Discovery of Benzotriazolo[4,3-d][1,4]diazepines as Orally Active Inhibitors of BET Bromodomains

Article abstract of DOI:10.1021/ml500411h

Inhibition of the bromodomains of the BET family, of which BRD4 is a member, has been shown to decrease myc and interleukin (IL) 6 in vivo, markers that are of therapeutic relevance to cancer and inflammatory disease, respectively. Herein we report substituted benzo[b]isoxazolo[4,5-d]azepines and benzotriazolo[4,3-d][1,4]diazepines as fragment-derived novel inhibitors of the bromodomain of BRD4. Compounds from these series were potent and selective in cells, and subsequent optimization of microsomal stability yielded representatives that demonstrated dose- and time-dependent reduction of plasma IL-6 in mice.

Full text of DOI:10.1021/ml500411h

Subscriber access provided by SUNY DOWNSTATE
Letter
Discovery of benzotriazolo[4,3-d][1,4]diazepines
as orally-active inhibitors of BET bromodomains
Alexander Merton Taylor, Rishi G Vaswani, Victor S Gehling, Michael Charles Hewitt, Yves Leblanc, James
E Audia, Steve Bellon, Richard T Cummings, Alexandre Cote, Jean-Christophe P. Harmange, Hariharan
Jayaram, Shivangi Joshi, Jose M Lora, Jennifer A Mertz, Adrianne Neiss, Eneida Pardo, Christopher G.
Nasveschuk, Florence Poy, Peter Sandy, Jeremy W Setser, Robert J Sims, Yong Tang, and Brian K Albrecht
ACS Med. Chem. Lett., Just Accepted Manuscript • Publication Date (Web): 25 Mar 2015
Downloaded from http://pubs.acs.org on March 26, 2015
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
ACS Medicinal Chemistry Letters is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 7
ACS Medicinal Chemistry Letters
1
2
3
4
5
6
7
8
9
Discovery of benzotriazolo[4,3-d][1,4]diazepines as orally-
active inhibitors of BET bromodomains.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Alexander M. Taylor*, Rishi G. Vaswani, Victor S. Gehling, Michael C. Hewitt, Yves Leblanc, James
E. Audia, Steve Bellon, Richard T. Cummings, Alexandre Côté, JeanꢀChristophe Harmange, Hari
Jayaram,† Shivangi Joshi, Jose M. Lora, Jennifer A. Mertz, Adrianne Neiss, Eneida Pardo, Christopher
G. Nasveschuk, Florence Poy, Peter Sandy, Jeremy W. Setser, Robert J. Sims, III, Yong Tang, Brian K.
Albrecht
Constellation Pharmaceuticals, 215 First St., Suite 200, Cambridge, Massachusetts 02142
BET, BRD4, bromodomain, IL-6
ABSTRACT: Inhibition of the bromodomains of the BET family, of which BRD4 is a member, has been shown to decrease myc
and interleukin (IL) 6 in vivo, markers that are of therapeutic relevance to cancer and inflammatory disease, respectively. Herein we
report substituted benzo[b]isoxazolo[4,5ꢀd]azepines and benzo[1,2,4][4,3ꢀd]triazolo[1,4]diazepines as fragmentꢀderived novel inꢀ
hibitors of the bromodomain of BRD4. Compounds from these series were potent and selective in cells, and subsequent optimizaꢀ
tion of microsomal stability yielded representatives that demonstrated doseꢀ and timeꢀdependent reduction of plasma ILꢀ6 in mice.
Scheme 1. Overview of the origin and development of different azepine-derived BET inhibitors.
