SUMMARY
Chromatin remodeling complexes instruct cellular differentiation and lineage specific transcription. However, the underlying mechanism is unclear. Here, using immunoprecipitation with mass spectrometry (IP-MS), we determined the dynamic composition of the BRG1/BRM associated factor (BAF) complexes during mammalian cardiac differentiation, and identified BAF60c (SMARCD3) and BAF170 (SMARCC2) as subunits enriched in cardiac precursors (CP) and cardiomyocytes (CM). The catalytic subunit Brg1 has a specific role in CPs to initiate cardiac gene expression programs and repress non-cardiac fates, a role shared by Baf60c and Baf170. In CMs, BAF60c and BAF170 are found in BRG1 and BRM-associated complexes, and are essential for maintaining the cardiac program, largely by repressing non-cardiac gene expression. BAF60c and BAF170 both modulate BAF complex composition and stoichiometry. In CM, BAF170 facilitates expulsion of a subset of BRG1-containing complexes near genes regulating post-transcriptional and epigenetic modification, as opposed to those evicted independent of BAF170 that regulate transcription, growth and development. Thus, BAF complexes use varied interdependent mechanisms to direct temporal gene expression programs in cardiogenesis.
INRODUCTION
Cell differentiation and organogenesis are regulated by the precise transcriptional output of a coordinated gene network. During mammalian development, gene expression programs are spatially and temporally controlled with specific sets of genes being expressed while others are poised or repressed in a developmental stage dependent manner. Delineating the factors that control these crucial decision points is essential to understand the developmental control of gene regulation.
Transcription factor (TF) activity regulates transcriptional output, and is intimately influenced by the underlying chromatin architecture. Chromatin remodeling complexes alter chromatin structure to regulate transcription. These are multi-subunit protein complexes that utilize energy from ATP hydrolysis to alter the histone DNA contact in nucleosomes, the basic unit of chromatin (Clapier and Cairns 2009; Bartholomew 2014; Zhou et al. 2016). The BRG1/BRM associated factor (BAF) chromatin remodeling complexes are 15-18 subunit protein complexes. Brahma (BRM) or Brahma-related gene 1 (BRG1) are the mutually exclusive catalytic subunits that associate with several other structural subunits and their isoforms, to from diverse BAF complexes and serve specific functions in widely different cell types and developmental processes (Wang et al. 1996b; Ho and Crabtree 2010; Hota and Bruneau 2016). It is proposed that this diversity in BAF complexes is tailored to provide a unique surface to bind TFs leading to tissue specific function. How BAF complexes regulate transcription of specific developmental processes is not clearly understood.
In the developing heart, various subunits of the BAF complex are involved in aspects of cardiac development including early cardiac specification, heart tube looping, cardiomyocyte commitment, ventricular chamber morphogenesis, interventricular septum formation and trabeculation (Devine et al. 2014; Lange et al. 2008; Cai et al. 2013; He et al. 2014; Takeuchi et al. 2011; Stankunas et al. 2008). It is not known if these subunits function individually, or assemble together as one or more specialized cardiac BAF complex(es) with shared or distinct function.
Here, we define dynamic BAF complex composition during mouse cardiomyocyte differentiation. We identify the BAF subunits BAF60c and BAF170 as enriched in cardiac precursors and cardiomyocytes. BRG1 initiates cardiac gene expression programs in precursor cells, a role shared by BAF60c and BAF170, which also maintain the cardiac program to facilitate cardiomyocyte differentiation. BAF60c and BAF170 also regulate BAF complex composition and stoichiometry. Further, we find that BAF170 facilitates dissociation of BRG1 complexes from their binding sites upon cardiomyocyte differentiation. These results reveal the instructive nature that specific combination of BAF subunits attain to dictate functional outcomes during lineage commitment and differentiation.
RESULTS
BRG1 complex shows dynamic composition during cardiomyocyte differentiation
To model early heart development, we differentiated mouse embryonic stem cells (mESCs) to cardiac troponin positive (cTnT+) beating cardiac myocytes (Kattman et al. 2011; Wamstad et al. 2012). We regularly obtained 75-90% cTnT+ beating cardiac myocytes as measured by flow cytometry. BAF complex members have been implicated in TF interaction and cardiac development (Hota and Bruneau 2016; Bevilacqua et al. 2014), however we have limited knowledge on how BAF subunits change during cardiogenesis (Singh and Archer 2014). In order to understand the composition of BAF complexes, we immunoprecipitated BRG1 using a strain that contains triple FLAG tag at the C-terminus of one of the endogenous Brg1 alleles (Attanasio et al. 2014), at five different stages of cardiac differentiation: ES cells (ESC), embroid bodies (EB), mesoderm (MES), cardiac progenitor (CP) and cardiac myocytes (CM) (Fig. 1A). BAF complexes isolated from ESCs and separated on SDS PAGE (Fig. 1B) shows protein profiles similar to other reports (Ho et al. 2009b). Mass-spectrometric analyses of protein complexes isolated from five different stages of cardiac differentiation identified the composition of the BAF complex at these stages (Fig. 1C; Supplemental Table S1). BAF complexes isolated form ES cells were enriched for subunits BRD9, GLTSCR1l, BCL7b&c, BAF155 and BAF60a, consistent with previous reports (Ho et al. 2009b; Kadoch and Crabtree 2013). Similarly, we identified proteins enriched at mesoderm (PDE4D, CRABP2), CP (RPL7a, Polybromo-1, BAF47, BRD7 and BAF45a) and CM stages (BAF170, BAF60c, BAF57, SS18l1, BAF45c and CC2D1B). Identification of these enriched subunits revealed that BAF complex changes its composition during cardiac differentiation and BAF subunits switch from one isoform to other (for example, BAF60a in ES BAF complexes is replaced by BAF60c in cardiac BAF complexes) or to a different protein (BAF155 in ES cells to BAF170 in cardiac cells). We confirmed the switch in BAF155 in ES BAF to more abundant BAF170 in cardiac cells and appearance of BAF60c only in cardiac cell lineages during differentiation by western blot (Fig. 1D). Thus, BAF complexes have dynamic subunit composition and enrich for specific subunits in cardiac lineages.
