Fig 2: Reciprocal expression of ID2 and TFCP2L1 in human BC cells.a–c ID2 and TFCP2L1 transcript (a) and protein (b, c) levels in basal (5637 and HT1197), luminal (HT1376), and mixed (T24 and UMUC3) subtypes of muscle-invasive BC (MIBC) as well as in RT4 non-muscle-invasive BC cells. b Western blot analysis of ID2, TFCP2L1, and CDK1 in the indicated human BC cell lines. β-Actin was used as the loading control. Molecular weight (MW) marker sizes (in kD) are shown on the left. c Representative images of double immunofluorescence staining of ID2 (green) and TFCP2L1 (red) proteins at 200× magnification are shown. Scale bars = 100 µm. Nuclei were stained with DAPI (blue). d ChIP–qPCR assay of Flag-TFCP2L1 binding to the promoter region of ID2 in basal HT1197 and luminal HT1376 muscle-invasive BC (MIBC) cells (n = 4). A schematic diagram of the ID2 locus and the sites recognized by primers used in the ChIP–qPCR assay are shown in the top panel. Quantitative data are expressed as the mean ± SEM values. Statistical analyses were performed using two-way ANOVA with the Bonferroni post hoc test. *p < 0.05, ***p < 0.001 compared with the Flag control group. e Protein expression of ID2 and TFCP2L1 in the indicated BC cells 4 days after infection with lentiviruses expressing the TFCP2L1 ORF or shRNA specific for TFCP2L1 (shTFCP2L1). The exact P-values and numbers of biological replicates can be found in the source data.

Fig 3: Reciprocal expression of CDK1 and ID2 in human BC cells.a–c Dot plots showing the expression of CDK1 in normal (N; n = 28) and bladder tumor (T; n = 404) samples (a), based on tumor grade (b), and pairwise gene expression correlations with CDK1 transcript expression (c) after analysis of TCGA datasets of BC patients using the GEPIA web server. Quantitative results are presented as the mean ± SEM values. b Statistical analyses were performed using the nonparametric Mann–Whitney U-test. ***p < 0.001. d Pairwise correlations between ID2 and CDK1 transcript levels in pT tumor stage subgroups in a TCGA dataset (Robertson et al., 2017) of BC patients. All gene expression data were obtained from the UCSC Xena Browser (http://xena.ucsc.edu/). e–g RT-qPCR (e) and western blot (f, g) analyses of the expression of ID2 in basal HT1197 or luminal HT1376 MIBC cells after exposure to 20 µM RO-3306 for 2 h or 40 µM apigenin for 24 h. ß-Actin was used as the loading control. Molecular weight (MW) marker sizes (in kD) are shown on the left. g Quantification of ID2 protein expression. e, g Quantitative data are expressed as the mean ± SEM values (n = 4). Statistical analyses were performed using one-way ANOVA with the Bonferroni post hoc test. ***p < 0.001 compared with the nontreated (NT) control group.

Fig 4: ID2 suppressed the TFCP2L1-mediated stemness features of BC cells.a Ectopic expression of the TFCP2L1 or ID2 protein was validated by western blot analysis in basal (HT1197) and luminal (HT1376) MIBC cell lines. b–g Cell proliferation (b, n = 4), tumor-sphere formation (c, d; n = 10), clonogenic limiting dilution (e, n = 3), and Matrigel invasion (f, g, n = 6) capacities of HT1197 and HT1376 BC cells 4 days after overexpression of TFCP2L1 alone or in combination with ID2 (TFCP2L1 + ID2). Representative images of the tumor-sphere formation and Matrigel invasion assays are shown at 40× (upper panel in c), 100× (lower panel in c), or 200× (f) magnification. Scale bars = 100 µm. Lentivirus expressing empty vector was used as the control. All quantitative data are expressed as the mean ± SEM values. Statistical analyses were performed using one-way (e–g) or two-way (b–d) ANOVA with the Bonferroni post hoc test. *p < 0.05, ***p < 0.001 compared with the empty control group; ##p < 0.001 compared with the TFCP2L1 group. The exact P values and numbers of biological replicates can be found in the source data.

Fig 5: Combination treatment of BC based on the reciprocal expression of CDK1 and ID2.a BC cells were treated with the indicated concentrations of apigenin and RO-3306, and cell growth was assessed (n = 3). NT indicates the nontreated control group. b, c BC cells were treated with suboptimal concentrations of apigenin (15 µM) and RO-3306 (3 µM) for up to 48 h. Cell death was assessed by (b) immunoblotting of cleaved caspase-3 and PARP and by (c) Annexin-V/PI staining and flow cytometry. The quantitative assessments of apoptotic cells (Annexin-V + /PI- population, %; n = 3) are shown in the panels to the right of the representative flow cytometry plots. Quantitative data are expressed as the mean ± SEM values. Statistical analyses were performed using one-way (c) or two-way (a) ANOVA with the Bonferroni post hoc test. *p < 0.05, **p < 0.01, ***p < 0.001 compared with the NT control group. d Experimental overview of the orthotopic BC xenograft model generated by transplantation of 1.0 × 106 human HT1376 BC cells through the outer layer of the bladder of nonobese diabetic (NOD) Cg-Prkdcscid Il2rgtm1Wjl/SzJ (NSG) immunodeficient mice. Treatment with RO-3306 alone or in combination with apigenin was performed by intraperitoneal injection according to the indicated schedules. e Representative images (left panel) and weights (right panel) of bladders bearing tumors after treatment with RO-3306 (4 mg/kg) alone or combined with apigenin (50 mg/kg) in duplicate experiments (five mice in each replicate). The data are shown as dot plots of the mean ± SEM values from ten independent animals in each group. ***p < 0.001 compared with the vehicle control group; ###p < 0.001, one-way ANOVA with the Bonferroni post hoc test. f Representative images of hematoxylin and eosin staining of bladder tissues from the indicated xenograft groups at 100× (upper panel) or 200× (lower panel) magnification. Scale bars = 100 µm. g Immunofluorescence assay for detecting ID2, CDK1, and TFCP2L1 (green) proteins in xenograft tumors. Representative merged images are shown at 200× magnification. Scale bars = 100 µm. Nuclei were stained with DAPI (blue).