Fig. 5: CircRACGAP1 acts as an endogenous sponge for miR-3657.

a RNA-seq analysis of circRNAs was performed in three paired tumor xenografts from the control vehicle group and apatinib group. b qRT-PCR was performed in vivo to confirm the circRNA-seq results. c–f qRT-PCR was used in BGC-823 cells and HGC-27 cells cultured with apatinib to detect the expression of circRACGAP1 (c, d) and RACGAP1 (e, f). NS: not significant. g CircRACGAP1 was validated by Sanger sequencing. h RNA FISH was performed to detect the localization of circRACGAP1 in BGC-823 cells. 18S rRNA was used as a positive control for the cytoplasmic fractions. Bar scale, 10 μm. i A schematic model showing the putative binding sites of miRNAs (fold change ≥2 or ≤ −2) and circRACGAP1 using miRanda. j RNA immunoprecipitation (RIP) for AGO2 in BGC-823 cells was performed. CircRACGAP1 and miR-3657 expression levels were detected by qRT-PCR. k The predicted binding sequences for miR-3657 within the human circRACGAP1 3′UTR by TargetScan. The mutation sequences of circRACGAP1 Mut are shown in red. l Luciferase activity was measured in BGC-823 cells transfected with miR-3657 mimics or control miRNA, and circRACGAP1 Wt or circRACGAP1 Mut. The luciferase activity was normalized to that in cells transfected with control miRNA. *P < 0.05; **P < 0.01.