Fig 2: SS18 interacts with BAF through its N-terminal 70 residues.a Representative images for EGFP fusions expressed in mESCs as indicated. Note the similar condensates formed by SS18, FUS and TAF IDRs as compared to those for MED1 and BRD4. Poly Y is an artificial 24 tyrosine residue polypeptide. FL full length. Scale bars, 5 µm. laser intensity 2.5%, detector gain 500 V, background subtraction treatment. Three biological replicates. b Rescue of Ss18KO by constructs in (a) to show that IDRs from TAF and FUS or even polyY can substitute that of SS18. Data are mean ± s.d., two-sided, unpaired t test; n = 3 independent experiments, ***p < 0.001, ****p < 0.0001. c Volcano plots of IP-MS results from flag tagged Ss18-FL, SS18(IDR), N terminal 70aa of Ss18 alone or fused with IDRs from FUS or Brd4 and polyY expressed in Ss18KO mESCs by Flag antibody with label-free quantification. Every point represents a single protein. SS18 and BAF components are marked in blue and red, respectively. IP-MS experiments were performed in triplicates and a two-sample t test was applied. P = 0.01 and fold change = 2 were used as threshold. d Summary table for various constructs for their abilities to condensate, interact with BAF and mediate PST. e A proposed structure and function relationship for SS18: N-terminal BAF interacting domain or BID and C-terminal condensate forming IDR domain (CFD).

Fig 3: Loss of ARID1A affects BAF complex targeting and regulation of enhancers.(A) Fold change (log2) in SS18 ChIP-seq signals in ARID1A-/- #1 (x value) or ARID1A-/- #2 (y value) relative to control (Ctrl) NGP cells at each BAF binding site. Sites with weakened (<2/3×) signals in both ARID1A-/- #1 and ARID1A-/- #2 are indicated by blue, while sites with strengthened (>3/2×) signals in both ARID1A-/- #1 and ARID1A-/- #2 are indicated by red; others are indicated by black. (B) Fold change (log2) in H3K27ac ChIP-seq signals in ARID1A-/- #1 (x value) or ARID1A-/- #2 (y value) relative to control (Ctrl) NGP cells at each promoter and enhancer. Promoters or enhancers with weakened (<1/2×) signals in both ARID1A-/- #1 and ARID1A-/- #2 are indicated by blue, while sites with strengthened (>2×) signals in both ARID1A-/- #1 and ARID1A-/- #2 are indicated by red; others are indicated by black. (C) Example of SS18, BRG1, H3K4me3, and H3K27ac ChIP-seq tracks and RNA sequencing (RNA-seq) tracks at TSS-distal BAF weakened leucine rich repeats and immunoglobulin like domains 3 (LRIG3) locus (left) and strengthened immunoglobulin superfamily containing leucine rich repeat 2 (ISLR2) locus (right) in control (red), ARID1A-/- #1 (blue), and ARID1A-/- #2 (black) NGP cells. (D) ChIP-seq profiles and heat maps of SS18, BRG1, H3K4me3, and H3K27ac in control (red), ARID1A-/- #1 (blue), and ARID1A-/- #2 (black) NGP cells and of ARID1A in control NGP cells (red), around all TSS-distal BAF binding sites. (E) Log2 fold change of RPKM in RNA-seq between ARID1A mutants and control NGP cells for TSS-distal BAF target genes. Boxes indicate the median (horizontal line), 25th percentile, and 75th percentile; whiskers, distances from the largest and smallest value to each end of the box that are within 1.5× box length; dots, outliers. Data were analyzed with the Mann-Whitney test.