Cardiac enriched BAF subunits form distinct cardiac BAF complexes
To further understand the composition of cardiac enriched BAF complexes, we immunoprecipitated endogenous FLAG tagged BAF170 or BAF60c at different stages of cardiac differentiation and analyzed by mass spectrometry. An untagged cell line processed identically served as negative control. BAF complexes isolated using BAF170 as the bait protein revealed that at each of the ES, EB and MES stages the protein abundance is too low (Fig. 1E). This is consistent with the expression of Baf170, which only peaks at the CP stage of differentiation (Supplemental Fig. S1A). Most of the BAF subunits enriched at the CP and CM stages using BRG1 as the bait protein were also enriched using BAF170 as the bait (Fig. 1E). In addition, we detected WDR5, PKP1 and BRM in the BAF170-FLAG purification. WDR5 was also present at lower enrichment in the BRG1 complexes (Supplemental Fig. S1B). We do not detect PKP1 or BRM in BRG1-FLAG purifications, indicating that BAF170 may specifically interact with these subunits, and consistent with BRG1 and BRM being mutually exclusive BAF complex catalytic subunits (Wang et al. 1996a). IP-MS revealed that BAF170 begins to associate with BRM at CP and BRM association with BAF170 increases in CM (Fig. 1E). BAF complexes isolated using FLAG-BAF60c were similar to those of BAF170-FLAG isolated complex (Fig. 1F). Again, we detected both BRG1 and BRM, indicating that BAF170 and BAF60c containing complexes swap out BRG1 for BRM as their primary catalytic subunit during cardiac differentiation.
To understand if changes in subunit abundance are due to transcriptional changes of that particular subunit or due to its association with the BAF complex, we plotted normalized FPKM values of these subunits at four stages of differentiation and found that while transcriptional changes correlate with BAF association in some cases (for example BAF60c, BAF170), in others it is independent of expression (for example, WDR5 is expressed mostly in ESC, while it associates with BAF complexes only in CMs) (Supplemental Fig. S1A,B).
Dynamic subunit composition does not affect BAF complex activity in vitro
We investigated if differentiation stages specific changes in composition affected biochemical activities of BAF complexes by performing ATP hydrolysis and nucleosome repositioning assays (Hota and Bartholomew 2012). In-vitro reconstituted mononucleosomes (Lusser and Kadonaga 2004; Lowary and Widom 1998) were incubated with equal amount of BRG1 isolated complex from all the five stages of differentiation with constant amount of ATP from 30s to 2hrs and resolved in a native polyacrylamide gel (Fig. 2A). Electrophoretic mobility shift assay revealed that BAF complexes isolated from ES cells (esBAF) could shift centrally positioned nucleosomes towards the end of the DNA fragment, as observed by the faster migration in native PAGE (Fig. 2A, lanes 1-5). BAF complexes isolated from other stages of differentiation (ebBAF, mesBAF, cpBAF and cmBAF) could also similarly shift centrally located nucleosome towards end of the DNA after incubation for various length of time (Fig. 2A, compare lanes 1-5 to 6-10, 11-15, 16-20 and 21-25 and Fig. 2B). This indicates that altered subunits in BAF complexes do not significantly contribute toward repositioning of in vitro reconstituted nucleosomes. We tested the abilities of these complexes to hydrolyze ATP in presence of nucleosome and found that BAF complexes isolated from different stages of differentiation could similarly hydrolyze ATP to ADP and free phosphate (Fig. 2C). Thus, although cell-type specific BAF complexes show alteration in subunit composition and enrichment of specific subunits, it appears that these subunits do not engage in regulating the ATP hydrolysis and nucleosome mobilizing activities of BAF complexes.
Brg1 initiates cardiac gene expression programs during cardiomyocyte differentiation
To understand how changing BAF complex subunit composition transcriptionally affects cardiac differentiation, we examined the role of the catalytic subunit BRG1 and cardiac enriched subunits BAF60c and BAF170 during cardiomyocyte differentiation. First, we conditionally deleted Brg1 (also known as Smarca4) using an ES cell strain that flanks exon 17 and 18 of Brg1 with loxP sites and expresses creER under β-actin promoter (Brg1flox/flox;βActin creER; Brg1KO) (Ho et al. 2011). Addition of 4-hydroxytamoxifen (4OHT) to culture medium for 48hrs effectively deleted these exons and reduced BRG1 protein levels to undetectable levels (Alexander et al. 2015). The effect of Brg1 deletion at CP and CM stages of differentiation on transcription was evaluated by RNA sequencing (RNASeq). Deletion of Brg1 deregulated a total of 545 genes at CP and 125 genes at CM stage of differentiation (p < 0.05, ± 1.5 fold), indicating a declining role of Brg1 during the cardiomyocyte stage, consistent with its reduced expression at this stage (Fig. 3; Supplemental Fig. S1A). In CP, of the 545 deregulated genes, 197 (36.1%) and 348 (63.9%) were significantly up or down regulated respectively, indicating a dual role of BRG1 in repression and activation of genes (Fig. 3A). Gene set enrichment analysis (GSEA) (Subramanian et al. 2005) revealed that genes down regulated in absence of BRG1 correlated more with genes upregulated in CP over mesoderm (Supplemental Fig. S2A, left panel), indicating that BRG1 is an activator of a CP gene expression program. No such correlation was found for BRG1-repressed genes in CPs (Supplemental Fig. S2A, right panel). Consistent with the GSEA results, gene ontology (GO) term enrichment analyses revealed that BRG1 activates genes involved in sarcomere organization and assembly and are essential components of cardiac cell fate establishment (Fig. 3B). Of note, BRG1 is essential for the immediate activation of these lineage-specific genes, in anticipation of the final differentiation status of the cells. Thus, BRG1-associated complexes prime the cell-type specific differentiation of cardiac precursors.
