DNMT1 promotes the proliferation and migration of gastric cancer cells by inducing microRNA-125a-5p methylation to promote SERPINE1 protein

Abstract

BACKGROUND

Gastric cancer (GC) is a malignant tumor originating from gastric mucosal epithelial cells that has high morbidity and mortality. microRNAs (miR) are important diagnostic markers and therapeutic targets in this disease.

AIM

To explore the mechanism of miR-125a-5p in the pathogenesis of GC.

METHODS

The expression levels of miR-125a-5p, SERPINE1 and DNMT1 in GC cells and tissues were detected by real-time polymerase chain reaction (PCR) and Western blotting. Methylation-specific PCR was used to detect the level of miR-125a-5p methylation. A cell counting kit 8 assay, scratch test, and a Transwell assay were performed to detect the proliferation, migration, and invasiveness of HGC27 cells, respectively. The expression of the epithelial mesenchymal transition (EMT)-related proteins E-cadherin, N-cadherin and vimentin in HGC27 cells was detected by Western blotting, while the expression of vimentin was detected by immunofluorescence.

RESULTS

This study revealed that miR-125a-5p was expressed at low levels in GC clinical samples and cells and that miR-125a-5p overexpression inhibited the proliferation, migration, invasiveness and EMT of GC cells. Mechanistically, miR-125a-5p can reduce GC cell proliferation, promote E-cadherin expression, inhibit N-cadherin and vimentin expression, and reduce the EMT of GC cells, thus constraining GC cells to a certain extent. Moreover, DNMT1 inhibited miR-125a-5p expression by increasing the methylation of the miR-125a-5p promoter, thereby promoting the expression of SERPINE1, which acts together with miR-125a-5p to exert antagonistic effects on GC.

CONCLUSION

Our study revealed that DNMT1 promoted SERPINE1 protein expression by inducing miR-125a-5p methylation, which led to the proliferation, migration and occurrence of EMT in GC cells.

Key Words: Gastric cancer; microRNA-125a-5p; DNA methyltransferase 1; SERPINE1; Methylation

Core Tip: This study revealed the tumor suppressive effect of microRNAs (miR)-125a-5p in gastric cancer (GC) through inhibition of cell proliferation, migration and invasion, and regulation of epithelial-mesenchymal transition related protein expression. The low expression of miR-125a-5p is associated with DNMT1-mediated promoter methylation and is negatively correlated with SERPINE1 expression. This discovery provides a new target for GC therapy.

INTRODUCTION

Gastric cancer (GC) is the fifth most common cancer and the third leading cause of cancer-related death worldwide[1]. Although the morbidity and mortality of GC have continued to decline in recent years, the mortality rate in China is still high[2]. The pathogenesis of GC is associated with bacterial infections, dietary habits, genetic factors and environmental factors[3]. Due to the lack of typical and specific symptoms, many patients are diagnosed at an advanced stage when distant metastasis has already occurred, and consequently, they are often not eligible for surgical resection. The only treatment option for these patients is chemotherapy, which generally leads to a poor prognosis, with a five-year survival rate of less than 10%[4]. Therefore, to improve the quality of life and survival of patients with GC, elucidation of the underlying molecular mechanisms is urgently needed to identify new effective therapeutic targets.

DNA hypermethylation of CpG sequences in promoters is a clear epigenetic feature of GC that leads to the transcriptional silencing of tumor suppressor genes and other cancer-related genes[5]. DNA methylation is regulated mainly by DNA methyltransferase (DNMT), which transfers a methyl group at the fifth carbon position of cytosine residues within DNA[6]. As an important member of the DNMT family, DNMT1 plays a role in maintaining methylation during DNA replication[7]. Recent studies have shown that high DNMT1 expression promotes the occurrence and development of GC[8,9]. However, the specific mechanism by which DNMT1 participates in GC is still poorly understood.

DNMT1 also regulates the expression of microRNAs (miRs)[10]. miRs are a large class of regulatory RNAs that repress target gene expression by base pairing with the complementary site of the 3’ untranslated region (UTR), thereby promoting mRNA decay and translation inhibition[11]. A growing body of evidence indicates that abnormal expression of miRs, such as miR-146a[12] and miR-452[13], is associated with GC progression. In addition, studies have shown that miR-125a-5p is expressed at significantly lower levels in GC patients and is associated with poor prognosis[14]. Moreover, miR-125a-5p can inhibit the invasiveness and metastasis of GC cells[15]. However, the specific mechanism by which miR-125a-5p affects the progression of GC has not been fully elucidated.

