Circ_0007031 Serves as a Sponge of miR-760 to Regulate the Growth and Chemoradiotherapy Resistance of Colorectal Cancer via Regulating DCP1A

Introduction

Colorectal cancer (CRC) is one of the most common cancers, and the number of CRC patients is increasing year by year.^1,2 Because the early symptoms are not obvious, most patients with CRC are already in the advanced stage when diagnosed, so this greatly increases the difficulty of treatment.³ At present, the combination of radiotherapy and chemotherapy is considered the common treatment for locally advanced CRC.⁴ 5-Fluorouracil (5-Fu) is a commonly used chemotherapeutic drug for CRC, but the development of chemoradiotherapy resistance (CRR) is a huge obstacle for the treatment of CRC.^5,6 Therefore, it is urgent to clarify the factors affecting the CRR of CRC.

Non-coding RNAs acting as non-protein-coding RNAs have received a lot of attention, while circular RNAs (circRNAs) have been favored by researchers in recent years due to their closed-loop structure.^7,8 CircRNAs have been shown to take part in the modulation of cancer progression and may also be involved in the CRR of cancer.^9–11 For example, circRNA Cdr1as inhibits the cisplatin resistance of ovarian cancer,¹² and circKDM4C participates in the regulation of the tumor progression and doxorubicin resistance of breast cancer.¹³ In CRC, Abu et al found that there were 773 up-regulated and 732 down-regulated circRNAs in the chemoresistant and chemosensitive HCT116 cells, indicating that the expression levels of circRNAs were crucial to the development of the chemical resistance of CRC.¹⁴ The microarray analysis by Xiong et al showed that 71 circRNAs were differentially expressed in the CRR of CRC cells, among which circ_0007031 (TUBGCP3 was its linear mRNA) was significantly highly expressed.¹⁵ However, the role of circ_0007031 in the growth and CRR of CRC has not been investigated.

The function of circRNAs as competitive endogenous RNAs (ceRNAs) for microRNAs (miRNAs) is a classic molecular mechanism of circRNAs and has been widely confirmed.^16,17 MiR-760 is lower expressed in many cancers and has been shown in many studies to suppress the CRR of many cancers, including pancreatic cancer, hepatocellular carcinoma and breast cancer.^18–20 It has been reported that mRNA-decapping enzyme 1a (DCP1A) is a decapping enzyme associated with mRNA degradation and translation inhibition, and may be involved in cell differentiation.^21,22 Studies have shown that DCP1A expression is up-regulated in CRC, and its elevated expression is associated with the reduced survival rate in CRC patients.^23,24 Therefore, it may be a key target to regulate CRC progress.

Our study was aimed to investigate the role of circ_0007031 in the growth and CRR of CRC and to elucidate its molecular mechanisms. Combined with the results of the background investigation, we focused on the interaction of circ_0007031, miR-760 and DCP1A, which might provide molecular therapeutic strategies for the treatment of CRC with CRR.

Materials and Methods

Cell Culture

CRC cell lines (HCT116 and Caco2) were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA) and cultured in RPMI-1640 medium (Gibco, Waltham, MA, USA). Human normal colon epithelial cells (NCM460) were purchased from Jining Shiye (Shanghai, China) and cultured in Dulbecco’s modified Eagle’s medium (DMEM; Gibco). All mediums were contained 10% fetal bovine serum (FBS; Gibco) and 1% penicillin/streptomycin (Invitrogen, Carlsbad, CA, USA), and all cells were incubated at 37°C with 5% CO₂ incubator.

Establishment of CRR Cell Model

HCT116 and Caco2 cells were seeded into 6-well plates and cultured until the cells reached 80–90% confluences. Then, the cells were treated with 10 μmol/L 5-Fu (Selleck, Shanghai, China) and simultaneously exposed to a dose of 4 Gy 6 Mv X-ray. After 24 h, the cell medium was replaced with 5-Fu-free medium, and the cells were placed in a normal environment. After 2 days, the surviving cells were transferred to fresh medium and further treated with 5-Fu and X-ray according to the above method. The whole process was repeated 8 times. After that, the last collected cells were HCT116/CRR and Caco2/CRR cell models.