In cardiomyocytes, loss of BRG1 deregulated only 125 genes with roughly equal number of genes being activated (48%) or repressed (52%) (Fig. 3C). Although BRG1 activated genes correlated with CM (Supplemental Fig. S2B) gene ontology (GO) biological processes did not significantly enrich for any term (Fig. 3D) indicating BRG1 has minimal role in cardiac myocytes, consistent with in vivo data (Hang et al. 2010).
Cardiac enriched BAF subunits BAF170 and BAF60c facilitate cardiomyocyte differentiation
As altered subunit composition during cardiac differentiation did not affect ATPase or nucleosome repositioning abilities of BAF complexes, we investigated how cardiac enriched subunits regulate gene expression programs. We created BAF60c or BAF170 KO ES cell lines by TALEN or CRISPR mediated disruption of each gene and confirmed the clones by PCR, DNA sequencing and western blot (Supplemental Fig. S3A, S4A). ES cells lacking BAF60c or BAF170 undergo normal cardiac differentiation as observed by beating cardiac cells, immunostaining and intracellular FACS analysis of cardiac troponin T (cTnT). To understand the contribution of these subunits towards cardiac cell function and maturation and to compare the gene expression profile of cells lacking BAF170 or BAF60c to those lacking BRG1, we collected cells at CP and CM stages and performed RNASeq.
At CP, in absence of BAF60c, a total of 474 genes (p-value < 0.05, ± 1.5-fold) were deregulated of which 215 (45.4%) were down regulated and 259 (54.6%) were upregulated (Supplemental Fig. S3B). while a total of 382 genes are deregulated in absence of BAF170, of which 192 (50.2%) were down regulated and 190 (49.8%) were upregulated (Supplemental Fig. S4B). Gene ontology analyses revealed that BRG1 and BAF60c share common function in activating genes involved in muscle tissue morphogenesis, cardiac cell development and cardiac muscle contraction while repressing skeletal system and limb morphogenesis in CPs (Fig. 3B; Supplemental Fig. S3C). Genes regulated (largely repressed) by BAF170 are mostly involved in skeletal system morphogenesis and pattern specification (Supplemental Fig. S4C). In contrast, BAF60C regulated a large number of genes in cardiomyocytes (2646), repressing 72% of genes and activating 28% of these genes (Supplemental Fig. S3D, S3E). Although BAF170 regulated 574 genes in cardiomyocytes, a large percentage (63%) were up-regulated in absence of BAF170 (Supplemental Fig. S4D, S4E). These results indicate a largely repressive function of these cardiac enriched subunits in gene regulation.
Overall comparison of the gene expression changes in CPs and CMs lacking BRG1, BAF60c, or BAF170 show that BRG1, BAF60c and BAF170 have shared functions in CP gene expression, but that in CPs and CMs BAF60c and BAF170 repress a cohort of genes independent of BRG1 (Fig. 4A,4B). This clearly indicates divergent role of these two subunits from BRG1, suggesting their association in a different BAF complex or potential independent function outside of BAF complex. In line with these observation, we observed decreasing overlap between genes regulated by BRG1, BAF60c and BAF170 in CMs (Fig. 4C, 4D). Thus, varied but regulated combinations of subunits in BAF complexes regulate the transcriptional outcome of specific stages of development.
Both BAF170 and BAF60c regulate BAF complex stoichiometry
As both BAF60c and BAF170 has critical roles in transcriptional regulation of cardiac differentiation, we wished to understand the nature of BAF complexes formed in absence of these cardiac enriched subunits. Toward this goal, we isolated BAF complexes via immunoaffinity purification of FLAG tagged BRG1 protein from strains lacking BAF60c or BAF170 at CP and CM. SDS PAGE revealed increased stability of BRG1-containing complexes in BAF170 KO CMs (Fig. 5A, compare lanes 8,9 and lanes 10,11 to Fig. 1B, lane 5 to lane 6). It appears that the decrease in levels of BRG1 band intensity from CP to CM is greatly minimized by lack of BAF170, indicating greater stability of BRG1 complex. Mass spectrometric analyses of these complexes revealed significant differences in subunit composition in the absence of BAF170 or BAF60c (Fig. 5B,5C). In CPs, BAF complexes lacking BAF170 had decreased association of BCL7a BAF45d, BAF60c, BAF47 and BAF53a, and were enriched for BAF45c, CREST and BAF180. In CMs, BAF complexes lacking BAF170 showed reduced association with BAF60c and increased association with BAF155 and BAF60b (Fig. 5B; Supplemental Fig S5A, S5B). Similarly, we observed alterations of subunits composition at CP and increased association of different subunits with BAF complex lacking BAF60c in cardiac myocytes (Fig. 5C; Supplemental Fig. S5A, S5B). The altered complexes formed are not due to changes in the transcriptional level of BAF subunits (Supplemental Fig. S5C).
These results suggest that a fine balance exists in the composition of the BAF complex and perturbation of one subunit extends to the stoichiometry of association of other subunits in the complex. It further indicates that different subunits or isoforms substitute for the absence of one or more subunits. For example, both BAF60a and BAF60b substitute for the lack of BAF60c, and depletion of BAF45c is balanced by enrichment of BAF45a (Supplemental Fig. S5B). These results emphasize that subunit switching and substitution in BAF complex composition could be an important mechanism in cardiac lineage specification.
