CircARVCF Contributes to Cisplatin Resistance in Gastric Cancer by Altering miR-1205 and FGFR1

Background: Chemoresistance is a major barrier to the treatment of human cancers. Circular RNAs (circRNAs) are implicated in drug resistance in cancers, including gastric cancer (GC). In this study, we aimed to explore the functions of circRNA Armadillo Repeat gene deleted in Velo-Cardio-Facial syndrome (circARVCF) in cisplatin (DDP) resistance in GC. Methods: The expression of circARVCF, microRNA-1205 (miR-1205) and fibroblast growth factor receptor 1 (FGFR1) was detected by quantitative real-time polymerase chain reaction (qRT-PCR), western blot assay or immunohistochemistry (IHC) assay. Cell Counting Kit-8 (CCK-8) assay and colony formation assay were performed to evaluate DDP resistance and cell colony formation ability. Transwell assay was conducted to assess cell migration and invasion. Flow cytometry analysis was done to analyze cell apoptosis. Dual-luciferase reporter assay and RNA immunoprecipitation (RIP) assay were manipulated to analyze the relationships of circARVCF, miR-1205 and FGFR1. Murine xenograft model was constructed to explore DDP resistance in vivo. Results: CircARVCF level was increased in DDP-resistant GC tissues and cells. CircARVCF silencing inhibited DDP resistance, colony formation and metastasis and induced apoptosis in DDP-resistant GC cells. CircARVCF directly interacted with miR-1205 and miR-1205 inhibition reversed circARVCF silencing-mediated effect on DDP resistance in DDP-resistant GC cells. FGFR1 served as the target gene of miR-1205. MiR-1205 overexpression restrained the resistance of DDP-resistant GC cells to DDP, but FGFR1 elevation abated the effect. In addition, circARVCF knockdown repressed DDP resistance in vivo. Conclusion: CircARVCF enhanced DDP resistance in GC by elevating FGFR1 through sponging miR-1205.


HIGHLIGHTS INTRODUCTION
Gastric cancer (GC) is a worldwide life-threatening malignant tumor with strong a ability of metastasis and proliferation (Guggenheim and Shah, 2013;Bray et al., 2018). The patients are often diagnosed at the late stage for lacking specific symptoms and diagnostic indicators in early GC, and then lead to poor prognosis (Song et al., 2017). Surgery and adjuvant chemotherapy are the major treatment methods for GC, although molecular targeted therapy has been developed. Cisplatin (DDP) is the main first-line drug for GC treatment (Wang et al., 2019a). However, drug resistance impedes the effectiveness of chemotherapy (Galluzzi et al., 2012;Marin et al., 2016). Based on the above reasons, studying the mechanism of chemoresistance is necessary for improving the survival of GC.
Circular RNAs (circRNAs) are newly discovered non-coding RNAs (ncRNAs) that are distinguished by signal-stranded closed loop (Chen and Yang, 2015). CircRNAs are dysregulated and play vital roles in the biological and pathological processes in human cancers (Wang et al., 2017). Moreover, circRNAs can act as the competitive endogenous sponge for microRNAs (miRNAs), which then directly targeted related mRNAs to alter gene expression (Zhong et al., 2018). CircRNAs are linked to the development of chemoresistance in GC (Cui et al., 2020). For instance, circ_0000144 restrained oxaliplatin (OXA) resistance of GC via altering miR-502-5p and ADAM9 (Gao et al., 2021). CircFN1 promoted the malignancy and DDP resistance of GC via decoying miR-182-5p (Huang et al., 2020). CircARVCF (hsa_circ_0092330) was formed by the exons of ARVCF gene and was found to be upregulated in GC (Ouyang et al., 2019). However, the functions of circARVCF in the carcinogenesis and drug resistance of GC are unclear.
The involvement of miRNAs in tumor progression and chemoresistance is widely characterized (Taheri et al., 2021). For example, miR-149 facilitated 5-FU resistance in GC through binding to TREM2 (Wang et al., 2020a). MiR-127-5p impeded the resistance of GC to Apatinib (Yu et al., 2021). MiR-1205 functioned as a tumor inhibitor in diverse cancers, such as colorectal cancer (Han et al., 2021), glioma , ovarian cancer (Wang et al., 2020b) as well as GC (Lin et al., 2020). Even so, it remains unclear whether miR-1205 is related to the drug resistance of GC. Moreover, fibroblast growth factor receptor 1 (FGFR1) was found to contain miR-1205 binding sites, but their relationship is still unknown.
In this paper, we elucidated the functions and relationships of circARVCF, miR-1205, and FGFR1 in regulating the chemoresistance of GC, attempting to find a novel target to relieve chemoresistance in GC.