The bromodomain and extra terminal (BET) family is made up
of BRD2, BRD3, BRD4, and BRDT, each of which contains
tandem bromodomains (BDꢀ1 and BDꢀ2) that recognize specific
acetylated lysine residues in the Nꢀterminal tails of histones.1
Members of the BET family modulate gene expression by
recruiting transcriptional regulators to specific genomic
locations.2 In 2009, the Mitsubishi Tanabe Pharma Corporation
reported the first small molecule inhibitors of this family of
chromatin adaptors,3 and publications followed on the use of BET
inhibitors to demonstrate antiꢀtumor activity4 and modulate
inflammation.5 Experiments applying genetic ablation or specific
inhibitor treatment have established the role of BET family
members in regulating genes crucial for normal cell growth.6,7,8
malignancies and demonstrated that BET inhibition regulates the
MYC oncogene, resulting in cell cycle arrest and apoptosis. In vivo
BET bromodomain inhibition leads to efficacy in xenograft
models representing multiple cancer types, and BET
bromodomains are also implicated in nonꢀoncology indications,
i.e. human immunodeficiency virus and human tꢀlymphotrophic
viral gene expression, DNA damage response, and cardiac
hypertrophy. Importantly, BET inhibitors have been shown to
selectively regulate a subset of multiple inflammatory cytokines,
including interleukins 6, 1β, 17, 21, and 23, and granulocyte
macrophage colonyꢀstimulating factor11,12 and have demonstrated
efficacy in several in vivo models of inflammation, including
endotoxic shock, experimental autoimmune encephalomyelitis
and arthritis.5,11
Subsequent studies9,10 have defined a mechanistic link between
potency of BET inhibition and activity against hematologic
Table 1. BRD4 BD-1 structure-activity relationships of benzo[b]isoxazolo[4,5-d]azepines
1
ACS Paragon Plus Environment

ACS Medicinal Chemistry Letters
Page 2 of 7
1
2
3
4
5
d
PPBf
(m/r/d/h)e
(%)
BRD4 BDꢀ1
AS IC50
ILꢀ6
EC50
(ꢀM)b
LM Clint
(m/r/d/h)e
compound
clogPc
(ꢀM)a
(ꢀL/min/mg)
6
7
8
9
R1
H
R2
(+)-JQ-1g
I-BET151g
4
0.024
0.018
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
H
H
0.55 ± 0.21
4.6 ± 0.45
4.3
4.1
22/24/18/17
>99/>99/>99/>99
>99/99/>99/>99
0.064 ± 0.012
0.036 ± 0.009
0.38 ± 0.039
101/195/165/86
5
6
Me
Me
0.078 ± 0.004
0.067 ± 0.023
4.7
3.5
44/210/240/62
44/62/22/33
99/98/>99/>99
96/97/99/99
0.031 ± 0.005
7
Me
0.032 ± 0.010
0.31 ± 0.086
0.094 ± 0.018
0.81 ± 0.027
4.3
2.6
64/71/65/52
37/22/69/27
99/97/98/98
96/94/98/99
8
9
CH2CONH2
aBinding was measured using an AlphaLISA assay with the isolated bromodomain. See supporting information for more details. N ≥ 3 bILꢀ
c
6 EC50 = measurement of the suppression of ILꢀ6 release from THPꢀ1 cells following stimulation with lipopolysaccharide. N = 2. clogP
calculated using ChemDraw 14. dLM Clint = intrinsic clearance of compound when incubated with liver microsomes. e(m/r/d/h) =
mouse/rat/dog/human. fPPB
=
plasma protein binding. gPotencies reported in refs.
4
and 5, respectively.
Because of the potential therapeutic application of BET inhibiꢀ
tion, we initiated a drug discovery program and recently reported
on a fragmentꢀbased effort that resulted in a series of isoxazoꢀ
lo[5,4ꢀc]thieno[2,3ꢀe]azepine BET inhibitors exemplified by CPI-
3 (Scheme 1).13 Concurrent with that effort we were interested in
exploring a different connectivity within the sevenꢀmembered ring
that would provide a unique basis for modulation of the physicoꢀ
chemical properties and BET potency of the resultant compounds.
