CircOMA1 promotes tumour growth and metastasis of bladder cancer by modulating IGF‐IR/MAPK/EMT pathway

Dear Editor, Although the prognostic results of patients with earlystage bladder cancer (BC) are very high, the 5-year survival rates of patients in advanced stages are greatly reduced to less than 6%.1,2 Thus, effective prognostic and therapeutic target biomarkers for early BC diagnostics are urgently needed. Unlike traditional linear RNA, circular RNA (circRNA), because of its covalently closed ring structure and excellent stability even exceeding GAPDH, has a potential application value as a clinical diagnosis and prognostic marker.3–5 Herein, we revealed the prognostic and diagnostic values of circOMA1 (hsa_circ_100242) and the critical role in promoting progression of BC by modulating insulin-like growth factor 1 receptor (IGFIR)/mitogen-activated protein kinase (MAPK)/epithelial mesenchymal transition (EMT) pathway. Altogether, our findings offer valuable and novel insights into the mechanism underlying BC regulation and new strategies for clinical practice. We analysed circRNA expression using CircRNA Microarray technology and identified 382 upregulated circRNAs between 6 pairs of BC tissues and corresponding adjacent noncancerous tissues6 (FC > 1.5, p < .05, Figure 1A). Our previous study indicated that miR-145-5p could inhibit tumour progression through regression of oncogenic IGF-IR in BC.7 Therefore, based on the prediction of Starbase, five upregulated circRNAs that can work as miR-145-5p molecular sponges have been screened (Figure 1B). Hsa_circRNA_100242 displayed the highest expression correlation with miR-145-5p and was selected for further study (Figure 1C). Hsa_circRNA_100242 is spliced from OMA1 exon two to exon eight and is referred to as circOMA1 (Figure 1D). To confirm the ring structure, we designed convergent and divergent primers and determined that circOMA1 exists as cDNA rather than gDNA (Figure 1E). RNase R digestion assay presented that linear OMA1 was degraded after RNase R treatment, whereas circOMA1 was resistant to RNase R (Figure 1F). Then we confirmed that circOMA1 was upregulated in

Dear Editor, Although the prognostic results of patients with earlystage bladder cancer (BC) are very high, the 5-year survival rates of patients in advanced stages are greatly reduced to less than 6%. 1,2 Thus, effective prognostic and therapeutic target biomarkers for early BC diagnostics are urgently needed. Unlike traditional linear RNA, circular RNA (circRNA), because of its covalently closed ring structure and excellent stability even exceeding GAPDH, has a potential application value as a clinical diagnosis and prognostic marker. [3][4][5] Herein, we revealed the prognostic and diagnostic values of circOMA1 (hsa_circ_100242) and the critical role in promoting progression of BC by modulating insulin-like growth factor 1 receptor (IGF-IR)/mitogen-activated protein kinase (MAPK)/epithelial mesenchymal transition (EMT) pathway. Altogether, our findings offer valuable and novel insights into the mechanism underlying BC regulation and new strategies for clinical practice.
We analysed circRNA expression using CircRNA Microarray technology and identified 382 upregulated circRNAs between 6 pairs of BC tissues and corresponding adjacent noncancerous tissues 6 (FC > 1.5, p < .05, Figure 1A). Our previous study indicated that miR-145-5p could inhibit tumour progression through regression of oncogenic IGF-IR in BC. 7 Therefore, based on the prediction of Starbase, five upregulated circRNAs that can work as miR-145-5p molecular sponges have been screened ( Figure 1B). Hsa_circRNA_100242 displayed the highest expression correlation with miR-145-5p and was selected for further study ( Figure 1C). Hsa_circRNA_100242 is spliced from OMA1 exon two to exon eight and is referred to as circOMA1 ( Figure 1D). To confirm the ring structure, we designed convergent and divergent primers and determined that circOMA1 exists as cDNA rather than gDNA ( Figure 1E). RNase R digestion assay presented that linear OMA1 was degraded after RNase R treatment, whereas circOMA1 was resistant to RNase R ( Figure 1F). Then we confirmed that circOMA1 was upregulated in BC tissue by fluorescence in situ hybridization (FISH) and qRT-PCR ( Figure 1G,H). Based on the median value of circOMA1 expression in BC tissues for patients whose survival data were available, we categorized these patients into the high-(n = 36) and low-expression group (n = 35). The Kaplan-Meier plot showed that high circOMA1 expression patients had lower overall survival (p = .042) ( Figure 1I). Furthermore, we found that the high expression of circOMA1 was highly associated with higher tumour grade (p = .001), higher pathological T stage (p = .036) and lymph node metastasis (p = .035) in BC ( Figure 1J). The KEGG enrichment analysis revealed that the IGF-IR/miR-145-5p axis most likely affects the downstream MAPK pathway ( Figure 1K).
