Exploring Selective Inhibition of the First Bromodomain of the Human Bromodomain and Extra-terminal Domain (BET) Proteins

Article abstract of DOI:10.1021/acs.jmedchem.5b01708

A midthroughput screening follow-up program targeting the first bromodomain of the human BRD4 protein, BRD4(BD1), identified an acetylated-mimic xanthine derivative inhibitor. This compound binds with an affinity in the low micromolar range yet exerts suitable unexpected selectivity in vitro against the other members of the bromodomain and extra-terminal domain (BET) family. A structure-based program pinpointed a role of the ZA loop, paving the way for the development of potent and selective BET-BRDi probes.

Full text of DOI:10.1021/acs.jmedchem.5b01708

Article
pubs.acs.org/jmc
Exploring Selective Inhibition of the First Bromodomain of the
Human Bromodomain and Extra-terminal Domain (BET) Proteins
Brigitt Raux,† Yuliia Voitovich,†,‡ Carine Derviaux,† Adrien Lugari,†,○ Etienne Rebuffet,† Sabine Milhas,†,§
Stephane Priet,§ Thomas Roux,∥ Eric Trinquet,∥ Jean-Claude Guillemot,§ Stefan Knapp,⊥,¶,∇
́
Jean-Michel Brunel,† Alexey Yu. Fedorov,‡ Yves Collette,† Philippe Roche,† Stephane Betzi,†
́
Sebastien Combes,*,† and Xavier Morelli*,†
†Centre National de la Recherche Scientifique (CNRS), Centre de Recherche en Cancer
́
́
ologie de Marseille (CRCM), UMR 7258;
́
UM105, 13273 Marseille, France
INSERM U1068; Institut Paoli-Calmettes; Aix-Marseille Universite,
‡Department of Organic Chemistry, Lobachevsky State University of Nizhni Novgorod, Gagarina av. 23, Nizhni Novgorod 603950,
Russia
§Screening Platform AD2P, CNRS, AFMB UMR 7257, Aix-Marseille Universite,
́
13288 Marseille, France
∥Cisbio Bioassays, R&D, Parc Marcel Boiteux, BP 84175, 30200 Codolet, France
⊥Target Discovery Institute, University of Oxford, NDM Research Building, Roosevelt Drive, Oxford OX3 7FZ, U.K.
¶Structural Genomics Consortium, University of Oxford, Old Road Campus Research Building, Roosevelt Drive, Oxford OX3 7DQ,
U.K.
∇Goethe-University, Institute for Pharmaceutical Chemistry and Buchmann Institute for Life Science, Campus Riedberg, Max-von
Laue Str. 9, 60438 Frankfurt am Main, Germany
S
* Supporting Information
ABSTRACT: A midthroughput screening follow-up program
targeting the first bromodomain of the human BRD4 protein,
BRD4(BD1), identified an acetylated-mimic xanthine deriva-
tive inhibitor. This compound binds with an affinity in the low
micromolar range yet exerts suitable unexpected selectivity in
vitro against the other members of the bromodomain and
extra-terminal domain (BET) family. A structure-based
program pinpointed a role of the ZA loop, paving the way
for the development of potent and selective BET-BRDi probes.
nucleosomes,6 and it has been suggested that BRD4 binds
INTRODUCTION
■
acetylated histones using primarily its first bromodomain
Bromodomains (BRD) are protein interaction modules that
preferentially bind ε-N-acetylated lysine residues through
structurally well-defined pockets. BRDs are found in eight
protein families, which include a total of 46 nuclear or
(BD1).7,8 The BD2 domain also recognizes and interacts
with the acetylated region of cyclin T1, which forms a complex
with the positive transcription elongation factor b and is crucial
for the sustained presence of Pol II in active genes and for
transcription initiation and elongation,9 thereby regulating the
expression of cell proliferation supporting genes, including c-
Myc and its target genes.10
BET proteins are often deregulated in disease, their
transcription-regulating activity being altered and thus affecting
numerous growth-promoting genes and cytokines. BET
proteins are known to be deregulated in cancer,11 and the
recent disclosure of pan-BET inhibitors that attenuate BRD
function has allowed the validation of these drug targets,
shedding light on their roles in disease. Interestingly, BRD4
cytoplasmic proteins in human with diverse structures and
functions, including chromatin-modifying enzymes, helicases,
chromatin remodelers, transcriptional coactivators and media-
tors, and the bromodomain and extra-terminal domain (BET)
family of proteins.1−3 BET proteins (BRD2, BRD3, BRD4, and
the testis-specific BRDT) have a conserved modular
architecture including two N-terminal tandem BRDs (BD1
and BD2). The BETs play a central role in chromatin biology
by acting as tissue-specific recruitment platforms that tether
complexes to acetylated histones and chromatin, facilitating the
assembly of the transcriptional machinery and controlling gene
expression in inflammation, viral infection, and cancer biology.
For example, BRD2 is specifically recruited to acetylated
histones H3 and H4, and this interaction is linked to active
transcription and mitosis.4,5 BRD2 and BRD3 are required for
permissive RNA polymerase II transcription through acetylated
Special Issue: Epigenetics
Received: November 2, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.jmedchem.5b01708
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 1. Identification of a selective BRD4(BD1) acetylated-mimic xanthine inhibitor 1. (A) Structure of 3-butyl-8-(6-butyl-5,7-dimethyl-
[1,2,4]triazolo[1,5-a]pyrimidin-2-ylsulfanylmethyl)-7-ethylxanthine, 1. (B) Isothermal titration calorimetry (ITC) thermogram (left) and nonlinear
least-squares fit model of the integrated data (right) at 35 °C (300 μM BRD4(BD1) vs 35 μM 1) as a function of the molar ratio of ligand (cell) to
the protein (syringe). (C) Homogeneous time resolved fluorescence (HTRF) selectivity assay of 1 on all BET members and ATAD2 as a non-BET
control after excitation at 337 nm, energy transfer at 620 nm, and fluorescence emission at 665 nm. Fluorescence data are normalized and plotted as
■
●
▲
;
a function of the ligand concentration. Curves are colored by bromodomains (BRD4(BD1), pink ; BRD3(BD1), purple ; BRD2(BD1), blue
◆
□
◇
○
△
BRDT(BD1), cyan ; BRD4(BD2), red ; BRD3(BD2), gray ; BRD2(BD2), orange ; ATAD2, gray ). (D) Effect of 1 on Jurkat cell viability
as a function of the compound concentration after 72 h incubation at 37 °C. (E) C-myc pro-oncogene downregulation profile in the presence of
different concentrations of 1 at 0.5% DMSO after 24 h incubation (50 μg of loaded protein).