Specifically, we hypothesized that changing the cyclic imine to a
substituted aniline would retain the concave shape important for
specific binding to the bromodomain while providing a different
metabolic profile for the series. We were also interested in evaluꢀ
ating a modified linker in combination with the reported triazole
binding element4,5 and in finding a replacement for the thiophene
group of CPI-3 because of its potential metabolic instability. Here
we describe the resulting benzo[b]isoxazolo[4,5ꢀd]azepine and
benzotriazolo[4,3ꢀd][1,4]diazepine series, which were potent inꢀ
hibitors of BET bromodomains in biochemical and cellular assays
and which allowed for optimization of metabolic stability and
demonstration of a pharmacodynamic effect in a mouse model of
ILꢀ6 suppression. (Scheme 1)
them, Nꢀarylation to form the 4ꢀcyano compound 4 (Table 1)
markedly improved biochemical potency and activity in a cellular
readout of ILꢀ6 suppression, a BETꢀrelated phenotype relevant to
modulation of the immune response in vivo (see supporting inꢀ
formation for details). Throughout our studies, BRD4 BDꢀ1 was
used as a surrogate for the entire BET family; tests of individual
compounds against the other bromodomains of the BET family
indicated that BRD4 BDꢀ1 inhibition was representative of inhibiꢀ
tion across the family (vide infra). A variety of substitutions
around the ring were explored; 4ꢀCNꢀPh was consistently the
most potent substitution and is the only one shown here, for simꢀ
plicity. Potency could be further improved with the addition of a
heterocycle at the 8 position, as in compound 5 (Table 1, see
numbering). Introduction of a methyl group on the azepine (Table
1, 6) maintained biochemical activity but improved translation of
that potency into a cellular context, perhaps by affording greater
conformational preorganization of the compound, which has been
shown to improve cell permeability.14,15 A variety of substituted
heterocycles were tolerated at the 8 position, as exemplified by
compounds 7 and 8, but any groups attached to the azepine ring
other than methyl were deleterious to activity (e.g. compound 9 in
Table 1, see Table S7 for further examples). The limitation to
methyl at the 4 position is in contrast to what we observed with
the isoxazolo[5,4ꢀc]thieno[2,3ꢀe]azepine scaffold (cf. CPI-3)13
and with what others have observed with related scaffolds.4,5 Subꢀ
stitutions at other positions of the benzo ring or at the 1 position
of the isoxazole diminished potency (data not shown).
Our investigations began with the preparation of benꢀ
zo[b]isoxazolo[4,5ꢀd]azepine fragment 1 (Scheme 1), which exꢀ
hibited a BRD4 BDꢀ1 binding affinity in an AlphaLISA (AS)
assay comparable to that of the 3ꢀmethylꢀ4ꢀphenylisoxazolꢀ5ꢀ
amine fragment from which CPI-3 (IC50 33 ꢀM)13 was derived
and which provided several vectors to use to improve potency. Of
Table 2. BRD4 BD-1 structure-activity relationships of benzo[1,2,4][4,3-d]triazolo[1,4]diazepines
2
ACS Paragon Plus Environment

Page 3 of 7
ACS Medicinal Chemistry Letters
1
2
3
4
5
d
PPBf
(m/r/d/h)e
(%)
BRD4 BDꢀ1
ILꢀ6 EC50
LM Clint
clogPc
AS IC50
(m/r/d/h)e
compound
(ꢀM)b
(ꢀM)a
(ꢀL/min/mg)
6
7
8
9
R1
H
R2
H
1.56 ± 0.48
2.3 ± 1.1
4.2
4.2
500/126/390/154
77/154/91/14
93/95/95/95
97/95/96/96
10
11
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
H
0.044 ± 0.019
0.13 ± 0.036
Me
0.026 ± 0.003
0.016 ± 0.002
4.8
ꢀ/34/94/6.4
98/97/97/98
12
H
H
Me
0.35 ± 0.16
2.1 ± 0.29
0.29 ± 0.042
1.8 ± 0.64
4.7
2.9
110/25/83/24
52/74/33/44
96/96/95/97
92/87/91/91
13
14
CH2CONHEt
RꢀMe
S-Me
0.015 ± 0.004
0.79 ± 0.24
0.013 ± 0.004
0.79 ± 0.19
4.8
4.8
16/26/19/5.1
ꢀ
98/96/96/98
ꢀ
3
15
1.2 ± 0.44
3.1g
4.7
ꢀ/65/ꢀ/25
ꢀ/97/ꢀ/98
16
aBinding was measured using an AlphaLISA assay with the isolated bromodomain. See supporting information for more details. N ≥ 3 bILꢀ
c
6 EC50 = measurement of the suppression of ILꢀ6 release from THPꢀ1 cells following stimulation with lipopolysaccharide. N = 2. clogP
calculated using ChemDraw 14. dLM Clint = intrinsic clearance of compound when incubated with liver microsomes. e(m/r/d/h) =
mouse/rat/dog/human.