BRG1 binding correlates well with sites of cardiac transcription factor binding
To understand the potential mechanisms by which cardiac enriched subunits BAF60c and BAF170 regulate gene expression, and to distinguish between the direct and indirect role of BRG1 in gene regulation, we evaluated genome-wide BRG1 chromatin occupancy by chromatin immunoprecipitation using an antibody against BRG1 followed by sequencing of immunoprecipitated DNA (ChIP-Seq) in WT and cells lacking BAF60c or BAF170. We could not detect BRG1 occupancy in CMs, presumably due to the low levels of protein at this stage. In CPs, BRG1 was bound to both transcriptional start sites (TSS) and active enhancer regions marked by acetylation of lysine 27 residue of H3 (Wamstad et al. 2012) (Fig. 6A). TSS of genes deregulated (up or down regulated) in the absence of BRG1 show significant binding of BRG1 (Fig. 6B). For example, the upstream regulatory regions of BRG1 activated genes Tbx5 and Myocd and promoter regions of the repressed gene Rdh10 show significant BRG1 binding over input control (Fig. 6C). We found 4791 significant BRG1 binding regions (Supplemental Fig. S6A). About 60% of BRG1 binding sites are associated with the promoter region and rest are in distal regulatory region and are associated with 5274 genes within one megabase of BRG1 binding sites (Supplemental Fig. S6B, S6C). Of the 545 Brg1KO deregulated genes in CPs, 21.9% of BRG1 activated and 20.9% of the BRG1 repressed genes overlapped with genes nearest to a BRG1 binding site indicating that these genes may be putative direct targets of BRG1 (Supplemental Fig. S6D). BRG1 activated direct targets enriched for biological processes like cardiac muscle and tissue development and contraction, and circulatory system processes. Target genes that are repressed by BRG1 are enriched for biological processes including embryonic hind limb and forelimb morphogenesis and skeletal system development (Fig. 6D). We selected active enhancer regions based on their enrichment of acetylation marks on lysine 27 residue of histone H3 in CPs (Wamstad et al. 2012). Several BRG1 regulated genes (23.6% of BRG1 activated and 16.8% of BRG1 repressed genes) intersected with genes nearest to enhancer marks (Supplemental Fig S6E). BRG1 activated genes near to the enhancer marks enriched for processes such as muscle contraction, acto-myosin structure organization and cardiac tissue morphogenesis whereas repressed genes near enhancer marks were enriched for negative regulation of endothelial cell proliferation, biosynthetic processes and regionalization (Fig. 6E). These results show that putative directs of BRG1 are involved in facilitation of cardiac gene expression programs while preventing other developmental programs including embryonic limb development.
Cardiac transcription factors (TFs) often recruit and use chromatin remodeling machinery to reorganize chromatin and regulate transcription (Lickert et al. 2004; Takeuchi and Bruneau 2009; Takeuchi et al. 2011). To evaluate a potential BRG1-TF collaboration during cardiac differentiation we searched for TF motifs in BRG1 binding regions associated with both BRG1 activated and repressed genes using Homer motif enrichment analysis (Heinz et al. 2010). BRG1 activated regions were found to be associated predominantly with GATA motifs whereas BRG1 repressed regions were associated with T-box motifs among others (Fig. 6F,6G). Consistent with these observations BRG1 binding sites correlated well with GATA4 and TBX5 binding sites in cardiac precursors (Fig. 6H,6I) (Luna-Zurita et al. 2016). These results indicate that BRG1 or other BRG1 associated factors likely interact with cardiac TFs to regulate gene expression during cardiac lineage commitment, as predicted from gain of function experiments (Lickert et al. 2004; Takeuchi and Bruneau 2009).
BAF170 facilitates temporal BRG1 dissociation from the genome
We explored the possibility that BAF60c or BAF170 help direct the genomic localization of BRG1. In CPs, BRG1 binds to a set of 4791 sites (Supplemental Fig. S6A) and are largely unaffected in absence of BAF170 (Fig. 7A). BRG1 binding at these site is severely reduced in cardiomyocytes in WT and cells lacking BAF60c, presumably due to the reduced predominance of a BRG1-containing complex. However, in the absence of BAF170, BRG1 binding at a large subset of these sites is retained in cardiomyocytes (Fig. 7B). This retention is reflected by stabilization of a BRG1 containing complex in CMs lacking BAF170 (Fig. 5A), perhaps due to posttranscriptional stabilization and/or retention of available BRG1 in a complex. Further, analysis revealed that genes near these sites are involved in chromosome organization, epigenetic modification of chromatin and post-transcriptional regulation of gene expression (Supplemental Fig S7A, S7B). On the other hand at a small subset of sites (21%, 1019 of 4791), BRG1 was evicted independent of BAF170 in CM (Supplemental Fig. S7A). Genes near these sites are involved in regulation of embryonic growth and morphogenesis and RNA pol II mediated transcription (Supplemental Fig. S7C). These surprising results suggest that recruitment of BRG1-inclusive complexes is regulated by cardiac specific subunits, and that concomitant dynamic expulsion of the complex is also highly regulated by similar processes.
DISCUSSION
BRG1 containing complexes respond to and modulate lineage decisions and differentiation in cardiac development by incorporating specific subunits at discrete stages. These specialized BAF complexes regulate distinct gene expression programs to drive cardiogenesis while simultaneously repressing alternate non-cardiac developmental programs. Our results suggest that BAF complexes utilize multiple interdependent mechanisms including switching subunits or isoforms, regulating subunit stoichiometry, altering BRG1 recruitment, and expulsion strategies, to modulate dynamics of cardiac differentiation. This unanticipated complexity of chromatin complex regulation emerges as a critical determinant of developmental trajectories.
Subunit composition switching of chromatin remodeling complexes have been reported during neural development (Lessard et al. 2007; Yoo et al. 2009; Nitarska et al. 2016). and skeletal myogenesis (Saccone et al. 2014; Goljanek-Whysall et al. 2014). During myogenic differentiation, BAF60c expression is facilitated over the alternative isoforms BAF60a and BAF60b, which are post-transcriptionally repressed by myogenic microRNAs. The subunit switches we describe for BAF complexes during cardiomyocyte differentiation involve different subunits than in neural cells but if such a microRNA mediated repression mechanism selects BAF60c is not known. Importantly, we show that BAF complexes at different stages of cardiac differentiation form specific complexes to activate or repress specific transcriptional programs. In addition, we determine that the presence of specific BAF subunits greatly dictates the overall composition, and therefore function, of the complex.
