Fig 2: Overexpressed YY1 induced tumor aggressive growth in breast cancer (BCa). (A) Overexpression of YY1 promoted MDA‐MB‐231 cell growth in mice xenograft models compared with that in the vector only group. (B) Tumor volumes were measured after injected MDA‐MB‐231 cells treated with the YY1 overexpression plasmid in the xenograft mouse model (n = 3). (C) qRT‐PCR analysis of the expression levels of YY1 mRNA in the xenograft tumors. (D) Western blotting analysis of the levels of YY1, KTN1, and epithelial‐to‐mesenchymal transition (EMT) markers. (E) Hematoxylin and eosin (HE) staining analysis of the effects on tumor cell aggression of YY1 overexpression in vivo; a black arrow represents the range of infiltrating of tumor cells. (F) Immunohistochemistry (IHC) staining of YY1, KTN1, and EMT marker proteins in xenograft tumors. (G) A model of the YY1/DDX3X/KTN1‐regulatory axis in BCa development. Error bars are shown with the s.d., n ≥ 3. *P < 0.05, **P < 0.01, and ***P < 0.001 compared with the negative control group. A two‐tailed t‐test or ANOVA was used to assess the P‐values. Scale bars, 100 μm

Fig 3: Depletion of YY1 blocked the invasive growth of breast cancer (BCa) cells in vitro and in vivo. (A) Knockdown of YY1 in MDA‐MB‐231 cells treated with siYY1_1 or siYY1_2 oligonucleotides compared with the siNC group as assessed by a qRT‐PCR assay. (B) CCK‐8 assay showing the proliferation of MDA‐MB‐231 cells treated with siNC and siYY1 oligonucleotides. (C) Monolayer colony formation assay showing the colony forming efficiencies of MDA‐MB‐231 cells treated with siNC and siYY1 oligonucleotides. (D) Transwell assay showing the migration and invasion of MDA‐MB‐231 cells treated with siNC and siYY1 oligonucleotides. (E) Western blotting analysis of protein levels in YY1‐overexpressing MDA‐MB‐231 cells treated with siKTN1 oligonucleotides. (F) Migration and invasion analysis in ‐overexpressing MDA‐MB‐231 cells treated with siKTN1 oligonucleotides using a Transwell assay. (G) Knockdown of YY1 with siYY1 oligonucleotides attenuated MDA‐MB‐231 cell growth in a mouse xenograft model compared with that in the siNC group. (H) Tumor volumes were measured after injection of MDA‐MB‐231 cells with siYY1 oligonucleotides in the xenograft mouse model; n = 6. (I) The expression levels of YY1 mRNA in the xenograft tumors. (J) Western blotting assay to detect the protein levels of YY1, KTN1, and epithelial‐to‐mesenchymal transition (EMT) markers. (K) Immunohistochemistry (IHC) staining of YY1, KTN1, and EMT marker in xenograft tumors. Error bars are shown with the s.d., n ≥ 3. * P < 0.05, ** P < 0.01 and *** P < 0.001 compared with the negative control groups. A two‐tailed t‐test or ANOVA was used to assess the P‐values. Scale bars, 100 μm

Fig 4: High YY1 expression was associated with poor clinical prognosis. (A) Immunohistochemistry (IHC) staining of YY1 in breast cancer (BCa) tissues. All data are shown as the means of the 95% confidence interval plus the s.d. (B) The expression level of YY1 mRNA was examined by qRT‐PCR in 36‐paired BCa tissues compared to adjacent normal tissues. (C) Western blotting analysis of the level YY1 protein in MCF10A, MCF‐7, T47D, ZR75‐1, MDA‐MB‐453, HCC38, BT20, MDA‐MB‐231, and BT549 cell lines. (D) Kaplan–Meier analysis of well differentiated, moderately differentiated, and poorly differentiated for relapse‐free survival (RFS) with high versus low expression levels of YY1 mRNA from the GEPIA database. (E) Correlation analysis between YY1 and KTN1 from the tissue microarray (R = 0.896, P = 0). (F) Knockdown of YY1 in MDA‐MB‐231 and BT549 cell lines treated with negative control (siNC), YY1_1 and YY1_2 siRNA oligonucleotides was determined using western blotting assays. (G) Overexpression of YY1 in both BCa cell lines transfected with empty vector or YY1 overexpression plasmid was detected using western blotting analysis. The P‐value of data showed significant differences as indicated using * P < 0.05, ** P < 0.01 and *** P < 0.001

Fig 5: YY1 regulated the downstream target gene KTN1 in a DDX3X‐dependent manner in BCa. (A) Chromatographic analysis of the proteins immunoprecipitated using anti‐YY1 antibodies or control IgG antibodies. (B) Black arrows indicate the DDX3X peptide peaks in the YY1‐pulldown samples from MDA‐MB‐231 and BT549 cells compared with those from control IgG samples. (C) Western blotting and co‐immunoprecipitation assays analysis of co‐factors of YY1. (D) Immunofluorescence chemistry assay analysis of the co‐location and co‐expression in both BCa cell lines. (E) The correlation analysis between DDX3X and YY1 expression from the GEPIA database (R = 0.52, P = 3.3e−95), and the correlation analysis between DDX3X and KTN1 expression from the GEPIA database (R = 0.47, P = 4.6e−76). (F) Activity of a reporter containing four canonical YY1‐binding sites and binding of YY1 to the KTN1 promoter in MDA‐MB‐231, BT549, and 293T cells transfected with the KTN1 promoter reporter, the full‐length DDX3X plasmid, and the full‐length YY1 plasmid (n = 4) as determined by a dual‐luciferase assays. (G) ChIP‐qPCR analysis showing the binding enrichment of YY1 at the binding sites on the promoter region of KTN1 detected after knockdown of DDX3X. Error bars are showed with the s.d., n ≥ 3. *P < 0.05, **P < 0.01, and ***P < 0.001. A two‐tailed t‐test or ANOVA was used to assess the P‐values