fPPB
=
plasma
protein
binding.
gN
=
1.
In the course of optimizing potency, we determined the in vitro
metabolic stability of several compounds. As shown in Table 1,
clearance in liver microsomes was generally high for the series
across species, which we surmised may have been due to the high
clogP values intrinsic to the class.14 Replacement of the isoxazole
with a triazole binding element presented an opportunity to lower
the clogP of the series; simple estimates of lipophilicity16 suggestꢀ
ed that the more polar heterocycle would consistently lower clogP
by one unit relative to the corresponding isoxazoles while mainꢀ
taining the overall shape and binding mode of the initial series,
allowing us to capitalize on similar structureꢀactivity relationships
(SAR).
methyl resulted in a roughly tenfold loss in potency, as exempliꢀ
fied by 14 (Table 2; further examples in Table S7). Separation of
the individual enantiomers of elaborated compounds established
that most of the binding activity resided within one enantiomer;
compound 3 (98% ee) was greater than 50ꢀfold more potent than
15 (>99% ee, Table 2). Gratifyingly, in vitro liver microsome
intrinsic clearance in this series was generally lower than that in
the corresponding isoxazoles (e.g. LM Clint of 6 vs. 12 across
species), even though incorporation of the chlorophenyl, which
was optimal for potency, returned the clogP of most compounds
to the same range as that of the isoxazole series. Finally, to deterꢀ
mine the impact of alternate methyl substitution on the azepine
ring, compound 16 was prepared. It demonstrated little change in
potency relative to unsubstituted 10 even though both 4ꢀ and 5ꢀ
methyl substituents would be expected to preorganize the azepine,
suggesting that the position of the methyl azepine susbstituent was
important beyond just its effect on the conformation of the azeꢀ
pine ring.
To explore whether the triazole could be combined with the
aniline azepine linker, we prepared the triazole fragment 2
(Scheme 1) and found that it, too, bound BRD4 with potency
comparable to 1. Even with the change in heterocycle, much of
the SAR from the benzo[b]isoxazolo[4,5ꢀd]azepine series could
be recapitulated; Nꢀarylation improved potency (Table 2, 10), as
did incorporation of certain heterocycles at the 8 position (Table
2, 11). In contrast with the benzo[b]isoxazolo[4,5ꢀd]azepine seꢀ
ries, 4ꢀClꢀPh was consistently the most potent Nꢀaryl substituent
for the benzotriazolo[4,3ꢀd][1,4]diazepines, although, again, a
variety of substitution patterns were tolerated. As before, subseꢀ
quent methylation at the 4 position of the azepine ring (Table 2,
12) provided compounds highly active in both the biochemical
and cellular assays. As before, azepine substitutions other than
3
ACS Paragon Plus Environment

ACS Medicinal Chemistry Letters
Page 4 of 7
To further explore that idea, we examined a number of resolved
1
2
3
4
5
6
7
8
compounds with a greater variety of substituents at position 8. As
shown in Table 3, heteroaromatic substituents such as pyridine
(17), pyrolidinone (18), and Nꢀlinked triazole (19) yielded
compounds with potent binding and cellular activity. A primary
carboxamide (20) also improved cellular potency roughly
threefold relative to unsubstituted 13 (Table 2). Interestingly,
compounds unable to align a π system with the edge of Trp81,
such as 21 and 22, lost potency relative to 13, consistent with the
model of an edge to face interaction driving the binding affinity at
the 8 position.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Based on its favorable combination of cellular activity and
ADME properties, compound
3 was chosen for in vivo
pharmacokinetic profiling in rat. As shown in Table 4, 3
demonstrated IV clearance ten percent below liver blood flow, a
reasonable halfꢀlife, and appreciable oral bioavailability.
