circ_0006988 promotes gastric cancer cell proliferation, migration and invasion through miRNA-92a-2-5p/TFAP4 axis

Abstract

Aim: To explore precise function and underlying mechanism of circ_0006988 in gastric cancer (GC).

Materials & methods: GC tissues were collected clinically, and GC cells were purchased from the company. Quantitative real-time polymerase chain reaction and western blot were used to detect mRNA and protein expression. Functional analysis was performed through CCK-8, Transwell and scratch experiment. Binding relationship was validated through dual luciferase reporter and RNA immunoprecipitation assays. HGC-27 cells were subcutaneously injected into mice to construct a xenograft tumor model.

Results: In GC tissues and cells, circ_0006988 overexpressed, promoting proliferation, migration and invasion. MiRNA-92a-2-5p downregulation or TFAP4 overexpression weakened effects of circ_0006988 silencing on GC progression.

Conclusion: circ_0006988 facilitates GC development through miRNA-92a-2-5p/TFAP4 axis.

Plain language summary

Article highlights.

• circ_0006988 was highly expressed in GC tissues and cells, and overexpression of circ_0006988 could promote the proliferation, migration and invasion of gastric cancer (GC) cells.

• miR-92a-2-5p was mainly located in the cytoplasm of GC cells, and circ_0006988 could interact with miR-92a-2-5p, acting as a sponge to downregulate the expression of miR-92a-2-5p.

• TFAP4 was upregulated in GC cells.

• Silencing circ_0006988 reduced the expression of TFAP4, and simultaneously inhibiting the expression of miR-92a-2-5p weakened this effect.

• Inhibiting miR-92a-2-5p or overexpressing TFAP4 could alleviate the inhibitory effect of silencing circ_0006988 on the malignant progression of GC cells.

• Silencing circ_0006988 could inhibit the progression of GC in mice.

1. Introduction

As one of the most prevalent malignancies in the digestive system, gastric cancer (GC) is the third primary cause of cancer related deaths around the world [1]. Family history, smoking, dietary factors, helicobacter pylori, alcohol consumption and Epstein–Barr virus infections are recognized risk factors contributing to the development of GC [2]. Currently, surgical resection is the major option for GC treatment [3]. With the continuous advancement in the diagnosis and treatment of GC, there has been a declining trend in the incidence and mortality rates of GC in recent years [4]. However, the prognosis of GC patients remains unfavorable because of tumor metastasis and recurrence [5,6]. GC is a progressive disease related to the activation of oncogenes or the inactivation of tumor suppressor genes [7]. Consequently, the identification and exploration of novel biomarkers can proffer new insights and approaches for GC treatment.

Circular RNAs (circRNAs), which have been increasingly recognized for their involvement in modulating various cellular functions, play important roles in human cancer progression [8,9]. Currently, several circRNAs, circSHKBP1, circMRPS35 and circ_0004872 are found to be closely linked with the progression of GC [10–12]. For instance, circNRIP1 is overexpressed in GC cells, and knocking out circNRIP1 successfully blocks the proliferation, migration, invasion and expression levels of AKT1 in GC cells. Furthermore, circNRIP1 can upregulate AKT1 expression through sponge of miRNA-149-5p, promoting malignant progression of GC [13]. circ_006100 has been found to enhance GC cells to grow and metastasize through the miRNA-195/GPRC5A signaling pathway [14]. As a member of circRNAs, circ_0006988, also known as circ-LDLRAD3, exerts oncogenic effects in GC [15], pancreatic cancer (PC) [16], and non-small cell lung cancer (NSCLC) [17]. Nevertheless, the specific mechanisms of circ_0006988 in GC development remain unclear.

According to previous studies, circRNAs can interact with microRNAs (miRNAs) through competitive endogenous RNA (ceRNA) mechanisms, thereby influencing gene expression [18]. miRNAs, a type of short non coding RNA molecules, have diverse biological functions, taking important parts in the tumorigenesis and progression of many tumors [19]. The extracellular vesicle miRNA-92a-5p secreted by macrophages increases the invasion of liver cancer cells by altering the AR/PHLPP/p-AKT/β-catenin signaling pathway. This regulatory axis is used to improve the tumor microenvironment and regulate liver cancer progression [20]. Renal cell carcinoma (RCC) [21], and GC [22], miRNA-92a-2-5p functions as a key biomarker and tumor suppressor. Nevertheless, the specific mechanisms of miRNA-92a-2-5p in GC and its interaction with circ_0006988 have not been elucidated. Unraveling the mechanism of miRNA-92a-2-5p and circ_0006988 in GC development may show new directions and strategies for the treatment and monitoring of GC.