Fig 4: Reciprocal regulation between BAF and MLL4 on active enhancers during adipogenesis.Mll3?/?Mll4f/f or Smarcb1f/f brown preadipocytes were infected with adenoviruses expressing GFP or Cre, followed by adipogenesis assays. Cells were collected before (D-3) and during (D2) adipogenesis for ChIP-Seq, ATAC-Seq and Western blot analyses. a, b Decreased SMARCA4 and SS18 binding and chromatin accessibility on adipogenic enhancers in Mll4 KO (Cre) cells during adipogenesis. Adipogenic enhancers were defined as active enhancers bound by either C/EBPß, C/EBPa, or PPAR? at D2 of adipogenesis. Published ChIP-Seq data sets for C/EBPß, C/EBPa, and PPAR? were used (GSE74189)18. Heat maps around SMARCA4+ SS18+ adipogenic enhancers (a) and average profiles around SMARCA4+ SS18+ MLL4+ adipogenic enhancers (b) are shown. Adipogenic enhancers shown in the heat maps were ranked by the binding intensity of SMARCA4 at the center in D2 control (GFP) cells. c–f Deletion of Smarcb1 reduces binding of BAF subunits (SMARCA4, SS18, and ARID1A) as well as chromatin opening, and impairs MLL4 binding and activation of adipogenic enhancers. Adipogenic enhancers at D2 of adipogenesis were re-defined using ChIP-Seq data sets obtained in Smarcb1f/f cells. c Heat maps for ChIP-Seq of BAF subunits (SMARCA4, SS18, and ARID1A) and ATAC-Seq around the center of SMARCA4+ SS18+ MLL4+ adipogenic enhancers. d Heat maps for ChIP-Seq of MLL4, CBP, H3K4me1 and H3K27ac on adipogenic enhancers. Adipogenic enhancers shown in the heat maps were ranked by the intensity of SMARCA4 at the center in D2 control (GFP) cells. e Fold changes of intensities between control (GFP) and Smarcb1 KO (Cre) cells on adipogenic enhancers (n = 3530) are shown in box plots. The box represents the first and third quartiles with the horizontal line showing the median, and whiskers indicating minimum and maximum in all box plots. Outliers were not included. Statistical significance levels are as follows (Wilcoxon signed rank test, one-sided): SMARCA4 (p = 0); SS18 (p = 0); ARID1A (p = 6.3E-101); ATAC (p = 0); MLL4 (p = 4.5E-281); CBP (p = 0); H3K4me1 (p = 0); H3K27ac (p = 0). Exact p-values cannot be computed for SMARCA4, SS18, ATAC, CBP, H3K4me1 and H3K27ac due to ties in the ranks. f Western blot analyses of H3K4me1 and H3K27ac in control and Smarcb1 KO cells during adipogenesis. Total histone H3 was used as a loading control. n = 3 biological replicates.

Fig 5: BAF co-localizes with LDTFs and MLL4 on active enhancers during adipogenesis.a Average binding profiles of SMARCA4, SS18, and ARID1A around the center of SMARCA4+ active enhancers (AEs) or promoters before (D-3) and during (D2) adipogenesis of brown preadipocytes. Normalized read counts are shown. b Motif analysis of the top 3,000 SMARCA4+ SS18+ AEs using SeqPos motif tool. c Pie chart depicting BAF-binding (SMARCA4+ SS18+) AEs at D2 of adipogenesis. Among the 10,813 BAF-binding AEs at D2, 6,403 are de novo BAF-binding AEs (BAF-de novo), while 4,410 AEs are already bound by BAF (BAF-prebound) at D-3. d, e BAF subunits SMARCA4, SS18 and ARID1A colocalize with LDTFs (C/EBPß, C/EBPa, and PPAR?), MLL4, CBP and open chromatin on AEs (H3K4me1+ H3K27ac+) at D2 of adipogenesis. The local epigenetic environment exhibits coordinated changes during differentiation. Heat maps (d) and average profiles (e) were aligned around the center of SMARCA4+ SS18+ AEs. Published ChIP-Seq data sets for MLL4, CBP, C/EBPß, C/EBPa, and PPAR? were used (GSE74189)18.