Selective inhibition of BET bromodomains

Abstract

Epigenetic proteins are intently pursued targets in ligand discovery. So far, successful efforts have been limited to chromatin modifying enzymes, or so-called epigenetic 'writers' and 'erasers'. Potent inhibitors of histone binding modules have not yet been described. Here we report a cell-permeable small molecule (JQ1) that binds competitively to acetyl-lysine recognition motifs, or bromodomains. High potency and specificity towards a subset of human bromodomains is explained by co-crystal structures with bromodomain and extra-terminal (BET) family member BRD4, revealing excellent shape complementarity with the acetyl-lysine binding cavity. Recurrent translocation of BRD4 is observed in a genetically-defined, incurable subtype of human squamous carcinoma. Competitive binding by JQ1 displaces the BRD4 fusion oncoprotein from chromatin, prompting squamous differentiation and specific antiproliferative effects in BRD4-dependent cell lines and patient-derived xenograft models. These data establish proof-of-concept for targeting protein-protein interactions of epigenetic 'readers', and provide a versatile chemical scaffold for the development of chemical probes more broadly throughout the bromodomain family.

Figure 1. Structure and selectivity of JQ1

a, Structure of the two JQ1 stereo-isomers. The stereocentre at C6 is indicated by an asterix (*). b, Assessment of inhibitor selectivity using Differential Scanning Fluorimetry (DSF). Shown are averaged temperature shifts (ΔT_m^obs) in °C upon binding of 10 μM (+)-JQ1. The temperature shifts are represented by spheres as indicated in the inset. Screened proteins are labelled and proteins not included in the selectivity panel are shown in grey. (−)-JQ1 did not reveal any significant temperature shifts to any of the screened bromodomains. c, Isothermal Titration Calorimetry (ITC). Raw injection heats are shown for a blank titration of (BRD4(1)) into buffer (A), and reverse titrations using the inactive isomer (−)-JQ1 (B) and the active isomer (+)-JQ1 (C). The inset shows normalized binding enthalpies corrected for the heat of dilution as a function of binding site saturation (symbols as indicated in the inset). Solid lines represent a nonlinear least squares fit using a single-site binding model. d, Thermal shifts (ΔT_m^obs) show good correlation to dissociation constants (K_d) determined by ITC for the BET bromodomains. The dotted red line represents a least squares fit with an R of 96 %. e, Competitive displacement of a histone peptide from human bromodomains is exhibited by JQ1 using a bead-based proximity assay. Alpha screen titrations monitoring the displacement of a tetra-acetylated histone H4 peptide by JQ1 isomers using the bromodomains BRD4(1), BRD4(2) or of an acetylated H3 peptide using CREBBP.

Figure 2. Characterization of BET complexes with (+)-JQ1

a, Superimposition of the mouse BRD4(1)/H3K14ac peptide complex with the human BRD4(1)/(+)-JQ1 complex structure. The hydrogen bond formed to the conserved asparagine (N140) in the peptide complex is shown as yellow dots. b, 2F_o-F_c map of (+)-JQ1 in complex with BRD4(1) contoured at 2σ. c, Electrostatic surface of BRD4(1) in complex with (+)-JQ1. The ligand is shown as a CPK model demonstrating the excellent shape complimentarity with the protein acetylated lysine receptor site. d, Ribbon diagram of the complex of human BRD4(1) with (+)-JQ1 in CPK representation. The main secondary structural elements and the conserved active site asparagine side chain (N140) are labelled.

Figure 3. Binding site comparison between N- and C- terminal bromodomains in complex with (+)-JQ1

a, The acetyl-lysine binding pocket of BRD4(1) is shown as a semitransparent surface with contact residues labelled and depicted in stick representation. Carbon atoms in (+)-JQ1 are coloured yellow to distinguish them from protein residues. b, The acetyl-lysine binding pocket of BRD2(2) is shown in identical representation and orientation as described in (a). c, Protein sequence alignment of the human BET subfamily highlighting conserved (red) and similar (yellow) residues. Major bromodomain structural elements are shown. The side-chain contacts with (+)-JQ1 are annotated with a black star. Contacts which differ between the N- and C-terminal BET bromodomains (red star) are highlighted. The family conserved asparagine is indicated by a blue star. d, Models of (+)-JQ1 (in yellow) and (−)-JQ1 (in green) docked into the BRD4(1) binding site. The steric clashes of the (−)-JQ1 stereo-isomer with Leu92 and Leu94 are highlighted in red. e, MD simulation demonstrating the flexibility of the ZA- and BC- loops of the BRD4(1) apo-structure. Shown are the backbone of BRD4(1) during a 20 ns simulation as snapshots separated by 1 ns intervals. The different structures are distinguished by colours changing from blue to green as indicated in the inset. f, MD simulation of the BRD4(1)/(+)-JQ1 complex depicted in 1 ns snapshots as described in (e).

