circRNA circSCAF8 promotes tumor growth and metastasis of gastric cancer by miR-1293/TIMP1signaling

doi:10.21203/rs.3.rs-3718138/v1

Download PDF

Article

circRNA circSCAF8 promotes tumor growth and metastasis of gastric cancer by miR-1293/TIMP1signaling

https://doi.org/10.21203/rs.3.rs-3718138/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

SCAF8 (SR-like CTD-associated factor 8) can regulate transcriptional termination, but the function of circSCAF8 is still unclear. In our study, we found circSCAF8 expression was significantly increased in gastric cancer, especially in gastric cancer tissues with lymph node metastasis. Kaplan-Meier curve indicated high circSCAF8 expression group possessed low overall survival time in gastric cancer. Moreover, circSCAF8 shRNA viruses could decrease gastric cancer proliferation, invasion, and migration in vitro. By using bioluminescence imaging (BLI) technology in vivo experiments, we also found circSCAF8 shRNA viruses could inhibit the growth of xenograft tumors and gastric cancer lung metastasis. RIP and circRNA pulldown assays confirmed circSCAF8 could directly bind to miR-1293, but circSCAF8 could not regulate the expression of miR-1293 in gastric cancer. Interestingly, the downstream gene TIMP1 of miR-1293 could be regulated by circSCAF8, and this view was further verified in gastric cancer tissues. Moreover, we confirmed that miR-1293 could directly inhibit TIMP1 expression. Subsequently, through rescue experiments, we found TIMP1 overexpression could reverse the effect of circSCAF8 shRNA viruses on gastric cancer. In conclusion, circSCAF8 expression was increased in gastric cancer, and circSCAF8 shRNA viruses could inhibit gastric cancer growth and metastasis by elevating TIMP1 expression via miR-1293.

Health sciences/Biomarkers

Health sciences/Diseases/Cancer/Gastrointestinal cancer/Gastric cancer

circRNA

miRNA

gastric cancer

tumorigenesis

Gastric cancer is one of the most prevalent cancers worldwide, with over 1 million cases annually, constituting 5.7% of all cancer diagnoses. It is the fourth most common cancer in males and the seventh most common in females[1]. The prognosis of cancer is poor, with a 5-year survival rate of only 42.9% due to the majority of cases having already metastasized at the time of diagnosis[2–4]. Helicobacter pylori infection, smoking and high salt diet were still the main risk factors of gastric cancer[5, 6]. In addition to those factors, upregulated oncogenes and downregulated tumor suppressor genes also played critical roles in gastric cancer progression. In recent years, myriad studies showed non-coding RNA was involved in gene regulation. Liu T et al found miR-124 overexpression can negatively modulate AKT2 and thus inhibit the progression of lung adenocarcinoma[7]. Miliotis C et al demonstrated miR-105-5p regulates PD-L1 expression and tumor immunogenicity in gastric cancer[8]. Zhou T et al revealed long noncoding RNA SPINT1-AS1 promotes breast cancer proliferation and metastasis by sponging let-7 a/b/i-5p[9].

Circular RNA (circRNA) is a non-coding RNA that can regulate tumor related genes. Most circRNAs are generated from the backsplicing of exon. With the help of Alu elements[10] and RNA binding protein (RBP)[11], splice-donor (SD) can covalently bind to splice acceptor (SA) to form a closed circular RNA loop[12, 13]. Recent studies illuminated the participation of circRNAs in tumorigenesis. For instance, circ_0007142 can promote colorectal cancer proliferation, invasion and migration by sponging miR-103a-2-5p[14]. CircNT5E enhances glioblastoma tumorigenesis by directly suppressing miR-422a expression[15]. Additionally, circFOXK2 have been demonstrated to promote pancreatic ductal adenocarcinoma tumorigenesis by sponging miR-142 and interacting with YBX1 and hnRNPK[16]. Although circRNA can directly bind to promoter region[17] or protein[18] to modulate gene expression, the main function of circRNA is to sponge miRNA and sequester miRNA function.

In our study, we identified several circRNA candidates by overlapping two GEO databases. Among those candidates, circSCAF8 as the 3rd highest expression circRNA arrested our attention. Using qRT-PCR, we conducted circSCAF8 expression was significantly increased in gastric cancer cells and gastric cancer tissues. Meanwhile, by RNA immunoprecipitation (RIP), circRNA pull down assay and luciferase reporter assay, we found circSCAF8 could directly bind to miR-1293. Finally, we demonstrated that although cirsSCAF8 could not directly regulate the expression of miR-1293, it could regulate the downstream gene TIMP1 of miR-1293, thereby promoting the occurrence of gastric cancer. Overall, our study unveiled a unique role of circSCAF8 in regulating gastric tumorigenesis.

Human tissue specimens and cell lines

Thirty-four gastric cancer tissues and paired adjacent normal tissues were excised from the Department of Surgery, Tongji Hospital of Tongji Medical College between 03/01/2017 and 06/01/2017. This project was approved by the Institutional Review Committee of Tongji Medical College of Huazhong University of Science and Technology, and all patients signed the informed consent forms.

The normal gastric mucosal cell line GES-1 was purchased from the Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China), and gastric cancer cell lines (AGS, MKN-45) were purchased from the American Type Culture Collection (ATCC; Manassas, VA, USA). Cell lines were authenticated by using Short Tandem Repeat (STR) analysis which has been uploaded as the additional files. All cell lines were cultured in RPMI 1640 (Cat. #SH30096.01, HyClone, Logan, UT, USA) containing 10% foetal bovine serum (FBS) (Cat. #16000044, Invitrogen, Carlsbad, CA, USA) in humidified incubator at 37°C with 5% CO₂.