Tissue Collection
The GC and matched adjacent non-tumor tissues were collected from 37 GC patients (20 males and 17 females, aged 25-71) at Huai'an First People's hospital. The enrolled patients received DDP-based chemotherapy before this surgery. According to the resistance of GC patients to DDP (Bagrodia et al., 2016), the patients were divided into two groups: chemoresistant group (n 20; tumor relapse during DDP-based chemotherapy) and chemosensitive group (n 17; no tumor recurrence during DDP-based therapy). All patients provided the written informed consents. The research granted approval by the Ethics Committee of Huai'an First People's Hospital.

Quantitative Real-Time Polymerase Chain Reaction
RNA isolation was manipulated with RNAiso Plus (Takara, Dalian, China). Then M-MLV Reverse Transcriptase reagent

Actinomycin D and RNase R Treatment
For the cyclization analysis of circARVCF, MKN-45/DDP, and AGS/DDP cells were interacted with Actinomycin D (Sigma-Aldrich) for indicated times. Total RNA was managed with RNase R (3 U/μg; Epicenter, Madison, WI, United States) at 37°C for 20 min. Afterward, the levels of circARVCF and ARVCF were detected via qRT-PCR.

Colony Formation Assay
MKN-45/DDP and AGS/DDP cells (300 cells/well) were planted into 6-well plates and maintained for 2 weeks. The medium was changed every 3 days. The colonies were fixed with 4% paraformaldehyde (Sigma-Aldrich) and then dyed with 0.1% crystal violet solution (Sigma-Aldrich). After that, the colonies were calculated.

Transwell Assay
By using the transwell insert chambers (Corning, Corning, NY, United States) with or without Matrigel (Corning) coverage, cell invasion and migration were assessed. In short, MKN-45/DDP and AGS/DDP cells (1 × 10 4 ) were resuspended in 200 μL serumfree medium was added into the top chamber. The 600 μL complete medium was filled into the lower chamber.
Following 24 h of incubation, the migrated or invaded cells were fixed with 4% paraformaldehyde (Sigma-Aldrich) anddyed with 0.1% crystal violet (Sigma-Aldrich). Next, the number of migrated or invaded cells was counted via a microscope (100×; Olympus, Tokyo, Japan) through the five random areas.

Flow Cytometry Analysis
After indicated transfection, MKN-45/DDP and AGS/DDP cells were resuspended in binding buffer and double-dyed with Annexin V-fluorescein isothiocyanate (FITC) and propidium iodide (PI) for 15 min without light based on the instructions of the Apoptosis Detection Kit (Beyotime, Shanghai, China). The apoptotic cells were detected using flow cytometry.

RNA Immunoprecipitation Analysis
The Magna RIP ™ reagent (Millipore, Bedford, MA, United States) was used for this analysis. In brief, MKN-45/ DDP and AGS/DDP cells were lysed in RIP buffer and then cell lysates were incubated with Anti-AGO2 or Anti-IgG coupled with magnetic beads. Then the immunoprecipitated RNA was extracted and qRT-PCR was used for circARVCF and miR-1205 levels.