In addition, SERPINE1, a member of the serine protease inhibitor family, is a key regulator of the plasminogen/plasmin system[16]. Previous studies have indicated that SERPINE1 has proangiogenic, growth, migration-stimulating and antiapoptotic effects; these activities are all targeted and promote the growth, survival and metastasis of cancer cells[17]. Studies have shown that SERPINE1 is highly expressed in GC tissues and that the overexpression of SERPINE1 can promote the malignant progression and poor prognosis of GC patients[18]. Similarly, Teng et al[19] reported that SERPINE1 can promote GC progression and angiogenesis by activating the epidermal growth factor receptor-2 signaling pathway. Therefore, this study aimed to identify potentially relevant biological pathways related to the role of SERPINE1 in the development of GC.

In summary, the purpose of this study was to investigate the mechanism of the DNMT1/miR-125a-5p/SERPINE1 signaling axis in GC, which will help us better understand the development of GC and provide new ideas for the treatment of this disease.

MATERIALS AND METHODS

Collection of clinical samples

Twenty-six pairs of GC tissues and adjacent normal tissues were collected from participants at the hospital, after which fresh tissues were subsequently snap-frozen in liquid nitrogen and stored at -80 °C until use. Informed consent was obtained from all individual participants prior to the start of all study-related procedures. This study was conducted in accordance with the Declaration of Helsinki (as revised in 2013) and was approved by the Ethics Committee of Yan’an Hospital of Kunming City/Yan’an Hospital Affiliated to Kunming Medical University (No. 2020-078-01).

Cell culture and transfection

The normal human gastric epithelial cell line GES-1 and the GC cell lines HGC27, MKN-45 and AGS were obtained from Otwo Biotech (Shenzhen, China). GES-1 cells were cultured in Dulbecco’s modified eagle medium (Gibco, United States), while HGC27, MKN-45 and AGS cells were cultured in RPMI 1640 medium (Gibco, United States) supplemented with 10% fetal calf serum (Gibco, United States). The cells were cultured in an incubator at 37 °C and 5% carbon dioxide (CO₂). Furthermore, to detect methylation regulation in GC cells, we cocultured HGC27 cells with different concentrations of the methylating agent 5-aza-2’-deoxycytidine (5-aza-dC) for 24 hours. HGC27 cells were cultured in 24-well plates overnight, and when the cell density reached 60%-70%, the cells were incubated with Lipofectamine 3000 reagents (Invitrogen, Grand Island, NY, United States) according to the manufacturer’s instructions. Negative control (NC) mimic, miR-125a-5p mimic, NC inhibitor, miR-125a-5p inhibitor, overexpression NC (OE-NC), OE-DNMT1, small interfering (si)-NC, si-DNMT1 and si-SERPINH1 were transfected into HGC27 cells. The cells were incubated at 37 °C in a 5% CO₂ incubator for 48 hours, after which the transfection efficiency was determined.

Real-time polymerase chain reaction

Total RNA from clinical samples and GES-1, HGC27, MKN-45, and AGS cells was extracted via TRIzol reagent (Invitrogen, 15596026). The RNA was reverse-transcribed into cDNA using a first strand cDNA synthesis kit (Genenode, China). Real-time polymerase chain reaction (PCR) was performed using the SYBR green real-time PCR kit (Solarbio, China) with U6 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as internal controls, and the results were calculated according to the 2^-ΔΔCt method. The primer sequences are shown in Table 1.

Table 1 Primer sequences.

Genes	Sequence
miR-125a-5p	F: 5´-GGAGGCGGGAGCCGGC-3´
	R: 5´-AGTGCAGGGTCCGAGGTATT-3´
DNMT1	F: 5´-TACCAGGGAGAAGGACAGGG-3´
	R: 5´-ACACCTCACAGACGCCACAT-3´
U6	F: 5´-TGCGGGTGCTCGCTTCGGCAGC-3´
	R: 5´-CCAGTGCAGGGTCCGAGGT-3´
GAPDH	F: 5´-GGGAAACTGTGGCGTGAT-3´
	R: 5´-AAAGGTGGAGGAGTGGGT-3´

F: Forward primer; R: Reversed primer.

Methylation-specific PCR

To detect the methylation level of the miR-125a-5p promoter region, DNA from GC tissues and GES-1, HGC27, MKN-45, and AGS cells was isolated via DNA-EZ Reagents V All-DNA-Out (Sangon Biotech, China). The genomic DNA was incubated with reagents from a Methylamp one-step DNA modification kit (Epigentek) according to the manufacturer’s instructions. PCR was then performed via HotStar Taq polymerase (Qiagen, Germany).