Importantly, this profile suggested that exposure of 3 would be
sufficient to modulate ILꢀ6 in an in vivo model.
(a)
Before proceeding with experiments to gauge the BET activity
3 in vivo, we determined its binding affinity to the
of
bromodomains of other members of the BET family. As
mentioned above, BRD4 BDꢀ1 proved representative of activity
against other BET proteins; IC50s ranged from 0.003 ꢀM to 0.017
ꢀM (see supporting information for individual values). And
because bromodomains are highly homologous across the class,19
we also assessed the activity of 3 against eight other nonꢀBET
bromodomains, as estimated by AlphaLISA or time resolved
fluorescence resonance energy transfer. With the exception of
CBP (AS IC50 = 2.2 ꢀM), 3 demonstrated an IC50 greater than 15
ꢀM on all of the bromodomains tested (see supporting
information), suggesting that its activity in vivo would be limited
to the BET family at the relevant concentrations.
(b)
Figure 1. (A) A 1.4 Å coꢀcrystal structure of 3 (green) bound to
BRD4 BDꢀ1. PDB code: XXXX. Key Hꢀbonding interactions are
indicated. (B) An overlay of 13 (green) and CPI-3 (gray) shows
the similarity in binding modes between the two series (RMSD =
Having established its BET activity, specificity, and potential
for meaningful in vivo exposure, 3 was dosed PO in mouse at 15
mg/kg at various time points prior to LPS stimulation. Plasma ILꢀ
6 levels were measured two hours later. As shown in Figure 2, ILꢀ
6 release was suppressed to baseline levels at higher plasma
concentrations of 3 and returned toward vehicle levels over the
time course of dosing, in conjunction with decreasing plasma
concentrations of 3. Substantial suppression of the ILꢀ6 response
in that experiment was observed only when the predicted free
plasma concentration of compound was above the cellular EC50,20
consistent with the free drug hypothesis.21 These results support
0.43 for common heavy atoms). PDB code: 4X2I.
A coꢀcrystal structure of 3 obtained in BRD4 BDꢀ1 established
the binding mode of these series (Figure 1a). Notably, it
complemented the binding site similar to our reported
isoxazolo[5,4ꢀc]thieno[2,3ꢀe]azepines, as exemplified by an
overlay of the crystal structures of 3 (chosen for simplicity) and
CPI-3; RMSD between the common heavy atoms of the two
series was only 0.43 (Figure 1b). As intended, 3 adopted a cupped
shape, filling the space between Asn140 and the WPF shelf
(residues 81ꢀ83) and making lipophilic contacts around
gatekeeper residue Ile146. The lone pair of the triazole nitrogen in
the 3 position served as an Hꢀbond acceptor in the crucial
the model of BET inhibition mediating
a reduction in
inflammatory response ILꢀ6 levels and further recommend BET
as a target for drug discovery.
Table 3. BRD4 BD-1 structure-activity relationship of 8-
substituted benzotriazolo[4,3-d][1,4]diazepines
interaction
with
Asn140
(highly
conserved
among
bromodomains), positioning the triazole methyl group in the same
location as the Nꢀacetyl of the natural acetylated lysine substrate
and orienting the nitrogen in position 2 to engage in a throughꢀ
water hydrogen bond with Tyr97. The 4ꢀmethyl azepine
substituent provided a second, nonꢀcanonical Hꢀbond donating
interaction17,18 with the Asn and enforced the curvature of the
azepine. The coꢀcrystal structure allowed for the unambiguous
assignment of the stereochemistry at that position (as did the coꢀ
crystal structure of the opposite enantiomer, 15, Figure S2).