Transcription factor activating enhancer-binding protein 4 (TFAP4) is a protein implicated in various biological functions such as tumor proliferation, metastasis, differentiation and angiogenesis [23]. Recently, as evidenced by a lot studies, TFAP4 overexpression may induce unfavorable prognosis in many cancers such as hepatocellular carcinoma (HCC) [24], NSCLC [25], prostate cancer [26] and colorectal cancer (CRC) [27]. Importantly, TFAP4 is also involved in GC occurrence and development. According to studies, lncRNA TRERNA1 is upregulated in GC cells, and TFAP4 specifically regulates TRERNA1 transcriptional activity by binding to the E-box motif in the TRERNA1 promoter, thereby promoting GC cell migration and invasion [28]. However, the molecular mechanism of the circ_0006988/miRNA-92a-2-5p/TFAP4 regulatory axis in the development of GC is unclear.

The main objective of this study was to investigate the impact of the circ_0006988/miRNA-92a-2-5p/TFAP4 axis on the progression of GC cells. This research contributed to a better understanding of the regulatory role of circRNA in the development of GC, providing new theoretical foundations and molecular targets for the early diagnosis and targeted therapy of GC. Additionally, the findings of this study could offer new insights and references for exploring the pathogenesis of other types of tumors and the development of novel therapeutic strategies.

2. Materials & methods

2.1. Bioinformatics analysis

To predict interactions among circ_0006988, miRNA-92a-2-5p and TFAP4, we employed CircNet2.0 in our analysis. The binding site prediction was performed by utilizing targetscan (http://www.targetscan.org/vert_71/) and miRanda (http://www.microrna.org/microrna/home.do).

2.2. Tissue sample collection

From Nanyang Central Hospital, we collected tumor tissues and adjacent normal tissues (n = 5) from GC patients (Supplementary Table S1) immediately frozen tissues at -80°C. We obtained written informed consent from all participants. Furthermore, the study protocol obtained approval by the Ethics Committee of Nanyang Central Hospital.

2.3. Cell culture & transfection

Human GC cell lines (HGC-27, AGS, SNU-16, MKN-45) and normal gastric epithelial cell line GES-1 were cultivated in RPMI-1640 medium with 10% fetal bovine serum (FBS) and 1% penicillin–streptomycin solution at 37°C with 5% CO₂. All cells were provided by the American Type Culture Collection (ATCC, USA).

pCD-ciR, circ_0006988, si-NC (empty plasmid for negative control), si-circ_000698, sh-NC, sh-circ_000698, TFAP4 and their corresponding control vectors, as well as miRNA-92a-2-5p mimics and their corresponding NC mimics, and miRNA-92a-2-5p inhibitor and its corresponding NC inhibitor were synthesized by Tsingke Biotech Co., Ltd (China). The transfection reagent kit UltraFection 3.0 high-efficiency transfection reagent was purchased from 4A BioTech (China), with transfection being performed according to the instructions of the transfection kit. Specifically, GC cells were seeded in a 6-well plate and the culture medium was replaced with serum-free medium when the cell density reached around 75–80%. After overnight incubation, the transfection reagent was diluted with serum-free medium. The transfection reagent was mixed evenly with the above plasmid and dropped into the cell culture medium. After 48 h of cell transfection, it was used for subsequent functional experiments.

2.4. Quantitative real-time polymerase chain reaction (qRT-PCR)

Total RNA was extracted from cells using TRIzol reagent purchased from Cwbio (China), and the RNA concentration and purity were measured using a NanoDrop 1000 spectrophotometer (Thermo Fisher, USA). Hifair^® II 1st Strand cDNA Synthesis Kit (Yeasn, China) and UltraSYBR Mixture (Cwbio, China) were used for reverse transcription of total RNA into cDNA and qPCR. The stem-loop primer sequence for miRNA-92a-2-5p was as follows: 5′-GTCGTATCCAGTGCAGGGTCCGAGGTATTCGCACTGGATACGACCCCACC-3′. The relative expression level of RNA was calculated by using the 2^-ΔΔCt method, and GAPDH served as an endogenous control for mRNA, while U6 served as an endogenous control for miRNA. Table 1 is the list of primer sequences.

Table 1.

The primer sequence in quantitative real-time polymerase chain reaction.

