Circular RNA hsa_circ_001895 serves as a sponge of microRNA-296-5p to promote clear cell renal cell carcinoma progression by regulating SOX12

Abstract

There is an urgent need to find novel potential therapeutic targets for the diagnosis and treatment of clear cell renal cell carcinoma (ccRCC) due to its highly invasive ability as a common urological malignant tumor. Circular RNAs (circRNAs) have been indicated as potentially critical mediators in various types of tumor progression. We first used qRT-PCR analysis to find dysregulated circRNAs in ccRCC. A novel circRNA, hsa_circ_001895, was upregulated in ccRCC specimens and associated with metastatic properties of ccRCC. However, the tumorigenic mechanism of hsa_circ_001895 on ccRCC is yet to be found. We first indicated that hsa_circ_001895 predicted a poor prognosis in ccRCC patients. Additionally, overexpression of hsa_circ_001895 not only promoted cell proliferation, invasion and migration of ccRCC, but also inhibited cell apoptosis, whereas hsa_circ_001895 knockdown reversed the effect on ccRCC progression. In vivo s.c. xenotransplanted tumor model also showed that silencing hsa_circ_001895 could suppress in vivo ccRCC growth. Mechanistically, hsa_circ_001895 directly binds with microRNA (miR)-296-5p and inhibits its expression. Moreover, sex determining region Y (SRY)-box 12 (SOX12) was identified as a target of miR-296-5p, the expression of which was suppressed by miR-296-5p. Notably, the inhibitory effect of hsa_circ_001895 on ccRCC progression was reversed by miR-296-5p inhibitor. In general, our findings indicated that hsa_circ_001895 may sponge miR-296-5p and promote SOX12 expression, which is the underlying mechanism of hsa_circ_001895-induced ccRCC progression.

Keywords: SOX12; clear cell renal cell carcinoma; hsa_circ_001895; miR-296-5p; progression.

Figure 1

Circular RNA (circRNA) hsa_circ_001895 was upregulated in both clear cell renal cell carcinoma (ccRCC) tissues and cell lines. A, Heat map shows dysregulated circRNAs between ccRCC tissues (T) and adjacent noncancer tissues (N). B, Expression of circRNA hsa_circ_001895 in ccRCC tissues and adjacent noncancer tissues detected by qRT‐PCR (N = 60). C, In situ hybridization analysis of hsa_circ_001895 expression in ccRCC tissues with different TNM stage. D, Overall survival analysis of ccRCC patients with high hsa_circ_001895 expression and low levels of hsa_circ_001895. E, Expression of hsa_circ_001895 in human ccRCC cell lines (786‐O, A498, OS‐RC‐2, 769‐P and ACHN) and HK2 detected by qRT‐PCR. F, Hsa_circ_001895 was resistant to RNase R digestion compared to linear CTBP1. **Tumor vs normal tissues, ccRCC cell lines vs HK2, RNase R vs MOCK, P < .01

Figure 2

Hsa_circ_001895 promoted clear cell renal cell carcinoma (ccRCC) proliferation. A, Expression of hsa_circ_001895 in cytoplasm or nucleus of 786‐O and A498 cells detected by qRT‐PCR. B, Subcellular localization of hsa_circ_001895 in 786‐O and A498 cells by RNA‐FISH. C, Transfection efficiency of sh‐hsa_circ_001895, sh‐CTBP1, pcDNA‐hsa_circ_001895 and pcDNA‐CTBP1 in 786‐O and A498 cells detected by qRT‐PCR. D, Influence of hsa_circ_001895 on cell viability of 786‐O and A498 cells detected by CCK‐8. E, Influence of hsa_circ_001895 on cell proliferation of 786‐O and A498 cells detected by colony formation assay. F, Influence of hsa_circ_001895 on cell proliferation of 786‐O and A498 cells detected by 5‐ethynyl‐2′‐deoxyuridine (EdU) staining assay. G, Influence of hsa_circ_001895 on cell cycle of 786‐O and A498 cells detected by flow cytometry. *, **sh‐hsa_circ_001895 vs sh‐NC or pcDNA‐hsa_circ_001895 vs EV, P < .05, P < .01. ^{# ##}sh‐CTBP1 vs sh‐NC or pcDNA‐CTBP1 vs EV, P < .05, EV, P < .01

