Design and Characterization of Novel Covalent Bromodomain and Extra-Terminal Domain (BET) Inhibitors Targeting a Methionine

Article abstract of DOI:10.1021/acs.jmedchem.8b00666

BET proteins are key epigenetic regulators that regulate transcription through binding to acetylated lysine (AcLys) residues of histones and transcription factors through bromodomains (BDs). The disruption of this interaction with small molecule bromodomain inhibitors is a promising approach to treat various diseases including cancer, autoimmune and cardiovascular diseases. Covalent inhibitors can potentially offer a more durable target inhibition leading to improved in vivo pharmacology. Here we describe the design of covalent inhibitors of BRD4(BD1) that target a methionine in the binding pocket by attaching an epoxide warhead to a suitably oriented noncovalent inhibitor. Using thermal denaturation, MALDI-TOF mass spectrometry, and an X-ray crystal structure, we demonstrate that these inhibitors selectively form a covalent bond with Met149 in BRD4(BD1) but not other bromodomains and provide durable transcriptional and antiproliferative activity in cell based assays. Covalent targeting of methionine offers a novel approach to drug discovery for BET proteins and other targets.

Full text of DOI:10.1021/acs.jmedchem.8b00666

Subscriber access provided by Kaohsiung Medical University
Article
DESIGN AND CHARACTERIZATION OF NOVEL
COVALENT BROMODOMAIN AND EXTRA-TERMINAL
DOMAIN (BET) INHIBITORS TARGETING A METHIONINE
Olesya Kharenko, Reena G Patel, S. David Brown, Cyrus Calosing, Andre White,
Damodharan Lakshminarasimhan, Robert K Suto, Bryan C Duffy, Douglas B. Kitchen,
Kevin G McLure, Henrik C. Hansen, Edward H van der Horst, and Peter R. Young
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b00666 • Publication Date (Web): 30 Aug 2018
Downloaded from http://pubs.acs.org on August 30, 2018
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 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 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.
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 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
DESIGN AND CHARACTERIZATION OF
NOVEL COVALENT BROMODOMAIN AND
EXTRAꢀTERMINAL DOMAIN (BET)
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
INHIBITORS TARGETING A METHIONINE
Olesya A. Kharenkoa*, Reena G. Patela, S. David Browna, Cyrus Calosinga, Andre Whiteb,
Damodharan Lakshminarasimhanb, Robert K. Sutob, Bryan C. Duffyc, Douglas B. Kitchenc,
Kevin G. McLurea, Henrik C. Hansena, Edward H van der Horsta, Peter R. Younga
a Zenith Epigenetics, Suite 300, 4820 Richard Road SW, Calgary, Alberta, T3E 6L1, Canada
b Xtal BioStructures, Inc., 12 Michigan Dr., Natick, MA 01760, USA
c AMRI, 26 Corporate Circle, PO Box 15098, Albany, NY 12212ꢀ5098, USA
KEYWORDS: BRD4, Bromodomains, BET Inhibitors, Covalent Inhibitor, Methionine
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 43
1
2
3
4
ABSTRACT
5
6
7
8
9
BET proteins are key epigenetic regulators which regulate transcription through binding to
acetylated lysine (AcLys) residues of histones and transcription factors through bromodomains
(BDs). The disruption of this interaction with small molecule bromodomain inhibitors is a
promising approach to treat various diseases including cancer, autoimmune and cardiovascular
diseases. Covalent inhibitors can potentially offer a more durable target inhibition leading to
improved in vivo pharmacology. Here we describe the design of covalent inhibitors of
BRD4(BD1) that target a methionine in the binding pocket by attaching an epoxide warhead to a
suitably oriented nonꢀcovalent inhibitor. Using thermal denaturation, MALDIꢀTOF mass
spectrometry and an Xꢀray crystal structure, we demonstrate that these inhibitors selectively
form a covalent bond with Met149 in BRD4(BD1) but not other bromodomains, and provide
durable transcriptional and antiꢀproliferative activity in cell based assays. Covalent targeting of
methionine offers a novel approach to drug discovery for BET proteins and other targets.
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
2
ACS Paragon Plus Environment