RNA preparation and real-time PCR

The primers of miRNA, circRNA, linear RNA, mRNA and internal reference genes were presented in Table 2. Total RNA and miRNA were extracted by mirVana™ miRNA Isolation Kit (Cat. #AM1560, ThermoFisher, Waltham, MA, USA) according to vendor’s protocol. MystiCq® microRNA cDNA Synthesis Mix kit (Cat. #MIRRT-25RXN, Sigma, St. Louis, MO, USA) was utilized to add poly-A tails to miRNA and perform reverse transcription to synthesize cDNA for miRNA following the manufacturer's procedures. cDNA for total RNA was synthesized using iScript™ Advanced cDNA Synthesis Kit (Cat. #1725038, Bio-Rad, Hercules, CA, USA). 20ng cDNA, 0.2µl forward primer (10µm/µl), 0.2µl reverse primer (10µm/µl), 5µl SYBR ® Premix Ex Taq™ II (Cat. #1725122, Bio-Rad, Hercules, CA, USA) and RNase free water was mixed together to total 10µl. Then the mix was pipetted into 384 plates to perform Real-time PCR in a CFX 384™ Real-Time PCR Detection System (Bio-Rad). The PCR procedure was 95°C for 30s, followed by 40 cycles of 95°C for 5s and 60°C for 30s. The miRNA data was normalized by U6 internal reference gene and circRNA, linear RNA or mRNA was normalized by GAPDH internal reference gene. In pulldown assays, relative RNA expression was normalized to input data. The gene expression levels were measured by Cycle threshold (Ct) values, and the relative expression level of gene was normalized to internal reference genes or input data. The 2^−ΔΔCt method was used to normalize the concentration of tissue or cell line samples.

Nucleic acid electrophoresis

To examine the cDNA and gDNA PCR products of circSCAF8 or linear SCAF8, 2% agarose (Cat. # 16500100, ThermoFisher, Waltham, MA, USA) gels were prepared. After aspirating DNA product into different wells, we separated PCR products for 30 m by electrophoresis at 100V. D0107 (Beyotime, China) was used as the DNA marker. Finally, circSCAF8 or linear SCAF8 expression was detected using UV irradiation.

RNase R digestion

The RNase R digestion assay was performed using RNase R kit (Cat. # M1228-500, Biovision, Milpitas, CA, USD). First, we mixed 4µg total RNA, 2µl 10×reaction buffer (20mM Tris-HCL, PH = 8.0, 0.1M KCl and 0.1mM MgCl₂), 12U (1.2µl) RNase R and RNase free water together to total 20µl. Then we incubated the mixture in water bath for 15 min at 37°C followed by a water bath for 15 min at 65°C to inactivate the RNase R. Finally, total RNA with or without RNase R treatment was converted into cDNA to perform qRT-PCR.

RNA isolation of nuclear and cytoplasmic fractions

To figure out the subcellular localization of circSCAF8, PARIS Kit (Cat. #AM1921, ThermoFisher, Waltham, MA, USA) was used according to vendor’s protocol.

Immunoprecipitation

First, we mixed Dynabeads™ Protein G (Cat. #10004D, Invitrogen, Camarillo, CA, USA) with Ago2 and IgG antibody. After incubating with rotation at room temperature for 10 min, the mixture was rinsed with TBST. Then the mixture was placed on the magnet and the supernatant was carefully discarded. Second, lysis buffer (Cat. #AM1560, ThermoFisher, Waltham, MA, USA) was used to lyse AGS cells. Then we added the AGS cells lysis into the mix containing beads and antibody. After incubating for 10 min in water bath at 37°C, we placed the complex on the magnet and then removed the supernatant to another tube as the input. Subsequently, we washed the complex 3 times by washing buffer on the magnet, then the tube was removed from the magnet and the complex was resuspended in washing buffer. After placing the tube on the magnet and discarding the supernatant, we added elution buffer into the tube. After incubating at room temperature for 2 min, the tube was placed on the magnet and supernatant was removed into a new tube to perform reverse transcription and qRT-PCR.

circRNA pull-down assay

We designed special circSCAF8 probe against the backsplicing junction of circRNA. The circRNA probe was made up of 2 parts which possessed 50nt fragments in each exon. To synthesize the special biotin-circRNA probe, T7 promoter sequences was added into the primers. Primers used for circRNA probe was present in Table 2. By using RT-PCR, DNA template was produced. Following this, special biotin-circSCAF8 probe was synthesized by using T7 High Yield Transcription Kit (Cat. #K0441, ThermoFisher, Waltham, MA, USA) with the existence of Biotinylated UTP (Cat. #AM8452, ThermoFisher, Waltham, MA, USA) according to the manufacturer’s instruction. After purifying by trizol and DNase I, biotin-circSCAF8 probe and control RNA probe (Geneseed, Guangzhou, China) was transfected into AGS cells by Lipofectamine 3000 (Cat. #L3000150, Invitrogen, Camarillo, CA, USA). After 24 h, AGS cells were harvested and lysed by lysis buffer. The washed magnetic beads were added into cell lysate. After incubating for 2 h, the tube was placed on low-speed rotator (10r/min) for 2 h at room temperature. Then the tube was positioned on the magnet and the supernatant was removed to a clean tube as the input. After washing 5 times by washing buffer, the tube was removed from magnet and elution buffer was added. After incubating at room temperature for 2 min, the tube was placed on the magnet and supernatant was removed into a new tube for subsequent reverse transcription and qRT-PCR.

