Fig 2: Sinensetin reduces mRNA expression of EMT marker. (A) Zeb1, Zeb2, E47 and β-actin protein expression in MCF7 and MDA-MB-231 cells were determined by Western blotting. (B) Quantitative analysis for Zeb1, Zeb2 and E47 protein expression in MCF7 and MDA-MB-231 cells. (C) Gene expression analysis to Zeb1, Zeb2 and E47 in MCF7 and MDA-MB-231 cells were performed by qRT-PCR. ***, P<0.001; **, P<0.01. Sin, sinensetin; EMT, epithelial-mesenchymal transition; qRT-PCR, quantitative real-time polymerase chain reaction.

Fig 3: Increased HOXA11-AS expression suppressed ATF3 transcription in GC cells.a Heatmap of altered lncRNAs (fold change > 4, p < 0.05) in ATF3low GC tissues compared with paired adjacent normal tissues. Red arrows indicate a significant increase in lncRNA HOXA11-AS expression in ATF3low GC tissues. b The expression level of HOXA11-AS in GC tissues compared with that in paired normal gastric tissues as detected by RT-qPCR in our GC samples. c Correlation between the mRNA levels of HOXA11-AS and ATF3 in the GEO GSE35809 dataset (i) and in our GC samples (ii). d The expression level of HOXA11-AS was detected by RT-qPCR in HOXA11-AS-knockdown MGC-803 and MKN-45 cells. e ATF3 mRNA (i) and protein levels (ii) were detected by RT-qPCR and western blot, respectively, in HOXA11-AS-knockdown MGC-803 and MKN-45 cells, with β-actin serving as an endogenous control. f RT-qPCR and western blotting were performed to detect the mRNA (i) and protein (ii) levels, respectively, of β-catenin in HOXA11-AS-knockdown MGC-803 and MKN-45 cells, with β-actin serving as an endogenous control. g RIP analyses of HOXA11-AS1 and its predicted binding partners in MGC-803 (i) and MKN-45 cells (ii). h The colocalization of HOXA11-AS and TCF3 was analyzed by concomitant FISH and IF in MGC-803 and MKN-45 cells. Red, HOXA11-AS; Green, TCF3; blue, DAPI. Scale bar: 10 μm. i Western blotting analyses of TCF3 expression in MKN-45 cells subjected to pull-down with full-length HOXA11-AS. j ChIP-qPCR analysis of TCF3, EZH2, and H3K27me3 occupancy in the ATF3 promoter in MGC-803 and MKN-45 cells. k ChIP-qPCR analysis of TCF3, EZH2, and H3K27me3 occupancy in the ATF3 promoter in HOXA11-AS-knockdown MGC-803 cells. *p < 0.05; **p < 0.01 and ***p < 0.001 by Student’s t test compared with the corresponding control group.

Fig 4: EZH2 promotes proliferation of EC cells by directly repressing TCF3.A Immunoblots of EZH2, TCF3, or p21 following knocking down of EZH2 or TCF3 individually or simultaneously in ISK (left) and KLE (right) cells. GAPDH was used as internal loading control. B, C Cellular proliferation measured by CCK8 (B) and colony formation (C) assays following knockdown of EZH2 or TCF3 individually or simultaneously in ISK and KLE cells (as indicated). **P < 0.01 vs. indicated control. D–F Photograph (D), tumor volumes (E), and tumor masses (F) of excised xenograft tumors from scrambled (Scr) control, EZH2-KD, TCF3-KD, and EZH2-KD + TCF3-KD groups. Data are presented as the means ± SD; ***P < 0.001, *P < 0.05 vs. indicated control. G Western blotting analysis of indicated proteins in excised xenograft tumors from Scr control, EZH2-KD, TCF3-KD, and EZH2-KD + TCF3-KD groups. GAPDH and histone H3 were used as endogenous loading controls. H Pearson correlation scatter plot of H scores (protein levels) of EZH2 and TCF3 (left) or p21 (middle), or H scores (protein levels) of TCF3 and p21 (right) in human EC (n = 26).

Fig 5: EZH2 directly regulates TCF3 which regulates p21 transcription in EC cells.A Relative mRNA expression levels of a panel of key transcription factors potentially regulating p21 normalized to GAPDH analyzed by quantitative real-time PCR in scrambled control (Scr) or EZH2 knockdown ISK and KLE cells. Data shown are mean ± SD (n = 3). **P < 0.01 compared with Scr control. B Immunoblots of EZH2 and TCF3 in Scr or PRC2 core subunit (EZH2, EED, and SUZ12) knockdown ISK and KLE cells. GAPDH served as a loading control. C Enrichment of H3K27me3 on TCF3 promoter in Scr or EZH2 knockdown ISK cells detected by ChIP-qPCR (bottom). Diagram shows positions of ChIP primers at the TCF3 promoter (top). Data shown are mean ± SD (n = 3). **P < 0.01, *P < 0.05 compared with indicated control. D Immunoblots of TCF3, p21, and EZH2 in negative control (NC, vector alone) or enforced expression of TCF3 variants (E12 or E47) in ISK and KLE cells. GAPDH served as a loading control. E, F Cellular proliferation were measured by CCK8 (E) and colony formation assay (F) after enforced expression TCF3 variants (E12 or E47) in ISK and KLE cells. **P < 0.01 vs. NC. G Effect of inhibition of EZH2 activity by GSK126 and EPZ-6438 on TCF3 and p21 protein levels in ISK (left) and KLE (right) cells. GAPDH served as a loading control. H Cell proliferation measured by CCK8 assay after GSK126 and EPZ-6438 treatment in ISK and KLE cells. **P < 0.01 vs. Vehicle treatment. I Immunohistochemical (IHC) staining of TCF3 protein in matched human normal and EC tissues (left). Representative micrographs are shown in original magnification (×400). Scale bars, 50 μm. Total IHC score of TCF3 in matched human normal and EC tissues (n = 26). ***P < 0.001 vs. Normal control (right).