Interestingly, the proximity observed between the aminopyridine
of 3 and Trp81 suggested that the increased potency of
compounds substituted at that position may have been due to a
productive edge to face interaction.
BRD4
ILꢀ6
BDꢀ1
EC50
AS IC50
cmpd
(ꢀM)b
(ꢀM)a
R
0.046 ±
0.045
0.10 ±
0.076
17
4
ACS Paragon Plus Environment

Page 5 of 7
ACS Medicinal Chemistry Letters
aLEAN = pIC50/#heavy atoms; LLE = pIC50 – clogP. bLMrat Clint
= intrinsic clearance of compound when incubated with rat liver
microsomes. PPBrat = rat plasma protein binding. IV formulaꢀ
0.025 ±
0.002c
0.036 ±
0.017
1
2
3
4
5
6
7
18
19
20
c
d
tion:
1
mL of
1
mg/mL solution of
3
in 5:25:70
0.074 ±
0.023
0.041 ±
0.010
DMA:PEG400:20% SBECD. N = 3 rats. eLBF = liver blood
flow.fPO formulation: 10 mL of 1 mg/mL suspension of 3 in 0.5%
methylcellulose. N = 3 rats.
0.065 ±
0.026
0.096 ±
0.028
8
9
10
11
12
13
0.28 ±
0.075
0.45 ±
0.028
21
22
0.67 ±
0.20
1.2 ± 0.49
aBinding was measured using an AlphaLISA assay with the isoꢀ
lated bromodomain; see supporting information for more details.
N ≥ 3. ILꢀ6 EC50 = measurement of the suppression of ILꢀ6 reꢀ
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
b
lease from THPꢀ1 cells following stimulation with lipopolysacꢀ
charide. N = ≥ 2. cN = 2.
In summary, we report the discovery of benzo[b]isoxazolo[4,5ꢀ
d]azepines and benzotriazolo[4,3ꢀd][1,4]diazepines as two new
series of BET inhibitors that combine design elements from our
previously reported isoxazolo[5,4ꢀc]thieno[2,3ꢀe]azepines with a
unique azepine connection pattern. Both series afforded
biochemically and cellularly potent compounds, some of which
could be optimized for metabolic stability and applied in vivo to
demonstrate doseꢀdependent modulation of ILꢀ6 suppression via
BET engagement. Further applications of these series will be
reported in due course.
Figure 2. Suppression of ILꢀ6 release in mouse correlates with
plasma exposure of 3 following 15 mg/kg dosing PO. Free
concentration of 3 calculated based on 98% PPBmouse. Dashed line
indicates cellular ILꢀ6 EC50.
Table 4. Pharmacokinetic and selected in vitro parameters
for 3 in rat
ASSOCIATED CONTENT
Supporting Information. Further references describing the bioꢀ
logical and pharmacological validation of BET as a drug discovꢀ
ery target; assay details; descriptions of the ILꢀ6 PD assay and
results; synthetic and characterization data for all compounds
reported; coꢀcrystal structures of 13 and 15; crystallographic
methods and data; additional SAR data; BET family and other
bromodomain potency data for 3. This material is available free of
charge via the Internet at http://pubs.acs.org.
Parameter
Compound 3
Cell potency
0.013
ILꢀ6 EC50 (ꢀM)
Binding
LEAN/LLEa
0.26/3.1
26
efficiency
b
LMrat Clint
AUTHOR INFORMATION
(ꢀL/min/mg)
PPBrat (%)c
Cl (L/h/kg)
Clunbound (L/h/kg)
VSS (L/kg)
t1/2 (h)
In vitro
Corresponding Author
96
* A. M. T. email: alex.taylor@constellationpharma.com.