Luciferase reporter assay

circSCAF8 full length sequences were acquired from AGS cDNA by Real-time PCR. After purifying by column (Cat. #AM1560, ThermoFisher, Waltham, MA, USA), DNA product was inserted into luciferase vector (Geneseed, Guangzhou, China). By using wildtype circRNA vector as the template and site-directed Gene Mutagenesis Kit (Beyotime, China), circSCAF8 mutant luciferase vector was constructed following the manufacturer’s protocol. Similarly, wild-type TIMP1 3’UTR vector and mutant TIMP1 3’UTR vector were constructed as previously described. The luciferase vector and miR-1293 mimic were co-transfected into AGS cells using Lipofectamine 3000. Subsequently, we examined the luciferase activity of cells by dual-luciferase reporter assay system (Promega) at 48 h after transfection. Renilla luciferase activity was normalized to the Firefly luciferase internal control, and relative luciferase activity was calculated. Primers for wild type and mutation type vector were shown in Table 2.

circSCAF8 Knockdown by RNA-interference

The BJS (backsplicing junction sites) of circSCAF8 was chosen for designing siRNA sequence. With the help of RNAi prediction tools(http://biodev.extra.cea.fr/DSIR/DSIR.html), circSCAF8 siRNA sequence was designed. The primers used for RNAi were shown in Table 2 and shRNA oligonucleotides were synthesized:

shRNA-a: Ccgg (21 nt siRNA) CTCGAG (21 nt antisence-siRNA) TTTTTg

shRNA-b: aattcaaaaa (21 nt siRNA) CTCGAG (21 nt antisence-siRNA)

circSCAF8 shRNA oligonucleotides were inserted into GV493 vector (Order NO. GIDL0225568, GeneChem Corporation, Shanghai, China). To package lentivirus, GV493, pHelper 2.0 and pHelper 1.0 were co-transfected into HEK293T cells using Lipofectamine 3000 (Cat. #L3000150, Invitrogen, Camarillo, CA, USA). After culturing for 48 h, viruses were harvested. Then concentration and purification for viruses were performed. Empty viruses were bought from GeneChem Corporation (Order NO. GIDL0225568, Shanghai, China). AGS cells, MKN-45 cells and AGS cells containing stable luciferase gene were infected by circSCAF8 shRNA or empty viruses. AGS cells and MKN-45 cells were used to perform in vitro experiments after 48 h. AGS cells containing stable luciferase gene were used for in vivo experiments after treating with puromycin (2µg/µl) (Cat. #A1113803, Invitrogen, Camarillo, CA, USA).

Oligonucleotide Transfection and Expression Plasmid Construction

miR-1293 mimic and negative control mimic were purchased from Sangon Biotechnology (Sangon, Shanghai, China). Using Lipofectamine 3000, miR-1293 mimic and negative control mimic was respectively transfected into AGS or MKN 45 cells for further experiments. The CDS fragments of TIMP1 were amplified from AGS cDNA via PCR. After PCR product purification by miRNAVana miRNA kit (Cat. #AM1560, ThermoFisher, Waltham, MA, USA), those CDS fragments were inserted into pCDNA3.1 as previously described. Lipofectamine 3000 (Cat. #L3000150, Invitrogen, Camarillo, CA, USA) was used to transfect the plasmid into AGS cells to perform the further experiments. The primers used for PCR amplification were shown in Table 2.

Antibodies and Immunoblotting

TIMP1 (#ab211926) and β-Actin (#ab179467), were purchased from Abcam company (Abcam, Cambridge, UK). HRP-conjugated goat anti-rabbit IgG (#sc-2004) was purchased from Santa Cruz Biotechnology (Santa Cruz Biotechnology, Santa Cruz, CA, USA). Briefly, using RIPA lysis buffer (Cat. # P0013E, Beyotime, China) cancer cells were lysed. 10% SDS polyacrylamide gel (Cat. # P0012A, Beyotime, China) electrophoresis was performed to separated proteins. After electrophoresing, the separated proteins were transferred onto PVDF membranes (Cat. # FFP24, Beyotime, China). After blocking in 5% non-fat milk in TBST for 1 h, PVDF membranes were steeped in primary antibodies overnight. After overnight, membranes were rinsed by TBST for three times. Then membranes were steeped in secondary antibodies for 1 h. Finally, the protein bands were detected by the enhanced chemiluminescence reagent (cat. # WBKLS0500, Millipore, Billerica, MA, USA).

Clonability and Cell Viability Assays

Colony formation assay: AGS and MKN-45 cells were cultured in 6-well plate at a very low density (500 cells/plate). After growing for 2 weeks, the plate was stained by Giemsa. Finally, the cell colonies were counted.

Cell viability assay: AGS and MKN-45 cells were cultured in 96-well plates at a density of 1×10⁴ cells/well. The cell viability was measured by the cell counting kit-8 (CCK-8) (Dojindo, Kumamoto, Japan) following vendor’s instruction. Briefly, 10µl of CCK-8 solution was aspirated into each well, and the 96-well plates were incubated at 37°C for 1 h. The absorbance of each well at 450 nm was measured using a microplate reader (Tecan, Switzerland).

Migration and Invasion Assays

For cell migration assays, 1×10⁵ AGS and MKN-45 cells were cultured on 24-well plate inserted with 8µm pores chamber. For cell invasion assays, 1×10⁵ AGS and MKN-45 cells were cultured on the top chamber with a Matrigel-coated membrane (Cat. # 353093, 24-well insert; 8 mm pore size; BD Biosciences). In the lower chamber, cells were cultured in medium containing 10% serum, while in the upper chamber, cells were cultured in serum-free medium. After 16 h, cotton-tipped swabs were used to erase cells from the upper sides of the membrane. In the end, migrated/invaded AGS or MKN-45 cells were stained by crystal violet and counted.