Murine Xenograft Model
Beijing Vital River Laboratory Animal Technology Co., Ltd. were subcutaneously injected into the flank of the mice. After 7 days, the mice were intraperitoneally administrated with 6 mg/kg of DDP (Sigma-Aldrich) or PBS every 3 days (Zhi et al., 2015). The xenograft tumors were monitored every 7 days and tumor volume (0.5 × length × width 2 ) was calculated. After 28 days, the mice were euthanized and tumors were weighed and preserved at −80°C for further use. This animal experiment was permitted by the Ethics Committee of Animal Research of Huai'an First People's Hospital.

Statistical Analysis
Data analysis was conducted using GraphPad Prism 7. The experiments were repeated three times. The data were exhibited as mean ± SD. Differences were compared through Student's t-test or one-way ANOVA. It was defined as significant when p < 0.05.

CircARVCF Was Upregulated in DDP-Resistant GC Tissues and Cells
As exhibited in Figures 1A,B, circARVCF was overexpressed in GC tissues and cell lines compared to normal tissues and cell lines. To explore the functions of circARVCF in the chemoresistance of GC, the expression of circARVCF in DDPresistant GC tissues and cells was detected by qRT-PCR. The results showed that circARVCF was upregulated in DDPresistant GC tissues compared to DDP-sensitive GC tissues ( Figure 1C). Compared to MKN-45 and AGS cells, circARVCF was increased in MKN-45/DDP and AGS/DDP cells ( Figure 1D). CCK-8 assay showed that the cisplatin resistance of MKN-45/DDP and AGS/DDP cells was enhanced for the elevation of IC50 of cisplatin ( Figures 1E,F). Moreover, Actinomycin D assay indicated that circARVCF possessed a longer half-life than linear ARVCF after Actinomycin D exposure ( Figures 1G,H). RNase R assay indicated that circARVCF was resistant to RNase R digestion, while linear ARVCF was digested by RNase R (Figures 1I,J).

CircARVCF Knockdown Repressed the Resistance of DDP-Resistant GC Cells to DDP
To explore the exact roles of circARVCF in DDP resistance in GC, si-circARVCF#1, si-circARVCF#2, or si-circARVCF#3 was transfected into MKN-45/DDP and AGS/DDP cells to silence circARVCF expression. As presented in Figure 2A, the transfection of si-circARVCF#1, si-circARVCF#2, or si-circARVCF#3 markedly reduced circARVCF expression in MKN-45/DDP and AGS/DDP cells compared to si-NC transfected cells. Then CCK-8 assay showed that circARVCF knockdown suppressed DDP resistance in MKN-45/DDP and AGS/DDP cells compared to si-NC control groups ( Figures  2B,C). Colony formation assay showed that circARVCF silencing suppressed the colony formation capacity of MKN-45/DDP and AGS/DDP cells compared to si-NC control groups ( Figure 2D). The results of transwell assay indicated that circARVCF knockdown repressed the migration and invasion of MKN-45/DDP and AGS/DDP cells in comparison with si-NC control groups (Figures 2E,F). Flow cytometry analysis exhibited that circARVCF deficiency led to a marked promotion in the apoptosis of MKN-45/DDP and AGS/DDP cells ( Figure 2G). In addition, our results showed that circARVCF interference decreased the levels of drug resistance-related genes (MRP1