Page 3 of 43
Journal of Medicinal Chemistry
1
2
3
4
INTRODUCTION
5
6
7
8
9
BET proteins are epigenetic readers, which regulate gene transcription by interacting with
acetylated lysine (AcLys) residues of histones and transcription factors through bromodomains,
evolutionarily conserved protein domains that have recently emerged as key epigenetic
regulators of gene transcription1. BET proteins consist of a family of four proteins, BRD2,
BRD3, BRD4 and BRDT that each contains two tandem bromodomains (BD1) and (BD2).
Inhibition of BET protein binding to acetylated histones can effect the transcription of genes
involved in various diseases including different types of cancer, cardiovascular and autoꢀimmune
diseases.2 Most BET inhibitors (BETi) are small molecules designed to reversibly interact with
the bromodomains of the BET proteins and disrupt their interaction with the acetylated lysine
residues, and several candidate BETi have advanced into clinical development for multiple
indications,3 including cancer indications such as acute myeloid leukemia, multiple myeloma,
nonꢀHodgkin’s lymphoma, and various solid tumors.4
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
Recently, proteolytic targeting chimera (PROTAC)ꢀinduced degradation of BRD4 has been
described as a novel and promising therapeutic strategy for BET inhibition.5 The principle of this
approach is a complete degradation of BRD4 through the PROTAC containing the BET protein
recognition unit and an ubiquitin ligase recognition module. Several reports show that complete
degradation of BRD4 through PROTAC results in superior potency, prolonged downregulation
of MYC, as well as lasting antiꢀproliferative effects and induction of cell death in multiple cancer
cell lines.6ꢀ10 However these molecules are large which may preclude oral delivery.
Covalent inhibitors have been widely used as successful drug therapies in various indications,12,
13
also offering similar advantages including increased duration of action, prolonged target
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 43
1
2
3
4
5
6
7
8
9
modulation, and potential pharmacological effect at lower concentrations (e.g. ibrutinib).14 Like
PROTAC technology, the durability of the pharmacological effect depends on the synthesis of
new protein but also includes a possible dominant negative effect due to the durability of the
drugꢀtarget protein. Target specificity is obtained by combining a selective nonꢀcovalent
inhibitor with a weakly reactive warhead that can target a proximal, typically nucleophilic amino
acid such as lysine, serine, threonine or cysteine, with the latter particularly attractive due to its
relatively low abundance leading to greater specificity. Examples include ibrutinib, afatinib and
rociletinib.15 Zhong et al. have described the development of irreversible BET inhibitors.11 A
metalꢀbased BET inhibitor was designed to bind irreversibly to histidine amino acid residues of
BRD4. Even though the approach of the irreversible covalent BET inhibition was novel, the Xꢀ
ray coꢀcrystal structure of the obtained adduct remains elusive and the exact interacting amino
acid residue has yet to be determined. Daguer at al. have described promiscuous covalent probes
capable of binding to cysteine amino acid residues across the family of bromodomains
suggesting the possibility of the covalent targeting of several bromodomains through this
approach.16
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
To date, methionine has not been considered an amino acid to target with a covalent inhibitor.
Here we describe the rational design and characterization of a novel series of covalent BET
inhibitors targeting amino acid residue Met149 in the binding site of BRD4, which demonstrate
promising durable inhibitory effects on proliferation in vitro and sustained effects on target
modulation. We show beneficial effects of our covalent approach for BET inhibition, which can
potentially lead to novel and more effective cancer therapy as well as providing an alternative
strategy for covalent inhibitor design.
4
ACS Paragon Plus Environment

Page 5 of 43
Journal of Medicinal Chemistry
1
2
3
4
RESULTS AND DISCUSSION
5
6
7
8
9
In order to assess the possibility of rationally designing a covalent BET inhibitor, we evaluated
various coꢀcrystal structures of inhibitors with BRD4(BD1) including the coꢀcrystal structure of
1 (ZEN-2759) bound to BRD4(BD1), which was obtained at 1.88 Å resolution (Figure 1). The
crystal structure shows that the dimethylꢀisoxazole group of 1 is involved in key interactions
within the binding site, by forming a Hꢀbond (3.1 Å) with the amide nitrogen of Asn140 and a
waterꢀmediated Hꢀbond to the hydroxyl group of Tyr97. While there was not a cysteine located
near the binding site, the benzyl’s 4ꢀposition of 1 was positioned in good proximity to the
Met149 residue (3.9 Å). While methionine has never been targeted before for covalent drug
design, Alferiev et al. have previously described epoxides with high reactivity towards
methionine and other aliphatic sulfides in physiological conditions, which resulted in the
formation of sulfonium complexes,17 suggesting that it might be possible to attach an epoxide to
a reversible inhibitor to obtain a covalent drug. We then docked proposed substitutions of 1 into
the binding site of BRD4(BD1). In those cases where the epoxide appeared close enough for
Met149 to react, the expected bond was formed and the complex minimized by a small sphere of
free atoms. Those compounds that were predicted to exert minimal strain in the binding site in
both noncovalent and reacted states were then considered for synthesis (Figure 2). A similar
process was conducted using in silico methods with 2 and 3.
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
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 43
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
Figure 1. The chemical structure of 1 (ZEN-2759) (top left). Coꢀcrystal structure of
BRD4(BD1) and 1, pdb code 6CZV.pdb, showing specific proteinꢀligand interactions (bottom
left) and a surface representation (right).
Figure 2. Chemical structures of epoxide containing analogues of BET inhibitors.
6
ACS Paragon Plus Environment