Figure 3

Hsa_circ_001895 promoted clear cell renal cell carcinoma (ccRCC) migration and invasion and inhibited cell apoptosis. A, Influence of hsa_circ_001895 on cell migration of 786‐O and A498 cells detected by wound‐healing assay. B, Influence of hsa_circ_001895 on cell invasion of 786‐O and A498 cells detected by Transwell assay. C, Influence of hsa_circ_001895 on cell apoptosis of 786‐O and A498 cells detected by flow cytometry. D, Influence of hsa_circ_001895 on protein expression of E‐cadherin, N‐cadherin, Bcl‐2, Bax and Cleaved caspase 3 in 786‐O and A498 cells detected by western blot. *, **sh‐hsa_circ_001895 vs sh‐NC or pcDNA‐hsa_circ_001895 vs EV, P < .05, P < .01

Figure 4

Negative correlation between hsa_circ_001895 and microRNA (miR)‐296‐5p. A, Potential binding targets of hsa_circ_001895 predicted by CircInteractome (circular RNA interactome) and starBase. B, Relative intensity of miR‐516a‐5p, miR‐520h and miR‐296‐5p in HEK293 cells. C, Influence of hsa_circ_001895 on miR‐516a‐5p, miR‐520h and miR‐296‐5p expression in 786‐O and A498 cells detected by qRT‐PCR. D, Subcellular localization of hsa_circ_001895 and miR‐296‐5p in 786‐O and A498 cells by RNA‐FISH. E, Wild‐type and mutant binding sites of miR‐296‐5p in hsa_circ_001895. F, Influence of miR‐296‐5p mimics on luciferase activities of pmirGLO‐wt‐hsa_circ_001895 or pmirGLO‐mut‐hsa_circ_001895 in 786‐O and A498 cells detected by qRT‐PCR. G, Enrichment of hsa_circ_001895 and miR‐296‐5p in Ago2‐containing beads of 786‐O and A498 cells. H, Expression of miR‐296‐5p in clear cell renal cell carcinoma (ccRCC) tissues and adjacent noncancer tissues detected by qRT‐PCR (N = 60). I, Negative correlation between miR‐296‐5p and hsa_circ_001895 in ccRCC patients. **sh‐hsa_circ_001895 vs sh‐NC, miR‐296‐5p mimics vs miR‐NC, Input vs Anti‐IgG, Tumor vs Normal tissues, P < .01. ^##Anti‐Ago2 vs Anti‐IgG, P < .01

Figure 5

SOX12 was a direct target of microRNA (miR)‐296‐5p. A, Potential miR‐296‐5p binding targets predicted by TargetScan, miRDB, miRTarBase, miRWalk and miRTargetLink Human. B, Relative intensity of RNF44, SOX12 and NFIC in HEK293 cells. C, Influence of hsa_circ_001895 on mRNA expression of RNF44, SOX12 and nuclear factor I‐C (NFIC) in 786‐O and A498 cells detected by qRT‐PCR. D, Influence of hsa_circ_001895 on protein expression of RNF44, SOX12 and NFIC in 786‐O and A498 cells detected by qRT‐PCR. E, Wild‐type and mutant binding sites of miR‐296‐5p in SOX12. F, Influence of miR‐296‐5p mimics on luciferase activities of pmirGLO‐wt‐SOX12 or pmirGLO‐mut‐SOX12 in 786‐O and A498 cells detected by qRT‐PCR. G, Transfection efficiency of miR‐296‐5p mimics or inhibitor in 786‐O and A498 cells detected by qRT‐PCR. H, Influence of miR‐296‐5p on protein expression of SOX12 in 786‐O and A498 cells detected by western blot. I, mRNA expression of SOX12 in clear cell renal cell carcinoma (ccRCC) tissues and adjacent noncancer tissues detected by qRT‐PCR (N = 60). J, In situ hybridization (ISH) analysis of SOX12 expression in ccRCC tissues with different TNM stage. K, Negative correlation between miR‐296‐5p and SOX12 in ccRCC patients. Positive correlation between hsa_circ_001895 and SOX12 in ccRCC patients. L, Protein expression of SOX12 in ccRCC tissues and adjacent noncancer tissues detected by western blot. **sh‐hsa_circ_001895 vs sh‐NC, miR‐296‐5p mimics vs miR‐NC, miR‐296‐5p inhibitor vs inh NC, Tumor vs Normal tissues, P < .01

