-498 mimic promoted the apoptosis of AGS and SGC-7901 cell lines, while transfected with the FOXK1 expressing lentivector the promotion was antagonized (Fig.  5D). Besides, the transwell chamber assays described miR-498 mimic inhibited the cell migration in gastric cancer cells, whereas the cell migration were increased in the miR-498 mimic combined the FOXK1 expressing lentivector group, when compared with miR-498 mimic combined with NC lentivector group (Fig.5 E). All the results suggested that the FOXK1 is the target of miR-498, which mediates the inhibitory effect of miR-498 on the gastric cancer progression in vitro.

Discussion
Due to its high morbidity and mortality, gastric cancer raises the broad concern all over the world and is considered to be the deadly malignancy [21] . Patients diagnosed with early gastric cancer could be treated by surgery. For advanced gastric cancer, chemotherapy and targeted therapy are the main methods. Although many treatments are developed to cure gastric cancer, the recurrence, metastasis, and no response to chemotherapy, as well as the drug resistance, cause the unsatisfactory prognosis [22] . Except for the main helicobacter pylori, the incidence of gastric cancer is relevant to many factors. Among which, miRNA plays its role as potential biomarkers and therapeutic targets in gastric cancer. For instance, miRNA-21 promotes the growth of gastric cancer cells by adjusting and controlling PTEN/Akt signal passage via regulation of PEG2 [23] . miRNA-103a-3p promotes human gastric cancer cell proliferation by targeting and suppressing ATF7 in vitro [24] . MiR-136 promotes apoptosis in gastric cancer cells by targeting AEG-1 and BCL2 [25] . The present study devoted to identify the roles of miR-498 on gastric cancer. Our results revealed that miR-498 acted as a repressor in gastric cancer cell proliferation, migration, invasion and apoptosis. The evidence emphasized the expression levels of miR-498 were lower in gastric cancer cell lines compared with the gastric epithelial cell line GES-1. The overexpression of miRNA-498 could inhibit proliferation, migration, invasion, and promotes apoptosis in AGS and SGC-7901 cells.
MiRNAs play essential roles by ne-tuning their target genes functionally associated with host genes [26] . In this study, we revealed FOXK1 is the target of miR-498. Forkhead box K1 (FOXK1) is a member of the FOX transcription factor family, which comprises diverse tissue-and cell type-speci c transcription factors that are generally considered to be important regulators in the physiological development and the driving force for evolution. FOX proteins are multifaceted transcription factors involved in the development of the nervous system, kidneys, lungs, hair and immune systems. At the meanwhile, they are also associated with cancer. The FOX family promotes the development, maintenance, progression and metastasis of cancer at different regulatory levels through a highly complex and extensive network [27] .
Previous works described FOXK1 played its important roles in proliferation, migration and invasion in several cancers [28] . For example, LINC01503 promotes colorectal cancer progression via acting as a competing endogenous RNA for miR-4492/FOXK1 [29] . FOXK1 facilitates cell proliferation through regulating the expression of p21 and promotes metastasis in ovarian cancer [30] . MCM3AP-AS1 promoted growth and migration through modulating miR-138-5p/FOXK1 axis in pancreatic cancer, providing insights into MCM3AP-AS1/miR-138-5p/FOXK1 axis as novel candidates for pancreatic cancer therapy [31] . In gastric cancer, several studies demonstrated that FOXK1 was a target for some certain miRNAs (e.g. MiR-646, miR-4492, miR-137) which could inhibit the gastric cancer progression by mediating the cell proliferation, migration and invasion [32][33][34] . Given that the gastric cancer progression was correlated with the expression levels of miR-498 and FOXK1, our study also demonstrated that the miRNA-498 targets on FOXK1 and negatively regulates its expression in the gastric cancer cells. More importantly, overexpression of FOXK1 partially retarded the inhibitory effect of miR-498 on the gastric cancer cell proliferation and migration, thus providing a novel mechanism through which miR-498 regulates the gastric cancer progression and improving the posttranscriptional network between miRNAs and target genes in gastric cancer.
In conclusion, the present study demonstrated that miR-498 could inhibit the proliferation and migration/invasion, while promote the apoptosis of gastric cancer cells via the inhibition of FOXK1. Thus, miR-498/FOXK1 axis might be used as a diagnosed biomarker or therapeutic target for the gastric cancer and its associated diseases.

Declarations
Ethics approval and consent to participate Not applicable.
Consent for publication   mimic. Asterisks indicated signi cant differences from the control (*P < 0.05, ** P<0.01, compared to GES-1group, #P < 0.05 and ## P < 0.01, compared to NC mimic group). All the data were presented as mean±SD. for cell cycle-related proteins. Asterisks indicated signi cant differences from the control (* P < 0.05, ** P<0.01, compared to Control ,# P < 0.05, ## P<0.01, compared to NC mimic group).). All the data were presented as mean±SD. for cell cycle-related proteins. Asterisks indicated signi cant differences from the control (* P < 0.05, ** P<0.01, compared to Control ,# P < 0.05, ## P<0.01, compared to NC mimic group).). All the data were presented as mean±SD. and invasion related proteins. Asterisks indicated signi cant differences from the control (* P < 0.05, ** P<0.01, compared to NC mimic group). All the data were presented as mean±SD.     expression level of FOXK1 in AGS and SGC-7901 cells. Asterisks indicated signi cant differences from the control (* P < 0.05, ** P < 0.01, compared to NC mimic group). All the data were presented as mean±SD.