CircARVCF Directly Targeted miR-1205 to Regulate miR-1205 Expression
It was widely documented that circRNAs can act as the sponge for miRNAs (Zhong et al., 2018). Thus, we further explored the underlying mechanism of circARVCF in the chemoresistance of GC through searching circinteractome (https://circinteractome. nia.nih.gov/api/v2/mirnasearch?circular_rna_query hsa_circ_ 0092330&mirna_query &submit miRNA+Target+Search) and found that miR-1205 was the target of circARVCF ( Figure 3A). MiR-1205 mimic transfection apparently increased miR-1205 expression and miR-1205 inhibitor (in-miR-1205) transfection markedly reduced miR-1205 expression in MKN-45/DDP and AGS/DDP cells ( Figure 3B). Then dual-luciferase reporter assay and RIP assay were performed to analyze the relationship between circARVCF and miR-1206. The results showed that miR-1205 overexpression restrained the luciferase activity of circARVCF WT in MKN-45/DDP and AGS/DDP cells, but had no effect on circARVCF MUT luciferase activity ( Figures 3C,D). RIP assay showed that circARVCF and miR-1205 enrichment was increased by Anti-AGO2 RIP compared to Anti-IgG and Input control groups ( Figures  3E,F). As expected, miR-1205 was lowly expressed in GC tissues and cells compared to normal tissues and cells ( Figures 3G,H). Furthermore, miR-1205 was downregulated in DDP-resistant GC tissues and cells in comparison with DDP-sensitive GC tissues and cells ( Figures 3I,J). In addition, it was found that circARVCF knockdown significantly increased miR-1205 expression in MKN-45/DDP and AGS/DDP cells, while in-miR-1205 transfection reversed the effect ( Figure 3K).  4A,B). Colony formation assay showed that circARVCF silencing repressed MKN-45/DDP and AGS/DDP cell colony formation ability, while miR-1205 inhibition reversed the effect ( Figure 4C). As illustrated by transwell assay, circARVCF deficiency hampered MKN-45/DDP and AGS/DDP cells to migrate and invade, whereas these impacts were weakened by reducing miR-1205 ( Figures 4D,E). Flow cytometry analysis indicated that circARVCF silencing facilitated the apoptosis of MKN-45/ DDP and AGS/DDP cells, with miR-1205 inhibition abated the impact ( Figure 4F). In addition, circARVCF silencing reduced the levels of MRP1, MDR1 and Bcl-2 and elevated the level of Bax in MKN-45/DDP and AGS/DDP cells, while miR-1205 downregulation reversed the effects (Figures 4G,H). These findings suggested that circARVCF inhibited DDP resistance in DDP-resistant GC cells by binding to miR-1205.

MiR-1205 Overexpression Suppressed DDP Resistance in DDP-Resistant GC Cells via Regulating FGFR1 Expression
To explore whether miR-1205 could regulate the resistance of DDP-resistant GC cells to DDP, MKN-45/DDP, and AGS/DDP cells were introduced with miR-NC, miR-1205, miR-1205 + pcDNA or miR-1205 + FGFR1. As indicated by CCK-8 assay, miR-1205 overexpression inhibited DDP resistance in MKN-45/ DDP and AGS/DDP cells, while FGFR1 elevation ameliorated the effect ( Figures 6A,B). The results of colony formation assay, transwell assay and flow cytometry analysis exhibited that miR-1205 overexpression restrained cell colony formation, migration and invasion and promoted apoptosis in MKN-45/DDP and AGS/DDP cells, while these impacts were abrogated by increasing FGFR1 expression ( Figures 6C-F). Additionally, miR-1205 overexpression reduced MRP1, MDR1 and Bcl-2 levels and elevated Bax level in MKN-45/DDP and AGS/DDP cells, with FGFR1 enhancement ameliorated the effects ( Figures 6G,H). To sum up, miR-1205 overexpression inhibited the resistance of DDP-resistant GC cells to DDP through interacting with FGFR1.

CircARVCF Knockdown Suppressed DDP Resistance in Vivo
Finally, the role of circARVCF in DDP resistance in vivo was investigated. Our results showed that circARVCF knockdown and DDP treatment repressed tumor growth (including tumor volume and tumor weight) ( Figures 7A,B). As determined by qRT-PCR, western blot assay and IHC assay, the levels of circARVCF and FGFR1 were decreased and the level of miR-1205 was increased in the tumors in sh-circARVCF + PBS and sh-circARVCF + DDP groups compared to sh-NC + PBS and sh-NC + DDP groups ( Figures 7C-F). These results suggested the suppressive role of circARVCF knockdown on DDP resistance in GC in vivo.