Cell counting kit 8 assay

HGC27 cells (5 × 10³ cells/well) were seeded in 96-well plates and cultured at 37 °C in a 5% CO₂ incubator for 24 hours. The cells in each group were treated according to each group’s designated treatment. After treatment, 10 μL of cell counting kit 8 (CCK-8) reagent (C0037, Beyotime, China) was added to each well, after which the absorbance of each well was measured at 450 nm with a microplate reader.

Scratch test

HGC27 cells in the logarithmic growth phase were used in this assay. After the cells were digested and routinely passaged, they were seeded in 24-well plates. When the cell density reached 90%, the cells were scratched with a pipette tip. After the cells were cultured for 24 hours, migrating cells were observed and imaged under a microscope.

Transwell assay

Cell invasion experiments were performed in Transwell plates (Corning, United States). The density of HGC27 cells in each group was adjusted in serum-free Roswell Park Memorial Institute (RPMI) 1640 medium to 1 × 10⁵ cells/mL. Two hundred microliters of cell suspension was added to the upper chamber of a Matrigel-coated Transwell chamber. In the lower chamber of a 24-well plate, 600 μL of RPMI 1640 medium supplemented with 10% fetal calf serum was added. After further culture for 24 hours, the cells in the lower chamber were fixed in 4% paraformaldehyde, stained with crystal violet (Solarbio, China), and examined under an inverted microscope. The number of cells at fixed positions in each well was observed, and five fields were selected for counting and imaging.

Western blot analysis

Protein was extracted from GES-1 and HGC27 cells using radio immunoprecipitation assay lysis buffer (Sigma-Aldrich, United States) supplemented with 1% protease inhibitors. The protein concentration was determined according to the instructions of the bicinchoninic acid detection kit (Thermo Scientific, United States). Total proteins were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis, transferred to polyvinylidene fluoride membranes (Millipore, United States) and then blocked with 5% skim milk powder at room temperature for 1.5 hours. The membranes were then incubated at 4 °C overnight with diluted primary antibodies against the following: E-cadherin (1:1000, ab231303, Abcam, United Kingdom), N-cadherin (1:2000, ab18203, Abcam, United Kingdom), vimentin (1:1000, ab92547, Abcam, United Kingdom), DNMT1 (1:2000, ab188453, Abcam, United Kingdom), SERPINH1 (1:1000, ab109117, Abcam, United Kingdom), and GAPDH (1:1000, ab181602, Abcam). Next, the membranes were incubated with the secondary antibody (1:4000, ab97051, Abcam, United Kingdom) at room temperature for 1 hour and then developed with an enhanced chemiluminescence kit (Millipore, United States). Finally, semiquantitative analysis of the bands was performed via Image J software.

Immunofluorescence staining

HGC27 cells were seeded in 24-well plates (2 × 10⁴ cells/well). After 24 hours, the cells were washed twice with phosphate-buffered saline, fixed in 4% paraformaldehyde for 30 minutes, permeabilized with 0.5% Triton X-100 for 10 minutes, and blocked with bovine serum albumin for 1 hour. Subsequently, the cells were incubated with primary antibodies against E-cadherin (1:200, ab231303, Abcam, United Kingdom), N-cadherin (1:200, ab18203, Abcam, United Kingdom), and vimentin (1:200, ab92547, Abcam, United Kingdom) at 4 °C overnight. The next day, the cells were incubated with the corresponding secondary antibody for 1 hour and stained with 4’,6-diamidino-2-phenylindole. Finally, the stained cells were observed, and images were captured under a fluorescence microscope (400857, Nikon, Japan).

Immunohistochemistry

Paraffin-embedded human GC tissues and adjacent tissues were deparaffinized and rehydrated, and antigen retrieval was subsequently performed in 0.01 M citrate buffer (potential of hydrogen = 6.0). The slides were incubated with antibodies against DNMT1 (1:500, ab188453, Abcam, United Kingdom) and SERPINH1 (1:200, ab109117, Abcam, United Kingdom) at 4 °C overnight, after which immunodetection was performed using a horseradish peroxidase secondary antibody and diaminobenzidine chromogenic reagent. The results were observed under a fluorescence microscope (400857, Nikon, Japan).