Figure 2. Molecular mode of action of 1 in the N-acetylated binding pocket of BRD4(BD1). (A) Three dimensional crystallographic structure of 1
(orange stick and ball representation) complexed with BRD4(BD1) obtained at 2.2 Å resolution (as a colored cartoon labeled by secondary
structures) showing the conserved N140 residue (pink), WPF shelf (green), I146 gatekeeper residue (blue), and four highly conserved water
molecules (red spheres) located at the bottom of the cavity. (B) Detailed view of 1/BRD4(BD1) interaction. BRD4(BD1) is shown as ball and stick
representation. The four conserved water molecules at the bottom of the cavity (red spheres marked with a ¤) are involved in an extensive hydrogen
bond network with BRD4(BD1) (red dashed lines). Direct hydrogen bond interactions between 1 and BRD4(BD1) are shown as black dashed lines,
while indirect water mediated interactions are shown as blue dashed lines. The L92 is displayed in transparency for clarity. (C) Detailed view of the
van der Waals interactions (gray dashed lines) of 1 stacked between Q85 and L92 of BRD4(BD1).
occupies “superenhancers” and its inhibition leads to significant
reduction of the transcript levels of only a few hundred genes,12
in a cell-, disease-, and context-specific manner. Preclinical
targeting of BETs has had initial successes, particularly in
oncology. For example, in a phase I acute leukemia study, (6S)-
4-(4-chlorophenyl)-N-(4-hydroxyphenyl)-2,3,9-trimethyl-6H-
thieno[3,2-f ][1,2,4]triazolo[4,3-a][1,4]diazepine-6-acetamide
(OTX015),13−16 a thienodiazepine, induced remissions, includ-
ing complete remission in two patients with refractory disease,
and 2-[4-(2-hydroxyethoxy)-3,5-dimethylphenyl]-5,7-dime-
thoxy-4(3H)-quinazolinone (RVX-208),8,17 a quinazolone
derivative of resveratrol that binds preferentially to the BD2
domain of BRD2 and BRD3, has already been tested in
hundreds of patients in phase II clinical trials for the treatment
of atherosclerosis, providing proof-of-concept that selective
inhibition within the BET family is feasible. However, pan-BET
inhibition might remain an issue regarding the impact on
numerous transcriptional pathways and the individual tissue
specific functions of BET members. The selective targeting of
individual BET and the discrimination between BD1 and BD2
present one opportunity to achieve more selective transcrip-
tional effects. Here, we follow-up on a previous midthroughput
screen (MTS) using the 2P2I3D chemical library, a structurally
diverse “protein−protein interaction inhibition (2P2I)-ori-
B
DOI: 10.1021/acs.jmedchem.5b01708
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Scheme 1. Synthesis of Substituted Xanthines
a
Reagents and conditions: (i) EtI, DMA, 60 °C, 12 h; (ii) HClaq 3 N, reflux, 1 h, then rt NH3aq 30%; (iii) NaNO2, H2O−AcOH, 50 °C, 1 h; (iv)
R2X, K2CO3, DMF, rt, 2 days.
ented” collection of compounds,18−23 by focusing on a selective
BRD4(BD1) acetylated-mimic xanthine inhibitor. Among the
17 hits that were identified following this screen, one xanthine
derivative was found to present a low micromolar IC50
measured by homogeneous time-resolved fluorescence
(HTRF) and an unforeseen selectivity profile among the
BET family members. A structure-based investigation program
around xanthine scaffold containing compounds was the
starting point of this observed selectivity with a key role
identified for the bromodomain ZA loop.
system is locked from one side by van der Waals contacts with
L92 and from the other side with Q85; the pyrimidine ring sits
tightly on this glutamine, which is bent by 90° to adopt this
complementary orientation (Figure 2C). This BD1-conserved
amino acid (except in BRDT(BD1)) is replaced by larger
residues such as arginine or lysine in the BD2 subfamily
(Supporting Information Figure S3). This difference could be
responsible for the observed selectivity profile of 1 toward
BRD4-, BRD3-, and BRD2(BD1) vs all the other BET proteins
tested. Crystallographic data collection statistics are summar-
ized in Supporting Information Table S1.
