Circ-RNF111 aggravates the malignancy of gastric cancer through miR-876-3p-dependent regulation of KLF12

Background The aberrant expression of circular RNAs (circRNAs) plays vital roles in the advancement of human cancers, including gastric cancer (GC). In this study, the functions of circRNA ring finger protein 111 (circ-RNF111) in GC were investigated. Methods Quantitative real-time polymerase chain reaction (qRT-PCR) assay was performed for the levels of circ-RNF111, microRNA-876-3p (miR-876-3p) and krueppel-like factor 12 (KLF12) mRNA. RNase R assay was conducted for the feature of circ-RNF111. Cell Counting Kit-8 (CCK-8) assay, colony formation assay, wound-healing assay, and transwell assay were applied for cell viability, colony formation, migration, and invasion, respectively. Flow cytometry analysis was used to analyze cell apoptosis and cell cycle process. The glycolysis level was examined using specific commercial kits. Western blot assay was carried out to measure the protein levels of hexokinase 2 (HK-2) and KLF12. Dual-luciferase reporter assay and RNA immunoprecipitation (RIP) assay were employed to verify the combination between miR-876-3p and circ-RNF111 or KLF12. Murine xenograft model was constructed for the role of circ-RNF111 in vivo. Immunohistochemistry (IHC) was used for KLF12 level. Results Circ-RNF111 was higher expressed in GC tissues and cells than normal tissues and cells. Silencing of circ-RNF111 restrained cell viability, colony formation, migration, invasion, cell cycle process and glycolysis and induced apoptosis in GC cells in vitro. Circ-RNF111 positively regulated KLF12 expression via absorbing miR-876-3p. MiR-876-3p downregulation reversed the impacts of circ-RNF111 silencing on GC cell malignant phenotypes. MiR-876-3p overexpression repressed GC cell growth, metastasis and glycolysis, inhibited apoptosis and arrested cell cycle, while KLF12 elevation weakened the effects. Besides, circ-RNF111 knockdown inhibited tumor growth in vivo. Conclusion Circ-RNF111 knockdown relieved the development of GC by regulating miR-876-3p/KLF12 axis. Supplementary Information The online version contains supplementary material available at 10.1186/s12957-021-02373-5.


Introduction
Gastric cancer (GC) is the fourth most usual cancer that endangers human health worldwide [1,2]. Despite various improvements have been made in diagnosis and treatment in the past few decades, the prognosis of GC is still very poor due to the metastasis and recurrence of tumors [3,4]. Hence, it is imperative to discover new targets for the diagnosis and therapy of GC.
MiRNAs are short ncRNAs and served as tumor promoters or suppressors in GC [18]. Yang et al. suggested that miR-876-3p enhanced the apoptosis and curbed the proliferation of pancreatic adenocarcinoma by targeting JAG2 [19]. Tang et al. declared that miR-876-3p directly targeted KIF20A to slow the carcinogenesis of glioma [20]. Moreover, the downregulation of miR-876-3p was related to the worse outcome of GC patients, and enhanced chemoresistance of GC by targeting TMED3 [21]. Nevertheless, the exact functions of miR-876-3p in GC remain unclarified.
Krueppel-like factor 12 (KLF12) is a member of the KLFs' family and plays essential roles in various human cancers, including GC [22,23]. In the study, bioinformatics analysis presented that miR-876-3p included the binding sequences of circ-RNF111 and KLF12, indicating the potential relationships of circ-RNF111, miR-876-3p, and KLF12.
Here, the expression pattern of circ-RNF111 in GC was determined. Moreover, the functions and relationships of circ-RNF111, miR-876-3p, and KLF12 in GC development were investigated.

Tissues acquisition
Thirty-one GC patients at the First People's Hospital of Xiaoshan were recruited in our research. After the study was permitted by the Ethics Committee of the First People's Hospital of Xiaoshan and written informed consents were offered by the participants, the tumor tissues and adjacent non-tumor tissues were acquired and stored at -80°C before use.

Cell counting kit-8 (CCK-8) assay
Firstly, 5 × 10 3 cells were seeded into each well of 96well plates. After 48 h of incubation, 10 μL CCK-8 (Sigma-Aldrich) was supplemented into the well and kept for an additional 2 h. At last, the absorption at 450 nm was measured.

Colony formation assay
Following indicated transfection, GC cells were seeded into 6-well plates for 14 days. When the colonies were visible, the culture was terminated. Next, the colonies were dyed with crystal violet (Sangon) and quantified utilizing a microscope (Olympus, Tokyo, Japan).