DISCUSSION
Chemoresistance is the main obstacle in the process of human cancer therapy (Marin et al., 2016). CircRNAs have been disclosed to play vital roles in tumor advancement and therapeutic resistance (Wei et al., 2020). Even so, there are still large amounts of circRNAs that have not been identified. In this study, the functions of circARVCF in DDP resistance were clarified. As a result, circARVCF inhibited the sensitivity of GC to DDP through a new regulatory axis of circARVCF/ miR-1205/FGFR1. Currently, the involvement of circRNAs and miRNAs in regulating GC chemoresistance has been reported. For example, Zhong et al. declared that circ_0032821 facilitated OXA resistance and malignant cell behaviors in GC partially via influencing miR-515-5p/SOX9 axis (Zhong et al., 2021). Zhang et al. demonstrated the suppressive role of circ_0026359/miR-1200/POLD4 axis in DDP sensitivity of GC . Circ-PVT1 enhanced the resistance of GC cells to paclitaxel by increasing ZEB1 via sponging miR-124-3p (Liu et al., 2019). Circ_0110805 aggravated the cisplatin resistance of GC by sponging miR-299-3p and elevating ENDOPDI (Yang et al., 2020a). CircAKT3/miR-198/PIK3RI played a promotional effect on cisplatin resistance of GC . These reports all indicated that chemoresistance could be modulated by circRNAs. In the present study, circARVCF was increased in DDP-resistant GC tissues and cells. Next, we performed experiments to explore the functions of circARVCF in DDP resistance of GC. It was found CircRNAs can be involved in the chemoresistance of cancers by binding to miRNAs and then alter miRNA target gene expression (Hansen et al., 2013). Herein, circARVCF could interact with miR-1205, which then targeted FGFR1 to alter FGFR1 expression. MiR-1205 has been reported to repress the metastasis and proliferation of GC cells via circCYFIP2/miR-1205/E2F1 pathway (Lin et al., 2020). Moreover, miR-1205 was associated with sorafenib resistance in hepatocellular carcinoma by interacting with circFN1 and E2F1 (Yang et al., 2020b).
Nonetheless, we firstly explored the functions of miR-1205 in DDP sensitivity of GC. It was found that miR-1205 was reduced in DDP-resistant GC. Inhibition of miR-1205 restored circARVCF knockdown-mediated inhibitory influence on DDP sensitivity of chemo-resistant GC cells. Overexpression of miR-1205 enhanced DDP sensitivity and restrained malignant cell phenotypes in DDP-resistant GC cells.
FGFR1 can be activated to influence the proliferation, survival, migration and differentiation of tumor cells (Neal et al., 1985;Guo et al., 2017). FGFR1 could serve as the target for diverse miRNAs, such as miR-198 (Gu et al., 2019), miR-497 (Xie et al., 2018), and miR-133b (Wen et al., 2013). However, FGFR1 was verified to be the target gene of miR-1205 for the first time. Moreover, FGFR1 participated in the regulation of chemosensitivity in several cancers Kong et al., 2020). In the research, FGFR1 could combine with miR- 1205 to alter DDP resistance in GC. In addtion, FGFR1 has been reported to regulate the biologocal processess, including wound repair, tissue regeneration, inflammation and angiogenesis, which are implcated in cancer progression and drug resistance (Liang et al., 2021;Yang et al., 2021). However, whether circARVCF/miR-1205/FGFR1 axis is involved in these biological processes to regulate DDP resistance is still unclear. Taken together, circARVCF contributed to DDP resistance and promoted tumor cell proliferation, migration, and invasion in GC by regulating miR-1205 and FGFR1 expression (Figure 8). This study deepened our understanding of the progression of chemoresistance and provided a novel insight to resist drug resistance in GC.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding authors.