Cell Counting Kit 8 (CCK8) Assay

CCK8 assay was used to evaluate the 5-Fu resistance and proliferation of cells. All cells were seeded in 96-well plates and cultivation for 24 h. For 5-Fu resistance assay, cells were treated with different concentrations of 5-Fu (0, 0.5, 1.0, 1.5, 2.0 and 2.5 μg/mL) for 48 h, followed by incubation with CCK8 solution (Genomeditech, Shanghai, China) for 4 h. The absorbance was detected at 450 nm, and half-maximal inhibitory concentration (IC50) was calculated to evaluate the 5-Fu resistance of cells. For proliferation assay, CCK8 solution was added into each well at a specific time point (0, 24, 48 and 72 h) for 4 h. The absorbance was determined at 450 nm to assess the proliferation ability of cells.

Clonogenic Assay

All cells were seeded in 6-well plates. After 24 h, cells were exposed to graded doses (0, 2, 4, 6 and 8 Gy) of X-ray. After 2 weeks, cells were fixed with methanol, stained with crystal violet, and the colony number (> 50 cells) was counted under a microscope (Novel, Ningbo, China). The survival fraction of cells was calculated as below: survival fraction = the colony number of the treatment group/control group.

Samples Collection

A cohort of 50 CRC patients who underwent surgical resection at The First Hospital of Hebei Medical University, Shijiazhuang were recruited. CRC tumor tissues and adjacent normal tissues were obtained and stored at −80°C. All patients signed informed consent, and our study was approved by the Ethics Committee of The First Hospital of Hebei Medical University, Shijiazhuang.

Quantitative Real-Time Polymerase Chain Reaction (qRT-PCR)

According to the instruction provided by the manufacturer, total RNA was extracted using the RNeasy Mini Kit (Qiagen, Valencia, CA, USA), and cDNA was synthesized using cDNA Synthesis SuperMix (Transgen, Beijing, China). Quantitative analysis was carried out using SYBR Green (Solarbio, Beijing, China) on Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). The thermocycling conditions were as below: 95°C for 10 min, 40 cycles of 95°C for 30 s, 55°C for 30 s, and 72°C for 42 s. The data were analyzed by the 2^−ΔΔCt method and normalized using 18S ribosomal RNA (rRNA) or U6. All primers were presented as follows: circ_0007031, F 5ʹ-ATCATTGCTGCACACGAGGT-3ʹ, R 5ʹ-AGGACCTTCCTAGACTGATCCA-3ʹ; TUBGCP3, F 5ʹ-TATGTGGAGCAGATCGAGAAG-3ʹ, R 5ʹ-TTGATGGACCTGAGGTGAG-3ʹ; 18S rRNA, F 5ʹ-ATCGGGGATTGCAATTATTC-3ʹ, R 5ʹ-CTCACTAAACCATCCAATCG-3ʹ; miR-760, F 5ʹ- TCAATCCACCAGAGCATGGATAT-3ʹ, R 5ʹ-CTCTACAGCTATATTGCCAGCCA-3ʹ; U6, F 5ʹ-CTCGCTTCGGCAGCACATATACT-3ʹ, R 5ʹ-CGCTTCACGAATTTGCGTGT-3ʹ.

Subcellular Fractionation and Localization

The cytoplasmic and nuclear RNA of HCT116/CRR and Caco2/CRR cells were isolated and extracted using the Cytoplasmic & Nuclear RNA Purification Kit (Norgen Biotek, Thorold, Ontario, Canada). QRT-PCR was employed to determine the expression of circ_0007031, U6 and 18S rRNA in cytoplasmic and nuclear of cells. U6 and 18S rRNA were used as nuclear control and cytoplasm control, respectively.

RNase R Treatment

Part of extracted RNAs from HCT116/CRR and Caco2/CRR cells were treated with Ribonuclease R (RNase R; Geneseed, Guangzhou, China) for 20 min and then continued for qRT-PCR analysis to detect the circ_0007031 and TUBGCP3 expression. Results without RNase R treatment (RNase R-) were used as negative controls, and TUBGCP3 was used as a representative linear control.

Cell Transfection

Transfection could be performed when the cell density rate reached 50%-60%. Circ_0007031 small interfering RNA and lentiviral short hairpin RNA (si-circ_0007031 and sh-circ_0007031) or their negative controls (si-NC and sh-NC), miR-760 mimic and inhibitor (miR-760 and anti-miR-760) or their negative controls (miR-NC and anti-miR-NC), DCP1A overexpression plasmid (DCP1A) and its negative control (vector) were purchased from RiboBio (Guangzhou, China). Cell transfection was performed using Lipofectamine 3000 (Invitrogen).

