a) ESRRB occupancy at ESRRB-binding sites that are activated during EpiSCs to 2i-ESCs reprogramming41 (Adachi et al.). ESRRB binding sites were classified into E2- and non-E2 type enhancers. P-values (Wilcoxon rank sum test; two-sided); left to right: p=0.007, p=0.35, p=2.7e-07, p=9.4e-29, p=8.9e-64. n=1,648 ESRRB binding sites from Adachi et al.41 that overlap E2 enhancers, n=1,648 sampled non-overlapping enhancers. b) Western blot analysis of Tamoxifen-inducible Esrrb-knockout ESCs cultured in serum or 2i-medium. Experiment was repeated twice with similar results. c) Representative microscopy pictures of Tamoxifen-inducible Nr5a2-knock out ES-cells cultured in serum or 72 h 2i-medium and treated with Tamoxifen (4-OH) for Cre-activation and LoxP excision. The experiment was repeated twice with similar results. Scale bar in the image=100 µm. d) Luciferase activity of representative E2 enhancers. Tamoxifen-inducible Esrrb-knockout ESCs were transfected with the reporter constructs and cultured in 2i-medium. After 12 h of transfection, medium was changed to culture-medium supplemented or not with Tamoxifen (4-OHT) and cells were harvested after 48 h and assayed for luciferase activity. Values represent the means of n = 3 biological replicates. Renilla-luciferase was employed for normalization and signal intensities represent Firefly-luciferase / Renilla-luciferase values. e) qRT-PCR analysis of Tamoxifen-inducible Esrrb-knockout ESCs cultured in serum or 2i-medium. Two biological replicates were used per condition. f) Average profile of H3K27ac at 2i-specific E2-enhancers in long term serum- or 2i-cultured ESCs. Cells were treated with Tamoxifen (4-OHT) for 48 h followed by 72 h culture in serum- or 2i-medium (similar to Fig. 8b). Signal in a ±5 kb window flanking the peak center is shown. Reads from two biological samples per condition were pooled. g), Boxplot showing the ChIP-seq levels of H3K27ac and ESRRB-occupancy at E2 mono-allelic enhancers in F1-ESCs. Biallelic enhancers were used for comparison. Reads in input DNA were measured to exclude the bias in allele mapping. n=882 mono-allelic enhancers with higher signal at 129 alleles, n=366 mono-allelic enhancers with higher signal in Castaneus alleles, n=3,527 bi-allelic enhancers. Wilcoxon rank-sum test (two-sided). From left to right: P-value < 2.2e-16, P-value < 2.2e-16, P-value < 2.2e-16, P-value = 5.9e-12. Box=5-75th percentile; bar=median; whiskers=5-95th percentile.