Primer sequence (5′-3′)
Gene	Forward	Reverse
circ_0006988	CAGAATGCGTCGGAAGTAGG	GAAGTCACAAACACCTGCCC
TFAP4	GACCCTGTGAAGTGAAGCAG	CGCTTGAGCTGTGTGTTCTG
miRNA-92a-2-5p	CGCGCATTACGTTGTTTAGG	AGTGCAGGGTCCGAGGTATT
GAPDH	ACATCGCTCAGACACCATG	TGTAGTTGAGGTCAATGAAGGG
U6	CGCTTCGGCAGCACATATACTAA	TATGGAACGCTTCACGAATTTGC

2.5. CCK-8 assay

Follow the instructions of the CCK-8 Cell Proliferation and Cytotoxicity Test Kit (Solarbio, China), cell proliferation ability was tested. In short, cells treated differently were seeded into a 96 well plate at a cell density of 200 cells per well and allowed to adhere to the wall. Ten microliter of CCK8 solution was added at 0, 24, 48, 72 and 96 h, and samples were incubated at 37°C for 2 h. A microplate reader (Bio-Rad, USA) was used to detect absorbance at 450 nm, which could reflect the proliferation of cells in each group.

2.6. Transwell assay

The migration and invasion abilities of HGC-27 cells were evaluated using Transwell apparatus. Specifically, a Transwell chamber (BD Biosciences, USA) with a size of 8 μm was added to a 24 well plate, and 50 μl of diluted matrix gel was added to it. After the matrix gel was completely solidified, HGC-27 cells incubated in serum-free medium were added to the upper chamber of the chamber, with a cell volume of 1 × 10⁴ cells/well, and complete medium was added to the lower chamber of the chamber. After 72 h, the chamber was removed, and the cells on the chamber membrane were fixed with 75% ethanol. Five percent crystal violet was used for staining for 10 min, and pictures were taken under a microscope for recording. The experimental steps for cell migration were the same as above, and there was no need to add matrix gel.

2.7. Scratch assay

GC cells transfected for 48 h were inoculated into a 6-well plate at a density of 1 × 10⁶ cells per well. When the cell confluence reached around 90%, a sterile 200 μl pipette tip was used to make a scratch in the middle of the plate, and the detached cells were washed off with PBS. RPMI-1640 medium without FBS was added to the plate. The wound closures at 0 or 24 h after scratching were recorded and the percentage of scratch healing was calculated. Healing ratio = (0 h scratch width- 24 h scratch width)/0 h scratch width × 100%.

2.8. Subcellular component analysis

Cellular cytoplasmic and nuclear RNA were isolated from 3 × 10⁶ cells by employing the PARIS Kit (Ambion, US). GAPDH and U6 were used as controls for cytoplasmic and nuclear fractions, respectively. We employed qRT-PCR to measure the subcellular distribution of circ_0006988 in the cytoplasm and nucleus.

2.9. Dual-luciferase reporter assay

Fragments of circ_0006988 or TFAP4 3′-UTR containing the binding sites for miRNA-92a-2-5p were cloned into the pmirGLO Vector (Promega, USA) to generate the luciferase reporter constructs circ_0006988 WT and TFAP4 3′-UTR WT. Mutations were introduced into the binding sites to generate circ_0006988 MUT and TFAP4 3′-UTR MUT. The plasmids were co-transfected with mimic NC/miRNA-92a-2-5p mimic into cells. With the application of the Dual-Luciferase Reporter Assay System (Promega, USA), the firefly luciferase intensity was determined.

2.10. Western blot (WB)

Cells from each treatment group were collected, seeded in a 6-well plate, and allowed to grow to maturity. After the cells matured, they were lysed using RIPA lysis buffer (Beyotime, China) containing protease and phosphatase inhibitors. The protein concentrations were quantitatively analyzed using a BCA protein concentration assay kit (Beyotime, China). The extracted protein samples were denatured at high temperatures for 5 min and then subjected to electrophoresis using a 10% sodium dodecyl sulfate-polyacrylamide gel (SDS-PAGE). Following electrophoresis, the proteins were transferred onto a polyvinylidene difluoride membrane and blocked with 5% skim milk for 1 h. Subsequently, the membranes were separately incubated overnight at 4°C with primary antibodies against TFAP4 (DF13624, 1:1000) and GAPDH (#AF7021, 1:3000), both purchased from Affinity (China). After washing with PBS, the membranes were incubated at room temperature for 2 h with secondary antibodies HRP Goat Anti-Rabbit IgG (H+L) (#S0001, 1:3000). Following another round of PBS washing, protein bands were visualized using the highly sensitive ECL chemiluminescence detection kit (Beyotime, China).

