Discovery of a hidden transient state in all bromodomain families

Abstract

Bromodomains (BDs) are small protein modules that interact with acetylated marks in histones. These posttranslational modifications are pivotal to regulate gene expression, making BDs promising targets to treat several diseases. While the general structure of BDs is well known, their dynamical features and their interplay with other macromolecules are poorly understood, hampering the rational design of potent and selective inhibitors. Here, we combine extensive molecular dynamics simulations, Markov state modeling, and available structural data to reveal a transiently formed state that is conserved across all BD families. It involves the breaking of two backbone hydrogen bonds that anchor the ZA-loop with the α_A helix, opening a cryptic pocket that partially occludes the one associated to histone binding. By analyzing more than 1,900 experimental structures, we unveil just two adopting the hidden state, explaining why it has been previously unnoticed and providing direct structural evidence for its existence. Our results suggest that this state is an allosteric regulatory switch for BDs, potentially related to a recently unveiled BD-DNA-binding mode.

Keywords: Markov models; allosteric effects; bromodomains; cryptic pockets; minor conformational states.

Fig. 1.

Bromodomains have two conserved backbone h-bonds that anchor the ZA-loop with the α_A helix. (A) Structure of CECR2 (PDB 3NXB) highlighting the two first helices, α_Z and α_A, and the long ZA-loop, detailing the ZA channel and the two conserved backbone h-bonds that anchor the loop with the α_A helix. (B) Sequence alignment of the ZA-loop region for the BDs analyzed in this work, including all families. Note the presence of a hairpin insertion in family VIII and the well-conserved Asp480 (yellow star). The orange star highlights lysine residues commented below.

Fig. 2.

A hidden conformational state in BDs. (A) Ensemble of structures of the “closed” crystal-like state (cyan) and the “open” state (yellow) for BRD4(1). A close view of the opening region is represented below, highlighting Asp106. (B) Distribution of the Asp106 solvent accessible surface area and closest distance between Gln84-Asp106 side chains for the two metastable states. (C) MSM reweighted free energy profile along the first independent component (TIC0, arbitrary units) and boxplot of a bootstrapped distribution of the closing and opening mean first passage times.

Fig. 3.

All bromodomain families share a hidden state that opens a cryptic pocket beneath the ZA-loop. Ensemble of structures of the “closed” crystal-like state (cyan) and the “open” state (yellow) for each BD family. MDpocket frequency maps are represented by blue isosurfaces at 0.25 (light) and 0.50 (intense) isovalues, highlighting different binding sites. Percentages refer to medians of MSM populations obtained from a bootstrapped distribution (SI Appendix, Table S2). Populations for TRIM28 have not been determined due to its complex conformational landscape and neither have the ones of ZMYND11 since all structures are in the open state.

Fig. 4.

Distribution of h-bonds in experimental structures reveal two BDs in the open state. (A) Projection of all BD structures from the Pfam database (black dots) on a MSM reweighted free energy landscape of BRD4(1) comprising the two conserved backbone h-bonds. Axes are given in a logarithmic scale and dashed lines indicate a distance of 0.35 nm as an upper bound for h-bond formation. The stars highlight the two crystal structures that are in the open state. (B) The structure of ZMYND11 (pale green, PDB 4N4G) is compared with the open state predicted for BRD4(1) (yellow). Pro199 is highlighted next to the conserved Asp. (C) The structure of PB1(6) (pale green, PDB 3IU6) is compared with the open state predicted for SMARCA2 (yellow). Thr789, in place of the conserved Asp, is highlighted together with an internal h-bond that is formed in the short helix of the ZA-loop.

Fig. 5.

Chemical shift predictions of SMARCA2 along the conformational change and experimental chemical shift changes of a BD-DNA binding mode. (A) CamShift ¹H_N predictions along the conformational change of SMARCA2 and experimental shifts upon its binding to DNA (5′-CTCAATTGGT-3′), obtained from Morrison et al. (32). The numbering of residues is that of PDB 5DKC, with a zoom in on residues 1,400 to 1,440. (B) Distribution of Leu1412 ¹H_N shift for the closed and open states (cyan and yellow, respectively). The cyan distribution can be split in two Gaussian distributions (shown in gray) that correspond to closed states with the Leu1412-Asp1430 h-bond formed or broken (panels 1 to 2). The yellow distribution has two clear modes that correspond to open states in which Leu1412 amide is placed or not in between the two backbone carbonyls (panels 3 to 4). The black arrow indicates the CamShift prediction for PDB 5DKC, and the red arrow indicates the mean prediction for PDB 2DAT. (C) Hypothesis of a closed-to-open population shift as a consequence of BD interactions with DNA.