Page 7 of 43
Journal of Medicinal Chemistry
1
2
3
4
Biochemical characterization of BET inhibitors
5
6
7
8
9
To evaluate whether the epoxide analogues retained the binding affinity of the unmodified
parent compounds to BRD4 bromodomains, we tested their ability to displace acetylated lysine
histone 4 peptide from BRD4 bromodomains using an AlphaScreen assay, which mimics the
displacement of acetylated H4 lysine from the BRD4 binding site by BET inhibitors. Table 1
demonstrates that the epoxide analogues strongly inhibited the interaction between the tetraꢀ
acetylated histone H4 Lys peptide and both individual and tandem dual bromodomains of BRD4
with comparable affinity to the parent compounds.
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
Table 1. Biochemical characterization of BET inhibitors. IC50s determined by AlphaScreen
assay measuring the doseꢀdependent inhibition of the interaction between BRD4 bromodomains,
either single or in tandem, and tetraꢀacetylated H4 peptide.
Average of IC50, µM
Compound number
BRD4(BD1)
0.06
BRD4(BD2)
0.04
BRD4(BD1BD2)
0.05
0.04
0.28
0.47
1.41
nd
JQ1
0.04
0.02
IBET
0.23
0.08
1 (ZEN-2759)
1A (ZEN-3219)
2 (ZEN-2906)
2A (ZEN-3862)
3 (ZEN-3212)
3A (ZEN-3411)
0.48
0.16
1.07
0.03
0.16
0.13
0.02
0.01
0.05
0.06
0.05
0.05
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 43
1
2
3
4
5
6
7
8
9
In order to evaluate the direct binding of the compounds to BET proteins, we performed a
thermal denaturation protein assay, which measures the increases in protein stability upon
compound binding to the protein of interest (Figure 3). We observed that after 30 min incubation
the interaction of the nonꢀcovalent 2 with BRD4(BD1) resulted in a moderate thermal shift in
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
o
Tm (ꢁTm) of ~ 5 C, which is indicative of an increased protein stability upon compound
binding (Figure 3, left panel). However, the incubation of BRD4(BD1) with its epoxide
o
o
analogue, 2A, resulted in two distinct thermal transitions with ꢁTms of 7 C and 22 C,
respectively. If incubation is continued for 4 h, the complex represented by the lower Tm is
converted to the higher Tm (Figure 3 right panel). Similar differences in Tm are seen when 1 is
compared to 1A and 3 is compared to 3A (Figure 3 top and middle panels). Because the lower
Tm of 2A is similar to that of 2, the data are consistent with a twoꢀstep process in which a
reversible complex is formed first followed by a reaction that yields an adduct with significantly
higher Tm (Figure 3 lower panel). Since the high Tm is higher than that seen with a very potent
inhibitor such as JQ1,18 it is very likely a covalent complex. Such twoꢀstep binding process was
observed for all epoxyꢀcontaining BET inhibitors.
8
ACS Paragon Plus Environment

Page 9 of 43
Journal of Medicinal Chemistry
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
Figure 3. Interaction of BET inhibitors with BRD4(BD1) measured by thermal denaturation
assay.
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 43
1
2
3
4
5
6
7
8
9
Even though the covalent compounds displayed comparable potencies on BD1 vs. BD2 domains
in the AlphaScreen assay, the high Tm shifts were observed with only BD1 domains. Incubations
of the covalent compounds with the BD2 domains led to Tm shifts comparable to the parent nonꢀ
covalent compounds with no higher Tm species forming (Figure S1). Because BD2 contains a
Met in a similar position to BD1, this suggests that the presence of Met alone is not a
determinant of covalent binding, and additional structural and/or reactivity considerations may
play a role. Incubation of covalent compounds or nonꢀcovalent controls with several nonꢀBET
bromodomains having methionine, cysteine or other amino acid in the corresponding position
also demonstrated no high melting Tm species at either 30 min or 4 h, suggesting that the
covalent inhibitors are selective for the BET bromodomain family (Figure S2).
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
To provide further evidence that the third epoxide analogue 3A bound irreversibly to
BRD4(BD1), we monitored its IC50 over time by competitive AlphaScreen assay (Table 2).
While the IC50 of its parent 3 did not change over 4 h, consistent with reversible inhibition, the
apparent IC50 of 3A decreased almost 18ꢀfold over 4 h, which is consistent with it making an
irreversible, covalent interaction with BRD4(BD1).19
10
ACS Paragon Plus Environment