In vivo tumour growth assays

All procedures were approved by the Committee on the Ethics of Animal Experiments of Tongji Hospital of Tongji Medical College, and this experiment was undertaken according to the guidelines for the Care and Use of Laboratory Animals of Tongji Medical College. Meanwhile, we used sodium pentobarbital anaesthesia to minimize suffering of nude mice throughout the experiment. During the experiment, the investigator was blinded to the group allocation. Male athymic nude mice (3–4 weeks of age), which bought from the Animal Experimental Center of Tongji Hospital of Tongji Medical College, were chosen to perform xenograft experiment. Before injecting stable gastric cancer AGS cells, nude mice were acclimated for 2 weeks. AGS cells (1×10⁶) containing empty virus or circSCAF8 shRNA virus were suspended in 100µl of PBS and then injected into the right or left flank of each mouse. To measure the xenograft tumour volume, formula was utilized: V = 1/2 a×b² (a: longest tumour axis; b: shortest tumour axis). Nude mice may be excluded out of group if the following situation occurs: death or lack of xenograft tumour growth. After 5 weeks, no mouse dies and xenograft tumour in each nude mouse grew up, and then all nude mice were sacrificed. Finally, tumours from mice were excised and preserved in liquid nitrogen. To conduct in vivo metastasis experiments, male SCID mice (3–4 weeks of age), also bought from the Animal Experimental Center of Tongji Hospital of Tongji Medical College, were picked up. We randomly split the nude mice into 2 groups (4 mice per group). Then we injected stable AGS cells (10⁶ cells in 200µL PBS) into the lateral tail veins of mice. By IVIS Spectrum system (Perkin-Elmer, Alameda, CA, USA), lung metastasis of each mouse was detected on day 15 and then sacrificed.

Statistical Analysis

The data were shown as mean ± standard deviation (s.d.). Difference between 2 groups were analysed by Student’s t-test. Spearman's rank test was performed to evaluate the relationships between the relative expression of circSCAF8, miR-1293 and TIMP1 in gastric cancer tissues. Kaplan-Meier (log-rank test) analysis was used to calculate overall survival time. Differences were considered significant at p < 0.05.

circRNAs selection and identification

GEO database can be utilized to screen different gene, miRNA, lncRNA and circRNA expression. Unfortunately, only a single GEO database dataset was insufficient to analyze the differential expression of circRNAs. To address this issue, two GEO databases (GSE89143ID and GSE78092ID), which detected different circRNAs expression between gastric cancer tissues and normal adjacent tissues, were collaborated. We found 79 circRNAs were up-regulated and 2 circRNAs were down-regulated by combining those two GEO databases together. Hsa_circ_0005273[19](the highest circRNA in database) and hsa_circ_0000981[20](the second highest circRNA in database) has been studied in other research papers. In this case, circSCAF8 (hsa_circ_0078362, the third highest circRNA in database) attracted our attention and was picked up to further investigate its’ function and role in gastric cancer. We examined circSCAF8 expression by divergent primer and linear SCAF8 expression by convergent primer in AGS cDNA and gDNA. In AGS cDNA, both circSCAF8 and linear SCAF8 could be detected. But in AGS gDNA, only linear SCAF8 could be detected (Fig. 1A). Since circRNA didn’t possess Poly-A tail, circSCAF8 could not be detected when total RNA was converted into cDNA using oligo (dT) primers. As we hypothesized, circSCAF8 expression was significantly lower in AGS cDNA which was converted using random primers, while the expression of linear SCAF8 remained relatively consistent across different AGS cDNA (Fig. 1B). Through RNase R digestion assay, we found circSCAF8 could be resistant to RNase R treatment but linear SCAF8 and GAPDH couldn’t (Fig. 1C). Finally, we determined circSCAF8 mainly expressed in cytoplasm by nuclear and cytoplasmic extraction assay (Fig. 1D).

The expression and prognostic significance of circSCAF8 in gastric cancer

circSCAF8 expression in different cell lines were detected by qRT-PCR and we found circSCAF8 expression was higher in gastric cancer cells when compared to normal gastric mucosal cells (Fig. 2A). To figure out the expression of circSCAF8 in gastric cancer tissues, circSCAF8 expression in 34 paired gastric cancer tissues and adjacent normal tissues were measured by using qRT-PCR. The clinical characters were presented in Table 1. The results showed circSCAF8 expression was significantly elevated in gastric cancer tissues compared with adjacent normal tissues (Fig. 2B). Additionally, we observed circSCAF8 expression was higher in lymph node metastasis group than non-lymph node metastasis group (Fig. 2C). To investigate prognostic significance of circSCAF8 in gastric cancer, Kaplan-Meier (log-rank test) analysis was performed. All 34 patients were enrolled, and follow-up information at 3 years was complete in 74% of the patients. The result showed high expression circSCAF8 group possessed lower overall survival time than low expression circSCAF8 group (Fig. 2D).

circSCAF8 promote gastric cancer proliferation, invasion, migration and metastasis