The cardiac enriched subunits BAF60c and BAF170 are primarily associated with BRG1 in CPs, but also associate with BRM in CMs, indicating a broader diversity of BAF complexes in later stages of differentiation. Paradoxically, the continued roles played by BAF60c and BAF170 are intriguing considering the reduced role of BRG1 in later stages of cardiac differentiation, and the apparent absence of a developmental or postnatal cardiac phenotype in BRM null mice (Reyes et al. 1998). It has been suggested that BAF60c may function independently from the BAF complex (Forcales et al. 2012; Wang et al. 2013), although there is no evidence for this in cardiac cells. It is also possible that the previously reported BRM deletion allele is not a complete null (Thompson et al. 2015). We are investigating both of these avenues.
BRG1 binding to genomic sites in CP and putative BRG1 direct targets have critical roles in promotion of cardiac gene expression programs while repressing non-cardiac fate. BRG1 binding is greatly reduced in CM indicating diminished BRG1 function. Our finding that, BAF170 largely helps BRG1 dissociation from genome indicates a potential mechanism of BAF170 mediated gene regulation during differentiation. How BAF170 regulate genomic expulsion of BRG1 is currently not understood, but association of BAF170 with BRM containing complexes in CM might evict BRG1 from BAF complexes as both these subunits are mutually exclusive. Further, concomitant increase in BAF170 and decrease of BAF155 association with BRG1 containing complexes during cardiac differentiation might make BRG1 prone to post-transcriptional modification and proteasome mediated degradation, as BAF155 is known to stabilize BRG1 containing complexes (Sohn et al. 2007). Consistently, we observe BRG1 complex stabilization in absence of BAF170 in CMs (Fig. 5A).
In conclusion, our study identifies morphing combinations of very specific BAF subunits that form and change subunit compositions during cardiac differentiation, and drive stage-specific cardiac gene expression programs. Genetic dissection of individual subunit contribution clearly unveils a profound and specific function of individual BAF complex subunits. How BAF chromatin remodelers with different subunit composition provide specificity to gene expression programs will be the focus of future studies.
Author Contribution
Project design and direction: B.G.B. ES cell engineering, in vitro differentiation, complex isolation, gene expression analysis, data analysis: S.K.H. BAF60c FLAG allele generation: X.S, under direction of B.G.B and J.R. Mass spectrometry and analysis: J.R.J, E.V., under direction of N.J.K. Manuscript writing: S.K.H and B.G.B with contribution from all authors.
Materials and Methods
Mouse ES cell culture
Mouse embryonic stem cells (ESCs) were cultured in feeder free condition in DMEM+ glutamax (ThermoFisher, 10566)media supplemented with 15% fetal bovine serum (Hyclone, SH30071.03), 103 U/ml leukemia inhibitory factor (Millipore, ESG1107), 2mM glutamine, 13mM 2-mercaptoethanol (Life Tech. 21985-023), 0.1mM non-essential amino acids (Life Tech, 11140-050), 0.1mM sodium pyruvate (LifeTech, 11360), and 1x Pencillin/Sreptomycin (LifeTech, 15140-122). ESCs were cultured in gelatinized tissue culture flasks. TrypLE (ThermoFisher, 12563011) was used to dissociate cells and followed by quenching in DMEM media with 10% FBS. Cells were passages every 2 days with daily media changes.
Knockout cell line generation
BAF60c was inactivated by targeting exon 2 of Baf60c gene with two pairs of TALENs following Sanjana et. al (Sanjana et al. 2012) except a CAGGS promoter replacing the CMV promoter (Giorgetti et al. 2014). The TALEN’s targets are as follows: GCCCCCTAAGCCCTCTCCAGAGAACATCCAAGCTA GAATGACTTGGTCGCTGCTAC and CCCGCCCCTCTCCAAGACCCTGGGTTGGTA ACCCTGCGCTGAGCGATGAGTGGGAG
BAF170 was targeted using CRISPR/Cas9 with sgRNA targeting Exon 2 of Baf170 gene following the protocol in Congo et al.(Cong et al. 2013). sgRNA were cloned to Bbs1 digested pX330 vector by annealing the following primers 5’ caccg CGCACCGCTTACTAAACTGC 3’ and 5’ aaac GCAGTTTAGTAAGCGGTGCG c 3’ (lowercase to create Bbs1 digestion site). Targeting vector was constructed by cloning 458 and 459bp of Baf170 DNA upstream and downstream of midpoint of sgRNA target site into Kpn1-Xho1 and BamH1-Not1 sites of pFPF (derivative of Addgene Plasmid #22687, Neomycin is replaced with Puromycin cassette).
Plasmids were prepared using maxiprep kit (Macherey-nagel), ethanol precipitated and DNA suspended in TE.
For transfection, 2.5ug of each of TALEN Baf60c plasmids were used. For BAF170 knockout, 2.5ug sgRNA plasmid and 20ug of BAF170 targeting constructs were used. Plasmids were transfected by microporation using the Neon system (Thermofisher) using a 100uL tip with 1 million cells at 1400V, 10ms and 3 pulses. Cells were grown for 48 hours and then serially diluted and grown either in absence (for BAF60c) or presence of puromycin (for BAF170) for 8-10days. Single colonies were picked manually and transferred to 96 well tissue culture plates coated with 0.1% gelatin and passaged 2-3 times and replica plated. Cells were then PCR genotyped from one replicate of plates to confirm loss of exon2 (Baf60c) or insertion of puromycin cassette (Baf170). Further, amplified DNA around the regions was sequenced to confirm mutation.