Flow Cytometry

Flow cytometry was performed to measure the cell cycle distribution and apoptosis of cells. HCT116/CRR and Caco2/CRR cells were harvested and collected into a centrifuge tube after transfection for 48 h. For cell cycle distribution assay, cells were fixed with 70% ethanol overnight at 4°C. Then, cells were incubated with RNase A (Beyotime, Shanghai, China) for 1 h at 37°C and stained with propidium iodide (PI, Beyotime) for 30 min. The cell cycle was analyzed using FACScan Flow Cytometer (BD Biosciences, San Jose, CA, USA). For apoptosis assay, cells were stained with Annexin V-fluorescein isothiocyanate (FITC) Apoptosis Detection Kit (Beyotime). The apoptosis of cells was also detected using FACScan Flow Cytometer.

Western Blot (WB) Analysis

Tissues and cells were lysed with RIPA lysis buffer (Beyotime) to extract total protein. After quantification with the BCA Protein Assay Kit (Beyotime), the same amount of protein was separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gel, transferred onto polyvinylidene fluoride (PVDF) membranes (Millipore, Billerica, MA, USA) and blocked with 5% nonfat milk. The membranes were incubated with primary antibodies against B-cell lymphoma-2 (Bcl-2; 1:1000, Abcam, Cambridge, MA, USA), Bcl2-associated X (Bax; 1:5000, Abcam), cleaved-caspase-3 (cleaved-cas-3, 1:1000, Abcam), DCP1A (1:1000, Abcam) or GAPDH (1:5000, Abcam) at 4°C overnight. After that, the membranes were interacted with secondary antibody (1:2000, Abcam) for 1 h and visualized using BeyoECL star (Beyotime).

Dual-Luciferase Reporter Assay

The fragments of circ_0007031-wild type (WT)/mutant type (MUT) or DCP1A 3ʹ-UTR-WT/MUT were inserted into the pGL3 reporter vector (Promega, Madison, WI, USA). HCT116/CRR and Caco2/CRR cells were co-transfected the reporter vectors and miR-760 mimic or miR-NC. After 48 h, the relative luciferase activity was determined using the Dual-Luciferase Reporter Assay Kit (Transgen).

RNA Immunoprecipitation (RIP) Assay

RIP Assay Kit (Millipore) was used to perform this assay. In brief, magnetic beads were incubated with antibodies against argonaute2 (Anti-Ago2) or immunoglobulin G (Anti-IgG) at 4°C overnight. HCT116/CRR and Caco2/CRR cells were lysed using RIP buffer. Then, the cell lysates were added into the magnetic bead-antibody complex at 4°C overnight. After that, the proteinase K buffer was used to purify RNA. The enrichment of circ_0007031 and miR-760 was determined using qRT-PCR.

Mice Xenograft Models

Animal experiments were authorized by the Animal Care Committee of The First Hospital of Hebei Medical University, Shijiazhuang and performed according to the Guide for the Care and Use of Laboratory Animals. Animal studies were performed in compliance with the ARRIVE guidelines and the Basel Declaration. All animals received humane care according to the National Institutes of Health (USA) guidelines. HCT116 cells transfected with sh-circ_0007031 or sh-NC were subcutaneously injected into the flank regions of nude mice (Vital River, Beijing, China). Then, the sh-circ_0007031 or sh-NC group mice were equally divided into 2 groups, one treated with 5-Fu and the other treated with X-ray radiation. For 5-Fu treatment, after 10 days of inoculation, mice in both the sh-NC and sh-circ_0007031 groups were injected with 5-Fu at 50 mg/kg/week, and tumor volume was monitored every 7 days and calculated as follows: volume = 0.5 × length × width². After 35 days, the mice were sacrificed, and the tumor weight was measured. For X-ray radiation, when the tumor volume was about 300 mm³, the mice had received 8 Gy X-ray radiation. Tumor volume was measured every 7 days until 35 days. The tumor was removed for further testing.

Statistical Analysis

Data were expressed as mean ± standard deviation and analyzed for the statistical difference using Student’s t-test or one-way analysis of variance in GraphPad Prism 5.0 software (GraphPad, La Jolla, CA, USA). P < 0.05 was considered statistically significant. All experiment was performed in triplicate, and all independent experiments were set for 3 times to take the average value.