Previous study confirmed the increased expression of circSCAF8 in gastric cancer, it remained unclear whether circSCAF8 could act as an oncogene to promote gastric tumorigenesis. To address this problem, following experiments were performed. In AGS and MKN-45 cells, circSCAF8 expression could be attenuated by circSCAF8 shRNA viruses, while linear SCAF8 expression remained unaffected (Fig. 3A). This attenuation of circSCAF8 expression led to the inhibition of AGS and MKN-45 cells proliferation (Fig. 3B). Meanwhile, gastric cancer metastasis ability could be curbed by circSCAF8 shRNA (Fig. 3C). Consistent with in vitro experiments, in vivo experiments also proved circSCAF8 shRNA could significantly mitigate the growth of xenograft tumour (Fig. 3D). Furthermore, the role of circSCAF8 in gastric cancer metastasis was also investigated. Using IVIS imaging system, we found circSCAF8 shRNA could inhibit gastric cancer lung metastasis (Fig. 3E). Overall, our in vitro and in vivo experiments provided strong evidence that gastric cancer tumorigenesis could be curbed by circSCAF8 shRNA viruses.

circSCAF8 directly bind to miR-1293

The main function of circRNA is to sponge miRNA. To figure out whether circSCAF8 can sequester miRNA function, RIP assay was performed. Ago2 protein, which can form RNA-induced silencing complex (RISC) by binding to miRNA, can degrade mRNA or prevent mRNA translation[21]. We utilized Ago2 antibody to pull down Ago2 protein complex and examined circSCAF8 expression. The results showed circSCAF8 expression could be detected in the Ago2 protein complex (Fig. 4A), indicating that circSCAF8 could bind to miRNA in gastric cancer. Using predictive tool, we found circSCAF8 could bind to 7 miRNAs (http://www.circbank.cn). To figure out the relationship between circSCAF8 and these miRNAs, we designed special biotin-circSCAF8 probe by using T7 in vitro transcription assay with the existence of Biotinylated UTP. After pulling down by biotin-circSCAF8, miRNAs expressions were detected by qRT-PCR. The results indicated only miR-1293 could be pulled down by circSCAF8 while other 6 miRNAs couldn’t (Fig. 4B). To further analyse the relation between circSCAF8 and miR-1293, luciferase reporter assay was performed. Wild type of circSCAF8 and mutation type of circSCAF8, which binding sites of miR-1293 were mutated, were cloned into luciferase vector (PsicheckTM-2 Vector). Then we examined the relative luciferase activity in AGS cells after transfecting miR-1293 mimics or control mimics. We found the relative luciferase activity was attenuated in AGS cells after transfecting with miR-1293 mimics and wild type of circSCAF8 luciferase vector, but the relative luciferase activity didn’t alter in AGS cells after transfecting miR-1293 mimics and mutation type of circSCAF8 luciferase vector (Fig. 4C). Those findings showed circSCAF8 could directly bind to miR-1293. To figure out if circSCAF8 could regulate miR-1293 expression, we detected miR-1293 expression in AGS cells after transfecting with circSCAF8 shRNA viruses. The results showed circSCAF8 couldn’t affect miR-1293 expression (Fig. 4D). In summary, circSCAF8 could sequester miR-1293 function but couldn’t inhibit miR-1293 expression.

circSCAF8 regulate TIMP1 expression by miR-1293

miRNA predicted tool (http://mirtarbase.mbc.nctu.edu.tw/) was utilized to find the downstream target of miR-1293, we found several target genes could bind to miR-1293 (supplementary table 2). Among those candidates, strong evidence supported only TIMP1 could bind to miR-1293. Consequently, we initiated to investigate the relationship between TIMP1 and miR-1293. miR-1293 expression was significantly increased in AGS cells after transfecting with miR-1293 mimic (Fig. 5A). And then, TIPM1 expression was detected by western blot, we found the expression of TIMP1 was downregulated due to miR-1293 overexpression (Fig. 5B). To further investigate the relationship between TIMP1 and miR-1293, luciferase reporter assay was performed. Wild type of TIMP1 3’UTR luciferase vector and mutation type of TIMP1 3’UTR luciferase vector, which binding sites of miR-1293 were mutated, were constructed. The relative luciferase activity was decreased in AGS cells after transfecting with miR-1293 mimics and wild type of TIMP1 3’UTR luciferase vector, but the relative luciferase activity didn’t alter in AGS cells after transfecting with miR-1293 mimics and mutation type of TIMP1 3’UTR luciferase vector (Fig. 5C). The results indicated miR-1293 could directly inhibit TIMP1 expression. We also discovered circSCAF8 shRNA viruses could downregulate TIMP1 expression (Fig. 5D). Similar results were also observed by using qRT-PCR (Fig. 5E). Using TCGA data, we found TIMP1 was significantly increased in gastric cancer tissues compared to normal tissues (supplementary Fig. 1). To further study the relationship between circSCAF8, miR-1293 and TIMP1 in gastric cancer, the expression of TIMP1 and miR-1293 in 16 gastric cancer tissues and normal adjacent tissues were detected. In consistent with TCGA database results, the expression of TIMP1 in gastric cancer tissues was higher than normal tissues while the expression of miR-1293 in gastric cancer tissues was lower than normal tissues (Fig. 5F). Then the correlations between circSCAF8, miR-1293 and TIMP1 were analysed by Spearman's rank test. Those results indicated there was no significant correlation between circSCAF8 and miR-1293, meanwhile, there was also no significant correlation between miR-1293 and TIMP1. But there was significant positive correlation between circSCAF8 and TIMP1(Fig. 5G). We have known that circSCAF8 only could affect the function of miR-1293 but could not affect the expression of miR-1293. Now, this conclusion was also proved by our clinical tissue data. In gastric cancer tissues, no significant correlation between circSCAF8 and miR-1293 means circSCAF8 couldn’t regulate the expression of miR-1293, and positive correlation between circSCAF8 and TIMP1 indicates the expression of TIMP1 might be affected by circSCAF8. To further explore the function of circSCAF8 and miR-1293 in gastric cancer tumorigenesis, rescue experiments were performed. We detected the expression of TIMP1 in AGS cells after transfecting with an exogenous TIMP1 plasmid by western blot, and the results indicated a significant upregulation of TIMP1 expression (Fig. 5H). We also detected the expression of TIMP1 in control group, shRNA circSCAF8 group and simultaneous overexpressed TIMP1 and shRNA circSCAF8 group. Those results showed TIMP1 expression could be curbed by shRNA circSCAF8 and restored by overexpressed TIMP1 (Fig. 5I). Using CCK-8, clonability and transwell assays, we found circSCAF8 shRNA viruses could inhibit gastric cancer tumorigenesis, but this effect was restored by overexpressing TIMP1 (Fig. 5J).