Cardiomyocyte differentiation
Cardiomyocyte differentiation were performed as described previously (Kattman et al. 2011; Wamstad et al. 2012). Briefly, ES cell (70-80% confluence) were harvested and resuspendend (7.5×105 cells/ml) in EB media (75% v/v Iscove’s modification of DMEM (Corning, 15-016-CV), 25% v/v Ham’s F-15 (Corning, 15-080-CV), 0.5% each of N2 supplement (ThermoFisher, 17502048), B27 without vitamin A (ThermoFisher, 12587010), 0.5% BSA (Sigma Aldrich, A9576-50ml), 1x glutamine (Gibco, 25030-081), 1x Penicillin/streptomycin (ThermoFisher, 15140-122) and supplemented just before use with ascorbic acid (1mM) and monothioglycerol (400uM)). The cells were grown into embroid bodies (EBs) for two days in ultra-low attachment suspension culture dishes (Corning, 430591). Embroid bodies (EBs) were dissociated using TrpLE and resuspendend in EB media containing 5ng/ml of hVEGF (R&D Systems, 293-VE-050), 8ng/ml Activin-A (R&D Systems, 338-AC-050) and 0.1 to 1.6ng/ml Bmp4 (R&D Systems, 314-BP-010, usually a range of Bmp4 concentrations optimized for each cell line) in ultralow suspension culture dishes for 40hrs. EBs were dissociated and cultured in in monolayers in StemProP34 media (ThermoFisher, 10639011) containing 0.2mM L-Glutamine, 0.1 mg/ml ascorbic acid, 5ng/ml VEGF, 10ng/ml FGF basic (233-FB-025) and 25ng/ml FGF 10 (R&D Systems 345-FG-025) for 6 days. Beating cardiac myocytes were observed day 6 onwards and efficiency of differentiation was measured by antibody staining of cardiac troponin T followed by immunofluorescence and fluorescence activated cell sorting (FACS). BRG1 was deleted as described earlier (Alexander et al. 2015). Briefly, Brg1fl/fl;Actin-CreER ESCs (Ho et al. 2009a; 2011) were differentiated and cultures were treated with 200 nM 4-hydroxytamoxifen (4-OHT) diluted from a 5 mg/ml stock solution in tetrahydrofuran (THF) or with only THF for control, 48 hrs. before the desired deletion.
Nuclear extract preparation, western blot and anti-FLAG immunoaffinity purification
Nuclear extract were prepared using protocol describe before (Abmayr et al. 2006). Briefly, cells were harvested, washed twice in ice cold PBS, washed once with hypotonic buffer (10 mM HEPES, pH 7.9, 1.5 mM MgCl2, 10 mM NaCl with 0.2 mM PMSF and 0.5 mM DTT added immediately before use) and incubated in 5 cell volume of hypotonic buffer for 15 min in ice. Cells were homogenized 15 times using a dounce, lysis checked with trypan blue staining and nuclei collected after centrifuging 3000g for 5 mins. Nuclei were suspended in 1 volume of nuclear lysis buffer (20 mM HEPES, pH 7.9, 25% glycerol, 1.5 mM MgCl2, 0.3 M NaCl, 0.02%, NP40, 0.2 mM EDTA, with 0.2 mM PMSF and 0.5 mM DTT added immediately before use) and incubated at 4°C in rotating condition. Supernatant were collected as nuclear extract after centrifuging 25,000g for 30min at 4°C. Western blotting was performed using standard techniques. Briefly, nuclear extract was diluted, boiled and resolved by electrophoresis. Protein was transferred to a PVDF membrane followed by blocking with Odessey blocking buffer (Licor, 927-40000) for 1 hr followed by desired primary antibody diluted in blocking buffer with 0.1% Tween 20, overnight at 4°C. Membranes were washed in TBST and stained with compatible secondary antibody in blocking buffer containing 0.1% Tween 20 and 0.02% SDS. After antibody staining, membranes were washed and imaged in a Odessy Fc imaging system (LI-COR). Primary antibodies used were anti-Brg1 (Abcam, ab110641, 1:1000), anti-FLAG (Sigma, F1804, 1:1000) and anti-BAF155 (Bethyl, A301-021A, 1:1000), anti-BAF170 (Bethyl, 1:1000, A301-39A), anti-BAF60c (Cell Signaling Technologies, 62265, 1:0000) and anti-H3 (SantaCruz,sc-86541:1000). Secondary antibodies used are Donkey anti-rabbit IRDye 800cw (Licor, 926-32213, 1:10,000), Donkey anti-mouse IRDye 800cw (Licor, 925-32212, 1:10,000), Donkey anti-goat IRDye 680cw (Licor, 925-68074-1:10,000).
For immunoaffinity purification of BAF complexes, cells were grown, differentiated and harvested at specific stages of cardiac differentiation. Nuclear extract was made from 108 cells as described above except the final clearing was at 100,000g for 1hr. Nuclear extract were incubated with 50ul of anti-FLAG M2 agarose gel (Sigma, A2220) at 4°C rotating overnight. Bead bound protein complexes were washed 10 times with 1ml of nuclear extract buffer supplemented with protease inhibitor (Sigma Aldrich, 5892970001). Proteins was competitively eluted by incubating with 0.1mg/ml FLAG peptides (ELIM Biopharmaceutical) four times in 50ul of nuclear extraction buffer containing 100mM NaCl. The protein complexes were resolved in 10% SDS PAGE and stained with SyproRuby protein gel stain (ThermoFisher, S12000).
Masspectrometry
Protein complexes were digested with trypsin for LC-MS/MS analysis. Samples were denatured and reduced in 2M urea, 10 mM NH4HCO3, 2 mM DTT for 30 min at 60° C, then alkylated with 2 mM iodoacetamide for 45 min at room temperature. Trypsin (Promega) was added at a 1:100 enzyme:substrate ratio and digested overnight at 37° C. Following digestion, samples were concentrated using C18 ZipTips (Millipore) according to the manufacturer’s specifications. Desalted samples were evaporated to dryness and resuspended in 0.1% formic acid for mass spectrometry analysis.