2.11. RNA immunoprecipitation (RIP)

The EZ-Magna RIP Kit (Millipore, USA) was utilized to conduct the RIP assay. In other words, we lysed GC cells in RIP buffer and then incubated them with magnetic beads conjugated with antibodies against Ago2 or IgG (Millipore, MA). Subsequently, we isolated RNA from the magnetic beads and detected the expression levels of circ_0006988, miRNA-92a-2-5p and TFAP4. According to the instructions provided in the EZ-Magna RIP Kit (Millipore, USA), RIP analysis was conducted as follows: GC cells were seeded in a 6-well plate at a density of 8 × 10⁵ cells per well. When the cells reached approximately 90% confluency, they were collected from each well. The cells were then lysed in RIP lysis buffer using magnetic beads coated with anti-Ago2 (Millipore, MA) or anti-IgG (Millipore, MA) antibodies. Finally, purified immunoprecipitated circ_0006988, miR-92a-2-5p, and TFAP4 were detected through qRT-PCR analysis.

2.12. Xenograft tumor experiment

Six female BALB/c nude mice were purchased from Shanghai SLAC Laboratory Animal Co., Ltd (China) company and each mouse was housed under disease-free conditions. HGC-27 cells transfected with sh-NC or sh-circ_0006988 were subcutaneously injected into nude mice to establish a xenograft tumor mouse model. The longest diameter (L) and longest transverse meridian (W) of the tumor were measured every 7 days using a vernier caliper, and the tumor volume was calculated according to the formula volume = (L × W²)/2. Then, the mice were euthanized after 28 days. We excised tumors from the euthanized mice to estimate tumor volume and weight. The Animal Research Ethics Committee of Nanyang Central Hospital approved animal experiments in this study.

2.13. Immunohistochemistry (IHC) staining

After the mouse tumor tissues from different treatment groups were fixed in formalin and embedded in paraffin, 4 μm thick tissue sections were prepared. The sections were deparaffinized twice in xylene for 10 min each time, followed by sequential dehydration in gradients of anhydrous ethanol at different concentrations for 5 min each. After washing three-times with PBS, the tissue sections were incubated overnight at 4°C with primary antibodies against Ki67 (HA721115; Huaxin, China) and TFAP4 (DF13624; Affinity, China) containing blocking agents. After PBS washing, the sections were incubated at room temperature for 2 h with a secondary antibody, Goat Anti-Rabbit IgG H&L conjugated with horseradish peroxidase (A0208; Beyotime, China). Following PBS washing of the tissue sections, the DAB Substrate Kit (Beijing Zhongshan Golden Bridge Bio, China) was used for staining the sections. The staining was terminated after rinsing with water. Subsequently, the sections were stained with hematoxylin for 2 min, and images were captured using a confocal microscope (Leica Microsystems, Wetzlar, Germany).

2.14. Data analysis

We carried out all experiments with three or more biological replicates. Data are presented as mean ± standard deviation (mean ± SD) and analyzed using Prism 8 software (GraphPad, USA) for unpaired t-test and one-way analysis of variance. p < 0.05: statistical significance.

3. Results

3.1. circ_0006988 overexpression enhances GC cell proliferation, migration & invasion

We found a significant upregulation of circ_0006988 in GC tissues compared with normal tissues by conducting qRT-PCR (Figure 1A). Consistently, compared with the normal gastric epithelial cell line GES-1, circ_0006988 was significantly elevated in GC cell lines (HGC-27, AGS, SNU-16, MKN-45) (Figure 1B). HGC-27 exhibited lower expression levels of circ_0006988 among the four GC cell lines. Therefore, HGC-27 was chosen for further investigations. The above experimental results showed that circ_0006988 was upregulated in GC tissues and cells.

Figure 1.

Effects of circ_0006988 overexpression on gastric cancer cell proliferation, migration and invasion.

(A) qRT-PCR analysis of circ_0006988 expression in GC tissues compared with normal tissues. (B) qRT-PCR analysis of circ_0006988 expression in GES-1, HGC-27, AGS, SNU-16 and MKN-45 cells. (C) qRT-PCR analysis of circ_0006988 expression in HGC-27 cells transfected with pCD-ciR or circ_0006988. (D) CCK-8 assay assessed the proliferation capacity of HGC-27 cells. (E & F) Transwell assay detected the migration and invasion of HGC-27 cells. Scale bar was 100 μm. (G) Scratch assay evaluated the migration ability of HGC-27 cells. Scale bar was 200 μm.