Results

The Sensitivity of HCT116/CRR and Caco2/CRR Cells to Chemoradiotherapy Was Verified

First, we examined the 5-Fu resistance and radiation resistance of HCT116/CRR and Caco2/CRR cell lines using CCK8 and clonogenic assays. As presented in Figure 1A and B, with the increase of 5-Fu concentration, the viabilities of HCT116/CRR and Caco2/CRR cells were significantly higher than that of HCT116 and Caco2 cells, and their IC50 value was also markedly enhanced compared with HCT116 and Caco2 cells, suggesting that the HCT116/CRR and Caco2/CRR cells had a strong resistance to 5-Fu. Besides, clonogenic assay results showed that with the increase of X-ray radiation, the survival fraction of HCT116/CRR and Caco2/CRR cells was increased compared to HCT116 and Caco2 cells (Figure 1C and D), and the statistical results on the colony number of cells also suggested that the colony number of HCT116/CRR and Caco2/CRR cells was remarkably higher than that of HCT116 and Caco2 cells (Figure 1E and F). These results indicated that HCT116/CRR and Caco2/CRR cells had good resistance to radiation. Therefore, our results confirmed the successful construction of 5-Fu-resistant and radiation-resistant HCT116/CRR and Caco2/CRR cells.

Figure 1 The verification of HCT116/CRR and Caco2/CRR cell resistance to 5-Fu and radiation. (A and B) The viabilities and IC50 values of HCT116, HCT116/CRR, Caco2 and Caco2/CRR cells were detected by CCK8 assay to reflect the 5-Fu resistance of cells. (C and D) The survival fraction of HCT116, HCT116/CRR, Caco2 and Caco2/CRR cells was measured by clonogenic assay to evaluate the radiation resistance of cells. (E and F) Representative pictures of colony formation were shown, and the colony number of HCT116, HCT116/CRR, Caco2 and Caco2/CRR cells were counted under exposed to graded doses of X-ray radiation. *P < 0.05.

The Upregulation of Circ_0007031 in CRC Tissues and Cells

For exploring the role of circ_0007031 in CRC, we first measured its expression in CRC tissues and cells. The results revealed that circ_0007031 had elevated expression in CRC tumor tissues compared with adjacent normal tissues (Figure 2A). The correlation analysis between circ_0007031 expression and the clinicopathologic characteristics of CRC patients showed that the high expression of circ_0007031 was markedly related to the tumor size, TNM stage and CEA of CRC patients (Table 1). Similarly, we also found that compared with NCM460, circ_0007031 was up-regulated in CRC cells, and was significantly higher expressed in HCT116/CRR and Caco2/CRR cells than in HCT116 and Caco2 cells (Figure 2B). By detecting the expression of circ_0007031 in the nucleus and cytoplasm of HCT116/CRR and Caco2/CRR cells, we found that circ_0007031 mainly accumulated in the cytoplasm (Figure 2C and D). Besides, circ_0007031 also showed strong resistance to RNase R digestion in HCT116/CRR and Caco2/CRR cells, indicating that circ_0007031 was a circular transcript (Figure 2E and F).

Table 1 Correlation Between circ_0007031 Expression and Clinicopathologic Characteristics of CRC Patients

Figure 2 The expression of circ_0007031 in CRC tissues and cells. (A) QRT-PCR was used to measure the expression of circ_0007031 in CRC tumor tissues (Tumor) and adjacent normal tissues (Normal). (B) The circ_0007031 expression was detected by qRT-PCR in NCM460 cells and CRC cells (HCT116, HCT116/CRR, Caco2 and Caco2/CRR). (C and D) The expression levels of circ_0007031, U6 and 18S rRNA in the nuclear and cytoplasm of HCT116/CRR and Caco2/CRR cells were assessed by qRT-PCR. (E and F) QRT-PCR was employed to test the expression of circ_0007031 and linear mRNA TUBGCP3 in HCT116/CRR and Caco2/CRR cells. *P < 0.05.

