Supplementary Figure 9: Characterization of EHF and functional effects of EHF depletion on chromatin.

(a) RT–qPCR of EHF and differentiation-specific genes in organotypic epidermis depleted of EHF using two independent siRNAs targeting EHF versus scrambled controls. Error bars represent s.d. across two biological replicates (independent organotypic tissues). (b) Quantification of differentiation protein immunofluorescence signal area within the epidermis across technical replicates for control and two biological replicate organotypic epidermal tissues depleted of EHF (Student’s t test, one-sided, Rep 1 versus Control P = 3.54 × 10⁻², Rep 2 versus Control P = 9.3 × 10⁻³; *P < 0.05, **P < 0.01). (c) Heat map of RNA-seq z scores for all differentially expressed genes in triplicate biological replicate EHF-depleted (EHFi) versus control (CTRi) organotypic epidermal tissue. GO enrichments correspond to genes exhibiting increased expression with EHF depletion. (d) ChIP–qPCR of EHF, day 0 and day 3, demonstrating enrichment at a panel of enhancers in contact with promoters of genes induced in differentiation. Genes are noted in the heat map in b. Error bars represent s.d. across biological replicates (independent samples of cells from a single culture). (e) Top motifs identified by HOMER in EHF ChIP–seq peaks relative to a random background. (f) Box-and-whisker plots representing the median and interquartile range of the change in contact q score between day 3 and day 0. Contact sets are defined by EHF binding and H3K27ac dynamics at the promoter locus (bait HindIII fragment) or enhancer (empirical FDR, *P < 0.01). (g) RT–qPCR of DSG3 in control and EHF-depleted cells. Error bars represent s.d. across technical replicates. (h) UMI-4C profile of interactions anchored by the DSG3 promoter in control and EHF-knockdown conditions. Error bands represent s.e.m. between biological replicates.