*p < 0.05.

GC: Gastric cancer; qRT-PCR: Quantitative real-time polymerase chain reaction.

The exact function of circ_0006988 in GC development was investigated. In Figure 1C, transfection with the circ_0006988 overexpression vector resulted in elevated levels of circ_0006988 in HGC-27 cells. Then, we investigated the impact of circ_0006988 overexpression on GC cell capability to proliferate, migrate and invade. CCK-8 assay results revealed that in HGC-27 cells, overexpression of circ_0006988 showed no significant change in cell viability at 0, 24 and 48 h, while it promoted cell viability at 72 and 96 h (Figure 1D). According to Transwell assay, circ_0006988 overexpression enhanced migration and invasion of HGC-27 cells (Figure 1E & F). Furthermore, results of the scratch assay showed that overexpression of circ_0006988 promoted the migration of HGC-27 cells after 24 h compared with the control group (Figure 1G). Collectively, our findings indicated that GC cell’s capability of proliferation, migration and invasion can be enhanced by circ_0006988 overexpression.

3.2. circ_0006988 is functioned as a sponge for miRNA-92a-2-5p

To explore the potential molecular mechanism of circ_0006988 in GC development, we verified it by the following experiments. Based on subcellular component analysis, circ_0006988 was greatly gathered in the cytoplasm of HGC-27 cells, indicating its potential as a miRNA sponge (Figure 2A). Analysis using starBase (http://starbase.sysu.edu.cn/starbase2/) determined binding sites between miRNA-92a-2-5p and circ_0006988 (Figure 2B). In contrast to the mimic NC group, transfection of miRNA-92a-2-5p mimic considerably increased the expression of miRNA-92a-2-5p in HGC-27 cells (Figure 2C). From dual-luciferase reporter assay, we found that miRNA-92a-2-5p overexpression repressed the luciferase activity of circ_0006988 WT in HGC-27 cells. However, there was no effect exerted on circ_0006988 MUT (Figure 2D). RIP experiments revealed enrichment of circ_0006988 and miRNA-92a-2-5p in the immunoprecipitated complex of the anti-Ago2 group in contrast to the anti-IgG control group (Figure 2E). All results indicated an interaction between miRNA-92a-2-5p and circ_0006988. Compared with normal cells, miRNA-92a-2-5p was low expressed in GC cells (Figure 2F). We also showed that circ_0006988 overexpression significantly reduced the expression of miRNA-92a-2-5p in HGC-27 cells (Figure 2G). In summary, circ_0006988 functioned as a sponge, sequestering miRNA-92a-2-5p and thus altering its expression.

Figure 2.

circ_0006988 acts as a sponge for miRNA-92a-2-5p. (A) Subcellular component analysis of circ_0006988 expression in the nucleus and cytoplasm of HGC-27 cells. (B) Binding sites of miRNA-92a-2-5p to circ_0006988. (C) qRT-PCR analysis of miRNA-92a-2-5p expression in HGC-27 cells transfected with mimic NC or miRNA-92a-2-5p mimic. (D & E) Dual-luciferase reporter gene assay and RIP experiment analyzing the binding of circ_0006988 to miRNA-92a-2-5p. (F) qRT-PCR analysis of miRNA-92a-2-5p expression in GES-1, HGC-27, AGS, SNU-16 and MKN-45 cells. (G) qRT-PCR analysis of miRNA-92a-2-5p expression after transfection of HGC-27 cells with pCD-ciR or circ_0006988.

*p < 0.05.

qRT-PCR: Quantitative real-time polymerase chain reaction; RIP: RNA immunoprecipitation.

3.3. TFAP4 is a target gene of miRNA-92a-2-5p

Through targetscan analysis, we identified TFAP4 as a target gene of miRNA-92a-2-5p (Figure 3A). To validate the predicted results, we carried out dual-luciferase reporter assay and RIP experiments. The dual-luciferase reporter gene assay demonstrated that miRNA-92a-2-5p overexpression inhibited the luciferase activity of TFAP4 3′-UTR WT in HGC-27 cells, with no obvious effect observed in TFAP4 3′-UTR MUT (Figure 3B). RIP experiments revealed that compared with the anti-IgG RIP group, miRNA-92a-2-5p and TFAP4 were more enriched in the anti-Ago2 group (Figure 3C). As expected, compared with GES-1 cells, HGC-27, AGS, SNU-16 and MKN-45 cells have elevated levels of TFAP4 mRNA and protein (Figure 3D & E). In HGC-27 cells, TFAP4 mRNA and protein levels were reduced by miRNA-92a-2-5p while were increased by miRNA-92a-2-5p inhibitor (Figure 3F & G). Importantly, our results suggested that silencing of circ_0006988 reduced TFAP4 mRNA and protein levels in HGC-27 cells, which was attenuated by miRNA-92a-2-5p (Figure 3H & I). To sum up, our findings unearthed that TFAP4 is a target gene of miRNA-92a-2-5p and has a negative regulatory relationship with miRNA-92a-2-5p.