In order to gain better understanding of this selectivity
profile, we undertook a structure-based program aiming to
define the structure/affinity and structure/selectivity relation-
ships by decomposing this model compound. First, the
xanthine 2 and the corresponding mono- and unsymmetrically
dialkylated derivatives 3−5 were prepared from guanosine,
taking advantage of an oriented N7-alkylation as previously
described in literature (Scheme 1).29
RESULTS AND DISCUSSION
■
A MTS of the 2P2I3D chemical library against BRD4(BD1)
identified 17 hits with affinities within the low micromolar
range. We setup isothermal titration calorimetry (ITC) as an
orthogonal assay and were able to validate direct binding for 7
out of the 17 compounds tested, with ligand efficiency (LE)
values ranging from 0.20 up to 0.23. Among them, 1 (Figure
1A), a xanthine derivative, presented a Kd of 1.4 μM, driven by
a large favorable entropy term (−TΔS of −7.1 kcal/mol) as
demonstrated by ITC (Figure 1B). This biophysical result
suggested conformational changes and water rearrangement
upon binding of the compound. Moreover, an HTRF
experiment against seven out of the eight BD1 and BD2 BET
domains (BRDT(BD2) was not available as single bromodo-
main from the company Cisbio) pinpointed a 10-fold lower
IC50 for 1 toward BRD4(BD1) (IC50 of 5 μM) compared with
its relative’s BD1 (IC50 > 50 μM), and no detectable inhibition
of the BD2 counterparts (Figure 1C and Table 3). In cell-based
assays, 1 reduced Jurkat T cell viability in a dose-dependent
manner with an EC50 of 27 μM. Also, 1 down-regulated c-Myc,
a pro-oncogene contributing to the pathogenesis of numerous
human cancers, in the same range of concentration (Figure
1D,E). We next solved the crystal structure of 1 in complex
with BRD4(BD1). The cocrystal revealed the globular domain
organization for BRD4(BD1)24 and a well-defined electron
density for 1 with the expected binding mode of an acetyl-lysine
mimetic (Figure 2A,B and see also Supporting Information
Figure S1),25 forming the canonical hydrogen bond with the
conserved asparagine N140 and a water mediated hydrogen
bond with Y97 also linking the inhibitor to the conserved water
network at the bottom of the binding pocket. Surprisingly, the
triazolopyrimidinyl moiety stacked against the ZA channel,
occupying the space at the rim of the acetyl-lysine binding site,
a binding mode that has also recently been reported for
benzimidazolone derivative inhibitors targeting BAZ2B bromo-
domain (Supporting Information Figure S2).26−28 This triazolo
moiety establishes a hydrogen bond interaction with the main
chain (NH amido group) of D88, a residue conserved
throughout the BET family. This interaction orientates the
triazolopyrimidinyl fragment in the ZA channel, while its ring
These compounds were further analyzed by HTRF, ITC, and
X-ray crystallography, when available (Table 1, Figure 3).
Table 1. Effect of Various Substituted Xanthine on
BRD4(BD1) Activity
substituent
R2
BRD4(BD1)
a
b
c
compd
R1
IC50 (μM)
KD (μM)
LE
2
3
4
5
H
H
d
d
d
Et
Et
Et
H
d
Bu
13.0 0.3
2.6 0.1
0.40
0.37
4-ClBn
1.8
a
Drug concentration that inhibits protein−protein interaction by 50%.
Data are the mean standard deviation (SD) of three experiments.
b
c
Kd determined by isothermal titration calorimetry. Ligand Efficiency
(LE) defined as the ratio of the log of the IC50 to the number (N) of
d
non-hydrogen atoms of the compound LE = 1.4(-log(IC50)/N). Not
applicable.
Chemical variations of the xanthine mimicking the reference
probe (6S)-4-(4-chlorophenyl)-2,3,9-trimethyl-6H-thieno[3,2-
f ][1,2,4]triazolo[4,3-a][1,4]diazepine-6-acetic acid 1,1-dime-
thylethyl ester (JQ1, PDB accession code 3MXF),24 such as a
substitution of the butyl group (compound 4) by a
chlorobenzyl one (compound 5), allowed an optimization of
the core with an IC50 of 2.6 μM as measured by HTRF (2-fold
increased potency) and a Kd of 1.8 μM as measured by ITC.
This optimized core presented a favorable enthalpy change of
−6.3 kcal/mol (Figure 3A). More importantly, the selectivity
profile of this optimized core 5 displayed a pan-BET inhibition
profile, similarly to 4 (Figure 3B). Analysis of the cocrystal
structures confirmed a characteristic recognition of the pocket
C
DOI: 10.1021/acs.jmedchem.5b01708
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 3. Studies of disubstituted xanthine derivatives as potential selective bromodomain inhibitors. (A) ITC thermogram (top) and integrated data
(bottom) at 35 °C of BRD4(BD1) (100 μM in the syringe) vs 5 (10 μM in the cell) as a function of the molar ratio of ligand to protein. (B) HTRF
selectivity assay of 4 (top) and 5 (bottom). The curve symbols and color scheme use the same pattern as in Figure 1. (C) Three dimensional
detailed view of 5 (purple)/BRD4(BD1) molecular mode of action obtained from the crystal structure of the complex (obtained at 1.60 Å
resolution) as a ball and stick representation as used for Figure 2. (D) Superimposed detailed view of the binding of 1 (orange) and 5 (purple) in the
N-acetylated lysine pocket. The gray trace illustrates the cavity surface profile. Hydrogen bond interactions with the N140 are shown in orange and
purple for 1 and 5, respectively.
a
Scheme 2. Design and Synthesis of Triazolosulfanylmethyl-Substituted Xanthine-Based BRD4(BD1) Inhibitors 1 and 6−8
a
Reagents and conditions: (i) glycolic acid, 120 °C, 30 min; (ii) NaOHaq., H2O−EtOH, 85 °C, 2 h; (iii) EtI, K2CO3, DMF, 0 °C to rt, 12 h; (iv)
SOCl2, CH2Cl2, reflux, 15 min; (v) DIPEA, DMF, rt, 30 min.
by this core, as exemplified by the superimposition of 1 with 5
(Figure 3C,D). The core was thus not responsible (as
expected) for the observed selectivity profiles of 1.
We next questioned whether the observed selectivity profile
of 1 was due to the triazolo or to the pyrimidinyl moiety of the
core extension. To address this purpose, a series of N-3, N-7
disubstituted triazoloxanthine derivatives 1 and 6−8 were
synthesized in ten steps following the synthetic route illustrated
in Scheme 2.