Flow cytometry analysis
The apoptosis and cell cycle process were examined with Annexin V-FITC/PI Apoptosis Kit (Beyotime, Shanghai, China) based on the manufacturers' instructions. To examine cell apoptosis, GC cells with various transfections were harvested, washed with PBS (Sangon), and then resuspended in binding buffer. Then, Annexin V-FITC and PI were adopted to dye the cells. For cell cycle process, after the transfected cells were washed, they were fixed with 70% ethanol and then kept with RNase (Solarbio, Beijing, China) in PBS (Sangon) for 1 h. Thereafter, the cells were interacted with PI. The apoptotic rate and cell cycle were assessed with a FACS flow cytometry (BD Biosciences, Franklin Lakes, NJ, USA).

Wound-healing assay
To determine the migration of GC cells, the transfected GC cells were added into 6-well plates and grown until 100% confluence. Then, the sterile pipette tip was utilized to make a scratch in the well. At 0 h and 24 h, the crosses were recorded.

Transwell assay
The invasion of GC cells was evaluated with the transwell insert chambers (BD Biosciences) pre-covered Matrigel (BD Biosciences). In short, the transfected AGS and SNU-638 cells in DMEM (Sigma-Aldrich) with serum-free were added into the upper chamber. The bottom chamber was filled with DMEM (Sigma-Aldrich) including 10% FBS (Sigma-Aldrich). Twenty-four hours later, the invaded cells were dyed with crystal violet (Sangon) and quantified utilizing a microscope (Olympus; × 100 magnification).

Measurement of glycolysis level
The lactate assay kit (Sigma-Aldrich), glucose assay kit (Sigma-Aldrich), and ATP assay kit (Sigma-Aldrich) were employed to test glucose uptake, lactate production, and ATP production levels according to the manufacturers' protocols.

RNA immunoprecipitation (RIP) assay
In brief, after AGS and SNU-638 cells were lysed in RIP buffer, the cell lysates were maintained with protein A/G sepharose beads conjugated with antibody IgG or Ago2. Next, total RNA in immunoprecipitates was subjected to qRT-PCR for the abundance of circ-RNF111, miR-876-3p, and KLF12.

Murine xenograft model
The BALB/c nude mice from Beijing Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China) were assigned into 2 groups (n = 5/group). Sh-NC or sh-circ-RNF111 transfected AGS cells (2 × 10 5 ) were suspended in 0.2 mL PBS (Sangon) and then subcutaneously introduced into the flank of the mice. After 7 days, tumor size was examined every 5 days and estimated via the formula: (LengthWidth 2 ) × 0.5. At 32 days, the mice were sacrificed and xenograft tumors were weighted. The in vivo study obtained permission from the Ethics Committee of Animal Research of the First People's Hospital of Xiaoshan.

Immunohistochemistry (IHC) assay
The expression of KLF12 and ki67 in the xenograft tumor tissues was examined by IHC assay, as previously described [24]. The antibodies against KLF12 (bs-16783R) and ki67 (bs-23103R) were provided by Bioss.

Statistical analysis
The results from three independent experiments were analyzed using GraphPad Prism 7 and exhibited as mean ± SD. The relationship between miR-876-3p level and circ-RNF111 level or KLF12 level in GC tissues was evaluated by Spearman's correlation coefficient. Student's t test or one-way analysis of variance was utilized for different analysis. P < 0.05 was thought to be significant.

Circ-RNF111 was upregulated in GC tissues and cell lines
To explore the effect of circ-RNF111 in GC development, the expression level of circ-RNF111 in GC tissues and normal tissues was determined. The results exhibited that circ-RNF111 was highly expressed in GC tissues compared to adjacent normal tissues (Fig. 1A). Moreover, we found that circ-RNF111 was conspicuously increased in AGS and SNU-638 cells relative to GES-1 cells (Fig. 1B). Besides, RNase R assay showed that GAPDH was digested by RNase R treatment, but circ-RNF111 was resistant to RNase R, indicating that circ-RNF111 was stable (Fig. 1C).