Dual-luciferase experiment

The targeted binding sites of miR-125a-5p and SERPINH1 were predicted using an online bioinformatics platform (http://starbase.sysu.edu.cn/). The SERPINH1 3’-UTR containing the miR-125a-5p binding site was cloned and inserted into the pGL3 vector (Promega) to construct the wild-type (WT) SERPINH1 vector. Mutant SERPINH1 (MUT) vectors were generated via a site-directed mutagenesis kit (Stratagene, United States). WT or MUT and the miR-125a-5p mimic or NC (NC mimic) were cotransfected into 293T cells using Lipofectamine 3000. After 48 hours, luciferase activity was detected with a dual-luciferase reporter assay system (Promega).

Statistical analysis

The data were analyzed via GraphPad Prism version 8 software (GraphPad, United States). All the experiments were repeated at least 3 times. Statistical analysis was performed via one-way analysis of variance and a t test. P < 0.05 indicated statistical significance.

RESULTS

Expression and methylation levels of miR-125a-5p in GC tissues and GC cell lines

We first detected the expression of and the methylation level of the miR-125a-5p promoter in GC. Compared with that in adjacent tissues, the expression level of miR-125a-5p was significantly lower in GC tissues (Figure 1A). The methylation level of the miR-125a-5p promoter was also significantly increased in GC tissues (Figure 1B). Similarly, compared with that in GES-1 cells, the expression level of miR-125a-5p in the GC cell lines HGC27, MKN-45 and AGS was significantly downregulated (Figure 1C), but its promoter methylation level was significantly upregulated (Figure 1D). These results indicated that in GC tissues and cells, the miR-125a-5p expression level was significantly reduced, but the methylation level was increased.

Figure 1 The expression and methylation levels of microRNA-125a-5p in gastric cancer tissues and gastric cancer cell lines. A: MicroRNA (miR)-125a-5p expression in clinical samples detected by real-time polymerase chain reaction (RT-qPCR); B: MiR-125a-5p promoter methylation level in clinical samples detected by methylation-specific polymerase chain reaction; C: RT-qPCR detection of miR-125a-5p expression in cells; D: Methylation-specific polymerase chain reaction detection of the miR-125a-5p promoter methylation level in cells. ^cP < 0.001. miR: MicroRNA; GC: Gastric cancer; M: Methylation; U: Unmethylated.

Overexpression of miR-125a-5p inhibits the proliferation, migration, invasiveness and epithelial mesenchymal transition of GC cells

Next, the effect of miR-125a-5p overexpression on the malignant biological behavior of GC cells was detected. We first overexpressed miR-125a-5p in HGC27 cells and found that the expression level of miR-125a-5p was significantly increased in the miR-125a-5p mimic group (Figure 2A). The CCK-8 assay revealed that, compared with the NC mimic, the overexpression of miR-125a-5p significantly inhibited the proliferation of HGC27 cells (Figure 2B). Furthermore, the results of the scratch and Transwell assays revealed that miR-125a-5p overexpression inhibited HGC27 cell migration and invasion (Figure 2C and D). The results of the Western blot analysis revealed that, compared with the NC mimic group, the miR-125a-5p mimic group presented significantly increased expression of E-cadherin and decreased expression of N-cadherin and vimentin (Figure 2E). Finally, immunofluorescence staining also revealed that, compared with that in the NC mimic group, the overexpression of miR-125a-5p significantly inhibited vimentin expression (Figure 2F). These results indicated that the overexpression of miR-125a-5p could significantly inhibit HGC27 cell proliferation, migration, invasion and epithelial mesenchymal transition (EMT).

Figure 2 Overexpression of microRNA-125a-5p inhibits gastric cancer cell proliferation, migration, invasion and epithelial mesenchymal transition. A: Transfection efficiency of the microRNA (miR)-125a-5p mimic in HGC27 cells, as detected by real-time polymerase chain reaction (RT-qPCR); B: Proliferation of HGC27 cells detected by cell counting kit 8 assay; C: Migration of HGC27 cells detected by scratch experiments; D: HGC27 cell invasion detected by Transwell assays; E: Western blot detection of the expression of the epithelial mesenchymal transition (EMT)-related proteins E-cadherin, N-cadherin and vimentin in HGC27 cells; F: Immunofluorescence detection of the expression of the EMT-related protein vimentin in HGC27 cells (scale bar = 10 μm). ^bP < 0.01. ^cP < 0.001. EMT: Epithelial mesenchymal transition; NC: Normal control; miR: MicroRNA; OD: Optical density; DAPI: 4’,6-diamidino-2-phenylindole; miR: MicroRNA; GAPDH: Glyceraldehyde-3-phosphate dehydrogenase.

