Additional file 1 of Circular RNA EIF4G3 suppresses gastric cancer progression through inhibition of β-catenin by promoting δ-catenin ubiquitin degradation and upregulating SIK1

Additional file 1: Figure S1. The expression and intracellular localization of circEIF4G3 in GC.(A) The common downregulated circRNAs in three GEO datasets were listed as indicated. (B) Nuclear/cytoplasm distribution of circEIF4G3 in GC cells.Actin and U6 were applied as positive controls. (C) qRT-PCR assays for the expression of circEIF4G3 in GC cell lines (HGC-27, AGS, BGC-823, SGC-7901, MGC-803, MKN-45,and NCI-N87) and a normal gastric mucosa epithelial cell line (GSE-1). (D) ROC curves for the diagnostic value of serum circEIF4G3 in GC. Data are shown as means±SD. ***P<0.001. Figure S2. CircEIF4G3 overexpression inhibits EMT in GC cells.(A) qRT-PCR was used to examine the efficiency of circEIF4G3 overexpression in GC cells. (B) Western blot and (C) qRT-PCR analyses of N-cadherin, E-cadherin, Vimentin, slug and cyclin D1 expression in control and circEIF4G3 overexpressing GC cells. Figure S3. CircEIF4G3 silencing promotes GC cell proliferation, migration and invasion in vitro.(A) Schematic illustration of specific circEIF4G3-targeting sites. (B) Efficiency of circEIF4G3 knockdown in GC cells by siRNAs was tested by qRT-PCR. (C) Cell counting assay,(D) Colony formation assay, and (E-F) Transwell migration and matrigel invasionassays for si-Scr and si-circEIF4G3 GC cells. (G) Western blot and (H) qRT-PCR assays to evaluate the expression of N-cadherin, E-cadherin, Vimentin and cyclin D1 mRNA and proteins in GC cells after circEIF4G3 knockdown. (I) Cell apoptosis assays for GC cells with or without circEIF4G3 knockdown. (J) Flow cytometry analyses of cell cycle distribution in si-Scr and si-circEIF4G3 GC cells. (K) Western blot analyses of β-catenin, c-Myc, and cyclin D1 expression in circEIF4G3 knockdown GC cells. Data are shown as means±SD (n = 3). *P<0.05, **P<0.01,***P<0.001; Scale bar=100 μm. Figure S4. δ-catenin overexpression promotes GC cell proliferation, migration, and invasion in vitro.(A) The protein level of δ-catenin overexpression in GC cells after transfection. (B) Cell counting assay, (C) Transwell migration, and (D) Matrigel invasion assays for GC cells with or without δ-catenin overexpression. Data are shown as means±SD. Scale bar=100 μm. Figure S5. δ-catenin partially rescues the inhibition of GC progression by circEIF4G3 overexpression.(A) Cell growth curve, (B) Colony formation,(C) Transwell migration, and (D) Matrigel invasion assays for circEIF4G3 overexpressing GC cells co-transfected with or without δ-catenin. Data are shown as means±SD. Scale bar=100 μm. (E) Western blot assays for protein levels of β-catenin and its downstream targets in circEIF4G3 overexpressing GC cells co-transfected with or without δ-catenin. (F) qRT-PCR analysis of circEIF4G3 expression and western blot assay for δ-catenin protein levels in paired tumor and non-tumor tissues. *P < 0.05,**P < 0.01. Figure S6. TRIM25 promotes ubiquitination and degradation of δ-catenin in a circEIF4G3-dependent manner. (A) Interaction of TRIM25 with circEIF4G3 was determined by TRAP assay and validated by western blot. (B) RNA FISH and immunofluorescence staining for the co-localization of circEIF4G3 (red) with TRIM25 (green) or δ-catenin (green) in GC cells. Scale bar=25 μm. (C) mRNA levels of δ-catenin in GC cells with TRIM25 overexpression. (D) The levels of ubiquitinated δ-catenin in TRIM25 overexpressing GC cells with circEIF4G3 knockdown. Data were expressed as means ± SD. Figure S7. SIK1 is a target of miR-4449. (A-B) The mRNA (A) and protein (B) levels of potential targets were detected in GC cells with circEIF4G3 (or miR-4449) overexpression and knockdown. (C) The correlation analysis between circEIF4G3 and HUNK or PINX.(D) The expression levels of β-catenin protein and downstream targets in GC cells with SIK1 overexpression were examined by western blot. (E) Relative luciferase activity of β-catenin was detected by dual-luciferase reporter assay.  (F-H) Cell counting (F), Transwellmigration (G), and Matrigel invasion (H) assays for GC cells with SIK1 overexpression. (I) The efficiency of SIK1 knockdown was determined by qRT-PCR.Scale bar =100 μm. **P< 0.01, ***P< 0.001. Data were expressed as means± SD. Figure S8. CircEIF4G3 overexpression inhibits δ-catenin while promotes SIK1 expression in vivo.The protein levels of δ-catenin and SIK1 in mouse tumor tissues were detected by western blot.