Silencing of circ_0007031 Repressed the Growth and CRR of CRC Resistant Cells

To investigate the function of circ_0007031 in CRC growth and CRR, we silenced its expression using si-circ_0007031. The detection of its expression by qRT-PCR indicated that si-circ_0007031 had a good inhibitory effect on circ_0007031 expression (Figure 3A and B). Subsequently, CCK8 results suggested that the silencing of circ_0007031 could inhibit the proliferation of HCT116/CRR and Caco2/CRR cells (Figure 3C and D). Besides, knockdown of circ_0007031 in HCT116/CRR and Caco2/CRR cells significantly caused the cell cycle arrest in the G0/G1 phase, with an obvious reduction in the number of HCT116/CRR and Caco2/CRR cells at the S phase (Figure 3E and F). Also, the apoptosis of HCT116/CRR and Caco2/CRR cells was promoted by circ_0007031 knockdown (Figure 3G and H). Consistent with this result, WB analysis revealed that the silencing of circ_0007031 suppressed the Bcl-2 protein level, while increased the Bax and Cleaved-cas-3 protein levels in HCT116/CRR and Caco2/CRR cells (Figure 3I and J). Further, silenced circ_0007031 also decreased the resistance of HCT116/CRR and Caco2/CRR cells to 5-Fu (Figure 3K and L). Moreover, the survival fraction of HCT116/CRR and Caco2/CRR cells was also reduced by circ_0007031 knockdown (Figure 3M and N), which could also be confirmed by the decrease in the colony number of HCT116/CRR and Caco2/CRR cells (Figure 3O and P). All data suggested that circ_0007031 might play pro-cancer and pro-resistance roles in CRC.

Figure 3 Effects of si-circ_0007031 on the growth and CRR of CRC resistant cells. HCT116/CRR and Caco2/CRR cells were transfected with si-circ_0007031 or si-NC. (A and B) The circ_0007031 expression in HCT116/CRR and Caco2/CRR cells was measured by qRT-PCR. (C and D) CCK8 assay was performed to test the proliferation of HCT116/CRR and Caco2/CRR cells. (E–H) The cell cycle distribution and apoptosis of HCT116/CRR and Caco2/CRR cells were assessed by flow cytometry. (I and J) The protein levels of Bcl-2, Bax and cleaved-cas-3 in HCT116/CRR and Caco2/CRR cells were determined by WB analysis. (K and L) The 5-Fu resistance of HCT116/CRR and Caco2/CRR cells was detected via measuring the viabilities and IC50 values of cells using CCK8 assay. (M–P) Clonogenic assay was used to measure the radiation resistance of HCT116/CRR and Caco2/CRR cells by detecting the survival fraction and colony number of cells. *P < 0.05.

Circ_0007031 Served as a Sponge of miR-760 in CRC Cells

To explore the mechanism by which circ_0007031 inhibited the growth and CRR of CRC resistant cells, we used the starBase v.2.0 tool (http://starbase.sysu.edu.cn/index.php) for bioinformatics analysis and found that miR-760 might directly bind to circ_0007031 (Figure 4A). Dual-luciferase reporter assay determined that miR-760 could restrain the luciferase activity of circ_0007031-WT reporter vector rather than circ_0007031-MUT reporter vector (Figure 4B and C). And RIP assay also revealed that circ_0007031 and miR-760 were enriched in Anti-Ago2 (Figure 4D). Besides, qRT-PCR showed that miR-760 was up-regulated by circ_0007031 knockdown in HCT116/CRR and Caco2/CRR cells (Figure 4E and F). Through detecting the miR-760 expression in CRC, we also found that miR-760 was under-expressed in CRC tumor tissues and cells compared with that in adjacent normal tissues or NCM460 cells, respectively (Figure 4G and H). Moreover, miR-760 was also lower expressed in HCT116/CRR and Caco2/CRR cells than in HCT116 and Caco2 cells (Figure 4H). These results indicated that miR-760 could be sponged by circ_0007031 in CRC.

Figure 4 MiR-760 could be directly targeted by circ_0007031. (A) The sequences of circ_0007031-WT and circ_0007031-MUT were shown. (B and C) Dual-luciferase reporter assay was used to verify the targeted binding relationship between miR-760 and circ_0007031 in HCT116/CRR and Caco2/CRR cells. (D) The enrichments of miR-760 and circ_0007031 in Anti-IgG or Anti-Ago2 were measured by RIP assay. (E and F) The expression of miR-760 was determined by qRT-PCR in HCT116/CRR and Caco2/CRR cells transfected with si-circ_0007031 or si-NC. (G) The miR-760 expression in CRC tumor tissues (Tumor) and adjacent normal tissues (Normal) was assessed by qRT-PCR. (H) QRT-PCR was performed to detect the expression of miR-760 in NCM460 cells and CRC cells (HCT116, HCT116/CRR, Caco2 and Caco2/CRR). *P < 0.05.