DNMT1 inhibits miR-125a-5p expression through methylation

DNA methylation is an important mechanism in the regulation of gene expression, and it has been reported that DNA methylation can regulate the expression of approximately 50% of miRs in the pathogenesis of many diseases[20]. Therefore, we explored whether the decreased expression of miR-125a-5p in GC was caused by DNMT1. First, DNMT1 expression in GC was observed to be upregulated in GC tissues compared with adjacent tissues (Figure 3A and B). Similarly, DNMT1 was significantly more highly expressed in HGC27 cells than in GES-1 cells (Figure 3C). Next, we confirmed that the low expression of miR-125a-5p in GC was related to methylation by adding the methylation inhibitor 5-aza-dC. First, the cytotoxicity of different concentrations of 5-aza-dC (0, 2.5, 5, 10, and 15 μmol/L) in HGC27 cells was tested. We found that when the concentration was greater than 10 μmol/L, the cells began to exhibit significant signs of cytotoxicity (Figure 3D). Therefore, HGC27 cells were treated with 5-aza-dC at a concentration not exceeding 10 μmol/L for 24 hours, after which the expression of miR-125a-5p was detected via real-time polymerase chain reaction. The results revealed that the inhibition of methylation could restore miR-125a-5p expression in a concentration-dependent manner (Figure 3E). DNMT1 was subsequently overexpressed or knocked down in HGC27 cells (Figure 3F). DNMT1 overexpression significantly increased the methylation level of the miR-125a-5p promoter, whereas DNMT1 knockdown decreased the methylation level of the miR-125a-5p promoter (Figure 3G). In addition, DNMT1 overexpression significantly inhibited miR-125a-5p expression, whereas DNMT1 knockdown promoted miR-125a-5p expression (Figure 3H). These results indicated that DNMT1 could repress miR-125a-5p expression in GC by increasing the methylation level of the miR-125a-5p promoter.

Figure 3 DNMT1 represses microRNA-125a-5p expression through methylation. A: DNMT1 expression in clinical samples detected by real-time polymerase chain reaction (RT-qPCR); B: Positive expression of DNMT1 in clinical samples detected by immunohistochemistry (scale bar = 50 μm); C: Western blot analysis of DNMT1 expression in HGC27 cells; D: Toxicity of 5-aza-2’-deoxycytidine in HGC27 cells; E: RT-qPCR detection of microRNA (miR)-125a-5p expression in HGC27 cells after treatment with different concentrations of 5-aza-2’-deoxycytidine; F: Western blot detection of DNMT1 transfection efficiency in HGC27 cells; G: Methylation of miR-125a-5p in HGC27 cells detected by methylation-specific polymerase chain reaction; H: RT-qPCR detection of miR-125a-5p expression in HGC27 cells. ^cP < 0.001. OE-NC: Overexpression negative control; OE-DNMT1: DNMT1 overexpression; si-NC: Small interfering-negative control; si-DNMT1: Small interfering DNMT1; 5-aza-dC: 5-aza-2’-deoxycytidine; miR: MicroRNA; GAPDH: Glyceraldehyde-3-phosphate dehydrogenase; GC: Gastric cancer; M: Methylation; U: Unmethylated.

DNMT1 affects GC cell proliferation, migration, invasion and EMT by mediating miR-125a-5p methylation

To explore whether DNMT1 affects the malignant biological behavior of GC cells by mediating miR-125a-5p methylation, we overexpressed DNMT1 in HGC27 cells and/or treated the cells with 10 μmol/L 5-aza-dC. Western blot analysis revealed that DNMT1 overexpression significantly promoted DNMT1 expression compared with the OE-NC, whereas further addition of 5-aza-dC inhibited DNMT1 expression (Figure 4A). Compared with the OE-NC, DNMT1 overexpression significantly increased the methylation level of the miR-125a-5p promoter, whereas the addition of 5-aza-dC decreased the methylation level of the miR-125a-5p promoter (Figure 4B). Moreover, the inhibitory effect of DNMT1 overexpression on miR-125a-5p expression was reversed by the addition of 5-aza-dC (Figure 4C). We then assessed HGC27 cell proliferation, migration and invasion. Compared with the OE-NC, the overexpression of DNMT1 significantly promoted HGC27 cell proliferation, migration and invasion, while the further addition of 5-aza-dC inhibited these malignant biological behaviors in HGC27 cells (Figure 4D-F). The results of the Western blot analysis revealed that, compared with that in the OE-NC group, the overexpression of DNMT1 significantly inhibited E-cadherin expression and promoted the expression of N-cadherin and vimentin. Further addition of 5-aza-dC partially reversed the effect of DNMT1 overexpression (Figure 4G). Finally, immunofluorescence was used to detect the expression of the EMT-related protein vimentin in HGC27 cells, and the results were consistent with those of the Western blot analysis (Figure 4H). These results indicated that DNMT1 could inhibit miR-125a-5p expression by increasing the methylation level of miR-125a-5p, thereby promoting the proliferation, migration, invasiveness and EMT of GC cells.

