Structure of the nucleosome-bound human BCL7A

Abstract

Proteins of the BCL7 family (BCL7A, BCL7B, and BCL7C) are among the most recently identified subunits of the mammalian SWI/SNF chromatin remodeler complex and are absent from the unicellular version of this complex. Their function in the complex is unknown, and very limited structural information is available, despite the fact that they are mutated in several cancer types, most notably blood malignancies and hence medically relevant. Here, using cryo-electron microscopy in combination with biophysical and biochemical approaches, we show that BCL7A forms a stable, high-affinity complex with the nucleosome core particle (NCP) through binding of BCL7A with the acidic patch of the nucleosome via an arginine anchor motif. This interaction is impaired by BCL7A mutations found in cancer. Further, we determined that BCL7A contributes to the remodeling activity of the mSWI/SNF complex and we examined its function at the genomic level. Our findings reveal how BCL7 proteins interact with the NCP and help rationalize the impact of cancer-associated mutations. By providing structural information on the positioning of BCL7 on the NCP, our results broaden the understanding of the mechanism by which SWI/SNF recognizes the chromatin fiber.

Graphical Abstract

Figure 1.

Sequence and disorder of BCL7 proteins. (A) Multiple sequence alignment of the three human BCL7 proteins performed using Muscle in JalView. The first 51 amino acids, highly homologous between the three proteins, are highlighted in a green box. They correspond to the BCL_N domain reported in the Pfam database of protein domains (PF04714). (B) CD analysis of the three BCL7 proteins. (C) Analysis of free BCL7A by NMR. ¹H-¹⁵N SOFAST-HMQC spectrum of the full length uniformly ¹⁵N labeled BCL7A. The assigned signals are labeled (labels in the central region are shown in the top-right corner inset). (D) Cα chemical shift index for all assigned residues.

Figure 2.

Binding of BCL7 proteins to the nucleosome. (A) EMSA gels using Cy5-labeled NCP and the indicated BCL7 proteins. Protein amounts were varied as indicated above each lane. (B) Quantification of the calculated amount of bound NCP and curve fitting using the Hill equation (red curve). Dissociation constants and Hill coefficients ‘n’ are indicated, along with their standard deviation from replicates of the experiment. The colors of the dots represent the various replicate titrations, with n ≥ 3. Black dots and brackets indicate the mean and standard error. (C) BLI response curves for the indicated proteins, with fitting of a heterogenous binding model curve in red. Dots of different color indicate the two biological replicates of the experiment. Black dots and brackets indicate the mean and standard deviation of the replicates. (D) Table of the calculated dissociation constants. Numbers rounded to the closest ten. N.D.: not determined.

Figure 3.

Cryo-EM structure of BCL7A in a nucleosome bound state. (A) Molecular surface representation of the nucleosome core particle colored according to the electrostatic potential (blue, positive [+5 kT]; white, neutral; red, negative [−5 kT]) in complex with BCL7A; BCL7A N-terminal region (aa 2–30) is shown in ribbon diagram representation and colored cyan. The panel below shows a zoomed view of BCL7A interaction with the nucleosome: amino acids 2–12 of the N-terminal α helix BCL7A are shown in stick representation where carbon atoms are colored cyan, oxygen atoms are red, nitrogen atoms are blue. The exact amino acid sequence of BCL7A N-terminus (aa 2–12) is indicated and the three Arg within the arginine anchor motif are colored blue. (B) Ribbon diagram representation of BCL7A in complex with the acidic patch of the nucleosome. BCL7A is colored cyan, DNA is colored orange and the histones H2A, H2B, H3, and H4 are colored yellow, red, light blue, and green, respectively. (C) Zoomed-in views of the interactions of BCL7A R4 (left panel), R7 (center panel), and R11 (right panel) with H2A and H2B. Residues of interest are shown in stick representation. BLC7A and histones are color coded as above. (D) Zoomed view of BAF47 C-terminal helix (PDB ID: 6LTJ) interaction with the acidic patch. BAF47 is colored purple and histones are colored as mentioned above. Residues directly involved in the interactions are in stick representation. Panels (A), (B), and (C) were rendered with PyMOL.

Figure 4.

Contact points between BCL7A and histones H2A and H2B. Residue numbers are indicated in boxes colored light blue for BCL7A, yellow for H2A, and red for H2B. The interactions are marked by dotted lines.

Figure 5.

