Fig. 7: Cohesin depletion impedes p53-mediates transcriptional response.

A Distribution of CHiCAGO scores of the 179 interactions found between p53-bound enhancers and distal target genes found at Nut 1 h for wild-type (WT) and RAD21 depleted (KD) conditions. B Number of significant promoter interactions (CHiCAGO score ≥ 5) lost and maintained between WT and KD conditions at Nut 1 h. C Logarithmic distribution of genomic distance of interactions lost (n = 135) and maintained (n = 68) at Nut 1 h between WT and KD conditions. D Control-normalized differential H3K27ac (log2FC) at promoters of p53 distal target genes between Nut 1 h and Nut 0 h time points in wild type (WT) and RAD21 depleted (KD) conditions. Control genes are defined as those not directly targeted by p53. Two-sided Mann-Whitney test was used to test whether the control-normalized mean H3K27ac log2FC differed significantly between groups using a p value cutoff of 0.05 (star, p value = 3.2e-6). Sample sizes are 191 promoters for WT and 184 promoters for KD. E Distribution of wild-type-normalized nascent mRNA expression levels (fold change) of 12 distal target genes along p53 activation in WT or KD conditions, from at least three replicates each. A two-sided Mann-Whitney test was used to test whether expression levels differed significantly between conditions for each time point (star) (p value: 0.6 at Nut 0 h, 1.4e-13 at Nut 1 h and 2.0e-15 at Nut 10 h). A, C–E The boxes show the interquartile range (IQR), the central line represents the median, the whiskers add 1.5 times the IQR to the 75 percentile (box upper limit) and subtract 1.5 times the IQR from the 25 percentile (box lower limit). Significance calculated with two-sided Mann–Whitney test. The star indicates p values below 0.05. F Relative nascent mRNA expression (fold change) of p53 distal target genes along p53 activation in the presence (WT) or absence of RAD21 (KD). A one-tailed Student’s t test was used to test whether relative expression differed significantly between conditions at each time point for each gene (star). A linear mixed model was used to test whether relative expression differed significantly between conditions of adjacent time points, globally. For this, relative expression was averaged between the three replicates at each condition and time point for each gene (p value: 0.3 at Nut 0 h, and 2.3e-21 at Nut 1 h and 4.3e-16 at Nut 10 h. linear mixed model see Methods). Error bars correspond to one standard deviation of the mean. G Interaction landscape of S100A1’s gene promoter (blue shade) showing interactions (arcs) with a p53-bound enhancer (red asterisk and pink shade) located 101 kb upstream. Arrows symbolize gene placement and orientation along the genomic window. H Interaction landscape of PCM1’s gene promoter (blue shade) showing interactions (arcs) with a p53-bound enhancer (red asterisk and pink shade) located 202 kb upstream. Arrows symbolize gene placement and orientation along the genomic window. I Levels of relative nascent mRNA expression at 0 and 1 h after p53 activation, measured by qRT-PCR, for S100A1 (left), in wild type cell and both clones of CRISPR deletions targeting the p53 binding site distally acting on S100A1 (p53BS^−/− S100A1); and for PCM1 (right), in wild type cells and both clones of CRISPR deletions targeting the p,53 binding site distally acting on PCM1 (p53BS^−/− PCM1). J Graphical representation of both mechanisms by which p53 controls transcription of distal genes: i) a stable and primed mechanism that relies on a non-dynamic 3D chromatin structure; and ii) a dynamic and non-primed mechanism associated with the formation and destruction of DNA loops along p53 activation.