Figure 6

Hsa_circ_001895 knockdown inhibited clear cell renal cell carcinoma (ccRCC) proliferation by sponging microRNA (miR)‐296‐5p. A, Influence of hsa_circ_001895 and miR‐296‐5p on cell viability of 786‐O and A498 cells detected by CCK‐8. B, Influence of hsa_circ_001895 and miR‐296‐5p on cell proliferation of 786‐O and A498 cells detected by colony formation assay. C, Influence of hsa_circ_001895 and miR‐296‐5p on cell proliferation of 786‐O and A498 cells detected by 5‐ethynyl‐2′‐deoxyuridine (EdU) staining assay. D, Influence of hsa_circ_001895 and miR‐296‐5p on cell cycle of 786‐O and A498 cells detected by flow cytometry. *, **sh‐hsa_circ_001895 + inh NC vs sh‐NC + inh NC, P < .05, P < .01. ^{#, ##}sh‐hsa_circ_001895 + miR‐296‐5p inhibitor vs sh‐NC + inh NC, P < .05, P < .01

Figure 7

Hsa_circ_001895 knockdown inhibited clear cell renal cell carcinoma (ccRCC) migration and invasion but induced cell apoptosis by sponging microRNA (miR)‐296‐5p. A, Influence of hsa_circ_001895 and miR‐296‐5p on cell migration of 786‐O and A498 cells detected by wound‐healing assay. B, Influence of hsa_circ_001895 and miR‐296‐5p on cell invasion of 786‐O and A498 cells detected by Transwell assay. C, Influence of hsa_circ_001895 and miR‐296‐5p on cell apoptosis of 786‐O and A498 cells detected by flow cytometry. D, Influence of hsa_circ_001895 and miR‐296‐5p on protein expression of E‐cadherin, N‐cadherin, Bcl‐2, Bax and Cleaved caspase 3 in 786‐O and A498 cells detected by Transwell assay. *, **sh‐hsa_circ_001895 + inh NC vs sh‐NC + inh NC, P < .05, P < .01. ^{#, ##}sh‐hsa_circ_001895 + miR‐296‐5p inhibitor vs sh‐NC + inh NC, P < .05, P < .01. PI, propidium iodide

Figure 8

Hsa_circ_001895 knockdown inhibited in vivo clear cell renal cell carcinoma (ccRCC) tumor growth. A, Influence of sh‐hsa_circ_001895 on hsa_circ_001895 and microRNA (miR)‐296‐5p expression in mice intratumorally injected with lentiviral vector with hsa_circ_001895 knockdown or the negative control (sh‐NC). B, Effect of sh‐hsa_circ_001895 on ccRCC tumor growth in xenograft tumor mice. C, Influence of sh‐hsa_circ_001895 on tumor volume and weight. D, H&E staining shows morphological features of ccRCC tissues, and immunohistochemical staining was used to determine expression of SOX12, Ki‐67, E‐cadherin, N‐cadherin and Cleaved caspase 3 affected by sh‐hsa_circ_001895. Black bar, 200 μm. *, **sh‐hsa_circ_001895 vs sh‐NC, P < .05, P < .01