Figure 3.

TFAP4 is a target gene of miRNA-92a-2-5p. (A) Binding sites of miRNA-92a-2-5p to TFAP4. (B & C) Dual-luciferase reporter gene assay and RIP experiment detected the relationship between miRNA-92a-2-5p and TFAP4. (D & E) qRT-PCR and WB analysis of TFAP4 expression in GES-1, HGC-27, AGS, SNU-16, and MKN-45 cells. (F & G) qRT-PCR and WB analysis of TFAP4 expression in HGC-27 cells transfected with mimic NC, miRNA-92a-2-5p mimic, inhibitor NC or miRNA-92a-2-5p inhibitor. (H & I) qRT-PCR and WB analysis of TFAP4 expression in HGC-27 cells transfected with si-NC, si-circ_0006988, si-circ_0006988 + inhibitor NC, or si-circ_0006988 + miRNA-92a-2-5p inhibitor.

*p < 0.05.

qRT-PCR: Quantitative real-time polymerase chain reaction; RIP: RNA immunoprecipitation; WB: Western blot.

3.4. TFAP4 overexpression or miRNA-92a-2-5p inhibition attenuates the impact of circ_0006988 silencing on GC malignant progression

In Figure 4A & B, transfection of the TFAP4 overexpression vector remarkably increased TFAP4 mRNA and protein levels in HGC-27 cells. Subsequently, we investigated the relationship between circ_0006988 and miRNA-92a-2-5p or TFAP4 in modulating GC progression. CCK-8 assay revealed that circ_0006988 had no significant effect on cell proliferation after 0, 24 and 48 h of transfection, while circ_0006988 silencing remarkably suppressed the capability of proliferation in HGC-27 cells at 72 and 96 h, which is attenuated by inhibition of miRNA-92a-2-5p or TFAP4 overexpression (Figure 4C). Transwell and scratch assay demonstrated that the inhibitory effect of circ_0006988 silencing on HGC-27 cell migration and invasion could be alleviated by the suppression of miRNA-92a-2-5p or the upregulation of TFAP4 (Figure 4D–F). Collectively, circ_0006988 regulates the development of GC cells by modulating miRNA-92a-2-5p or TFAP4.

Figure 4.

Inhibition of miRNA-92a-2-5p or TFAP4 overexpression attenuates the impact of circ_0006988 silencing on the proliferation, migration and invasion of gastric cancer. (A & B) qRT-PCR and WB analysis of TFAP4 expression levels in HGC-27 cells transfected with TFAP4 or vector. (C) CCK-8 assay assessed the proliferation of HGC-27 cells. (D–F) Transwell assay and scratch assay evaluated migration and invasion capabilities in HGC-27 cells.

*p < 0.05.

qRT-PCR: Quantitative real-time polymerase chain reaction; WB: Western blot.

3.5. Knockdown of circ_0006988 inhibits tumor growth in vivo

We investigated the role of circ_0006988 in tumor development in vivo. Compared with the sh-NC control group, the knockdown of circ_0006988 suppressed tumor volume and weight (Figure 5A–C). Furthermore, the sh-circ_0006988 group exhibited decreased levels of circ_0006988, TFAP4 mRNA, and TFAP4 protein, along with increased levels of miRNA-92a-2-5p, in the transplanted tumors compared with the sh-NC group (Figure 5D–G). Additionally, IHC experiments demonstrated that circ_0006988 knockdown repressed Ki67 and TFAP4 in the transplanted tumors compared with the sh-NC group (Figure 5H & I). In summary, the silencing of circ_0006988 represses tumor growth in vivo.

Figure 5.

Knockdown of circ_0006988 inhibits tumor growth in vivo. (A–C) Measurement of tumor volume and weight. (D–F) qRT-PCR analysis of circ_0006988, miRNA-92a-2-5p and TFAP4 levels in the transplanted tumors. (G) WB analysis of TFAP4 protein levels in the transplanted tumors. (H & I) IHC staining of Ki67 and TFAP4 expression levels in the transplanted tumor tissues. Scale bar was 50 μm.