Digested peptide mixtures were analyzed by LC-MS/MS on a Thermo Scientific LTQ Orbitrap Elite mass spectrometry system equipped with a Proxeon Easy nLC 1000 ultra high-pressure liquid chromatography and autosampler system. Samples were injected onto a C18 column (25 cm x 75 um I.D. packed with ReproSil Pur C18 AQ 1.9um particles) in 0.1% formic acid and then separated with a one-hour gradient from 5% to 30% ACN in 0.1% formic acid at a flow rate of 300 nl/min. The mass spectrometer collected data in a data-dependent fashion, collecting one full scan in the Orbitrap at 120,000 resolution followed by 20 collision-induced dissociation MS/MS scans in the dual linear ion trap for the 20 most intense peaks from the full scan. Dynamic exclusion was enabled for 30 seconds with a repeat count of 1. Charge state screening was employed to reject analysis of singly charged species or species for which a charge could not be assigned.
Raw mass spectrometry data were analyzed using the MaxQuant software package (version 1.3.0.5)(Cox and Mann 2008). Data were matched to the SwissProt mouse protein sequencess (downloaded from UniProt on 7/19/2016). MaxQuant was configured to generate and search against a reverse sequence database for false discovery rate (FDR) calculations. Variable modifications were allowed for methionine oxidation and protein N-terminus acetylation. A fixed modification was indicated for cysteine carbamidomethylation. Full trypsin specificity was required. The first search was performed with a mass accuracy of +/−20 parts per million and the main search was performed with a mass accuracy of +/−6 parts per million. A maximum of 5 modifications were allowed per peptide. A maximum of 2 missed cleavages were allowed. The maximum charge allowed was 7+. Individual peptide mass tolerances were allowed. For MS/MS matching, a mass tolerance of 0.5 Da was allowed and the top 6 peaks per 100 Da were analyzed. MS/MS matching was allowed for higher charge states, water and ammonia loss events. The data were filtered to obtain a peptide, protein, and site-level FDR of 0.01. The minimum peptide length was 7 amino acids. Results were matched between runs with a time window of 2 minutes for technical duplicates. All precursor (MS1) intensities of valid peptide matches were quantified by the Maxquant LFQ algorithm using the match between runs option to minimize missing values.
For statistical analysis, the quantitative change of peptides that were uniquely assigned to protein isoforms across all IPs were compared with the MSstats R package (v. 2.3.4)(Choi et al. 2014). Briefly, all peak intensities were Log2-transformed and their distributions were median-centered across all runs using the scale option. All remaining missing intensity values were imputed by setting their value to minimal intensity value per run, as an estimate for the MS Limit Of Quantitation. The normalized dataset was then analyzed by fitting a mixed effects model per protein using the model without interaction terms, unequal feature variance and restricted scope of technical and biological replication. The average change (Log2-Fold-Change) of the model-based abundance estimate was computed by comparing replicates of each differentiation stage against the ES undifferentiated pool. Proteins with a greater than four fold change (Log2 Fold Change > 2) and test p-value < 0.05 were determined as significantly altered during differentiation.
Nucleosome reconstitution, repositioning and ATPase assay
Recombinant Xenopus laevis histone octamers were constituted on a 601-nucleosome positioning DNA (Lowary and Widom 1998) as described (Hota et al. 2013). Briefly, 601 DNA containing mononucleosomes were assembled at 37C with 8-10ug of recombinant octamers, 100-200 fmol of 32P-labelled DNA, 10ug of unlabeled homogenous DNA and 1.8M NaCl in a 10ul reaction. Samples were serially diluted to a final concentration of ~300mM NaCl. Nucleosome assemblies were analyzed on a 4% native polyacrylamide gel (acrylamide:bisacrylamide; 35.36:1) in 0.5x Tris-Borate-EDTA buffer at 4°C. For nucleosome repositioning assay, 5.8nM homogenous nucleosomes with 30 and 60bp overhangs (30-nuc-60) were incubated with 1nM BAF enzymes and excess ATP (1mM) for increasing amount of time (0, 2, 6, 18 and 54min) at 30°C. Reaction was quenched in presence of 16.6mM γ-S-ATP. Repositioned nucleosome was separated from starting nucleosomes in a 5% native polyacrylamide gel in 0.2x Tris-Borate EDTA buffer at 4°C. The gel was dried and exposed and imaged acquired in a Typhoon Phosphorimager (GE) and quantified using Optiquant (GE). ATPase assays were performed by adding 800nM (γ-32P) ATP to bound nucleosome-BAF complexes and incubating for increasing times (0 to 128 mins). ATP and free phosphate were separated by thin layer chromatography on polyetheleneimine-cellulose (Thomas Scientific, 2738E80).
RNA-Sequencing
ES cell-derived cardiac precursors or cardiomyocytes (n=3 for each genotype) were harvested for RNA extraction using miRNeasy micro kit with on-column DNase I digestion (Qiagen). Total RNA was isolated from 1.5-2×106 cells using TRIzol reagent. RNA-seq libraries were prepared with the Ovation RNA-seq system v2 kit (NuGEN). With this method, mRNA is reverse transcribed to synthesize first-strand cDNA with random hexamers and a poly-T chimeric primer. Following second strand synthesis double-stranded DNA is then amplified using single primer isothermal amplification (SPIA). Libraries from the SPIA amplified cDNA were made using the Ultralow DR library kit (NuGEN). RNA-seq libraries were analyzed by Bioanalyzer and quantified by QPCR (KAPA). High-throughput sequencing was done using a HiSeq 2500 instrument (Illumina). RNA reads were aligned with TopHat2 (Kim et al. 2013), counts per gene calculated using featureCounts (Liao et al. 2014)and edgeR (Robinson et al. 2010) was used for the analysis of differential expression. Differential expressed RNAs were analyzed at cardiac precursor (D5.3) and cardiac myocyte (D10) stages of differentiation. RNAs that showed significant differential expression between wild type and nulls (p-value < 0.05) and also changed more than 1.5-fold in nulls over wild type at a specific stage of differentiation were selected for analysis, avoiding duplicate and redundant entries. K-means clustering and pheatmap functions in R were used to cluster FKM values of selected RNAs and generate heat maps respectively.