Figure 4. JQ1 binds BRD4 competitively with chromatin resulting in differentiation and growth arrest of NMC cells

a, Fluorescence recovery after photobleaching (FRAP) of GFP-BRD4 demonstrates enhanced recovery in the presence of JQ1. Nuclei are false-colored in proportion to fluorescence intensity. White circles indicate target regions of photobleaching. b–c, JQ1 accelerates fluorescence recovery in FRAP experiments performed with transfected (b) GFP-BRD4 and (c) GFP-BRD4-NUT. d, Quantitative comparison of time to half-maximal fluorescence recovery for FRAP studies (b–c, Supplementary Fig. 3a). Data represent the mean ± s.d. (n = 5), and are annotated with p-values as obtained from a two-tailed t-test. e, Differentiation of NMC cells by JQ1 (500 nM) is prompt and characterized by a marked increase in cytokeratin expression (AE1/AE3; 10x, scale bar is 50 μm). f, Comparative gene expression studies of (+)-JQ1 (red; 250 nM, 48 h) versus (−)-JQ1 (gray; 250 nM, 48 h) and vehicle (black) confirm squamous differentiation. Data represent the mean ± s.d. (n = 3), and are annotated with p-values as obtained from a two-tailed t-test. g, Growth effects of BRD4 inhibition on BRD4-NUT dependent cell lines. Cells were incubated with (+)-JQ1 (red circles) or (−)-JQ1 (black circles) and monitored for proliferation after 72 hours. (+)-JQ1 uniquely attenuates proliferation by NMC cell lines. Data is presented as mean ± s.d. (n = 3). Curve fit was calculated by logistic regression. h, Flow cytometry for DNA content in NMC 797 cells. (+)-JQ1 (250 nM, 48 h) induces a G1 arrest compared to (−)-JQ1 (250 nM) and vehicle control. i, Flow cytometric analysis of NMC 797 squamous carcinoma cells treated with vehicle, JQ1 or staurosporine (STA), as indicated. PI, propidium iodide. AV, annexin-V.

Figure 5. JQ1 promotes differentiation, tumor regression and improved survival in murine models of NMC

a, PET imaging of murine NMC 797 xenografts. FDG uptake in xenograft tumors is reduced by 50 mg kg ⁻¹ JQ1 treatment compared to vehicle control. b, Tumor volume is reduced in mice with established disease (NMC 797 xenografts) treated with 50 mg kg ⁻¹ daily JQ1 compared to vehicle control. A significant response to therapy is observed by two-tailed t-test at 14 days (p = 0.039). Data represent the mean ± s.d. (n = 7). c, Histopathological analysis of NMC 797 tumors excised from animals treated with JQ1 reveals induction of keratin expression (AE1/AE3, 40x) and impaired proliferation (Ki67, 40x), as compared to vehicle-treated animals (scale bar is 20 μm). d, Viability of patient-derived NMC 11060 xenografts was confirmed by PET imaging. e, Therapeutic response of primary 11060 NMC xenografts to (+)-JQ1 (50 mg kg ⁻¹ daily for four days) was demonstrated by PET imaging. f, Histopathological analysis of primary NMC 11060 tumors excised from animals treated with (+)-JQ1 reveals induction of keratin expression (AE1/AE3, 20x; scale bar is 20 μm), compared to vehicle-treated animals. Quantitative analysis of keratin induction was performed using image masking (f, right panel) and pixel positivity analysis (g). A significant response to therapy is observed by two-tailed t-test (p = 0.0001). Data represent the mean ± s.d. of three independent wide microscopic fields. Comparative images of stained excised tumors and quantitative masks are provided in Supplementary Figure 14. h–k, (+)-JQ1 (50 mg kg ⁻¹ daily for 18 days) produces a decrease in tumor volume (h, j) and promotes improved survival (i, k) in patient-derived 11060 (h, i) and Per403 (j, k) NMC xenograft models (n=10 in all groups). A significant response to therapy is observed for tumor volume by two-tailed t-test (p < 0.0001) and for overall survival by a log-rank test (p < 0.0001).