Page 11 of 43
Journal of Medicinal Chemistry
1
2
3
4
Table 2. Timeꢀdependence of biochemical IC50 values of BET inhibitors
5
6
7
AlphaScreen (average of IC50, µM)
8
9
IC50 fold change
Compound number
5 min
0.017
0.274
30 min
0.016
0.115
1 h
4 h
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
3
0.008
0.075
0.019
0.015
1
(non-covalent)
3A
18
(covalent)
1A Binds Covalently to BRD4(BD1)
To provide definitive evidence that 1A forms a covalent linkage to BRD4(BD1), we prepared the
complex of BRD4(BD1) with 1A based on the kinetics observed in Figure 3. This complex was
o
stable under various pH conditions (pH=5, 7.4, 9), incubation at room temperature and 37 C,
and multiple freezeꢀthaw cycles, as evidenced by retention of the single high Tm peak BRD4
(Figure S3). Similarly, incubation in the presence of excess free methionine or cysteine suggests
that the covalent complexes are stable and not reversible (Figure S4).
MALDIꢀTOF mass spectrometry was conducted on the complex to confirm the covalent
complex formation. In addition to a peak corresponding to the mass of unbound BRD4(BD1), we
observed a higher mass peak which differed in mass by +325 Da, which is consistent with the
formation of a covalent sulfonium linkage as expected from Aferiev et al17. (Figure 4).
Furthermore, there was a peak with lower mass than BRD4(BD1), suggesting that it results from
the loss of a chemical fragment upon ionization in the mass spectrometer. The difference in mass
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 43
1
2
3
4
5
6
of 52 Da is consistent with the parent and degradation ions being products from cleavage on
either side of the sulfonium ion of the covalent adduct.
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
Figure 4. MALDIꢀTOF spectrometry of 1A/BRD4(BD1) complex vs. apo BRD4(BD1).
The Xꢀray coꢀcrystal structure of BRD4(BD1) with 1A was obtained and resolved at 1.47 Å
resolution (Figure 5). 1A has an epoxide linked to the benzyl’s 4ꢀposition. The epoxide forms a
covalent link with side chain Sγ atom of Met149, resulting in a sulfonium ion. The distance
between the carbon (named C18) of the isopropanol carbon linker connecting with the Met149 Sγ
atom is 1.87 Å, which is consistent with a typical sulfurꢀcarbon covalent bond.17 The SꢀC
covalent bond is elongated by 0.05 Å from the ideal value (1.82 Å), which is well within the
range of literature precedents for this type of bond.19 The electron density around the isopropyl
chain (near the covalent bond) clearly indicates that there is only one enantiomeric form of the
compound present at the binding pocket. The oxygen substitution off the central pyridone ring
forms a waterꢀmediated Hꢀbond with the Lys57 residue side chain and this water molecule
12
ACS Paragon Plus Environment

Page 13 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
further mediated an interaction with the nitrogen atom of the Gln85 residue side chain amide
nitrogen. This scaffold also forms van der Waals interactions with Trp81, Pro82 and Leu92. The
dimethylꢀisoxazole accepts an Hꢀbond (3.1 Å) from Asn140 amide nitrogen and also forms a
waterꢀmediated Hꢀbond to Tyr97 hydroxyl group. This isoxazole ring is positioned between
Val87 and Ile146 side chains. The resolved coꢀcrystal structure demonstrates that 1A binds
covalently to Met149 in the binding pocket as predicted. Overall, the coꢀcrystal Xꢀray structure
clearly demonstrates the covalent bond formation between 1A and Sγ atom of Met149 in the
BRD4 binding site.
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
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 43
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
Figure 5. XꢀRAY coꢀcrystal structure of 1A (ZEN-3219) with BRD4(BD1), pdb code
6CZU.pdb. A. Interacting amino acid residues of BRD4(BD1) with 1A. B. Surface
representation of 1A binding mode in BRD4(BD1) binding pocket. C. Electron density map
(SigmaAꢀweighted 2FoꢀFc in teal colour at 1.0 σ contour level) corresponding to compound 1A
bound to BRD4(BD1) and a covalent link between the compound and Sγ atom of Met149. D.
Ligand Interaction Diagram (LID) demonstrating the covalent interaction of 1A and Met149 of
BRD4(BD1). Hydrogen bonds are designated with a dashed line and van der Waals interactions
are shown by a half “sun” arrangement.
14
ACS Paragon Plus Environment

