Fig. 1

CircSTX6 is verified and characterized in BCa cells and clinical samples. A Heatmap revealed the differentially expressed circRNAs in the four paired BCa tissues and NATs (GSE92675). B Volcano plots showed significantly different expression of circRNAs in paired five BCa tissues and NATs in research of Shen et al. |log2(FC)| > 2, P < 0.05. C Schematic diagram showed the screening of co-upregulated circRNA in both cohorts. D The back-splicing junction of circSTX6 was identified by Sanger sequencing. The red arrow showed the “head-to-tail” splicing sites of circSTX6. E The existence of circSTX6 was validated in two bladder cancer tissues and EJ, UMUC3 cell lines by RT-PCR and gel electrophoresis. Divergent primers amplified circSTX6 in cDNA but not genomic DNA (gDNA). GAPDH was used as a control. F qRT-PCR analysis depicted the circSTX6 and linear STX6 expression in BCa cells treated with or without RNase R. G qRT-PCR analysis revealed the levels of circSTX6 in SV-HUC-1, EJ, UMUC3, T24, 5637 and RT4 cells. H qRT-PCR analysis revealed the levels of circSTX6 in our 16 paired samples of BCa and NAT. N noncancerous adjacent tissues, T tumorous tissue. P values are calculated by paired two-sided t-test. I RNA levels of circSTX6, β-actin, and U1 in the nuclear and cytoplasmic fractions of bladder cancer cells. U1 was considered as a nuclear control and GAPDH was applied as positive controls in the cytoplasm. J RNA FISH analysis revealed that circSTX6 was predominantly localized to the cytoplasm. Nuclei were stained with DAPI. Scale bar, 10 μm. The data are presented as the means ± S.D. of at least three independent experiments. *P < 0.05. P values are calculated by Student’s t test in F and one-way ANOVA in G