Figure 4 DNMT1 affects gastric cancer cell proliferation, migration, invasion and epithelial mesenchymal transition by mediating microRNA-125a-5p methylation. A: DNMT1 expression in HGC27 cells detected by Western blot; B: Methylation-specific polymerase chain reaction detection of the microRNA (miR)-125a-5p promoter methylation level in HGC27 cells; C: MiR-125a-5p expression in HGC27 cells detected by real-time polymerase chain reaction; D: Detection of HGC27 cell proliferation by cell counting kit 8 assay; E: HGC27 cell migration detected by scratch test; F: HGC27 cell invasion determined by Transwell assay; G: Western blot detection of the expression of the epithelial mesenchymal transition-related proteins E-cadherin, N-cadherin and vimentin in HGC27 cells; H: Immunofluorescence detection of the expression of the epithelial mesenchymal transition-related protein vimentin in HGC27 cells (scale bar = 10 μm). ^aP < 0.05. ^bP < 0.01. ^cP < 0.001. OE-NC: Overexpression negative control; OE-DNMT1: DNMT1 overexpression; 5-aza-dC: 5-aza-2’-deoxycytidine; miR: MicroRNA; GAPDH: Glyceraldehyde-3-phosphate dehydrogenase; OD: Optical density; M: Methylation; U: Unmethylated; DAPI: 4’,6-diamidino-2-phenylindole; PBS: Phosphate-buffered saline.

DNMT1 regulates SERPINH1 expression through miR-125a-5p

Previous studies have shown that, as a member of the serine protease inhibitor family, SERPINE1 also promotes the malignant progression and poor prognosis of GC[18]. Therefore, using immunohistochemistry, we found that SERPINE1 was upregulated in GC tissues compared with adjacent noncancerous tissues (Figure 5A). Similarly, SERPINE1 expression was higher in HGC27 cells than in GES-1 cells (Figure 5B). In addition, we predicted the targeted binding site of miR-125a-5p using an online bioinformatics platform (http://starbase.sysu.edu.cn/) and found that SERPINE1 was a downstream target of miR-125a-5p (Figure 5C). A dual-luciferase gene reporter experiment confirmed that miR-125a-5p could regulate SERPINE1 expression (Figure 5D). Next, miR-125a-5p was knocked down in HGC27 cells, which significantly reduced miR-125a-5p expression (Figure 5E). Moreover, the Western blot analysis revealed that miR-125a-5p overexpression significantly inhibited SERPINE1 expression and that miR-125a-5p knockdown significantly promoted SERPINE1 expression (Figure 5F). Overall, compared with the OE-NC, DNMT1 overexpression significantly inhibited miR-125a-5p expression and promoted SERPINE1 expression, while further overexpression of miR-125a-5p partially reversed the effect of DNMT1 overexpression (Figure 5G and H). These results indicated that DNMT1 could promote the expression of SERPINH1 through the inhibition of miR-125a-5p expression.

Figure 5 DNMT1 regulates SERPINH1 expression through microRNA-125a-5p. A: Positive expression of SERPINH1 in clinical samples detected by immunohistochemistry (scale bar = 50 μm); B: Western blot detection of SERPINH1 expression in cells; C: A bioinformatics platform (http://starbase.sysu.edu.cn/) was used to predict the target binding site of microRNA (miR)-125a-5p; D: Real-time polymerase chain reaction (RT-qPCR) verification of the targeted binding of miR-125a-5p and SERPINE1; E: RT-qPCR was used to detect the efficiency of miR-125a-5p knockdown in HGC27 cells; F: Western blot detection of SERPINH1 expression in HGC27 cells; G: RT-qPCR detection of miR-125a-5p expression in HGC27 cells; H: Western blot detection of SERPINH1 expression in HGC27 cells. ^aP < 0.05. ^bP < 0.01. ^cP < 0.001. WT: Wild-type; MUT: Mutant; NC: Negative control; miR: MicroRNA; GC: Gastric cancer; OE-NC: Overexpression negative control; OE-DNMT1: DNMT1 overexpression.

