Background
Recent studies have identified that circular RNAs (circRNAs) have an important role in cancer via their well-recognized sponge effect on miRNAs, which regulates a large variety of cancer-related genes. However, only a few circRNAs have been well-studied in renal cell carcinoma (RCC) and their regulatory function remains largely elusive.
Conclusions
Based on the experimental data, in RCC, circNTNG1/miR-19b-3p/HOXA5 axis can regulate the epigenetic silencing of Slug, thus interfering EMT and metastasis of RCC. Together, our findings provide potential biomarkers and novel therapeutic targets for future study in RCC.
Methods
Bioinformatics approaches were used to characterize the differentially expressed circRNAs in our own circRNA-sequencing dataset, as well as two public circRNA microarray datasets. CircNTNG1 (hsa_circ_0002286) was identified as a potential tumor-suppressing circRNA. Transwell assay and CCK-8 assay were used to assess phenotypic changes. RNA pull-down, luciferase reporter assays and FISH experiment were used to confirm the interactions among circNTNG1, miR-19b-3p, and HOXA5 mRNA. GSEA was performed to explore the downstream pathway regulated by HOXA5. Immunoblotting, chromatin immunoprecipitation, and methylated DNA immunoprecipitation were used to study the mechanism of HOXA5.
Results
In all three circRNA datasets, circNTNG1, which was frequently deleted in RCC, showed significantly low expression in the tumor group. The basic properties of circNTNG1 were characterized, and phenotype studies also demonstrated the inhibitory effect of circNTNG1 on RCC cell aggressiveness. Clinically, circNTNG1 expression was associated with RCC stage and Fuhrman grade, and it also served as an independent predictive factor for both OS and RFS of RCC patients. Next, the sponge effect of circNTNG1 on miR-19b-3p and the inhibition of HOXA5 by miR-19b-3p were validated. GSEA analysis indicated that HOXA5 could inactivate the epithelial-mesenchymal transition (EMT) process, and this inactivation was mediated by HOXA5-induced SNAI2 (Slug) downregulation. Finally, it was confirmed that the Slug downregulation was caused by HOXA5, along with the DNA methyltransferase DNMT3A, binding to its promoter region and increasing the methylation level. Conclusions: Based on the experimental data, in RCC, circNTNG1/miR-19b-3p/HOXA5 axis can regulate the epigenetic silencing of Slug, thus interfering EMT and metastasis of RCC. Together, our findings provide potential biomarkers and novel therapeutic targets for future study in RCC.