Silencing of circ-RNF111 suppressed the malignant biological behaviors of GC cells
To explore the functions of circ-RNF111 in GC, AGS and SNU-638 cells with circ-RNF111 silencing were constructed by transfecting si-circ-RNF111#1 or si-circ-RNF111#2 into AGS and SNU-638 cells. As a result, the transfection of si-circ-RNF111#1 or si-circ-RNF111#2 led to a distinct suppression in circ-RNF111 expression in AGS and SNU-638 cells compared to si-NC control groups ( Fig. 2A). The results of CCK-8 assay showed that circ-RNF111 knockdown repressed the viability of AGS and SNU-638 cells relative to control groups (Fig.  2B). Colony formation assay indicated that the colony formation ability of AGS and SNU-638 cells was inhibited by the downregulation of circ-RNF111 in Fig. 1 High level of circ-RNF111 in GC tissues and cells. A The expression of circ-RNF111 in tumor tissues and normal tissues was determined by qRT-PCR assay. B The expression of circ-RNF111 in GES-1, AGS, and SNU-638 cells was detected by qRT-PCR assay. C Following total RNA in AGS and SNU-638 cells was treated with or without RNase R, the expression levels of circ-RNF111, and GAPDH were detected by qRT-PCR assay. ***P < 0.001 comparison with si-NC control groups (Fig. 2C). Flow cytometry analysis indicated that circ-RNF111 silencing facilitated the apoptosis of AGS and SNU-638 cells in comparison with control groups (Fig. 2D). As illustrated by wound-healing assay and transwell assay, the migration and invasion of AGS and SNU-638 cells were restrained by circ-RNF111 knockdown relative to control groups (Fig. 2E, F). Moreover, circ-RNF111 silencing arrested cell cycle in G0/G1 phase, as analyzed by flow cytometry analysis (Fig. 2G, H). Circ-RNF111 knockdown reduced the levels of cell cycle-related protein CyclinD1 and cell metastasis-related protein MMP9 (Fig.  2I, J). Besides, the impact of circ-RNF111 deficiency on glycolysis was explored. It was found that circ-RNF111 deficiency reduced the levels of glucose consumption, lactate production, ATP production and HK-2 protein in AGS and SNU-638 cells compared to control groups, suggesting the repression of glycolysis in AGS and SNU-638 cells after circ-RNF111 knockdown (Fig. 2K-N). Taken together, circ-RNF111 knockdown suppressed GC cell growth, metastasis and glycolysis, promoted apoptosis, and arrested cell cycle.