The Inhibition of circ_0007031 Silencing on the Growth and CRR of CRC Resistant Cells Could Be Partially Reversed by miR-760 Inhibitor

To explore whether the role of circ_0007031 in CRC was mediated by miR-760, we co-transfected si-circ_0007031 and anti-miR-760 into HCT116/CRR and Caco2/CRR cells. The detection results of miR-760 expression showed that si-circ_0007031 had a good promoting effect on the expression of miR-760, while this effect could be reversed by anti-miR-760 (Figure 5A and B). CCK8 assay revealed that miR-760 inhibitor could partially invert the suppression of circ_0007031 knockdown on the proliferation of HCT116/CRR and Caco2/CRR cells (Figure 5C and D). Also, the arresting effect of circ_0007031 knockdown on the G0/G1 phase of HCT116/CRR and Caco2/CRR cells could also be partially recovered by miR-760 inhibitor (Figure 5E and F). Besides, the promotion effect of circ_0007031 silencing on the apoptosis of HCT116/CRR and Caco2/CRR cells also could be reversed by miR-760 inhibitor (Figure 5G and H), which was also reflected by anti-miR-760 changing the protein levels of Bcl-2, Bax and cleaved-cas-3 of HCT116/CRR and Caco2/CRR cells (Figure 5I and J). Meanwhile, we measured the resistance of HCT116/CRR and Caco2/CRR cells to 5-Fu and radiation. The results showed that miR-760 inhibitor partially inverted the reduction effect of silenced circ_0007031 on the 5-Fu resistance and radiation resistance of HCT116/CRR and Caco2/CRR cells (Figure 5K–N). Furthermore, the inhibition effect of circ_0007031 silencing on the colony number of HCT116/CRR and Caco2/CRR cells also could be reversed by miR-760 inhibitor (Figure 5O and P). Therefore, our results confirmed that the effect of circ_0007031 on the growth and CRR of CRC cells was mediated by miR-760.

Figure 5 Circ_0007031 regulated the growth and CRR of CRC resistant cells through miR-760. HCT116/CRR and Caco2/CRR cells were transfected with si-NC, si-circ_0007031, si-circ_0007031 + anti-miR-NC or si-circ_0007031 + anti-miR-760, respectively. (A and B) The miR-760 expression in HCT116/CRR and Caco2/CRR cells was determined by qRT-PCR. (C and D) The proliferation of HCT116/CRR and Caco2/CRR cells was measured by CCK8 assay. (E–H) Flow cytometry was employed to test the cell cycle distribution and apoptosis of HCT116/CRR and Caco2/CRR cells. (I and J) WB analysis was used to detect the protein levels of Bcl-2, Bax and cleaved-cas-3 in HCT116/CRR and Caco2/CRR cells. (K and L) The viabilities and IC50 values of HCT116/CRR and Caco2/CRR cells were determined using CCK8 assay to evaluate the 5-Fu resistance of cells. (M–P) The survival fraction and colony number of HCT116/CRR and Caco2/CRR cells were detected using the clonogenic assay to assess the radiation resistance of cells. *P < 0.05.

DCP1A Was a Target of miR-760 in CRC Cells

At the same time, we also used the starBase v.2.0 tool (http://starbase.sysu.edu.cn/index.php) to predict the targets of miR-760. As shown in Figure 6A, DCP1A was predicted to have binding sites for miR-760. Then, miR-760 overexpression could reduce the luciferase activity of DCP1A 3ʹ-UTR-WT reporter vector without affecting that of the DCP1A 3ʹ-UTR-MUT reporter vector, confirming the interaction between them (Figure 6B and C). Further, we also found that the protein level of DCP1A could be suppressed by miR-760 overexpression in HCT116/CRR and Caco2/CRR cells (Figure 6D and E). Moreover, DCP1A expression was up-regulated in CRC tumor tissues compared with that in adjacent normal tissues (Figure 6F). Similarly, its expression was also increased in CRC cells compared to NCM460 cells, and markedly higher in HCT116/CRR and Caco2/CRR cells than that in HCT116 and Caco2 cells (Figure 6G). In addition, we also discovered that DCP1A expression was decreased by circ_0007031 knockdown, while miR-760 inhibitor could reverse this effect (Figure 6H and I). Our data confirmed that DCP1A could be targeted by miR-760.

