Nonchromatin regulatory functions of the histone variant H2A.B in SWI/SNF genomic deposition

Abstract

The replacement of canonical histones with their variant forms enables the dynamic and context-dependent regulation of the mammalian genome. Histone variants also play key roles in various pathological processes including malignancies. Among these, the aberrant expression of the testis-specific histone variant H2A.B contributes to the pathogenesis of Hodgkin lymphoma. The multifunctionality of histone variants is regulated by their posttranslational modifications (PTMs). However, the PTMs of H2A.B and their functional implications are unknown. Here, we demonstrate that the Amino terminus of H2A.B serves as a central hub for a diverse range of gene regulatory protein-protein interactions, orchestrated by phosphorylation and arginine methylation. This includes a mechanism whereby non-chromatin-bound H2A.B associates with SWI/SNF, which limits its access to the genome. Last, we identify phosphorylated H2A.B as a previously uncharacterized marker of active RNA polymerase II transcription start sites. These findings elucidate a central role for H2A.B in genome regulation and highlight the importance of its PTMs in modulating its multifunctional roles.

Fig. 1.. H2A.B undergoes arginine methylation by PRMT1 in HL cells.

(A) An N-terminal sequence alignment of H2A.B orthologs across various eutherian mammals. The N terminus of human H2A is also shown. Conserved residues R7, S9, and R15 are highlighted in red, while residues identical to human H2A.B are shaded in gray. (B) H2A.B PTMs identified in L1236 and L428 cells. These include arginine monomethylation (me), arginine asymmetric dimethylation (me2a), and serine phosphorylation (ph). (C) Western blot analyzing H2A.B immunoprecipitates using anti-sDMA and anti-aDMA antibodies. (D) Western blot analyzing R7me2a, R15me2a, and S9ph modifications in H2A.B immunoprecipitates using custom-made modification-specific antibodies. (E) Western blot analyzing PRMT1 protein levels in nontargeting control (NTC) and PRMT1 shRNA–transduced L1236 and L428 cells. β-Actin served as a loading control. The levels of PRMT1, normalized to β-actin, were quantified and displayed below the blot. (F) Western blot analyzing R7me2a levels in H2A.B immunoprecipitates from NTC or PRMT1 shRNA–transduced L1236 and L428 cells. The levels of R7me2a were normalized to total H2A.B and displayed below the blot. (G) Proximity-dependent BioID cellular assay confirming that the interaction between PRMT1 and H2A.B. L1236 cells (right panel) and L428 cells (left panel) was transduced with either an empty vector (vector-only), a BirA*-H2A.B vector, or an H2A.B-BirA* vector. Following biotin labeling, streptavidin affinity purification of cell lysates was analyzed by Western blot using an anti-PRMT1 antibody. The positive control was loaded as 1% (v/v) input. The Western blot signal for H2A.B runs below 15 kDa (the molecular weight of H2A.B is 12.7 kDa). The asterisk (*) designates non-H2A.B antibody binding.

Fig. 2.. Interaction between hSWI/SNF and H2A.B independent of CHR.

(A) A Western blot detecting SMARCC1, SMARCE1, and TRA2B binding to unmodified or R7me2a-modified H2A.B peptides. The peptide pull-downs were performed using nuclear extracts from L428 cells. (B) BioID assays demonstrating that the proximity of SMARCC1 to H2A.B. L428 cells (upper panel) and L1236 cells (lower panel) were transduced either with an empty vector (vector-only), a BirA*-H2A.B vector, or an H2A.B-BirA* vector. Streptavidin affinity purification of the cell lysate was analyzed by Western blot using an anti-SMARCC1 antibody. (C) Co-IP analysis of H2A.B with SMARCC1, SMARCD1, BRG1, and BRM. (D) Co-IP analysis of SMARCD1 with H2A.B, ARID1A, and ARID2. (E) Co-IP analysis of SMARCC1 with H2A.B, ARID1A, and ARID2. (F) Co-IP analysis of BRG1 (upper panel) and BRM (lower panel) with H2A.B. (G) Co-IP analysis of SMARCC1 with H2A.B, ARID1A, and ARID2. (F) Co-IP analysis of ARID1A (left panel) or ARID2 (right panel) with H2A.B. All IP experiments were performed using CN fractions. The 1% (v/v) input sample was included as a positive control.

Fig. 3.. hSWI/SNF preferentially binds to the N-terminal tail of H2A.B.

(A) H2A, H2A.Z, H2A.B, H2A.BR7me2a, H2A.BR7me, or H2A.BR15me biotin-tagged N-terminal peptides were incubated with purified recombinant CBAF (BRG1 and BRM) or PBAF (BRG1 and BRM). Peptide-bound hSWI/SNF complexes were subjected to a Western blot analysis, and relative peptide binding affinities were determined using a combination of different common and subunit-specific CBAF and PBAF antibodies (see Materials and Methods). As an additional control, biotinylated recombinant human nucleosomes were also examined as an hSWI/SNF binding substrate. Error bars represent the standard error of the mean. (B) Western blots analyzing the levels of CN H2A.B coimmunoprecipitated BRG1 and SMARCC1 from NTC or PRMT1 shRNA–transduced L428 cells.

Fig. 4.. H2A.BR7me2 is not present at the TSS of RNA Pol II genes.

