IκBα is highly expressed in pluripotent cells.

A. Schematic of pluripotent states in mouse embryonic stem cells (mESCs). mESCs were maintained regularly in naïve pluripotency by culturing them in serum/LIF medium, or they were polarized towards the ground state of naïve pluripotency by culturing them for two passages in LIF plus Gsk3β (CHIR99021 or Chiron) and MEK (PD0325091) inhibitors (2i/LIF). mESCs were further differentiated towards the primed pluripotent state (epiblast-like cells) by culturing mESCs in a medium containing 20ng/mL Activin A and 10ng/mL Fgf2 for 120h. B-C. Relative mRNA levels (based on qPCR experiments) of the canonical IκB genes (Nfkbia, Nfkbib, Nfkbie) (B) or Nf-κB effector genes (Rela, c-Rel, Relb, Nfkb1, Nfkb2) (C) across the different states of pluripotency. Expression was normalized using the house-keeping gene Tbp relative to 2i/LIF. Data from three independent experiments. Dots indicate mean values and error bars refer to ± standard deviation (SD). D. Western blot analysis of IκBα protein distribution in cytoplasm, nucleoplasm (Nucleus) or chromatin in cells at the ground-state of naïve pluripotency (2i/L), naïve pluripotency (S/L) or primed pluripotency (Epi). E. Transcriptome and proteome differences across naïve (2i/LIF) and primed (Epiblast-like cells or EpiSCs) pluripotent states. Enriched transcripts and proteins in naïve state are highlighted in red (Log2 fold-change < -1). Enriched transcripts and proteins in the primed state are marked in dark gray (Log2 fold-change > 1). Data from 33. F. Heatmap showing the expression levels of pluripotency (Dppa4, Pou5f1, Sox2, Prdm14, Klf4, Zfp42, Nanog), differentiation (Mesp1, Gata4, Gata6, Lhx1, Wnt3a, Fgf5, Nes, Axin2, Wnt3, Eomes, Foxa2, T, Mixl1, Lef1) NF-κB (Nfkbia, Nfkbib, Nfkbie, Nfkbiz, Rela, c-Rel, Relb) and Polycomb (Ezh1, Ezh2, Eed, Epop, Jarid2, Suz12, Mtf2, Aebp2) genes from the RNAseq samples at mESCs (Serum/LIF) and differentiating cells (embryoid bodies) at 48h and 96h of IκBα-WT cells. Normalized counts based on z-score are represented.

The absence of IκBα hinders the exit from the pluripotent state and prevents the activation of differentiation programs.

A. Representative immunofluorescence of 216h EBs stained with the pluripotency markers NANOG and OCT3/4. B. Quantification of total number of alkaline phosphatase (AP) colonies after 216h EB differentiation and additional culture of 96h in Serum/LIF (S/L) medium. Dots represent values from two independent experiments from three different IκBα-WT or IκBα-KO clones. Unpaired two-sided t-test applied. Representative images from AP colonies are included in the upper panel. C. PC1 and PC2 from Principal Component Analysis (PCA) of normalized RNA-seq data from 3 different time points (mESCs, 48h EBs and 96h EBs) of IκBα-WT and IκBα-KO cells (three independent clones from each genotype). D. Heatmap showing the expression levels of naïve pluripotency genes (Gab1, Sox2, Esrrb, Zfp42, Dppa4, Klf2, Prdm14, Klf4, Tfcp2l1) from the RNAseq samples of IκBα-WT and IκBα-KO EBs at 96h. Normalized counts based on z-score are represented. E-G. Z-score values for gene sets referring to Endoderm (E), Mesoderm (F) and Ectoderm (G) formation from The Gene Ontology database (GO:0001706, GO:0001707 and GO:0001705 terms, respectively) and obtained from corresponding expression levels at mESCs, 48h and 96h EBs IκBα-WT or IκBα-KO cells (three independent clones of each genotype). Loess curves are also represented with 95% confidence intervals (shadowed area surrounding the curves). H. Schematic of teratoma formation assay. 5x105 IκBα-WT or IκBα-KO mESCs cultured in Serum/LIF were injected intramuscularly in the leg of immunocompromised NSG mice. 6 weeks after the transplant, teratomas were formed, and mice were euthanized for further teratoma analysis. I. Representative images of immunohistochemistry for OCT3/4 in IκBα-WT (left panel) or IκBα-KO (right panel) teratomas. Teratomas were derived from three independent clones of mESCs of both genotypes. J. Percentage of OCT3/4+ cells in teratomas. Positive cells were counted from five different microscope fields from each IκBα-WT and IκBα- KO clone. Unpaired two-sided t-test was performed. Bars indicate mean values and error bars refer to ± SD.