3.8 (90% LBFe)
Present Addresses
†Editas Medicine, 300 Third St., Cambridge, Mass. 02142.
IVd
95
2.2
0.67
0.29
1.8
1.4
23
Author Contributions
(1 mg/kg)
The manuscript was written through contributions of all authors.
All authors have given approval to the final version of the manuꢀ
script.
ACKNOWLEDGMENT
Cmax (ꢀM)
t1/2 (h)
We are grateful to Ted Peters and Christina Lee in the Lead Disꢀ
covery group for plating compounds used in these studies; PPD
BioDuro for synthesizing some of the compounds presented here;
WuXi AppTec and Custom NMR Services for compound characꢀ
terization.
POf
(10 mg/kg)
AUClast (ꢀM.h)
F (%)
REFERENCES
5
ACS Paragon Plus Environment

ACS Medicinal Chemistry Letters
Page 6 of 7
(1) Wu, S. Y.; Chiang, C. M. The double bromodomainꢀ
(11) Mele, D. A.; Salmeron, A.; Ghosh, S.; Huang, H. R.; Bryant,
1
2
containing chromatin adaptor Brd4 and transcriptional regulation.
J. Biol. Chem. 2007, 282, 13141ꢀ45.
B. M.; Lora, J. M. BET bromodomain inhibition suppresses
TH17ꢀmediated pathology. J. Exp. Med. 2013, 210, 2181ꢀ90.
3
4
5
6
7
8
9
(2) Dey, A.; Chitsaz, F.; Abbasi, A.; Misteli, T.; Ozato, K. The
double bromodomain protein Brd4 binds to acetylated chromatin
during interphase and mitosis. Proc. Natl. Acad. Sci. U. S. A.
2003, 100, 8758ꢀ63.
(12) Khan, Y. M.; Kirkham, P.; Barnes, P. J.; Adcock, I. M. Brd4
is essential for ILꢀ1βꢀinduced inflammation in human airway
epithelial cells. PLoS One 2014, 9, e95051.
(13) Gehling, V. S.; Hewitt, M. C.; Vaswani, R. G.; Leblanc, Y.;
Côté, A.; Nasveschuk, C. G.; Taylor, A. M.; Harmange, J. C.;
Audia, J. E.; Pardo, E.; Joshi, S.; Sandy, P.; Mertz, J. A.; Sims, R.
J. 3rd; Bergeron, L.; Bryant, B. M.; Bellon, S.; Poy, F.; Jayaram,
H.; Sankaranarayanan, R.; Yellapantula, S.; Bangalore Sriniꢀ
vasamurthy, N.; Birudukota, S.; Albrecht, B. K. Discovery, deꢀ
sign, and optimization of isoxazole azepine BET inhibitors. ACS
Med. Chem. Lett. 2013, 4, 835ꢀ40.
(3) Miyoshi, M.; Ooike, S.; Iwata, K.; Hikawa, H.; Sugahara, K.
Antitumor Compounds, WO 2009/084693A1, Published Sept. 9,
2009.
(4) 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.; Bradner, J. E. Selective inhibition of BET bromodomains.
Nature 2010, 468, 1067ꢀ73.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(14) Veber, D. F.; Johnson, S. R.; Cheng, H.ꢀY.; Smith, B. R.;
Ward, K. W.; Kopple, K. D. Molecular properties that influence
the oral bioavailability of drug candidates. J. Med. Chem. 2002,
45, 2615ꢀ23.
(5) Nicodeme, E.; Jeffrey, K. L.; Schaefer, U.; Beinke, S.; Dewell,
S.; Chung, C. W.; Chandwani, R.; Marazzi, I.; Wilson, P.; Coste,
H.; White, J.; Kirilovsky, J.; Rice, C. M.; Lora, J. M.; Prinjha, R.