Figure 6 DCP1A was a target of miR-760. (A) The sequences of DCP1A 3ʹ-UTR-WT and DCP1A 3ʹ-UTR-MUT were presented. (B and C) The interaction between miR-760 and DCP1A was verified using the dual-luciferase reporter assay. (D and E) The DCP1A protein level in HCT116/CRR and Caco2/CRR cells transfected with miR-760 mimic or miR-NC was measured by WB analysis. (F) The protein level of DCP1A in CRC tumor tissues (Tumor) and adjacent normal tissues (Normal) was determined by WB analysis. (G) WB analysis was employed to assess the protein level of DCP1A in NCM460 cells and CRC cells (HCT116, HCT116/CRR, Caco2 and Caco2/CRR). (H and I) HCT116/CRR and Caco2/CRR cells were transfected with si-NC, si-circ_0007031, si-circ_0007031 + anti-miR-NC or si-circ_0007031 + anti-miR-760, respectively. The protein level of DCP1A in HCT116/CRR and Caco2/CRR cells was detected by WB analysis. *P < 0.05.

MiR-760 Regulated the Growth and CRR of CRC Resistant Cells Through Targeting DCP1A

To verify the conclusion of the above, we co-transfected miR-760 mimic and DCP1A overexpression plasmid into HCT116/CRR and Caco2/CRR cells. The inhibition effect of miR-760 mimic on DCP1A protein level and the recovery effect of DCP1A plasmid on DCP1A protein level confirmed the success of both transfection (Figure 7A and B). So, we detected the growth of CRC cells. CCK8 results showed that miR-760 overexpression repressed the proliferation of HCT116/CRR and Caco2/CRR cells, while DCP1A overexpression could invert this effect (Figure 7C and D). Also, miR-760 overexpression could cause the cell cycle arrest in the G0/G1 phase, reducing the proportion of S phase in HCT116/CRR and Caco2/CRR cells, whereas DCP1A overexpression could recover this effect (Figure 7E and F). Besides, the promotion of miR-760 overexpression on the apoptosis of HCT116/CRR and Caco2/CRR cells could be reversed by DCP1A overexpression (Figure 7G and H), which could also be confirmed by reversing the protein levels of Bcl-2, Bax and cleaved-cas-3 in HCT116/CRR and Caco2/CRR cells (Figure 7I and J). On the other hand, DCP1A overexpression inverted the inhibition of miR-760 overexpression on the 5-Fu resistance and radiation resistance of HCT116/CRR and Caco2/CRR cells (Figure 7K–N). In addition, the recovery effect of DCP1A overexpression on the colony number of HCT116/CRR and Caco2/CRR cells also indicated DCP1A overexpression could reverse the inhibitory effect of miR-760 overexpression on the resistance of HCT116/CRR and Caco2/CRR cells to radiation (Figure 7O and P). Hence, we confirmed that the regulation of miR-760 on the growth and CRR of CRC cells was achieved by targeting DCP1A.

Figure 7 MiR-760 regulated the growth and CRR of CRC resistant cells through DCP1A. HCT116/CRR and Caco2/CRR cells were transfected with miR-NC, miR-760, miR-760 + vector or miR-760 + DCP1A, respectively. (A and B) The protein level of DCP1A in HCT116/CRR and Caco2/CRR cells was measured by WB analysis. (C and D) The proliferation of HCT116/CRR and Caco2/CRR cells was determined by CCK8 assay. (E–H) The cell cycle distribution and apoptosis of HCT116/CRR and Caco2/CRR cells were tested using flow cytometry. (I and J) WB analysis was employed to assess the protein levels of Bcl-2, Bax and cleaved-cas-3 in HCT116/CRR and Caco2/CRR cells. (K and L) The viabilities and IC50 values of cells were detected by CCK8 assay to assess the 5-Fu resistance of HCT116/CRR and Caco2/CRR cells. (M–P) The survival fraction and colony number of cells were tested by clonogenic assay to evaluate the radiation resistance of HCT116/CRR and Caco2/CRR cells. *P < 0.05.

Knockdown of circ_0007031 Enhanced the Sensitivity of CRC Tumor to 5-Fu and Radiation in vivo

To further confirm the function of circ_0007031 in CRC, we performed the in vivo experiments. The results showed that under the 5-Fu or radiation condition, the tumor volume of mice in the circ_0007031 knockdown group was markedly smaller than that in the control group (Figure 8A and E). Besides, the tumor weight was also remarkably decreased in the circ_0007031 knockdown group under either 5-Fu treatment or radiation treatment compare with that in the sh-NC group (Figure 8B and F). Through the detection of the circ_0007031 and miR-760 expression, we confirmed that in the sh-circ_0007031 group, circ_0007031 expression was indeed inhibited and miR-760 expression was significantly higher than that in the sh-NC group (Figure 8C and G). At the same time, compared to the sh-NC group, the protein level of DCP1A was also hindered in the sh-circ_0007031 group (Figure 8D and H). Therefore, we confirmed that interference of circ_0007031 could restrain the resistance of CRC tumor to 5-Fu and radiation through regulating the miR-760/DCP1A axis, thereby aggravating the inhibitory effect of 5-Fu and radiation on CRC tumor growth.