Page 15 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
We then overlaid the BRD4(BD1)/1A complex with the BRD4(BD1)/1 complex and observed
that the overall orientation of 1A is shifted in comparison to the parental compound (Figure 6).
The anchoring dimethylꢀisoxazole ring of 1A was shifted ~0.7 Å compared to bound 1. The
pyridone ring is also significantly shifted 1.5 Å and rotated by ~17° degrees in the
1A/BRD4(BD1) complex probably due to the pulling effect of the SꢀC covalent bond formation.
The phenyl ring substituent is also significantly shifted and rotated by 1.6 Å and ~ 2ꢀ7 °,
respectively. The pulling effect of the covalent bond caused the long range shift in the Trp81,
Asp145 and Phe79 residues pulling the amino acid residues to ~1.0 Å, ~0.9 Å and ~1.3 Å away
from the compound.
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
Figure 6. Structural overlay of BRD4(BD1)/1A (pink) and 1 (green).
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 43
1
2
3
4
5
6
Cellular effects of the covalent inhibition: Sustained effect on proliferation in AML cell line
MV4-11
7
8
9
An important benefit of the covalent inhibitors may be an extended inactivation of the protein
target, potentially leading to prolonged duration of action and sustained biological effects. This
could be mediated by covalently bound BET protein acting in a dominant negative fashion to
sequester key transcriptional components which is offset by the speed by which newly
synthesized BET protein becomes available to replenish its function after inhibitor is no longer
present. To determine whether BET protein had a long halfꢀlife to mediate durable inhibition, we
determined the halfꢀlife of BET proteins by cycloheximide treatment of the acute myeloid
leukemia (AML) cell line OCIꢀAML3 for 3, 6, 24, 48, 72 hours. The estimated halfꢀlife of all
BET bromodomains was ≥ 48h which is sufficiently long to be able to form a sustained covalent
interaction with the irreversible BET inhibitors (Figure S5). To elucidate whether the covalent
BET inhibition results in durable antiꢀproliferative effects, the AML cell line MV4ꢀ11 was
treated with covalent inhibitor vs. nonꢀcovalent control in continuous and washout treatments,
respectively (Figure 7). Continuous treatment of either covalent 2A or nonꢀcovalent 2 control
showed similar effects on inhibition of proliferation after 48 hours. However, 2A demonstrated
longer durability of antiꢀproliferative effects in MV4ꢀ11 cells when the compound was removed
after 24 h and cells incubated for a further 48 h without inhibitor. A similar effect on durability
of proliferation was observed with 3A covalent inhibitor (Figure S6).
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
16
ACS Paragon Plus Environment

Page 17 of 43
Journal of Medicinal Chemistry
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
Figure 7. Effect of covalent binding on durability of proliferation in MV4ꢀ11 cells. Left panel:
Proliferation of MV4ꢀ11 was measured after 48 h of continuous treatment with ZEN compounds.
Right panel: MV4ꢀ11 cell line was treated with 2 (reversible inhibitor) and 2A (irreversible
inhibitor) for 24 h before compounds were removed, replaced with media and proliferation was
measured at 48 h.
Next, we then evaluated whether the inhibition of proliferation by a covalent inhibitor and its
nonꢀcovalent control is due to apoptosis in MV4ꢀ11 (Figure 8). Continuous treatment for 72 h
with either covalent inhibitor 3A or its nonꢀcovalent control 3 led to the significant induction of
apoptosis of 90 and 83%, respectively. In contrast, neither compound had any cytotoxic effect on
viability of normal PBMCs (Figure S7).
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 43
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
Figure 8. Covalent inhibitor 3A and its nonꢀcovalent analogue 3 induce apoptosis in AMLꢀcell
line MV4ꢀ11.
Irreversible inhibitor elicits sustained repression of PD markers in hematological cancer
cell lines
Furthermore, we evaluated whether the treatment with a covalent inhibitor would lead to
sustained repression of previously identified target genes MYC, BCL2, CCR1, GRP183, and
IL1RN, which are currently being used to measure target engagement to BET inhibitor ZENꢀ
3694 in the ongoing clinical trial (NCT02711956). We treated MV4ꢀ11 for 4 hours with
irreversible inhibitor 2A or with the control compound 2, after which compounds were removed.
Postꢀwashout durability of mRNA repression of MYC, BCL2, CCR1, GRP183, and IL1RN was
then timeꢀdependently assessed (Figure 9). 2A repressed all PD markers significantly longer
when compared to the nonꢀcovalent 2 control treatment. These data are consistent with 2A
forming an irreversible, covalent linkage with BRD4 in cells.
18
ACS Paragon Plus Environment