By a yeast-two-hybrid screen for mammalian, SCAF8 (SR-like CTD-associated factors 8) was firstly discovered[22]. The carboxyl-terminal domain (CTD) of RNA polymerase II (Pol II) plays a pivotal role in transcription cycle[23–26]. The affinity of the CTD is elevated upon phosphorylation, but SCAF8 can disrupt the phosphorylation pattern on the CTD[27]. Additionally, SCAF8 acts as an anti-terminator, inhibiting alternative polyadenylation (polyA) sites. In other words, SCAF8 as the RNA PII elongation factor is involved in transcriptional termination[28]. circSCAF8 (hsa_circ_0078362), formed by splicing of SCAF8 pre-mRNA, located on chromosome 6q25.2 on the positive strand and spans exons 2–3. In our study, we found circSCAF8 expression was upregulated in gastric cancer cells and gastric cancer tissues compared to normal gastric mucosa cells and normal gastric tissues. Furthermore, we also found circSCAF8 expression was elevated in lymph node metastasis group than non-lymph node metastasis group. Those results demonstrated circSCAF8 was involved in gastric cancer metastasis. By Kaplan-Meier curve, we demonstrated high circSCAF8 possessed low survival time. It means circSCAF8 could act as a prognostic factor for gastric cancer.

Previous research has shown that when the termination of RNA Pol II was inhibited by depleting the polyadenylation/cleavage machinery, circRNA expression was significantly increased[29]. It means SCAF8 as the anti-terminator may be involved in the regulation of circRNA. We don’t know whether circSCAF8 can regulate the biogenesis of circRNA, but we found circSCAF8 can directly bind to miRNA by performing RIP assay. To figure out which miRNA could bind to circSCAF8, predicted tool and circRNA pull down assay was utilized. The results showed circSCAF8 could directly bind to miR-1293. To further confirm the relationship between circSCAF8 and miR-1293, dual luciferases report assay was employed. The results verified circSCAF8 could directly bind to miR-1293. Intriguingly, circSCAF8 knockdown vector couldn’t affect miR-1293 expression. We inferred although circSCAF8 could sponge miR-1293, the quantity of miR-1293 in gastric cancer cells didn’t alter. To further figure out if circSCAF8 could regulate downstream gene of miR-1293, we performed following experiments. By online tool miRbase, several downstream genes of miR-1293 were predicted. Among these predicted genes, only TIPM1 was supported by strong evidence[30]. TIPM1 expression was indeed attenuated by miR-1293 mimics. Dual luciferases report assay also confirmed miR-1293 could directly inhibit TIPM1 expression. Furthermore, circSCAF8 could increase TIMP1 expression. Those relationship between circSCAF8, miR-1293 and TIMP1 was also studied in gastric cancer tissues. We found no correlation was found between circSCAF8 and miR-1293, but strong positive correlation was found between circSCAF8 and TIMP1. Those clinical data mean circSCAF8 only can sequester miR-1293 function but can’t affect miR-1293 expression. Combining those results, we concluded circSCAF8 increased TIMP1 expression by sponging miR-1293 function.

miR-1293 is identified as a tumor-suppressive miRNA by performing function-based screening in 10 cancer cell lines. Moreover, it exhibits the ability to inhibit DNA repair gene RPA1, POLD4 and APEX1[31]. In other study, miR-1293 has demonstrated its tumor-suppressive function by suppressing the viability, invasion, and migration of renal cell carcinoma cells through binding to HAO2[32]. In summary, miR-1293 is a tumor-suppressive miRNA in various types of cancer. In our previous study, we have proved circSCAF8 can directly upregulate TIMP1 expression by sponging miR-1293 in gastric cancer. Since many studies indicated TIMP1 as the oncogene could promote gastric cancer proliferation, invasion and migration[33, 34], we inferred circSCAF8 may facilitate gastric cancer growth and metastasis by upregulating TIMP1 expression via binding to miR-1293. Then we conducted a series of in vitro and in vivo experiments to examine the function of circSCAF8 in gastric tumorigenesis, and those results indicated circSCAF8 shRNA viruses could significantly inhibit gastric cancer growth and metastasis. As we known, miRNA could regulate multiple genes expression. It is still unknow whether circSCAF8 promoted gastric tumorigenesis by regulating TIMP1 expression via changing miR-1293 function. To prove it, rescue experiments were performed. Those results showed overexpressed TIMP1 could restore the effect of circSCAF8 shRNA viruses to gastric cancer. So, we confirmed circSCAF8 could promote gastric tumorigenesis by regulating TIMP1 expression.

In conclusion, circSCAF8 expression was increased in gastric cancer cells and tissues. Meanwhile, circSCAF8 could promote gastric tumorigenesis by upregulating TIMP1 expression via binding to miR-1293 (Fig. 6). Our study on the molecular mechanism of circSCAF8 may contribute to the development of new therapeutic strategies for gastric cancer.

Data Availability Statement

The data are available if reasonable request.