The 3-alkyl-8-chloromethyl-7-ethylxanthine substrates 26
and 27 for reaction were prepared from corresponding 5,6-
diaminouracils,30 18 and 19, by condensation with glycolic acid
followed by base induced ring closure acylation. Starting from
alkylamine, the condensation with potassium cyanate afforded
alkylurea 10 and 11 converted into corresponding 6-amino-
uracil 14 and 15 by treating with cyanoacetic acid and sodium
hydroxide successively (Supporting Information Scheme S1).
Subsequent nitrosylation with sodium nitrite in aqueous acetic
acid, then reduction with sodium dithionite led to expected
diaminouracils 18 and 19 in good overall yields (42% and 20%,
respectively).
After evaluation, compound 7 revealed an IC50 of 26.4 μM in
HTRF, which represents a 10-fold decrease in potency
compared with 5 (Tables 2 and 3). Compound 7, however,
exhibited a shift of selectivity toward BET-BD1 domains
(Figure 4A and Table 3). The introduction of the triazolo
fragment thus brought a decrease of potency yet drove a
commencement of selectivity among BET bromodomains
(BD1 vs BD2).
A potential hydrogen bond between the triazolo fragment
and D88, conserved in all the BET bromodomains, could be
stabilized with certain bromodomains and not with others, due
to differences in the dynamics of the ZA loop, which has been
demonstrated to be crucial in the binding kinetics of several
BET-BRDi.24,31,32 Superimposition of BRD4 with or without 1
illustrates structural rearrangements with displacement of the
ZA loop up to 2.7 Å (Supporting Information Figure S4).
These rearrangements are not observed with the “pan inhibitor”
D
DOI: 10.1021/acs.jmedchem.5b01708
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
interaction with the side chain of Q85. A steric clash prevents
this interaction with the R54, its equivalent residue in
BRDT(BD1) numbering, or the K present in the other
BRDX(BD2) bromodomains. These findings suggest a
potential mechanism for the selectivity of BET bromodomains.
However, a more comprehensive evaluation of the role of the
first bromodomain in cancer cell proliferation and chromatin
biology will require the development of more selective probes
with improved selectivity.
Table 2. Effect of Various Triazolo-Substituted Xanthine
Derivatives on BRD4(BD1) Activity
CONCLUSION
■
a
Drug concentration that inhibits protein−protein interaction by 50%.
A permanent wavering in drug discovery is related to the
development of pan- (multitargeted) or selective (single
targeted) inhibitors. In this study, we have investigated the
key structural feature responsible for the selectivity of a
xanthine-based BRD4(BD1) inhibitor identified through MTS.
This compound represents the first low micromolar selective
inhibitor targeting BRD4(BD1) with a >10-fold ratio in binding
affinity toward any other BET bromodomain tested, yet
presenting low but dose−response down regulation of c-Myc
levels in cell-based assay. The cocrystal structure revealed an
original orientation of the triazolopyrimidinyl moiety, expand-
ing into the ZA channel, setting the basis of a structure−
selectivity relationship inside the BET family. Even though it is
conserved throughout the BET family, the different dynamic
behavior of the ZA loop could be exploited to tune the
selectivity of BET inhibitors. Optimization of the selective
BRD4(BD1) inhibitor to a probe has not been achieved yet.
However, the structural features identified in this study will
allow further optimization of more selective BRD4(BD1)
inhibitors. The generation of such probes will allow a genuine
evaluation of the biological role for each bromodomain and a
validation (or not) of the interest toward the development of
selective vs pan-BET inhibitors in the clinic.
Data are the mean standard deviation (SD) of three experiments.
b
c
Kd determined by isothermal titration calorimetry. Ligand efficiency
(LE) defined as the ratio of the IC50 to the number (N) of non-
hydrogen atoms of the compound LE = 1.4(−log IC50)/N.
5, thus validating this assumption. The role of this loop and its
associated variable dynamics would also explain the IC50
observed for 1 toward BRD4(BD1) vs BRD3(BD1) and
BRD2(BD1), this hydrogen bond being more stable with
BRD4(BD1). Analysis of the protein−ligand(s) interactions
including hydrophobic and hydrogen-bonding contacts was
carried out using LigPlot+,33 as shown in Supporting
Information Figure S5. This weak interaction prevented us
from solving a nonambiguous 3D structure of 7 in complex
with BRD4(BD1). However, crystals with 50% occupancy of 7
highlighted that this compound was still interacting with
BRD4(BD1) and that hydrogen bond between the triazolo
moiety and the main chain (NH amido group) of D88 was
present, with BRD4(BD1) (Figure 4B,C). The triazolo was
thus explaining only partly the observed selectivity profiles of 1.
An explanation for the selectivity profile of 1 was finally
confirmed with the synthesis of 6, the N-chlorobenzyl analogue
of 1. Of note, this compound was found to be less active than
the original compound 1 (IC50 values of 20 μM vs 5 μM,
respectively). However, the selectivity profile of this compound
was similar to that of 1 (Figure 4A), that is, selective toward
BRD4(BD1). Altogether, these data suggest that the N3,N7-
dialkylated xanthine core (see 4 and 5) is responsible for the
pan-BET inhibition, while condensed with triazolo moiety (see
7) it allows the preferential inhibition of BET-BD1 domains
forming a hydrogen bond with the conserved D88 residue.
Further condensation with pyrimidinyl moiety drives the
corresponding triazolopyrimidinyl xanthine derivatives (1 and
6) to be selective toward BRD4(BD1) due to van der Waals
EXPERIMENTAL SECTION
■
General Methods. All solvents were purified according to
reported procedures, and the commercially available reagents were
used as received. Separation by column chromatography was
performed using SDS Kieselgel (70−230 mesh). Petroleum refers to
the fraction with distillation range 40−65 °C. 1H and 13C NMR
spectra were recorded using a Bruker AC400 or AC250 spectrometer.
Chemical shifts, δ, are reported in ppm and coupling values, J, in hertz.