*p < 0.05.

IHC: Immunohistochemistry; qRT-PCR: Quantitative real-time polymerase chain reaction; WB: Western blot.

4. Discussion

GC remains one of the most common malignant tumors worldwide, and due to its subtle onset symptoms and low early screening rate, most patients are diagnosed as advanced stage [29]. Systemic chemotherapy, radiotherapy, surgery, immunotherapy and targeted therapy have been proven to have good effects on the treatment of GC [30], but conventional treatment methods for GC are limited, and further exploration and research are needed on the pathogenesis of GC [31].

Many current studies are focusing on the correlation between circRNAs and human cancers. As a member of circRNAs, circ_0006988 has been studied in PC [16], NSCLC [32] and HCC [33]. circ_0006988 is highly expressed in the blood of CRC patients and can serve as a biomarker for this type of cancer [34]. circ_0006988 is highly expressed in NSCLC tissues and cells, with low expression of miR-137. circ_0006988 can promote malignant progression of NSCLC cells by downregulating miR-137 and upregulating SLC1A5 expression [17]. In the context of GC research, circ_0006988 is reported to participate in various biological processes of GC by modulating different signaling pathways. For instance, circ_0006988 is overexpressed in cisplatin (DDP) resistant GC tissues and cells. Knockdown expression of circ_0006988 can significantly reduce the IC₅₀ value of DDP resistant cells, inhibit cell proliferation, survival, and invasion, and this inhibitory effect is achieved by regulating the miRNA-588/SOX5 pathway [35]. Furthermore, circ_0006988 enhances cells to grow, migrate and invade as well as inhibits cell apoptosis in GC through the modulation of the miRNA-224-5p/NRP2 axis [15]. Being consistent with former findings, the present study also revealed that circ_0006988 overexpression significantly promotes GC cells to proliferate, invade and migrate. Moreover, to better understand the influence of circ_0006988 on GC tumor development, we further established a xenograft model to demonstrate that silencing of circ_0006988 suppressed tumor growth in vivo. Obviously, the experimental results of this study are consistent with previous studies, that is, circ_0006988 is upregulated in GC cells, and its overexpression can promote the malignant progression of GC.

The argument that circRNA regulates cancer progression by downregulating miRNA expression through sponge-like transformation has been fully confirmed. To explore the potential downstream targeting miRNA of circ_0006988 in GC cells, we validated it using bioinformatics web tools. The analysis results showed that miRNA-92a-5p can be adsorbed by circ_0006988 sponge. We are aware that circ_0006948 can bind to several miRNAs, but the reason we chose miRNA-92a-2-5p as the target miRNA in this study is twofold: first, through bioinformatics predictions and experimental validation, we confirmed that circ_0006948 can sponge miRNA-92a-2-5p with binding sites between them, and second, previous literature has reported miRNA-92a-2-5p as a biomarker in cancers, including GC and as a tumor suppressor [20,22]. The abnormal expression of miRNAs in tumors is closely related to the occurrence and development of tumors. In esophageal squamous cell carcinoma, overexpression of miRNA-92a-2-5p inhibits esophageal squamous cell carcinoma proliferation and migration, and this inhibitory effect is achieved through the AKT/mTOR and Wnt3a/β-catenin signaling pathways [36]. In previous studies, miRNA-92a-2-5p serves as a regulatory target of circRNAs and plays important roles in various cancers. For example, circRNA circAMOTL1L is downregulated in RCC tissues and cells. circAMOTL1L can act as a sponge for miRNA-92a-2-5p, enhancing the signal transduction of circAMOTL1L/miRNA-92a-2-5p/KLLN axis and to some extent inhibiting the growth of RCC in mice [21]. In our study, we found that circ_0006988 acts as a molecular sponge for miRNA-92a-2-5p. When we inhibited miRNA-92a-2-5p, the suppressive effect of circ_0006988 silencing on GC malignant progression was remarkably attenuated. This suggested that circ_0006988 changes the function of miRNA-92a-2-5p in GC by sequestering it, affecting the growth and metastasis of GC. A previous study has also shown that miRNA-92a-2-5p is highly expressed in GC and can serve as a diagnostic biomarker for precancerous lesions of GC [22]. However, this study did not further confirm the effect of miRNA-92a-2-5p on tumor cell progression through cellular functional experiments, so our research has a certain degree of innovation based on this foundation.