To investigate the function of circOMA1 in BC cells, T24 and 5637 cells were separately transfected with sh-circOMA1, sh-NC, OV (overexpression)-circOMA1 and OV-NC. We detected the lentivirus transfection efficiency as high as 70% in BC cells by using qRT-PCR ( Figure S1A). In vitro, the proliferation of BC cells was determined by CCK-8 and EdU assays, respectively. The results indicated that circOMA1 positively regulated T24 and 5637 cells proliferation (Figures 2A,B andS1B-D). Meanwhile, similar results were found on BC xenograft nude mice (Figures 2C and S1E,F). In animal tumour tissues, lower levels of circOMA1 and Ki67 in sh-circOMA1 xenograft tumours were detected by FISH and immunohistochemistry staining, respectively (Figures 2D and S1G,H). In vitro assays showed that circOMA1 could positively regulate migration (Figures 2E and S1I,J), invasion (Figures 2F and S1K,L) and angiogenesis in BC (Figures 2F and S1M-P). In vivo imaging, the fluorescence intensities of metastatic lung nodules in mice were weaker in sh-circOMA1 group. Similarly, HE staining results showed that knock-down of circOMA1 reduced quantity and size of pulmonary metastases ( Figures 2G,H and S1Q,R).
Double-labelling immunofluorescence demonstrated that circOMA1 co-expressed with IGF-IR ( Figure 3A). Moreover, in protein and RNA level, circOMA1 could   Figure S2D,E. Based on the previous study, 8,9 these results indicated that circOMA1 may activate MAPK signalling pathway via regulating IGF-IR expression to regulate EMT progression. The results of qRT-PCR revealed that circOMA1 negatively regulated the expression of miR-145-5p ( Figure S3A). Confocal images showed that circOMA1 was colocalized with miR-145-5p prominently in cytoplasm ( Figure 3E). Furthermore, higher circOMA1 and miR-145-5p levels are found in anti-AGO2 RIP ( Figure 3F). Meanwhile, the results of dual-luciferase reporter assay showed that miR-145-5p decreased the luciferase activity of circOMA1-WT and IGF-IR-WT group, with no effect in circOMA1-MUT and IGF-IR-WT group ( Figure 3G,H). Knock-down of circOMA1 decreased the luciferase activity in IGF-IR-WT group ( Figure 3I). Afterwards, we also verified the effect of the expression of miR-145-5p and IGF-IR on the prognosis of BC patients and confirmed that the circOMA1/miR-145-5p/IGF-IR regulatory network can affect the prognosis of BC patients ( Figure S2F,G). These results indicated that circOMA1 positively regulates IGF-IR expression via sponging miR-145-5p in BC. In the reversal experiment, we silenced miR-145-5p in sh-circOMA1-transfected cell and overexpressed miR-145-5p in OV-circOMA1-transfected cell. The IGF-IR expression of 'sh-circOMA1+ inhibitor-miR-145-5p' group is higher than that of 'sh-circOMA1+ inhibitor-NC' group, whereas that of 'OV-circOMA1+ mimic-miR-145-5p' group is lower than that of 'OV-circOMA1+ mimic-NC' group ( Figure 4A). Meanwhile, silencing or overexpressing miR-145-5p markedly reversed the inhibition or activation of proliferation ( Figure Figure 4K).
In summary, our study is the first to reveal that cir-cOMA1 positively regulates IGF-IR expression by sponging miR-145-5p and activates the MAPK signalling pathway, thus playing a carcinogenic role in the pathogenesis of BC. Furthermore, we propose a ceRNA network, circOMA1/miR-145-5p/IGF-IR, as a potential regulatory mechanism that mediates the occurrence and development of BC. Cumulatively, circOMA1 is a powerful BC regulator and a new therapeutic target for BC treatment that warrants further investigation. 10 Lianghao Zhang 1 Yonghao Zhan 1 Longqing Li 2 Haotian Deng 1 Jiange Wang 3 Zhaowei Zhu 1 Xuepei Zhang 1