Abbreviations for peaks are br = broad, s = singlet, d = doublet, t =
triplet, q = quadruplet, quint = quintuplet, sex = sextuplet, and m =
multiplet. Reaction monitoring and purity of compounds were
recorded by using analytical Agilent Infinity high performance liquid
Table 3. Selectivity Profiles of Xanthine Derivatives toward BET Family Members
a
IC50 (μM)
c
bromodomain
1
4
5
6
7
BRD4(BD1)
BRD3(BD1)
BRD2(BD1)
BRDT(BD1)
BRD4(BD2)
BRD3(BD2)
BRD2(BD2)
5.0 0.1
62.0 7.8
59.0 2.9
13.0 0.3
4.8 0.3
18.8 0.7
7.9 0.4
8.1 0.3
5.8 0.2
6.6 0.3
d
2.6 0.1
1.8 1.1
4.7 0.1
2.3 0.2
1.9 0.1
1.8 0.2
1.6 0.2
d
20.4 3.6
26.4 0.7
31.7 1.1
30.0 0.9
d
d
d
d
d
d
d
>100
>100
d
d
d,e
d,e
d,e
d
d
b
ATAD2
>100
a
b
Drug concentration that inhibits protein−protein interaction by 50%. Data are the mean standard deviation (SD) of three experiments. ATAD2
c
is used as a non-BET family member bromodomain control. Data above 25 μM have been excluded due to precipitation and fluorescence
interference at 620 nm. Not converged. Fluorescence interference at 620 nm with BD2.
d
e
E
DOI: 10.1021/acs.jmedchem.5b01708
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 4. Structure−activity relationship of some triazolo-substituted xanthine derivatives. (A) HTRF selectivity assay results of 7 (top) and 6
(bottom). The curve symbols and color scheme use the same pattern as in Figure 1. For compound 6, data above 25 μM have been excluded due to
precipitation and fluorescence interference at 620 nm. (B) Three dimensional detailed view of 7 (cyan)/BRD4(BD1) molecular mode of action
obtained from the crystal structure of the complex (1.05 Å resolution) as a ball and stick representation as used for Figure 2. (C) Superimposed
detailed view of the binding of 1 (orange) and 7 (cyan) in the N-acetylated lysine pocket. Hydrogen bond interactions with the N140 are shown in
orange and cyan for 1 and 7, respectively.
chromatography (Column Zorbax SB-C18 1.8 μM (2.1 mm × 50
mm); mobile phase A 0.1% FA H2O, B 0.1% FA MeCN, time/%B 0/
10, 4/90, 7/90, 9/10, 10/10; flow rate 0.3 mL/min; diluent MeOH)
with DAD at 230 nM. All tested compounds yielded data consistent
with a purity of ≥95%. Low-resolution mass spectra were obtained
with Agilent SQ G6120B mass spectrometer in positive electrospray
mode. High-resolution mass spectra were performed at the Spectro-
pole Analytical Laboratory of the University Paul Cezanne, Marseille.
General Procedure for the Synthesis of 8-Triazolosulfanyl-
methylxanthine Derivatives 1 and 6−8. To a solution of
mercapto triazole derivative (0.15 mmol) in dimethylformamide (2
mL) was injected diisopropylethylamine (26 μL, 1 equiv). After 5 min
under stirring, a solution of appropriate 3-alkyl-8-chloromethyl-7-
ethyl-3,7-dihydro-purine-2,6-dione (1 equiv) in dimethylformamide (2
mL) was added dropwise. The resulting mixture was stirred at room
temperature (30 °C) for 30 min. The solvent was distilled off under
reduced pressure, and the residue was purified by column
chromatography (CH2Cl2−MeOH) to afford the corresponding 3-
alkyl-8-([1,2,4]triazolo-3-yl-sulfanylmethyl)-7-ethyl-3,7-dihydro-pu-
rine-2,6-dione.
8-(5-Amino-1H-[1,2,4]triazol-3-ylsulfanylmethyl)-3-(4-chloroben-
zyl)-7-ethyl-3,7-dihydro-purine-2,6-dione, 7. Yield 62% as yellow
solid. Rf = 0.18 (EtOAc). 1H NMR (250 MHz, MeOD) δ 7.38 (2H, d,
J = 8.5 Hz), 7.29 (2H, d, J = 8.5 Hz), 5.12 (2H, s), 4.42 (2H, s), 4.38
(2H, q, J = 7.2 Hz), and 1.42 (3H, t, J = 7.1 Hz). 13C NMR (63 MHz,
MeOD) δ 156.3, 152.7, 151.5, 150.7, 136.8, 134.5, 131.1, 129.6, 108.8,
45.9, 42.2, 31.7, 16.6. LCMS C17H17ClN8O2S Rt = 4.838 min, m/z =
432.5, purity >96%. HRMS (ESI+) for C17H18ClN8O2S (M + H)
calcd, 433.0956; found, 433.0956.
8-(5-Amino-1H-[1,2,4]triazol-3-ylsulfanylmethyl)-3-butyl-7-ethyl-
3,7-dihydro-purine-2,6-dione, 8. Yield 82% as light yellow powder. Rf
= 0.40 (EtOAc-MeOH 6:1). 1H NMR (250 MHz, DMSO-d6) δ 12.08
(1H, sbroad), 11.07 (1H, sbroad), 6.15 (2H, sbroad), 4.41 (2H, s),
4.28 (2H, q, J = 7.1 Hz), 3.85 (2H,t, J = 7.3 Hz), 1.58 (2H, quint, J =
7.3 Hz), 1.34−1.19 (5H, m) and 0.89 (3H, t, J = 7.3 Hz). 13C NMR
(100 MHz, DMSO-d6) δ 154.6, 150.8, 149.6, 149.1, 106.9, 41.6, 40.6,
29.9, 26.8, 19.6, 16.3, and 13.9. LCMS C14H20N8O2S Rt = 4.841 min,
m/z = 364.6, purity >96%. HRMS (ESI+) for C14H21N8O2S (M + H)
calcd, 365.1503; found, 365.1512.
