Fig. 5.

EZH2 regulates CC-IC self-renewal through control of IHH expression. a RT-qPCR for IHH of UNC1999-treated cells. Cells were treated with UNC1999 at 3 µM for 7 days and processed for RT-qPCR. Data shown are mean (POP66/POP92 n = 6, LS174T n = 4,  ± SD, Student’s t test). b–d mRNA Levels on POP92 CC-ICs of Hedgehog target genes Gli1 and BMP4 (b), differentiation markers FABP2 and CDX2 (c), and the stem cell markers OCT4, KLF4, NANOG (d) monitored by RT-qPCR following UNC1999 or UNC2400 treatment (3 µM), or recombinant IHH (5 µg/mL) for 7 days. Data shown are mean (n = 4,  ± SEM, one-way ANOVA). e LDA on POP92, POP66 and LS174T cells pretreated with recombinant IHH (5 µg/mL) for 10 days. Data shown are n = 3 and represent mean  ± 95% confidence interval, frequency and probability were computed using ELDA software. f POP92 spheroids treated with DMSO or UNC1999 (3 µM) in the presence of SMO inhibitor Vismodegib (10 µM) or DMSO control for 7 days. Viable cell count was measured and samples normalised to their respective control. Data shown are mean (n = 5,  ± SD, two-way ANOVA). g IHH knocked down in POP92 using two different hairpins, and sensitivity towards UNC1999 was assessed using viable cell count via flow cytometry. Data are normalised to DMSO control performed in n = 3,  ± SD, two-way ANOVA. h POP92 with IHH knockdown were treated with UNC1999 (3 μM) or DMSO control for 7 days prior to seeding in LDA. Data shown are n = 2 and represented as mean  ± 95% confidence interval. Frequency and statistics were computed using ELDA software. i POP92 spheroids infected with either shRNA against Luciferase (shLuc) or shRNA targeting IHH (shIHH) were injected into mice and treated with vehicle or UNC1999 (300 mg/kg). Tumour volume was measured over time. Data are n = 20 tumours (shLuc  ± UNC1999), n = 19 tumours (shIHH ± UNC1999),  ± SEM, two-way ANOVA with Tukey multiple test correction. *P < 0.05, **P < 0.01, ***P < 0.001. Source data are provided as a Source Data file