Circ-RNF111 directly targeted miR-876-3p
Through analyzing circular RNA interactome, miR-876-3p was found to share the binding sites with circ-RNF111 (Fig. 3A). Dual-luciferase reporter assay was then performed to estimate the relationship between circ-RNF111 and miR-876-3p. The results exhibited that the luciferase activity of WT-circ-RNF111 in AGS and SNU-638 cells was restrained by miR-876-3p transfection, while the luciferase activity of MUT-circ-RNF111 Fig. 2 Knockdown of circ-RNF111 repressed the progression of GC cells. A The expression of circ-RNF111 in AGS and SNU-638 cells transfected with si-NC, si-circ-RNF111#1, or si-circ-RNF111#2 was detected by qRT-PCR assay. B The viability of si-NC, si-circ-RNF111#1, or si-circ-RNF111#2 transfected AGS and SNU-638 cells was assessed by CCK-8 assay. C-L AGS and SNU-638 cells were transfected with si-NC or si-circ-RNF111. C The colony formation, D apoptosis, E migration, and F invasion of AGS and SNU-638 cells were investigated by colony formation assay, flow cytometry analysis, wound-healing assay, and transwell assay, respectively. G, H Cell cycle process in AGS and SNU-638 cells was analyzed by flow cytometry analysis. I, J The protein levels of CyclinD1 and MMP9 in AGS and SNU-638 cells were measured by western blot assay. K-M The levels of glucose uptake, lactate production, and ATP production in AGS and SNU-638 cells were measured with commercial kits. N The protein level of HK-2 in AGS and SNU-638 cells was measured by western blot assay. **P < 0.01, ***P < 0.001 was not affected (Fig. 3B). Thereafter, RIP assay further demonstrated the interaction between circ-RNF111 and miR-876-3p for the abundance of circ-RNF111 and miR-876-3p was elevated in Ago2 pellets relative to IgG control groups (Fig. 3C). Indeed, miR-876-3p level was decreased in GC tissues and cells compared to normal tissues and cells (Fig. 3D, E). Moreover, there was an inverse correlation between miR-876-3p level and circ-RNF111 level in GC tissues (Fig. 3F). Besides, si-circ-RNF111 or circ-RNF111 was successfully transfected into AGS and SNU-638 cells to reduce or elevate circ-RNF111 expression, which was demonstrated by qRT-PCR assay (Fig. 3G). Of note, our results exhibited that circ-RNF111 silencing increased miR-876-3p expression in AGS and SNU-638 cells, while circ-RNF111 overexpression showed the opposite results (Fig. 3H). To summarize, circ-RNF111 directly targeted miR-876-3p to regulate miR-876-3p expression.
Overexpression of miR-876-3p inhibited GC cell growth, metastasis and glycolysis and facilitated apoptosis and cell cycle arrest by targeting KLF12 To investigate the relationship between miR-876-3p and KLF12 in regulating GC progression, rescue experiments were conducted. As shown in Fig. 6A, B, miR-876-3p overexpression reduced the mRNA and protein levels of KLF12 in AGS and SNU-638 cells, while KLF12 overexpression vector transfection rescued the impacts. CCK-8 assay and colony formation assay indicated that miR-876-3p overexpression suppressed the viability and colony formation of AGS and SNU-638 cells, with KLF12 elevation reversed the impacts (Fig. 6C, D). As demonstrated by flow cytometry analysis, miR-876-3p overexpression facilitated AGS and SNU-638 cell apoptosis, while the effect was abrogated by elevating KLF12 (Fig.   Fig. 4 Silencing of circ-RNF111 suppressed the progression of GC cells by targeting miR-876-3p. AGS and SNU-638 cells were transfected with si-NC, si-circ-RNF111, si-circ-RNF111 + anti-miR-NC, or si-circ-RNF111 + anti-miR-876-3p. A The expression of miR-876-3p in AGS and SNU-638 cells was detected by qRT-PCR assay. B Cell viability, C colony formation, D apoptosis, E migration, and F invasion in AGS and SNU-638 cells were evaluated using CCK-8 assay, colony formation assay, flow cytometry analysis, wound-healing assay, and transwell assay, respectively. G Cell cycle process of AGS and SNU-638 cells was analyzed by flow cytometry analysis. H, I The protein levels of CyclinD1 and MMP9 in AGS and SNU-638 cells were measured by western blot assay. J-L The levels of glucose consumption, lactate production and ATP production in AGS and SNU-638 cells were examined with specific kits. M, N The protein level of HK-2 in AGS and SNU-638 cells was measured by western blot assay. *P < 0.05, **P < 0.01, ***P < 0.001 6E). The results of wound-healing assay and transwell assay, miR-876-3p upregulation suppressed the capacities of AGS and SNU-638 cells to migrate and invade, whereas these effects were ameliorated by KLF12 overexpression (Fig. 6F, G). The cell cycle process of AGS and SNU-638 cells was arrested by miR-876-3p overexpression, with KLF12 elevation abrogated the effect (Fig.  6H). Western blot assay showed that miR-876-3p overexpression reduced CyclinD1 and MMP9 protein levels in AGS and SNU-638 cells, with KLF12 elevation abrogated the effects (Fig. 6I, J). Besides, overexpression of miR-876-3p reduced the levels of glucose uptake, lactate production, ATP production, as well as HK-2 protein in AGS and SNU-638 cells, while KLF2 elevation abated the impacts (Fig. 6K-N). Taken together, miR-876-3p overexpression suppressed the malignancy of GC cells by targeting KLF12.
Circ-RNF111 knockdown suppressed KLF12 expression by targeting miR-876-3p As exhibited in Fig. 7A, B, circ-RNF111 silencing led to an apparent suppression in KLF12 protein level in both AGS and SNU-638 cells, while miR-876-3p inhibition restored the impact. The findings indicated that circ-RNF111 positively regulated KLF12 expression in GC cells by adsorbing miR-876-3p.

Knockdown of circ-RNF111 blocked tumor formation in vivo
At last, the functional role of circ-RNF111 in tumor progression in vivo was explored. Our results presented that tumor size and tumor weight were inhibited after circ-RNF111 knockdown compared to sh-NC control groups ( Fig. 8A-C). Moreover, the levels of circ-RNF111, KLF12 mRNA, and KLF12 protein were declined and  3p. B, C The interaction between miR-876-3p and KLF12 was demonstrated by dual-luciferase reporter assay and RIP assay. D, E The mRNA expression level of KLF12 in GC tissues and cells was detected by qRT-PCR assay. F The correlation between the levels of miR-876-3p and KLF12 mRNA in GC tissues was evaluated. G, H The protein level of KLF12 in GC tissues and cells was tested by western blot assay. I-K The levels of miR-876-3p and KLF12 protein in AGS and SNU-638 cells transfected with miR-NC, miR-876-3p, anti-miR-NC, or anti-miR-876-3p were detected by qRT-PCR assay and western blot assay, respectively. *P < 0.05, **P < 0.01, ***P < 0.001 the level of miR-876-3p was enhanced in the tumors in sh-circ-RNF111 groups compared to control groups (Fig. 8D, E). IHC assay also showed that reduced KLF12 and Ki-67 levels in tumors in sh-circ-RNF111 groups compared to sh-NC groups (Fig. 8F). Collectively, circ-RNF111 contributed to tumorigenesis in vivo.

