Sik4b

RCC4 and HCT116 cell lines were cultured in DMEM (Hyclone) with 10% FBS and 1% pen‐strep (P/S, Hyclone). SKOV3ip.1 cells were cultured as described previously.¹³ RCC4 and HCT116 cell lines were kindly provided by Amato Giaccia Lab (Stanford University). For short‐term KDM4B knockdown experiments, cells were transiently transfected with siRNA (Dharmacon siGenome Smartpool) specifically targeting KDM4B (siK4B, Thermo Fisher Scientific) using DharmaFECT 1 transfection reagent (Thermo Fisher Scientific) following the manufacturer's protocol. Scrambled siRNA (Dharmacon siControl 2) was used as the control group (siCon, Thermo Fisher Scientific). For hypoxia treatment, cells were cultured for 16 h in Ruskinn InVivo300 glove‐box hypoxic incubators (Baker), where oxygen levels were set at either 2% or 0.5%.

All three cell lines were transfected with siK4B (Thermo Fisher Scientific) or siCon (Thermo Fisher Scientific) for 32 h (transfection was repeated after the first 24 h) before culturing in 21% O₂ (all three cell lines) or 0.5% O₂ (SKOV3ip.1 and HCT116 cell lines) for 16 more h. Each RNA sample (100 ng) was profiled using the GeneChip® Human Exon 1.0 ST Arrays (Affymetrix, Thermo Fisher Scientific). Normalization and differential gene expression analysis was performed in the Partek Genomic suite (v 6.5, Partek Inc.). These exon arrays were RMA (Robust
Multi‐Array Average)‐background‐corrected, quantile‐normalized, and gene‐level‐summarized with the Median Polish algorithm.³⁵ The resulting expression values (log₂‐transformed signal intensities) were used in a two‐way analysis of variance model to calculate differential expression. p‐values were corrected for multiple hypothesis testing using the Benjamini and Hochberg method.³⁶ The analysis was done on samples obtained from biological triplicates. Genes with ≥1.4‐fold significant up‐ or down‐regulation by siKDM4B knockdown (p < 0.05), compared to the siControl cells, were considered as potential KDM4B downstream targets. Genes with ≥1.4‐fold significant up‐regulation when comparing siControl in hypoxia (0.5% O₂) to siControl in normoxia (21% O₂) were considered as hypoxia‐induced genes.