Figure 8 Effects of silenced circ_0007031 on the tumor growth of CRC in vivo. HCT116/CRR cells transfected with sh-NC or sh-circ_0007031 were injected into nude mice and then treated with 5-Fu or radiation for 35 days. (A and E) Tumor volume was calculated at the indicated time point. (B and F) Tumor weight was detected in mice. (C and G) The expression levels of circ_0007031 and miR-760 were detected by qRT-PCR. (D and H) The protein level of DCP1A was evaluated by WB analysis. *P < 0.05.

Discussion

CRC is a malignant tumor, because it spreads to distant organs.²⁵ Preventing the occurrence of CRR is an important step to improve the CRC treatment response and has important clinical significance. To our knowledge, although many circRNAs have been shown to be differentially expressed before and after CRR, there is no specific evidence for their function in the CRR of CRC. In our system, we discovered that circ_0007031 was remarkably highly expressed in CRC tumor tissues and cells, indicating that circ_0007031 might play a vital function in CRC. Then, we constructed the CRR of CRC cell lines and focused on the role of circ_0007031 in the CRR of CRC. The high expression of circ_0007031 in CRC resistant cells was consistent with previous studies.¹⁵ On the other hand, we also confirmed that circ_0007031 was mainly involved in post-transcriptional regulation and had cyclic properties.

Studies have confirmed that the development of CRR is a complex process that may involve DNA damage related to mitochondria and pathways related to mitochondrial metabolism, thus regulating cell survival.^26,27 Therefore, the study of cell proliferation and apoptosis can well explain the sensitivity of cells to radiation and drugs. Here, we found that silenced circ_0007031 repressed the proliferation, caused cell cycle arrest in the G0/G1 phase, and accelerated the apoptosis of CRC resistant cells. Also, circ_0007031 knockdown reduced the resistance of CRC resistant cells to 5-Fu and radiation. Further, sh-circ_0007031 also enhanced the sensitivity of CRC tumors to 5-Fu and radiation, thereby notably inhibiting CRC tumor growth in vivo. This suggested that circ_0007031 knockdown might be an important step to improve the chemoradiotherapy sensitivity of CRC.

Previous studies have shown that miR-760 is associated with CRC tumor generation, and its high expression can inhibit the proliferation of CRC.²⁸ Also, miR-760 expression was related to the poor prognosis and malignant clinicopathologic features of CRC patients.²⁹ Besides, Xian et al also reported that miR-760 was involved in the modulation of long-noncoding RNA KCNQ1OT1 on the methotrexate resistance of CRC,³⁰ suggesting that miR-760 might be related to CRC resistance. In our study, we confirmed that miR-760 could be sponged by circ_0007031, its low expression in CRC and the restoring effect of its inhibitor on the silenced circ_0007031 function again confirmed that it could be targeted by circ_0007031. Moreover, we also showed that miR-760 could target DCP1A. Current studies had shown that high expression of DCP1A promoted the development of many cancers, such as gastric cancer and melanoma.^31,32 Herein, the reversing effect of DCP1A on the miR-760 function confirmed that DCP1A was also involved in the regulation of circ_0007031 on the growth and CRR of CRC. Therefore, the targeting of DCP1A by miR-760 also helped us better understand the role of circ_0007031 in the growth and CRR of CRC.

Currently, molecular targeted therapy is considered as an effective treatment strategy for cancer.³³ The elucidation of the mechanism of CRR also points the way to reduce the occurrence of CRR in cancer treatment. Our study suggests a novel mechanism by which circ_0007031 regulates CRC resistance through the miR-760/DCP1A pathway in vitro and in vivo. However, miR-760 is not the only miRNA that can be targeted by circ_0007031, and there may be other circ_0007031 targets in CRC that are still worthy of further investigation.

In conclusion, we demonstrated that circ_0007031 could increase DCP1A to promote the growth and CRR of CRC via regulating miR-760. The findings of this study might provide a reference for the study of circ_0007031 in other diseases and a solution for the occurrence of CRR in CRC.