Discussion
Recently, circRNAs have attracted researchers' attention for their potential in cancer biology [25]. Substantial evidence implied that circRNAs are engaged in enhancing or repressing GC development. For instance, circ_001653 promoted GC malignancy by Fig. 6 MiR-876-3p overexpression repressed GC cell progression by interacting with KLF12. AGS and SNU-638 cells were administrated with miR-NC, miR-876-3p, miR-876-3p + pcDNA, or miR-876-3p + KLF12. A, B The mRNA and protein levels of KLF12 in AGS and SNU-638 cells were detected by qRT-PCR assay and western blot assay, respectively. C Cell viability, D colony formation, E apoptosis, F migration, G invasion, and H cell cycle process in AGS and SNU-638 cells were evaluated. I, J The protein levels of CyclinD1 and MMP9 in AGS and SNU-638 cells were measured via western blot assay. K-M The levels of glucose consumption, lactate production and ATP production in AGS and SNU-638 cells were examined with kits. N The protein level of HK-2 in AGS and SNU-638 cells was measured via western blot assay. **P < 0.01, ***P < 0.001 upregulating NR6A1 via decoying miR-377 [26]. Circ_ 0027599 directly targeted miR-101-3p.1, leading to the suppression in GC cell growth and metastasis [14]. Herein, we clarified the functions of circ-RNF111 in GC development. As a result, the upregulation of circ-RNF111 triggered the malignant phenotypes of GC cells. Furthermore, we discovered a novel pathway of circ-RNF111/miR-876-3p/KLF12 in regulating GC development.
Tang et al. manifested that circ_0001982 was overexpressed in breast cancer, and promoted tumor cell growth and invasion through adsorbing miR-143 [15]. Deng et al. demonstrated the oncogenic role of circ_ 0001982 in colorectal cancer development through Fig. 7 Circ-RNF111 regulated KLF12 expression by sponging miR-876-3p. A, B After AGS and SNU-638 cells were transfected with si-NC, si-circ-RNF111, si-circ-RNF111 + anti-miR-NC, or si-circ-RNF111 + anti-miR-876-3p, the protein level of KLF12 was measured with western blot assay. **P < 0.01, ***P < 0.001 Fig. 8 Circ-RNF111 silencing repressed tumor growth in vivo. A Tumor volume was monitored. B Xenograft tumors were presented. C Tumor weight was examined after 32 days. D The levels of circ-RNF111, miR-876-3p, and KLF12 mRNA in collected tumors were detected by qRT-PCR assay. E The protein level of KLF12 in collected tumors was measured with western blot assay. F The levels of KLF12 and ki67 in collected tumors were evaluated by IHC assay. *P < 0.05, **P < 0.01 sponging miR-144 [16]. Moreover, Wang et al. uncovered that circ-RNF111 sponged miR-27b-3p to aggravate GC cell growth and metastasis and repress apoptosis [17]. Corresponding to the previous studies, we also demonstrated the promotional effect circ-RNF111 in GC. In the present research, circ-RNF111 was elevated in GC. Functionally, circ-RNF111 interference curbed GC cell viability, colony formation, motility, triggered apoptosis and blocked cell cycle process. Besides, tumor cells prefer to obtain energy to meet the need for their rapid growth rather than oxidative phosphorylation and the suppression of glycolysis plays a vital role to hamper tumor progression [27,28]. Therefore, we explored glycolysis level in GC cells and found that circ-RNF111 silencing decreased glucose uptake, lactate production, ATP synthesis and HK-2 levels, thereby suppressing glycolysis. In addition, to further explore the function of circ-RNF111, the murine xenograft model was established. It was demonstrated that circ-RNF111 silencing restrained tumor formation in vivo. All these findings demonstrated the promotional effect of circ-RNF111 on GC malignancy.
In conclusion, circ-RNF111 was abnormally increased in GC. Moreover, circ-RNF111/miR-876-3p/KLF12 axis could deteriorate the progression of GC by triggering tumor cell growth, metastasis, and glycolysis and curbing apoptosis. Our results might offer key clues to develop an effective treatment method for GC.