Detailed view of the BCL7A–NCP cryo-EM density map at the acidic patch filled with BCL7A coordinates displayed at two different map thresholds. (A) Atomic models derived from the density maps show histones H2A (yellow) and H2B (red) in ribbon representation (left panel) and in stick representation (right panel) at a threshold of σ = 0 077. (B) Same as in A but the map threshold is set at σ = 0,0223. BCL7A density (displayed in cyan) is visible and positioned on the top of H2A and H2B and showed in ribbon representation (left panel) and stick representation (right panel). The figure was prepared using ChimeraX software in order to adjust the threshold and color the different histone and BCL7A densities as needed.

Figure 6.

Analysis of BCL7A–NCP complex by NMR. (A) Overlay of the ¹H-¹⁵N spectra of uniformly ¹⁵N-labeled BCL7A alone (in blue, as in Fig. 1E) and after addition of 0.33 equivalent of unlabeled NCP (in orange). (B) Zoom of the ¹H-¹⁵N spectra of uniformly ¹⁵N-labeled BCL7A alone (in blue) and after addition of 0.2, 0.33, and 1 molar equivalent of unlabeled NCP (in yellow, orange, and red, respectively). (C) Mapping BCL7A–NCP interaction region with the ratio of intensities (I/I0), where I and I0 are the intensities of isolated signals of the ¹H-¹⁵N spectra before and after addition of NCP at different ratio (with the same color code as in B).

Figure 7.

SAXS analysis of the BCL7A–NCP complex. (A) Resulting SAXS curves with Guinier plots and their respective gyration radii (Rg) in inset. (B) Pair distance distribution functions with respective maximum dimensions Dmax. (C) Normalized Kratky plots: BCL7A shows the characteristic plot of a flexible and mainly unstructured protein, while NCP and NCP:BCL7A complex show the plot of particles with compact core with a peak at √3. (D) Summary of biophysical parameters derived from the SAXS analysis.

Figure 8.

Nucleosome binding by BCL7A is impaired by mutations at the arginine anchor. (A) NCP binding experiments performed by EMSA for WT and various mutants of BCL7A. Representative replicates are shown. N ≥ 4. (B) Fitting of the experimental data to the Hill equation (red curves). The blue curves indicate the curve fit for the WT proteins (same as shown in Fig. 2). Dots of different color represent biological replicates of the experiment.

Figure 9.

BCL7 proteins contribute in vitro to remodeling by mSWI/SNF. (A) Diagram of the NCP remodeling substrate. A 601 DNA sequence is flanked by the indicated lengths of exposed DNA and labeled at both extremities with Cy5 dye. Prior to remodeling, an MfeI restriction site is occluded by the histone octamer. Sliding or eviction of the octamer leads to exposure of the restriction site and allows cleavage by the enzyme. (B) Results of restriction enzyme accessibility assay performed with two replicate preparations of the substrate NCP and mSWI/SNF complexes (parental cell line or TKO for the BCL7 genes), each in technical triplicates. Identical amounts of complex are incubated for time periods ranging from 5 to 40 min. The two bands produced by MfeI digestion are indicated. (C) Graphs of the quantified data. A slight horizontal jitter was added to the individual points to prevent overlap. Error brackets indicate the standard error and are centered at the mean value. P-values are by repeated measures two-way ANOVA, for the factor “genotype”.

Figure 10.

Identification of mSWI/SNF target loci dependent on BCL7 proteins for DNA accessibility. (A) Heatmap of ATAC-seq signal at regions bound by BRG1 and where BRM014 treatment leads to a significant decrease in DNA accessibility. Rows are clustered based on whether ATAC-seq signal also significantly changes in the BCL7 TKO cell line. (B) Boxplot showing log₂ fold change in ATAC signal for the indicated contrasts, separated by clusters. (C) Example of ATAC-seq signal at loci from cluster “DA down in TKO”. (D) Example of other loci not affected in any of the conditions. (E) Distribution of position and distances between DA regions in each of the three clusters and genes. (F) A selection of the ChIP-seq datasets from K-562 cells showing a significant enrichment in one of the two differential DA clusters when compared to the other. Rows labeled in green are for datasets that showed a positive log₂ enrichment ratio, indicating that they are enriched in the regions where accessibility decreased in BCL7 TKO cells when compared to the regions that increased in accessibility in BCL7 TKO cells. Rows labeled in red show enrichment in the opposite direction. The fractions in each bar indicate the proportion of the regions in the cluster that overlap with the indicated ChIP-seq dataset peaks. All TFs shown have adjusted P-values lower than .001. (G) A selection of TF binding motifs enriched in the DA down or DA up clusters.