IκBα-KO mESCs retain the ground-state of naïve pluripotency under Serum/LIF culture.

A. (Upper panel) Schematics of mESCs differentiation towards Epiblast Stem Cells (EpiSCs). mESCs cultured in Serum/LIF were then seeded in N2B27 medium supplemented with 10ng/mL Fgf2 and 20ng/mL Activin A for 120h (see Materials and Methods section for further information). (Bottom panel) Relative RNA levels (based on qPCR experiments) of naïve pluripotency genes (Nanog, Sox2, Rex1, Klf4, Gbx2, Tbx3, Dppa3) in IκBα-WT or IκBα-KO EpiSCs at 120h. RNA levels were normalized based on Tbp expression relative to IκBα-WT. Each dot represents an independent clone either from IκBα-WT or IκBα-KO genotypes. Unpaired two-sided t-test applied. Horizontal bars indicate mean values and error bars refer to ± SD. B. GSEA results comparing IκBα-KO vs IκBα-WT mESCs cultured in Serum/LIF against ground (2i/LIF) and naïve (Serum/LIF) state pluripotency signatures retrieved from 35. Adjusted p-value by Benjamini-Hochberg procedure and Normalized Enrichment Score (NES) indicated. C. Relative RNA levels of the naïve pluripotency genes Rex1 (Zfp42), Klf2 and Tbx3 upon culture of IκBα-WT either in Serum/LIF or two passages in 2i/LIF. 2i/LIF IκBα-WT mESCs were compared with IκBα-KO mESCs cultured in Serum/LIF. Unpaired two-sided t-test was applied to calculate statistical significance. Each dot represents an independent clone from each of the two genotypes. D. Representative immunofluorescence images of DNA-methylation-related mark 5-methylcytosine (5mC) in IκBα-WT (left) and IκBα-KO (right) mESCs cultured in Serum/LIF. E. DNA methylation profile (using DNA methylation arrays 36) in IκBα-WT vs IκBα-KO mESCs (upper) and 96h EBs (bottom panel) showing the distribution density of mean ß-values from all 261,220 CpGs under test per condition.

Absence of IκBα reinforces the pluripotency program at the epigenetic and transcriptional levels.