SERPINH1 knockdown attenuates the effects of DNMT1 on GC cell proliferation, migration, invasion and EMT

Finally, we detected the effects of SERPINH1 on GC cell proliferation, migration, invasion and EMT. We first knocked down SERPINH1 in HGC27 cells (Figure 6A) and observed that DNMT1 overexpression significantly promoted SERPINH1 expression, while SERPINH1 knockdown inhibited SERPINH1 expression (Figure 6B). The CCK-8 assay revealed that, compared with the OE-NC, overexpression of DNMT1 significantly increased the proliferation of HGC27 cells, and further knockdown of SERPINH1 reduced the proliferation of HGC27 cells (Figure 6C). Compared with the OE-NC group, DNMT1 overexpression promoted HGC cell migration and invasion, and further SERPINH1 knockdown inhibited HGC cell migration and invasion (Figure 6D and E). The Western blot analysis revealed that, compared with that in the OE-NC group, the overexpression of DNMT1 significantly inhibited E-cadherin expression and promoted the expression of N-cadherin and vimentin; further knockdown of SERPINH1 partially reversed the effect of DNMT1 overexpression (Figure 6F). Finally, immunofluorescence staining revealed that, compared with that in the OE-NC group, overexpression of DNMT1 significantly promoted the expression of vimentin, and further knockdown of SERPINH1 inhibited vimentin expression (Figure 6G). These results indicated that SERPINH1 knockdown could attenuate the promoting effect of DNMT1 on GC cell proliferation, migration, invasion and EMT.

Figure 6 SERPINH1 knockdown attenuates the effects of DNMT1 on gastric cancer cell proliferation, migration, invasion and epithelial mesenchymal transition. A: Western blot detection of SERPINH1 transfection efficiency in HGC27 cells; B: Western blot detection of SERPINH1 expression in HGC27 cells; C: Proliferation activity of HGC27 cells detected by cell counting kit 8 assay; D: HGC27 cell migration detected by scratch experiments; E: HGC27 cell invasion detected by Transwell assay; F: Expression of the epithelial mesenchymal transition-related proteins E-cadherin, N-cadherin and vimentin in HGC27 cells, as detected by Western blot; G: Expression of the epithelial mesenchymal transition-related protein vimentin in HGC27 cells detected by immunofluorescence (scale bar = 10 μm). ^aP < 0.05. ^bP < 0.01. ^cP < 0.001. OE-NC: Overexpression negative control; OE-DNMT1: DNMT1 overexpression; si-NC: Small interfering-negative control; si-SERPINH1: Small interfering SERPINH1; GAPDH: Glyceraldehyde-3-phosphate dehydrogenase; OD: Optical density; DAPI: 4’,6-diamidino-2-phenylindole.

DISCUSSION

GC is a malignant tumor originating from gastric mucosal epithelial cells that has a wide global incidence and is a serious threat to human life and health[21]. Therefore, the identification of reliable diagnostic markers of GC is still an important research focus. Recent studies have shown that, as key regulators of various biological and pathological processes, miRs are extensively involved in the occurrence and development of GC[22,23]. miR-125a-5p is an important tumor suppressor whose expression is decreased in many types of human cancers, including lung cancer[24] and cervical cancer[25]. Importantly, Cao et al[15] reported that miR-125a-5p can inhibit the invasiveness and metastasis of GC cells by promoting BRMS1 expression. Similarly, Xu et al[26] reported that miR-125a-5p could inhibit the proliferation, migration and invasiveness of GC cells by regulating E2F3 expression. Therefore, we also investigated miR-125a-5p expression in GC and its potential regulatory mechanisms. We found that the expression of miR-125a-5p was downregulated in GC tissues and cells and that the overexpression of miR-125a-5p could reduce the proliferation of GC cells, promote the expression of E-cadherin, and inhibit the expression of N-cadherin and vimentin. E-cadherin is an epithelial cell adhesion protein involved in cancer cell invasion and tumor metastasis[27]. Inhibition of E-cadherin leads to an increase in the mesenchymal cell markers N-cadherin and vimentin[28]. This dysregulation of E-cadherin stimulates the initiation of EMT signaling, which can induce the development and progression of cancers, including GC[29]. In addition, studies have reported that downregulation of E-cadherin in GC can lead to gastric epithelial cell dysfunction and the development of GC[30]. It can be concluded that miR-125a-5p can reduce the occurrence of EMT in GC cells by promoting expression of the E-cadherin protein, thereby alleviating GC progression. The detection of abnormal changes in miR-125a-5p and E-cadherin is expected to be clinically applicable to the diagnosis, prognosis and treatment of GC. These genes can be used as potential targets for GC treatment and can serve as biomarkers for prognostic evaluation. Therefore, we continued to explore the specific mechanisms underlying the decrease in miR-125a-5p expression and the inhibition of the malignant biological behavior of GC cells by miR-125a-5p.