Acknowledgments

We acknowledge TCGA and other public databases for providing their plat‑forms and contributors for uploading their meaningful datasets.

Author Contributions

YP designed and supervised the study; BM and JC performed the experiments; BM and JC collected and analysed the data; BM and JC prepared the manuscript; YP revised the manuscript; All authors read and approved the final manuscript.

Compliance with ethical standards

Conflict of interest the authors declare that they have no conflict of interest.

Funding

This work was supported by the National Natural Science Foundation of China (81602134).

SEXTON R E, AL HALLAK M N, DIAB M, AZMI AS. Gastric cancer: a comprehensive review of current and future treatment strategies [J]. Cancer Metastasis Rev. 2020;39(4):1179–203.
WONG M C S, HUANG J, CHAN P S F, CHOI P, LAO X Q, CHAN S M, et al. Global Incidence and Mortality of Gastric Cancer, 1980–2018 [J]. JAMA Netw Open. 2021;4(7):e2118457.
XU Z-Y, RUAN S-M YUANL, QIN J-J MOS. CHENG X-D. Long non-coding RNAs towards precision medicine in gastric cancer: early diagnosis, treatment, and drug resistance [J]. Mol Cancer. 2020;19(1):96.
LI Y, FENG A, ZHENG S, CHEN C. Recent Estimates and Predictions of 5-Year Survival in Patients with Gastric Cancer: A Model-Based Period Analysis [J]. Cancer Control. 2022;29:10732748221099227.
THRUMURTHY S G, CHAUDRY M A, HOCHHAUSER D. The diagnosis and management of gastric cancer [J]. BMJ. 2013;347:f6367.
COLLATUZZO G, PELUCCHI C, LóPEZ-CARRILLO NEGRIE, TSUGANE L, HIDAKA S. Exploring the interactions between Helicobacter pylori (Hp) infection and other risk factors of gastric cancer: A pooled analysis in the Stomach cancer Pooling (StoP) Project [J]. Int J Cancer. 2021;149(6):1228–38.
LIU T, ZHU J, DU W, NING W, ZHANG Y, ZENG Y, et al. AKT2 drives cancer progression and is negatively modulated by miR-124 in human lung adenocarcinoma [J]. Respir Res. 2020;21(1):227.
MILIOTIS C, SLACK F J. miR-105-5p regulates PD-L1 expression and tumor immunogenicity in gastric cancer [J]. Cancer Lett. 2021;518:115–26.
ZHOU T, LIN K, NIE J, PAN B, HE B. LncRNA SPINT1-AS1 promotes breast cancer proliferation and metastasis by sponging let-7 a/b/i-5p [J]. Pathol Res Pract. 2021;217:153268.
DONG R, MA X K, CHEN L L YANGL. Increased complexity of circRNA expression during species evolution [J]. RNA Biol. 2017;14(8):1064–74.
JIANG H, TIAN Y, ZHAO X, WU ZHANGL. A circular RNA derived from FAT atypical cadherin 3 promotes lung cancer progression via forming a regulatory loop with oncogenic ELAV like RNA binding protein 1 [J]. J Biochem. 2022;171(5):519–28.
RYBAK-WOLF A, STOTTMEISTER C, GLAZAR P, JENS M, PINO N, GIUSTI S, et al. Circular RNAs in the Mammalian Brain Are Highly Abundant, Conserved, and Dynamically Expressed [J]. Mol Cell. 2015;58(5):870–85.
HOU L-D ZHANGJ. Circular RNAs: An emerging type of RNA in cancer [J]. Int J Immunopathol Pharmacol. 2017;30(1):1–6.
ZHU C L, SHA X, WANG Y, LI J, ZHANG M Y, GUO Z Y et al. Circular RNA hsa_circ_0007142 Is Upregulated and Targets miR-103a-2-5p in Colorectal Cancer [J]. J Oncol. 2019; 2019: 9836819.
WANG R, ZHANG S, CHEN X, LI N, LI J, JIA R, et al. CircNT5E Acts as a Sponge of miR-422a to Promote Glioblastoma Tumorigenesis [J]. Cancer Res. 2018;78(17):4812–25.
WONG C H, LOU U K, LI Y, CHAN S L, TONG J H M, TO K F et al. CircFOXK2 Promotes Growth and Metastasis of Pancreatic Ductal Adenocarcinoma by Complexing with RNA Binding Proteins and Sponging MiR-942 [J]. Cancer Res. 2020.
CHEN N, ZHAO G, YAN X, LV Z, YIN H. A novel FLI1 exonic circular RNA promotes metastasis in breast cancer by coordinately regulating TET1 and DNMT1 [J]. Genome Biol. 2018;19(1):218.
LOU J, HAO Y, LIN K, LYU Y, CHEN M, WANG H, et al. Circular RNA CDR1as disrupts the p53/MDM2 complex to inhibit Gliomagenesis [J]. Mol Cancer. 2020;19(1):138.
YANG H, LI X, MENG Q, SUN H, HU WUS. CircPTK2 (hsa_circ_0005273) as a novel therapeutic target for metastatic colorectal cancer [J]. Mol Cancer. 2020;19(1):13.
LIANG HF, ZHANG X Z, LIU B G, JIA G T, LI WL. Circular RNA circ-ABCB10 promotes breast cancer proliferation and progression through sponging miR-1271 [J]. Am J Cancer Res. 2017;7(7):1566–76.
LI J, CHEN K, DONG X, XU Y, SUN Q, WANG H, et al. YTHDF1 promotes mRNA degradation via YTHDF1-AGO2 interaction and phase separation [J]. Cell Prolif. 2022;55(1):e13157.
YURYEV A, PATTURAJAN M, LITINGTUNG Y, JOSHI R V, GENTILE C, GEBARA M, et al. The C-terminal domain of the largest subunit of RNA polymerase II interacts with a novel set of serine/arginine-rich proteins [J]. Proc Natl Acad Sci U S A. 1996;93(14):6975–80.
MCCRACKEN S, FONG N, ROSONINA E, YANKULOV K, BROTHERS G, SIDEROVSKI D, et al. 5'-Capping enzymes are targeted to pre-mRNA by binding to the phosphorylated carboxy-terminal domain of RNA polymerase II [J]. Genes Dev. 1997;11(24):3306–18.
HIROSE Y, OHKUMA Y. Phosphorylation of the C-terminal domain of RNA polymerase II plays central roles in the integrated events of eucaryotic gene expression [J]. J Biochem. 2007;141(5):601–8.
DESCOSTES N, HEIDEMANN M, SPINELLI L, SCHüLLER R, MAQBOOL M A, FENOUIL R, et al. Tyrosine phosphorylation of RNA polymerase II CTD is associated with antisense promoter transcription and active enhancers in mammalian cells [J]. Elife. 2014;3:e02105.
ETCHEGARAY J-P ZHONGL, LI C, HENRIQUES T, ABLONDI E, NAKADAI T et al. The Histone Deacetylase SIRT6 Restrains Transcription Elongation via Promoter-Proximal Pausing [J]. Mol Cell. 2019; 75(4).
BECKER R, LOLL B. Snapshots of the RNA processing factor SCAF8 bound to different phosphorylated forms of the carboxyl-terminal domain of RNA polymerase II [J]. J Biol Chem. 2008;283(33):22659–69.
MITTER GREGERSENLH, UGALDE A P R, NOJIMA T, PROUDFOOT N J, AGAMI R et al. SCAF4 and SCAF8, mRNA Anti-Terminator Proteins [J]. Cell 2019; 177(7): 1797 – 813 e18.
LIANG D, TATOMER D C, LUO Z, WU H, YANG L. CHEN L L, The Output of Protein-Coding Genes Shifts to Circular RNAs When the Pre-mRNA Processing Machinery Is Limiting [J]. Mol Cell. 2017; 68(5): 940 – 54 e3.
LI P, MA Y, WANG Y, CHEN T, WANG H, CHU H, et al. Identification of miR-1293 potential target gene: TIMP-1 [J]. Mol Cell Biochem. 2013;384(1–2):1–6.
TAKAGAWA Y, GEN Y, MURAMATSU T, TANIMOTO K, INOUE J, HARADA H et al. miR-1293, a Candidate for miRNA-Based Cancer Therapeutics, Simultaneously Targets BRD4 and the DNA Repair Pathway [J]. Mol Ther. 2020.
LIU X L, PAN W G, LI K L, MAO Y J, LIU S D, ZHANG R M. miR-1293 Suppresses Tumor Malignancy by Targeting Hydrocyanic Oxidase 2: Therapeutic Potential of a miR-1293/Hydrocyanic Oxidase 2 Axis in Renal Cell Carcinoma [J]. Cancer Biother Radiopharm. 2020.
OMAR O M SOUTTOM, BHAT N S, BHAT A A, LU H, CHEN Z, et al. TFF1 antagonizes TIMP-1 mediated proliferative functions in gastric cancer [J]. Mol Carcinog. 2018;57(11):1577–87.
LAITINEN A, HAGSTROM J, MUSTONEN H, KOKKOLA A, TERVAHARTIALA T, SORSA T, et al. Serum MMP-8 and TIMP-1 as prognostic biomarkers in gastric cancer [J]. Tumour Biol. 2018;40(9):1010428318799266.