A. ChIP-seq Enrichment Analysis (ChEA) of differentially expressed genes (DEGs; adjusted p-value<0.05) in IκBα-KO mESCs. P-value was calculated using Fisher exact test. B. Number of differentially bound regions (adjusted p-value <0.05, FDR) for H3K4me3 and H3K27me3 histone marks in IκBα -/- vs IκBα +/+ mESCs cultured in Serum/LIF. Enriched or reduced regions in IκBα -/- mESCs are distinguished (log2 Fold-change >0 or <0, respectively). Overrepresented WikiPathways, for (left) H3K4me3 and (right) H3K27me3 are indicated. Annotated genes to differential regions showing H3K4me3 enrichment or H3K27me3 reduction were considered. A one-sided hypergeometric test was conducted. C. Venn Diagram of genes contained in regions with either increased H3K4me3 or reduced levels of H3K27me3. Statistical significance of the overlapping was calculated using the chi-squared method. Overrepresented pathways for genes containing increased H3K4me3 and reduced H3K27me3 levels are indicated (only pathways with an adjusted p-value<0.05 were considered). D-E. Representative genomic regions of naïve pluripotency genes having either differential enriched H3K4me3-only (D) or reduced H3K27me3-only (E) levels in IκBα-KO vs IκBα-WT mESCs cultured in Serum/LIF. Shadowed regions highlight the differential levels of the histone marks. F. Number of regions with differential enhancer activity in IκBα-KO vs IκBα- WT mESCs cultured in Serum/LIF. Increased/Reduced activity for (left) poised enhancers based on the H3K4me1 differential binding and absence of H3K27ac overlapping peaks in IκBα-KO mESCs and, (right) active enhancers based on the differential binding of H3K27ac and presence of H3K4me1 overlapping peaks in IκBα-KO. Differential binding based on adjusted p-value <0.05 (FDR). Increased or reduced regions in IκBα-KO are distinguished by log2 Fold-change >0 or <0, respectively, compared to IκBα-WT mESCs. Overrepresented WikiPathways for annotated genes to (left) poised enhancers with reduced activity and (right) active enhancers with increased activity in IκBα-KO vs IκBα-WT mESCs are indicated (adjusted p-value < 0.05, FDR). A one-sided hypergeometric test was conducted. G. Representative genomic regions of naïve pluripotency genes having an increase in active enhancer status (H3K27Ac enrichment) in IκBα-KO mESCs cultured in Serum/LIF. Shadowed regions highlight the differential levels of the histone mark. H. Heatmap showing the expression levels of the naïve pluripotency genes from the RNAseq samples of IκBα-KO and IκBα-WT mESCs cultured in Serum/LIF. Normalized counts (z-score) from genes are represented.

IκBα prevents the hyperactivation of naïve pluripotency program in an NF-κB-independent manner.

A. Boxplots, violin plots and individual values representing the normalized gene expression levels in IκBα-WT or IκBα-KO mESCs, EBs 48h and EBs 96h. Considered genes are those annotated to NF-κB signaling pathway (KEGG PATHWAY mmu04064). Box Plots: center line, median; lower and upper hinges, first and third quartiles; whiskers, 1.5x interquartile range (IQR) extended to the largest/smallest value within 1.5xIQR; no outliers present. Unpaired two-sided Wilcoxon test applied. B. Schematic of chromatin-specific/NF-κB-deficient (IκBαΔNF-κB) or NF-κB-specific/chromatin-deficient (IκBαΔChrom) IκBα mutants (Separation-Of-Function or SOF mutants). C. Pull-down of GST-H2A and GST-P50 with cell lysates from different SOF IκBα mutant mESCs. D. Western blot analysis of cytoplasmic protein extracts of doxycycline-inducible IκBα forms (i-IκBαΔWT, i-IκBαΔNF-κB, i-IκBαΔChrom) after 16h of doxycycline induction. E. Western blot analysis of chromatin protein extracts of doxycycline-inducible IκBα forms (i-IκBαΔWT, i-IκBαΔNF-κB, IκBαΔChrom) after 16h of doxycycline induction. F. Relative RNA levels of naïve pluripotency genes (Zfp42, Klf2, Sox2, Tbx3) in i-IκBαWT, i-IκBαΔNF-κB and i-IκBαΔChrom mESCs after doxycycline induction. Unpaired two-sided t-test was performed. Bars indicate mean values and error bars refer to ± SEM. G. Representative immunofluorescence images of DNA-methylation-related mark 5-methylcytosine (5mC) in i-IκBαΔWT (upper panel), i-IκBαΔNF-κB (medium panel), IκBαΔChrom (bottom panel) mESCs cultured in Serum/LIF. H. Flow cytometry analysis of the pluripotency surface marker SSEA-1 at 216h EBs in i-IκBαWT, i-IκBαΔNF-κB and i-IκBαΔChrom cells. I. Quantification of Alkaline Phosphatase (AP) staining in 216h EBs upon reconstitution with the different IκBα forms. Each dot represents an independent experiment. One-way ANOVA was applied. Bars indicate mean values and error bars refer to ± SEM. Representative images of Alkaline Phosphatase (AP) staining in 216h EBs are included in the upper panel. J. Representative immunofluorescence for NANOG and OCT3/4 at 216h EBs in i-IκBαWT (upper panel), i-IκBαΔNF-κB (medium panel) and i-IκBαΔChrom (bottom panel) cells.