3-Butyl-8-(6-butyl-5,7-dimethyl-[1,2,4]triazolo[1,5-a]pyrimidin-2-
ASSOCIATED CONTENT
■
ylsulfanylmethyl)-7-ethyl-3,7-dihydro-purine-2,6-dione, 1. Yield
1
S
* Supporting Information
88% as white solid. Rf = 0.29 (CH2Cl2−MeOH 10:0.5). H NMR
(250 MHz, CDCl3) δ 7.88 (1H, sbroad), 4.73 (2H, s), 4.49 (2H, q, J =
7.1 Hz), 4.04 (2H,t, J = 7.4 Hz), 2.76 (3H, s), 2.73−2.69 (2H, m),
2.69 (3H, s), 1.73 (2H, quint, J = 7.4 Hz), 1.56−1.34 (6H, m), 1.40−
1.28 (2H, m), 1.45 (3H, t, J = 7.4 Hz), 1.01 (3H, t, J = 7.1 Hz) and
0.95 (3H, t, J = 7.4 Hz). 13C NMR (100 MHz, CDCl3) δ 165.0, 164.1,
154.1, 153.8, 150.6, 150.1, 149.2, 143.5, 121.8, 42.7, 41.4, 31.9, 30.2,
28.0, 26.8, 23.8, 22.9, 20.0, 16.6, 14.0, 13.9, and 13.8. LCMS
C23H32N8O2S Rt = 6.755 min, m/z = 484.6, purity >99%. HRMS (ESI
+) for C23H33N8O2S (M + H) calcd, 485.2442; found, 485.2442.
8-(6-Butyl-5,7-dimethyl-[1,2,4]triazolo[1,5-a]pyrimidin-2-ylsulfa-
nylmethyl)-3-(4-chlorobenzyl)-7-ethyl-3,7-dihydro-purine-2,6-
dione, 6. Yield 37% as light yellow solid. Rf = 0.35 (EtOAc). 1H NMR
(400 MHz, CDCl3) δ 7.48 (2H, d, J = 8.5 Hz), 7.23 (2H, d, J = 8.5
Hz), 5.15 (2H, s), 4.71 (2H, s), 4.47 (2H, q, J = 7.2 Hz), 2.75 (3H, s),
2.73−2.70 (2H, m), 2.69 (3H, s), 1.54−1.49 (4H, m), 1.45 (3H, t, J =
7.2 Hz) and 1.00 (3H, t, J = 7.2 Hz). 13C NMR (100 MHz, CDCl3) δ
164.8, 164.0, 153.8, 153.7, 150.4, 149.5, 149.2, 143.4, 134.7, 133.8,
130.7, 128.6, 121.8, 107.6, 45.1, 41.3, 31.8, 27.9, 26.7, 23.7, 22.8, 16.4,
13.9, and 13.8. LCMS C26H29ClN8O2S Rt = 6.728 min, m/z = 552.5,
purity >99%. HRMS (ESI+) for C26H30ClN8O2S (M + H) calcd,
553.1895; found, 553.1898.
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jmed-
chem.5b01708.
Additional figures, schemes, and tables and detailed
experimental procedures for biochemistry, biology, and
chemistry including spectral data and structures of
synthesized compounds (PDF)
SMILES representations of compounds with associated
data (CSV)
Accession Codes
PDB codes for BRD4(BD1) with bound 1, 5 and 7,
respectively, are 5EGU, 5EIS, and 5EI4.
AUTHOR INFORMATION
■
Corresponding Authors
*S.C. E-mail: sebastien.combes@univ-amu.fr. Phone: (+33)
491-835-551.
F
DOI: 10.1021/acs.jmedchem.5b01708
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(8) Picaud, S.; Wells, C.; Felletar, I.; Brotherton, D.; Martin, S.;
Savitsky, P.; Diez-Dacal, B.; Philpott, M.; Bountra, C.; Lingard, H.;
*X.M. E-mail: xavier.morelli@inserm.fr. Phone: (+33)486-977-
́
331. Mailing address: Centre de Recherche en Cancerologie de
Fedorov, O.; Muller, S.; Brennan, P. E.; Knapp, S.; Filippakopoulos, P.
̈
Marseille (CRCM); INSERM U1068; CNRS UMR7258; Aix-
Marseille Universite UM105 & Institut Paoli Calmettes, 27
RVX-208, an inhibitor of BET transcriptional regulators with
selectivity for the second bromodomain. Proc. Natl. Acad. Sci. U. S.
A. 2013, 110, 19754−19759.
́
̈
Boulevard Lei Roure CS30059, 13273 Marseille Cedex 9,
France.
(9) Schroder, S.; Cho, S.; Zeng, L.; Zhang, Q.; Kaehlcke, K.; Mak, L.;
̈
Present Address
Lau, J.; Bisgrove, D.; Schnolzer, M.; Verdin, E.; Zhou, M.-M.; Ott, M.
̈
○A.L.: Bioaster; 40 Avenue Tony Garnier, 69007 Lyon, France.
Two-pronged binding with bromodomain-containing protein 4
liberates positive transcription elongation factor b from inactive
ribonucleoprotein complexes. J. Biol. Chem. 2012, 287, 1090−1099.
(10) Rahl, P. B.; Lin, C. Y.; Seila, A. C.; Flynn, R. A.; McCuine, S.;
Burge, C. B.; Sharp, P. A.; Young, R. A. c-Myc regulates transcriptional
pause release. Cell 2010, 141, 432−445.
Author Contributions
B.R. and Y.V. contributed equally to this work as first author.
S.C. and X.M. contributed equally to this work as senior author.
Notes
The authors declare no competing financial interest.