Epigenetic changes, which play an important role in the silencing of key tumor suppressor genes and the activation of oncogenes that lead to carcinogenesis, are not accompanied by changes in the DNA sequence[31]. Various biochemical pathways essential for tumorigenesis are regulated by epigenetic phenomena, such as histone modifications, nucleosome remodeling and DNA methylation[32,33]. Furthermore, DNA methylation of promoter CpG islands has been shown to be a key mediator of the downregulation of miRNA expression[34]. DNMT1 is a key gene involved in methylation maintenance during DNA replication[35]. Qiao et al[36] reported that DNMT1 represses miR-30b-5p expression by increasing the methylation of the miR-30b-5p promoter, thereby promoting the migration of GC cells and accelerating the development of GC. Importantly, the MethPrimer bioinformatics results revealed that the miR-125a promoter contains CpG islands[37]. Therefore, we first detected the methylation level of the miR-125a-5p promoter and found that its methylation level was significantly increased in GC tissues and cells. These results indicated that DNMT1-mediated promoter methylation might be responsible the downregulation of miR-125a-5p expression. Further investigation revealed that DNMT1 was significantly highly expressed in GC. The addition of 5-aza-dC or DNMT1 knockdown reduced the methylation level of the miR-125a-5p promoter and promoted the expression of miR-125a-5p. DNMT1 overexpression promoted GC cell proliferation, migration, invasion, and EMT by inhibiting miR-125a-5p expression. Our study was similar to previous studies, but for the first time, we demonstrated that the regulation of miR-125a-5p expression by DNMT1 was a key factor in the progression of GC.

What is the downstream mechanism by which miR-125a-5p regulates GC progression? Recent studies have confirmed that SERPINH1, an important collagen-specific gene, is a core gene that regulates the progression of GC[38]. The levels of both SERPINH1 mRNA and protein are significantly upregulated in GC tissues compared with normal tissues, and inhibition of SERPINH1 can significantly inhibit the migration and invasiveness of cancer cells[39,40]. SERPINH1 expression is also regulated by miRs. For example, Li et al[41] reported that miR-378a-5p can regulate immune cell infiltration through the inhibition of SERPINH1 expression, thereby inhibiting the carcinogenesis and progression of GC. Similarly, Kawagoe et al[39] reported that miR-148a-5p could suppress the invasiveness of GC cells by inhibiting SERPINH1 expression, thereby alleviating the development of GC. In this study, we found that SERPINH1 expression was upregulated in GC tissues and cells, which was consistent with the findings of previous studies. More importantly, through bioinformatics analysis, dual-luciferase gene reporter experiments and Western blot analysis, we confirmed that miR-125a-5p can target and negatively regulate the expression of SERPINH1 and that the overexpression of DNMT1 can promote SERPINH1 expression by inhibiting the expression of miR-125a-5p. In addition, SERPINH1 knockdown attenuates the promoting effects of DNMT1 overexpression on GC cell proliferation, migration, invasion, and EMT. These results indicate that SERPINH1 acts as an oncogenic factor in GC and is negatively regulated by miR-125a-5p.

CONCLUSION

In summary, our study demonstrated that DNMT1 can promote the expression of SERPINH1 through the inhibition of miR-125a-5p expression, which promotes GC cell proliferation, migration, invasion and EMT and accelerates the carcinogenesis and development of GC. Our study revealed that miR-125a-5p plays a key role in the progression of GC and that miR-125a-5p may be an effective target for GC treatment. This study not only deepens our understanding of GC development but also provides potential therapeutic targets for the precise treatment of GC. However, this study has several limitations. We have demonstrated the proposed mechanism using only in vitro cell experiments, and thus validation in vivo is lacking. Future studies will focus on in vivo experiments using xenograft models to evaluate the effects of DNMT1 and miR-125a-5p on tumor growth.