Table 1 and 2 are available in the Supplementary Files section.

There is NO conflict of interest to disclose.

supplementtable1.xlsx
supplementtable2.xlsx
supplementaryFigure1.tif
Table1.xlsx
Table 1. Clinical characteristics of the patient cohort.
Table2.xlsx
Table 2. Different primers used in the experiments.

Download PDF

Editorial decision: revise
15 May, 2024
Review #2 received at journal
15 Apr, 2024
Review #1 received at journal
18 Mar, 2024
Reviewer #2 agreed at journal
17 Mar, 2024
Reviewer #1 agreed at journal
29 Feb, 2024
Reviewers invited by journal
07 Dec, 2023
Editor assigned by journal
07 Dec, 2023
Submission checks completed at journal
07 Dec, 2023
First submitted to journal
06 Dec, 2023

You are reading this latest preprint version

circRNA circSCAF8 promotes tumor growth and metastasis of gastric cancer by miR-1293/TIMP1signaling

Status:

Version 1

Abstract

Figures

Introduction

Materials and Methods

Human tissue specimens and cell lines

RNA preparation and real-time PCR

Nucleic acid electrophoresis

RNase R digestion

RNA isolation of nuclear and cytoplasmic fractions

Immunoprecipitation

circRNA pull-down assay

Luciferase reporter assay

circSCAF8 Knockdown by RNA-interference

Oligonucleotide Transfection and Expression Plasmid Construction

Antibodies and Immunoblotting

Clonability and Cell Viability Assays

Migration and Invasion Assays

In vivo tumour growth assays

Statistical Analysis

Results

circRNAs selection and identification

The expression and prognostic significance of circSCAF8 in gastric cancer

circSCAF8 promote gastric cancer proliferation, invasion, migration and metastasis

circSCAF8 directly bind to miR-1293

circSCAF8 regulate TIMP1 expression by miR-1293

Discussion

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1