Furthermore, we determined TFAP4 as a target gene of miRNA-92a-2-5p. We also acknowledge that miRNAs can target many genes. In this study, we selected TFAP4 as the target mRNA for two reasons: first, through bioinformatics predictions and experimental validation, we confirmed the binding relationship between miRNA-92a-2-5p and TFAP4, and second, TFAP4 has been reported to promote the migration and invasion of tumor cells in GC [28]. The molecular mechanisms of the circ_0006988/miR-92a-2-5p/TFAP4 regulatory axis in the development of GC remain unclear. As an important transcription factor, TFAP4 acts as an oncogene when targeted by miRNAs in various cancers, such as miRNA-371b-5p [37], and miRNA-608 [38]. In addition, the long noncoding RNA TTN-AS1 (TTN-AS1) promotes the proliferation, migration, invasion and EMT process of osteosarcoma cells by downregulating miRNA-16-1-3p and upregulating TFAP4 expression [39]. miRNA-373-3p is downregulated in HCC, and its targeted negative regulation of TFAP4 inhibits the malignant progression of HCC cells and promotes apoptosis [40]. However, in our study, we first discovered the interaction between miRNA-92a-2-5p and TFAP4. Further research showed that circ_0006988 can modulate TFAP4 through miRNA-92a-2-5p. It is worth noting that our findings indicate that TFAP4 overexpression attenuates the suppressive effects of circ_0006988 silencing on GC malignant development. However, the function of TFAP4 in modulating GC progression mediated by miRNA-92a-2-5p needs further elucidation. Although the regulatory relationship between miRNA-92a-2-5p and TFAP4 has not been elucidated in previous studies, it can be confirmed that TFAP4 plays a significant procancer role in various cancers [24,41], which is consistent with our research results. To provide evidence to support the arguments of this study, we analyzed the expressions of circ_0006988, miRNA-92a-2-5p, and TFAP4 in GC tissues using The Cancer Genome Atlas (TCGA) database. However, the results were not satisfactory. Circ_0006988 was not found in the TCGA database, so we were unable to analyze its expression in GC tissues using this database. Additionally, the expression of miRNA-92a-2-5p in GC tissues was not significant. Taking all these factors into consideration, we did not mention the analysis results of the TCGA database. Although the results from the TCGA database analysis were not ideal, bioinformatics data analysis serves as a predictive tool, and the actual expression results need to be verified through cellular functional experiments.

In summary, the present study confirmed that circ_0006988 motivates the malignant progression of GC through the miRNA-92a-2-5p/TFAP4 axis, showing that circ_0006988 is promising to be a target for GC treatment. However, we also acknowledged certain limitations in our study. First, our research did not cover all the miRNAs regulated by circ_0006988. In future investigations, in order to learn more about the regulatory network of circ_0006988 in GC cells, it will be important to explore other miRNAs associated with circ_0006988. Second, we did not extensively investigate the signaling pathways regulated by the TFAP4. We plan to design and incorporate a positive control group in experiments to further support the conclusions of this study. Deeper studies are required to clarify the mechanistic function of TFAP4 in GC and to identify the regulatory pathways that interact with circ_0006988.

5. Conclusion

This study confirms that circ_0006988 acts as a sponge for miRNA-92a-2-5p, upregulating the expression of TFAP4, thereby promoting the malignant progression of GC in vivo and in vitro. This discovery provides new insights into the mechanisms underlying GC development, offering a theoretical basis for identifying novel therapeutic targets for GC. Moreover, this regulatory network may not only be applicable to GC but also play a crucial role in other cancers such as lung cancer and CRC. Therefore, further exploration of the role of this regulatory network in various malignancies could pave the way for precision therapies in cancer treatment.

Supplemental material

Supplemental data for this article can be accessed at https://doi.org/10.1080/17501911.2024.2410697

Author contributions

Y Mu conceived of the study, and participated in its design and interpretation and helped to draft the manuscript. J Lu and K Yue participated in the design and interpretation of the data and drafting/revising the manuscript. S Yin, R Zhang and C Zhang performed the statistical analysis and revised the manuscript critically. All the authors read and approved the final manuscript.

Financial disclosure

This paper was not funded.

Competing interests disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Writing disclosure

No writing assistance was utilized in the production of this manuscript.

Ethical conduct of research

The animal study was reviewed and approved by the Animal Ethics Committee of Nanyang Central Hospital. The studies involving human participants were reviewed and approved by the Ethics Committee of Nanyang Central Hospital. The patients/participants provided their written informed consent to participate in this study.

Data availability statement

The data and materials in the current study are available from the corresponding author on reasonable request.