Depletion of IκBα in mESCs compromises the exit from pluripotency. Related to Figure 2.

A. Schematic of the approach followed to deplete IκBα protein in mESCs (upper panel) and western blot analysis of three independent IκBα-WT and IκBα-KO mESC clones (bottom panel). B. Flow cytometry plots of SSEA-1 from IκBα-WT or IκBα-KO EBs 216h. C. Relative RNA levels (based on qPCR experiments) of naïve pluripotency genes (Pou5f1, Gbx2, Klf2, Sox2, Rex1, Nanog) in IκBα-WT or IκBα-KO 216h EBs. RNA levels were normalized based on Tbp expression relative to IκBα-WT. Each dot represents an independent clone either from IκBα-WT or IκBα-KO genotypes. Unpaired two-sided t-test was applied. D. Schematic of Alkaline Phosphatase (AP) staining carried out in EBs. IκBα-WT or IκBα-KO 216h EBs were dissociated and seeded in Serum/LIF medium for another 96h. Attached cells were screened for AP+ total cell number. E. Violin plot and individual values showing the quantification of relative size of 168h EBs. Each dot represents an embryoid body. Unpaired two-sided t-test was applied. F. Volcano plot from differential expression analysis between IκBα-KO vs IκBα- WT 96h EBs. Dashed vertical lines represent an absolute shrunken log2 Fold Change = 1 and dashed horizontal line represents an adjusted p-value = 0.05 (FDR).

Absence of IκBα favors the ground-state of naïve pluripotency in Serum/LIF conditions. Related to Figure 3.

A. Brightfield images of IκBα-WT or IκBα-KO EpiSCs at 120h. Each picture corresponds to an independent clone from either IκBα-WT or IκBα-KO genotypes. B. Bright-field microscopy images of IκBα-WT and IκBα-KO mESCs cultured in Serum/LIF conditions. Two independent clones from each genotype are shown. C. DNA methylation levels (mean ß-values) across chromosomes in IκBα-WT vs IκBα-KO mESCs.

Epigenetic rewiring characterization in IκBα-KO mESCs cultured in Serum/LIF. Related to Figure 4. A-D.

Volcano plots from differential binding analysis of H3K4me3 (A), H3K27me3 (B), H3K4me1 (C) and H3K27Ac (D) in IκBα-WT vs IκBα-KO mESCs cultured in Serum/LIF. Representative genes with enriched or reduced levels of H3K4me3 (A), H3K27me3 (B), H3K4me1 (C) and H3K27Ac (D) marks in IκBα-KO mESCs are highlighted.

Reversion of IκBα-KO-related hyper pluripotent status is NF-κB independent. Related to Figure 5.

A. Relative expression level of Nfkbia gene after overnight doxycycline induction in i-IκBαΔWT, i-IκBαΔNF-κB and IκBαΔChrom mESCs. B. Schematic representation of reconstitution of IκBα-KO mESCs with either of the two doxycycline-inducible SOF mutants (i- IκBαΔNF-κB or IκBαΔChrom). mESCs were then kept in Serum/LIF medium supplemented with 1μg/mL Doxycycline for 4 consecutive passages. mESCs were characterized at the pluripotent level and then their potential to be further differentiated was addressed. C. Percentage of differentiated cells (determined as SSEA-1.) at 216h EBs 216h derived from i-IκBαΔWT, i-IκBαΔNF-κB and IκBαΔChrom mESCs. D. Quantification of EB differentiation status (percentage of pluripotent cells) at 216h after induction of SOF IκBα mutants. Data from three independent experiments.