Fig 2: BAF60c and BAF170 regulate both common and distinct gene expression programs compared with BRG1. (A) Loss of BAF60c and BAF170 proteins in respective KO cells detected by western bot in CPs. BRG1 levels are shown as loading controls. (B) Immunofluorescence of cardiac troponin T protein in wild-type, BAF60cKO and BAF170 KO cells in CMs. (C) Schematics showing the onset of beating in differentiating CMs (an average of multiple wells and differentiations). (D) Assessments of beating properties of wild-type, BAF60c KO and BAF170 KO CMs obtained by measuring the number of beats per minute and amplitude of beating represented by peak heights. Boxplots show data from the 1st quartile to the 3rd quartile with horizontal lines indicating the median. Whiskers are 1.5 times the interquartile range, with outliers shown outside of the whiskers. (E) Venn diagram shows an overlap of significantly affected genes (±1.5-fold, P<0.05) in the absence of Brg1, Baf60c and Baf170 at CP and CM stages. (F,G) Heat maps comparing genes significantly deregulated in any of the three genotypes over their respective wild-type counterpart at CP (F) and CM (G) stages. Clusters are shown in vertical colored bars on the left and representative genes involved in various biological processes are shown on the right of the heat map.

Fig 3: BRG1, BAF60c and BAF170 regulate temporal chromatin accessibility. (A,E) Heat maps of significantly affected ATAC-seq peaks (FDR <0.05, twofold change) in BRG1 conditional, BAF60c KO and BAF170 KO over their respective wild-type counterparts in CPs (A) or CMs (E). Enriched biological processes of the two nearest genes (within 100 kb) to the ATACseq peaks were analyzed using GREAT (McLean et al., 2010) and are shown to the right with representative genes. (C,H) Browser tracks showing delayed opening of Tnnt2 promoter chromatin (H) and loss of Scn10a enhancer in absence of BAF170 (C). (B,D,F,G) Chromatin opening and closing near Tbx5 (B) and Myf5 (D) in the absence of BAF60c and BAF170 in CPs, and chromatin closing in Tpm1 (F) and Mef2c (G) in the absence of BAF60c in CMs are shown.

Fig 4: Baf155 ablation does not affect the stability of components of SWI/SNF complex in the skeletal muscle.(A) RT-qPCR analysis of the expression of Baf155 in hind limb skeletal muscles (TA, EDL, GA, SOL, Q) of Baf155WT and Baf155ΔMF mice (n = 3 mice per genotype). (B, C) Representative immunoblotting analysis of Baf155 (B) and the densitometric quantification of relative Baf155 protein level in the Q muscle of Baf155ΔMF mice compared to Baf155WT mice (n = 3 mice per genotype) (C). (D, E) Representative immunoblotting analysis of Brg1, Snf5, and Baf170 (D) and the densitometric quantification of relative level of each indicated protein in Q muscle of Baf155ΔMF mice compared to Baf155WT mice (n = 3 mice per genotype) (E). (F) RT-qPCR analysis of the expressions of Brg1, Snf5, and Baf170 in Q muscle of Baf155WT and Baf155ΔMF mice (n = 3 mice per genotype). (G, H) Representative immunoblotting analysis of Arid1a, Brm, Baf60a, Baf60b, and Baf60c (G) and the densitometric quantification of relative protein level of each indicated protein in Q muscle of Baf155ΔMF mice compared to Baf155WT mice (n = 3 mice per genotype) (H). Each lane in immunoblotting analysis (B, D, and G) indicates each mouse (biological replicate), and each dot in the graphs (A, C, E, F, and H) represents each mouse (biological replicate). Data are presented as mean ± SEM of biological replicates. Statistical analyses were performed using unpaired Student’s t test (n.s., not significant; *P < 0.05, **P < 0.01, ***P < 0.001 versus Baf155WT control). The data underlying this figure can be found in S1 Data. Arid1a, AT-rich interaction domain 1a; Baf60, Brg1/Brm-associated factor 60; Baf155, Brg1/Brm-associated factor 155; Baf170, Brg1/Brm-associated factor 170; Brg1, Brahma-related gene 1; Brm, Brahma; EDL, extensor digitorum longus; GA, gastrocnemius; MF, myofiber; RT-qPCR, reverse transcription quantitative real-time PCR; SEM, standard error of the mean; Snf5, sucrose non-fermentable 5; SOL, soleus; TA, tibialis anterior; WT, wild type; Q, Quadriceps.

Fig 5: Morphology of MKN74 cells overexpressing SMARCD3. Microscopy was used to compare the shape of the cells (A), and fluorescence microscopy was used to analyze the shape of E-cadherin (B). The shape of a stable cell line overexpressing SMARCD3 in MKN74 cells (C) and the shape of E-cadherin in the stable cell line (D). Statistical analysis was performed by Student’s t-test (A) or one-way ANOVA (C). ** p < 0.01, and **** p < 0.0001. (A,C) are at 200× magnification, (B,D) are at 400× magnification, and the yellow box in (A,C) represents 100 µm on one side.