GO enrichment analysis were performed in GO RILLA (http://cbl-gorilla.cs.technion.ac.il/) and visualized in REVIGO (http://revigo.irb.hr/).
ChIP-Seq
Chromatin immunoprecipitation of BRG1 in WT, ΔAF60c and ΔBAF170 were performed according to O’Geen et al. (O’Geen et al. 2011) with modifications, in biological duplicate for CP and one replicate for CM but with samples combined from two biological replicates. Briefly, cells at CP and CM of cardiac differentiation were harvested and crosslinked with 2mM disuccinimidyl gulatarate (DSG) for 45 mins, washed twice with PBS, followed by crosslinking with 1% formaldehyde for 15 mins. Cells were quenched with 0.125M glycine for 5 mins, washed in PBS thrice and stored at −80°C. Frozen pellet of dual crosslinked cells (5×107) were thawed, washed in PBS and lysed in PBS containing 1% triton X for 3 mins in ice followed by 25 cycles of dounce with a tight pestle. Nuclei were collected and washed once with MNase digestion buffer (50mM Tris.cl, pH7.6, 1mM CaCl2, 0.2% Triton X, 5mM Sodium Butyrate with 1x protease inhibitor and 0.5mM PMSF added just before use) at 350g for 5mins. Nuclei were digested with 400U of micrococcal nuclease (ThermoFisher, 88216) for exactly 5 mins at 37°C. MNase digestion was stopped by adding stop buffer (10mM EDT, 10mM EGTA and 0.1% SDS). Chromatin were sonicated for short time (2 cycles, 30s ON, 1 min OFF at output 4 in a VirSonic sonicator), centrifuged at 10,000g for 10mins and supernatant stored at −80°C. Extent of MNase digestion was measured by agarose gel electrophoresis. Chromatin (40ug) was diluted to 5 fold in IP dilution buffer (50mM Tris.Cl, pH7.4, 150mM NaCl, 1% NP-40, 0.25% Sodium deoxycholate, 1mM EDTA) without EDTA and pre-cleared with 25ul of M-280 goat anti-rabbit IgG dyna beads (ThermoFisher, 11203D) for 2 hrs followed by addition with 1 ug of anti-Brg1 antibody (Abcam, 110641) for 12-16hrs. 5% of samples were set aside as input before antibody addition. Antibody bound BRG1-DNA complexes were Immunoprecipitated using 25ul of M-280 goat anti-rabbit IgG dyna beads for 2hrs, washed twice with IP dilution buffer, five times with IP wash buffer (100 mM Tris.Cl, pH 9.0, 500mM Lithium chloride, 1% NP-40 and 1% Sodium deoxycholate) and thrice with IP wash buffer containing 150mM NaCl. DNA was eluted with 200ul of elution buffer (10mM Tris.Cl, pH 7.5, 1mM EDTA and 1%SDS), NaCl was added to both ChIP and input samples (0.52M) and crosslinking reversed for 6 – 12 hrs at 65°C. Samples were digested with RNase A, and proteinase K in presence of 3ug of glycogen, and DNA purified with Ampure beads (Beckman Coulter, A63881) and eluted in 50ul of TE. To prepare libraries for ChIP Sequenceing, DNA were end repaired, A-tailed followed by adapter ligation (Illumina TrueSeq) and 14 cycles of PCR amplification. PCR amplified DNA was size selected (200-400bp) followed by gel extraction (Qiagen) and eluted in 20ul TE. The concentration and size of eluted libraries was measured (Qubit and Bioanalyzer) before sequencing in a NEBNextSeq sequencer.
Reads (single end 75bp) were trimmed using fastq-mcf and aligned to mouse genome mm9 assembly using Bowtie (Langmead et al. 2009). Minimum mapping quality score was set to 30. Statistically enriched bins with a P-value threshold set to 1×106 were determine with input DNA serving as the background model (Thomas et al. 2017). UCSC genome browser was used to generate the browser tracks. Galaxy (https://usegalaxy.org/) was used to pool multiple replicates to generate browser tracks and tornado plots. GREAT (http://bejerano.stanford.edu/great/public/html/) was used to generate gene lists near BRG1 binding sites.
Acknowledgements
We thank A. Williams and S. Thomas (Gladstone Bioinformatics Core) for data analysis, L. Ta and J. McGuire (Gladstone Genomics Core) for RNAseq library preparation, and G. Howard for editing. RNAseq (258R1 and 420R) and ChIPSeq (451R) data are publicly available at https://b2b.hci.utah.edu/gnomex/ and GEO (accession number). This work was supported by grants from the NIH/NHLBI (R01HL085860, P01HL089707, Bench to Bassinet Program UM1HL098179), the California Institutes of Regenerative Medicine (RN2-00903), and the Lawrence J. and Florence A. DeGeorge Charitable Trust/American Heart Association Established Investigator Award (B.G.B), and postdoctoral fellowships from the American Heart Association (13POST17290043) and Tobacco Related Disease Research Program (22FT-0079) to S.K.H. L.A.P. was supported by NHLBI grant R24HL123879, and NHGRI grants R01HG003988, and UM1HG009421, and research was conducted at the E.O. Lawrence Berkeley National Laboratory was performed under Department of Energy Contract DE-AC02-05CH11231, University of California. This work was also supported by an NIH/NCRR grant (C06 RR018928) to the J. David Gladstone Institutes and by The Younger Family Fund (B.G.B.).