(11) Belkina, A. C.; Denis, G. V. BET domain co-regulators in
obesity, inflammation and cancer. Nat. Rev. Cancer 2012, 12, 465−477.
(12) Loven, J.; Hoke, H. A.; Lin, C. Y.; Lau, A.; Orlando, D. A.;
́
ACKNOWLEDGMENTS
■
Vakoc, C. R.; Bradner, J. E.; Lee, T. I.; Young, R. A. Selective inhibition
of tumor oncogenes by disruption of super-enhancers. Cell 2013, 153,
320−334.
(13) Boi, M.; Bonetti, P.; Ponzoni, M.; Tibiletti, M. G.; Stathis, A.;
Cvitkovic, E.; Inghirami, G.; Zucca, E.; Bertoni, F. The brd-Inhibitor
OTX015 shows pre-clinical activity in anaplastic large T-cell
lymphoma (ALCL). ASH Annu. Meet. Abstr., Blood 2012, 120, 4872.
(14) Bonetti, P.; Boi, M.; Ponzoni, M.; Tibiletti, M. G.; Stahis, A.;
Inghirami, G.; Noel, K.; Zucca, E.; Bertoni, F. ASH Annu. Meet. Abstr.,
Blood 2012, 120, 1657.
(15) Braun, T.; Coude, M.; Berrou, J.; Bertrand, S.; Riveiro, E.;
Herait, P.; Baruchel, A.; Dombret, H.; Gardin, C. Preclinical study of
the bromodomain inhibitor OTX015 in acute myeloid (AML) and
lymphoid (ALL) leukemias. ASH Annu. Meet. Abstr., Blood 2013, 122,
4218.
(16) Noel, J. K.; Iwata, K.; Ooike, S.; Sugahara, K.; Nakamura, H.;
Daibata, M. Development of the BET bromodomain inhibitor
OTX015. Mol. Cancer Ther. 2013, 12, C244.
This work was supported by grants from the French
Infrastructure for Integrated Structural Biology (FRISBI)
ANR-10-INSB-05-01, from the “Region PACA”, and from
“fond FEDER” 1860-38025 CPER 2007-2013. Sabine Milhas
was the recipient of a Ph.D. fellowship from Region PACA and
from the “Fondation pour la Recherche Medicale” (FRM).
Yuliia Voitovich was a recipient of a “Metchnikov” Ph.D.
fellowship from French government. The authors acknowledge
the European Synchrotron Radiation Facility for provision of
synchrotron radiation facilities (under the beam time
application MX1696) and thank Antoine Royant and David
Flot for assistance in using beamline ID29 and ID23-2,
respectively. We also thank Bruno Canard for access to
́
facilities at UMR7257 CNRS-AMU, Marseille, and Valerie
Depraetere-Ferrier for manuscript editing.
ABBREVIATIONS
■
(17) Jahagirdar, R.; Zhang, H.; Azhar, S.; Tobin, J.; Attwell, S.; Yu, R.;
Wu, J.; McLure, K. G.; Hansen, H. C.; Wagner, G. S.; Young, P. R.;
Srivastava, R. A. K.; Wong, N. C. W.; Johansson, J. A novel BET
bromodomain inhibitor, RVX-208, shows reduction of atherosclerosis
in hyperlipidemic ApoE deficient mice. Atherosclerosis 2014, 236, 91−
100.
BRD, bromodomain; BET, bromodomain and extra-terminal
domain; MTS, midthroughput screening; HTRF, homoge-
neous time-resolved fluorescence; ITC, isothermal titration
calorimetry
(18) Bourgeas, R.; Basse, M.-J.; Morelli, X.; Roche, P. Atomic analysis
of protein-protein interfaces with known inhibitors: the 2P2I database.
PLoS One 2010, 5, e9598.
(19) Basse, M.-J.; Betzi, S.; Bourgeas, R.; Bouzidi, S.; Chetrit, B.;
Hamon, V.; Morelli, X.; Roche, P. 2P2Idb: a structural database
dedicated to orthosteric modulation of protein-protein interactions.
Nucleic Acids Res. 2013, 41, D824−327, http://2p2idb.cnrs-mrs.fr,.
(20) Morelli, X.; Bourgeas, R.; Roche, P. Chemical and structural
lessons from recent successes in protein-protein interaction inhibition
(2P2I). Curr. Opin. Chem. Biol. 2011, 15, 475−481.
(21) Hamon, V.; Bourgeas, R.; Ducrot, P.; Theret, I.; Xuereb, L.;
Basse, M.-J.; Brunel, J.-M.; Combes, S.; Morelli, X.; Roche, P. 2P2I
HUNTER: a tool for filtering orthosteric protein-protein interac
interaction modulators via a dedicated support vector machine. J. R.
Soc., Interface 2014, 11, 20130860.
(22) Hamon, V.; Brunel, J. M.; Combes, S.; Basse, J.-M.; Roche, P.;
Morelli, X. 2P2Ichem: focused chemical libraries dedicated to
orthosteric modulation of protein-protein interactions. MedChem-
Comm 2013, 4, 797−809.
(23) Zhang, X.; Betzi, S.; Morelli, X.; Roche, P. Focused chemical
libraries–design and enrichment: an example of protein-protein
interaction chemical space. Future Med. Chem. 2014, 6, 1291−1307.
(24) Filippakopoulos, P.; Qi, J.; Picaud, S.; Shen, Y.; Smith, W. B.;
Fedorov, O.; Morse, E. M.; Keates, T.; Hickman, T. T.; Felletar, I.;
Philpott, M.; Munro, S.; McKeown, M. R.; Wang, Y.; Christie, A. L.;
West, N.; Cameron, M. J.; Schwartz, B.; Heightman, T. D.; La
Thangue, N.; French, C. A.; Wiest, O.; Kung, A. L.; Knapp, S.;
REFERENCES
■
(1) Filippakopoulos, P.; Knapp, S. The bromodomain interaction
module. FEBS Lett. 2012, 586, 2692−2704.