Genomic annotations are categorized as follows: upstream-10K, regions encompassing 2 to 10 kb downstream of the TSS; promoter-TSS, regions encompassing 2 kb upstream to 500 bp (base pairs) downstream of the TSS; TTS, regions encompassing 100 bp upstream to 1 kb downstream of the transcription termination site; UTR, untranslated regions. (A) Genomic annotation of pan H2A.B CUT&RUN peaks in L428 cells. (B) Genomic annotation of H2A.BR7me2a CUT&RUN peaks in L428 cells. (C) Genomic annotation of pan H2A.B CUT&RUN peaks in L1236 cells. (D) Genomic annotation of H2A.BR7me2a CUT&RUN peaks in L1236 cells. (E) Venn diagram displaying the overlap between pan H2A.B and H2A.BR7me2a CUT&RUN peaks in promoter-TSS regions. (F) Venn diagram displaying the overlap between pan H2A.B and H2A.BR7me2a CUT&RUN peaks in all other genomic regions. (G) H2A.B mean CUT&RUN coverage aligned between −1 and +1 kb from the TSS ranked according to the level of gene expression in L428 cells. (H) H2A.BR7me2a mean CUT&RUN coverage aligned between −1 and +1 kb from the TSS ranked according to the level of gene expression in L428 cells. (I) H2A.B mean CUT&RUN coverage aligned between −1 and +1 kb from the TSS ranked according to the level of gene expression in L1236 cells. (J) H2A.BR7me2a mean CUT&RUN coverage aligned between −1 and +1 kb from the TSS ranked according to the level of gene expression in L1236 cells.

Fig. 5.. H2A.B antagonizes the genomic deposition of the hSWI/SNF complex.

(A) Volcano plot of the differential occupancy of BRG1 at H2A.B peaks in L1236 cells transduced with an H2A.B-targeting shRNA (shH2A.B) lentiviral vector normalized to an shNTC. Peaks with altered BRG1 levels (FDR < 0.2) are highlighted (red and blue peaks have an increase or a decrease in BRG1 occupancy, respectively). (B) UpSet plots showing individual and shared BRG1 peaks at promoter-TSS regions in shH2A.B- and shNTC-transduced L1236 cells. (C) UpSet plots showing individual and shared BRG1 peaks at non–promoter-TSS genomic regions in shH2A.B- and shNTC-transduced L1236 cells (FDR < 0.1). (D) Genomic annotation of the H2A.B peaks with decreased ATAC-seq accessibility following H2A.B knockdown (FDR < 0.1). (E) Genomic annotation of the H2A.B peaks with increased ATAC-seq accessibility following H2A.B knockdown (FDR < 0.1).

Fig. 6.. Phosphorylation of H2A.B is correlated with high levels of transcription.

(A) Genomic annotation of H2A.BS9ph CUT&RUN peaks in L428 cells. (B) Genomic annotation of H2A.BS9ph CUT&RUN peaks in L1236 cells. (C) Venn diagram displaying the overlap between pan H2A.B and H2A.BS9ph CUT&RUN peaks in promoter-TSS regions. (D) Venn diagram displaying the overlap between pan H2A.B and H2A.BS9ph CUT&RUN peaks in non–promoter-TSS genomic regions. (E) H2A.BS9ph mean CUT&RUN coverage aligned between −1 and +1 kb from the TSS ranked according to the level of gene expression in L428 cells. (F) H2A.BS9ph mean CUT&RUN coverage aligned between −1 and +1 kb from the TSS ranked according to the level of gene expression in L1236 cells.

Fig. 7.. The N terminus of H2A.B regulates its incorporation into CHR.

(A) Western blot analysis of the subcellular location of H2A.BS9ph. H2A.B was immunoprecipitated from L428 CN, LB, and CHR fractions. The Western blot was then probed with anti-H2A.BS9ph antibodies. (B) N-terminal sequences of H2A.B and the different serine-to-alanine and arginine-to-lysine amino acid residue substitutions and the complete replacement of the N terminus of H2A.B, exogenously expressed in HL cell lines. (C) Western blot analyses of endogenous H2A.B (Endo) and exogenously expressed wild-type H2A.B construct (Exo) in different subcellular fractions of L428 cells (upper panel) and L1236 cells (lower panel). (D) Western blot analyses of endogenous H2A.B (Endo) and exogenously expressed mutant S9A-H2A.B construct (Exo) in different subcellular fractions of L428 cells (upper panel) and L1236 cells (lower panel). (E) Western blot analyses of endogenous H2A.B (Endo) and exogenously expressed mutant S9AS10A-H2A.B construct (Exo) in different subcellular fractions of L428 cells (upper panel) and L1236 cells (lower panel). (F) Western blot analyses of endogenous H2A.B (Endo) and exogenously expressed mutant R-K-H2A.B construct (Exo) in different subcellular fractions of L428 cells (upper panel) and L1236 cells (lower panel). (G) Western blot analyses of endogenous H2A.B (Endo) and exogenously expressed mutant nH2A-H2A.B construct (Exo) in different subcellular fractions of L428 cells (upper panel) and L1236 cells (lower panel). (H) Schematic illustrating the effects of N-terminal PTMs and N-terminal mutations on H2A.B CHR incorporation.