K.; Lee, K.; Tarakhovsky, A. Suppression of inflammation by a
synthetic histone mimic. Nature 2010, 468, 1119ꢀ23.
(15) Bogdan, A. R.; Davies, N. L.; James, K. Comparison of difꢀ
fusion coefficients for matched pairs of macrocyclic and linear
molecules over a drugꢀlike molecular weight range. Org. Biomol.
Chem. 2011, 9, 7727ꢀ33.
(6) Belkina, A. C.; Denis, G. V. BET domain coꢀregulators in
obesity, inflammation and cancer. Nat. Rev. Cancer 2012, 12,
465ꢀ77.
(16) clogP values were generated using ChemBioDraw Ultra 14.0.
(17) Horowitz, S.; Trievel, R. C. Carbonꢀoxygen hydrogen bondꢀ
ing in biological structure and function. J. Biol. Chem. 2012, 287,
41576ꢀ82.
(7) Mertz, J. A.; Conery, A. R; Bryant, B. M.; Sandy, P.; Balꢀ
asubramanian, S.; Mele, D. A.; Bergeron, L.; Sims, R. J. 3rd.
Targeting MYC dependence in cancer by inhibiting BET bromoꢀ
domains. Proc. Nat. Acad. Sci. 2011, 108, 16669ꢀ74.
(18) Bissantz, C.; Kuhn, B.; Stahl, M. A Medicinal Chemist’s
Guide to Molecular Interactions. J. Med. Chem. 2010, 53, 5061ꢀ
84.
(8) Zuber, J.; Shi, J.; Wang, E.; Rappaport, A. R.; Herrmann, H.;
Sison, E. A.; Magoon, D.; Qi, J.; Blatt, K.; Wunderlich, M.; Tayꢀ
lor, M. J.; Johns, C.; Chicas, A.; Mulloy, J. C.; Kogan, S. C.;
Brown, P.; Valent, P.; Bradner, J. E.; Lowe, S. W.; Vakoc, C. R.
RNAi screen identifies Brd4 as a therapeutic target in acute myeꢀ
loid leukaemia. Nature 2011, 478, 524ꢀ8.
(19) Muller, S.; Filippakopoulos, P.; Knapp, S. Bromodomains as
therapeutic targets. Expert Rev. Mol. Med. 2011, 13, e29.
(20) The sequence homology between mouse and human BRD4
(100% for BDꢀ1 and 97% for BDꢀ2), and our previous observaꢀ
tion that inhibition of mouse or human BRD4 has the same effect
on myc expression, supports this analysis (see ref. 7).
(9) Shi, J.; Vakoc, C. R. The mechanisms behind the therapeutic
activity of BET bromodomain inhibition. Mol. Cell. 2014, 54,
728ꢀ36.
(21) Smith, D. A.; Di, L.; Kerns, E. H. The effect of plasma proꢀ
tein binding on in vivo efficacy: misconceptions in drug discovꢀ
ery. Nature Rev. Drug Disc. 2010, 9, 929ꢀ39.
(10) An extensive list of references describing specific biological
and pharmacological studies is including in the supporting inforꢀ
mation.
For TOC use only
Discovery of benzotriazolo[4,3-d][1,4]diazepines as orally-
active inhibitors of BET bromodomains.
Alexander M. Taylor*, Rishi G. Vaswani, Victor S. Gehling, Michael C. Hewitt, Yves Leblanc, James
E. Audia, Steve Bellon, Richard T. Cummings, Alexandre Côté, JeanꢀChristophe Harmange, Hari
Jayaram,† Shivangi Joshi, Jose M. Lora, Jennifer A. Mertz, Adrianne Neiss, Eneida Pardo, Christopher
6
ACS Paragon Plus Environment

Page 7 of 7
ACS Medicinal Chemistry Letters
G. Nasveschuk, Florence Poy, Peter Sandy, Jeremy W. Setser, Robert J. Sims, III, Yong Tang, Brian K.
Albrecht
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
7
ACS Paragon Plus Environment