(2) Filippakopoulos, P.; Picaud, S.; Mangos, M.; Keates, T.; Lambert,
J.-P.; Barsyte-Lovejoy, D.; Felletar, I.; Volkmer, R.; Muller, S.; Pawson,
T.; Gingras, A.-C.; Arrowsmith, C. H.; Knapp, S. Histone recognition
and large-scale structural analysis of the human bromodomain family.
Cell 2012, 149, 214−231.
(3) Hewings, D. S.; Rooney, T. P.; Jennings, L. E.; Hay, D. A.;
Schofield, C. J.; Brennan, P. E.; Knapp, S.; Conway, S. J. Progress in
the development and application of small molecule inhibitors of
bromodomain-acetyl-lysine interactions. J. Med. Chem. 2012, 55,
9393−9413.
(4) Kanno, T.; Kanno, Y.; Siegel, R. M.; Jang, M. K.; Lenardo, M. J.;
Ozato, K. Selective recognition of acetylated histones by bromodo-
main proteins visualized in living cells. Mol. Cell 2004, 13, 33−43.
(5) Huang, H.; Zhang, J.; Shen, W.; Wang, X.; Wu, J.; Wu, J.; Shi, Y.
Solution structure of the second bromodomain of Brd2 and its specific
interaction with acetylated histone tails. BMC Struct. Biol. 2007, 7, 57.
(6) LeRoy, G.; Rickards, B.; Flint, S. J. The double bromodomain
proteins Brd2 and Brd3 couple histone acetylation to transcription.
Mol. Cell 2008, 30, 51−60.
(7) Baud, M. G.; Lin-Shiao, E.; Cardote, T.; Tallant, C.; Pschibul, A.;
Chan, K. H.; Zengerle, M.; Garcia, J. R.; Kwan, T. T.; Ferguson, F. M.;
Ciulli, A. Chemical biology. A bump-and-hole approach to engineer
controlled selectivity of BET bromodomain chemical probes. Science
2014, 346, 638−641.
G
DOI: 10.1021/acs.jmedchem.5b01708
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Bradner, J. E. Selective inhibition of BET bromodomains. Nature
2010, 468, 1067−1073.
(25) Vollmuth, F.; Blankenfeldt, W.; Geyer, M. Structures of the dual
bromodomains of the P-TEFb-activating protein Brd4 at atomic
resolution. J. Biol. Chem. 2009, 284, 36547−36556.
(26) Drouin, L.; McGrath, S.; Vidler, L. R.; Chaikuad, A.; Monteiro,
O.; Tallant, C.; Philpott, M.; Rogers, C.; Fedorov, O.; Liu, M.; Akhtar,
W.; Hayes, A.; Raynaud, F.; Muller, S.; Knapp, S.; Hoelder, S.
̈
Structure enabled design of BAZ2-ICR, a chemical probe targeting the
bromodomains of BAZ2A and BAZ2B. J. Med. Chem. 2015, 58, 2553−
2559.
(27) Bennett, J.; Fedorov, O.; Tallant, C.; Monteiro, O.; Meier, J.;
Gamble, V.; Savitsky, P.; Nunez-Alonso, G. A.; Haendler, B.; Rogers,
C.; Brennan, P. E.; Muller, S.; Knapp, S. Discovery of a chemical tool
̈
inhibitor targeting the bromodomains of TRIM24 and BRPF. J. Med.
Chem. 2015, DOI: 10.1021/acs.jmedchem.5b00458.
(28) Palmer, W. S.; Poncet-Montange, G.; Liu, G.; Petrocchi, A.;
Reyna, N.; Subramanian, G.; Theroff, J.; Yau, A.; Kost-Alimova, M.;
Bardenhagen, J. P.; Leo, E.; Shepard, H. E.; Tieu, T. N.; Shi, X.; Zhan,
Y.; Zhao, S.; Barton, M. C.; Draetta, G.; Toniatti, C.; Jones, P.; Geck
Do, M.; Andersen, J. N. Structure-guided design of IACS-9571, a
selective high-affinity dual TRIM24-BRPF1 bromodomain inhibitor. J.
Med. Chem. 2015, DOI: 10.1021/acs.jmedchem.5b00405.
(29) Vidal, A.; Giraud, I.; Madelmont, J.-C. Improved synthesis of 7-
(alkyl/aryl)guanine. Synth. Commun. 2004, 34, 3359−3365.
(30) Kalla, R. V.; Elzein, E.; Perry, T.; Li, X.; Gimbel, A.; Yang, M.;
Zeng, D.; Zablocki, J. Selective, high affinity A2B adenosine receptor
antagonists: N-1 monosubstituted 8-(pyrazol-4-yl)xanthines. Bioorg.
Med. Chem. Lett. 2008, 18, 1397−1401.
(31) Kuang, M.; Zhou, J.; Wang, L.; Liu, Z.; Guo, J.; Wu, R. Binding
Kinetics versus Affinities in BRD4 Inhibition. J. Chem. Inf. Model. 2015,
55, 1926−1935.
(32) Ran, T.; Zhang, Z.; Liu, K.; Lu, Y.; Li, H.; Xu, J.; Xiong, X.;
Zhang, Y.; Xu, A.; Lu, S.; Liu, H.; Lu, T.; Chen, Y. Insight into the key
interactions of bromodomain inhibitors based on molecular docking,
interaction fingerprinting, molecular dynamics and binding free energy
calculation. Mol. BioSyst. 2015, 11, 1295−1304.
(33) Laskowski, R. A.; Swindells, M. B. LigPlot+: multiple ligand-
protein interaction diagrams for drug discovery. J. Chem. Inf. Model.
2011, 51, 2778−2786.
H
DOI: 10.1021/acs.jmedchem.5b01708
J. Med. Chem. XXXX, XXX, XXX−XXX