Page 19 of 43
Journal of Medicinal Chemistry
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
Figure 9. Effect of covalent binding of 2A on durability of repression of PD markers in a
washout time course versus its nonꢀcovalent analogue 2. MV4ꢀ11 cells were treated for 4 h with
irreversible or reversible inhibitor. After 4 h compounds were removed, cells washed, replaced
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 43
1
2
3
4
5
6
with media and cells were harvested at 4, 6 and 24 h postꢀwashout and expression of PD markers
was determined.
7
8
9
CONCLUSIONS
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
Here, we report for the first time the design and characterization of covalent protein inhibitors
that target methionine, and illustrate the concept through the design of BET inhibitors that
selectively target Met149 in the BRD4(BD1) binding pocket. Lin et al. have recently described a
similar approach for enabling methionine bioconjugation using an oxaziridine group, though
their approach is not focused on inhibitor design.21 This covalent interaction was confirmed by
thermal shift assay, MALDIꢀTOF spectrometry and highꢀresolution Xꢀray crystal structure of the
protein complex. Our findings demonstrate that covalent inhibition leads to durable antiꢀ
proliferative effects and decreased mRNA expression of several BETꢀdependent genes when
compared to reversible BET bromodomain inhibitors. Covalent BET inhibitors may offer a novel
and potentially superior approach to BET inhibition with long lasting duration of action, durable
target modulation, and potential pharmacological effects at lower concentrations in future
oncology applications. Targeting methionine may provide an alternative strategy for the design
of covalent inhibitors for target evaluation and drug design.
20
ACS Paragon Plus Environment

Page 21 of 43
Journal of Medicinal Chemistry
1
2
3
4
EXPERIMENTAL
5
6
7
1. Synthetic Materials and Methods
8
9
Unless otherwise noted, reagents and solvents were used as received from commercial suppliers.
Proton nuclear magnetic resonance spectra were obtained on a Bruker ADVANCE 300, 400 or
500 spectrometer. Spectra are given in ppm (δ) and coupling constants, J values, are reported in
hertz (Hz). Mass spectra analyses were performed on a Waters Aguity UPLC, Agilent 6130A,
Applied Biosystems APIꢀ150EX, or a Shimadzu 2020 instrument in ESI or APCI mode when
appropriate. The purity of the compounds was analyzed by HPLC, ESIꢀMS and was >95 %. The
HPLC traces for purity are shown in the supplemental information (Figure S8). Molecular
Formula Strings are shown in Table S1.
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
1-Benzyl-5-(3,5-dimethylisoxazol-4-yl)pyridin-2(1H)-one 1 (ZEN-2759).
To a solution of 5ꢀbromoꢀ1,2ꢀdihydropyridinꢀ2ꢀone (150 mg, 0.86 mmol) in acetonitrile (3 mL),
benzyl bromide (1.03 mmol) and potassium carbonate (237 mg, 1.72 mmol) was added. The
o
reaction mixture was heated at 80 C for 16 h, the reaction mixture was cooled to room
temperature, concentrated and purified by chromatography (silica gel, 0–20% ethyl aceꢀ
tate/hexanes) to give 1ꢀbenzylꢀ5ꢀbromoꢀ1,2ꢀdihydropyridinꢀ2ꢀone to which 3,5ꢀdimethylꢀ4ꢀ
(4,4,5,5ꢀtetramethylꢀ1,3,2ꢀdioxaborolanꢀ2ꢀyl)isoxazole (170 mg, 0.76 mmol), 2 M aq. sodium
carbonate (0.48 mL, 0.95 mmol) and tetrakis(triphenylphosphine)palladium(0) (37 mg, 0.032
o
mmol) were added. The reaction mixture was purged with nitrogen and heated at 80 C for 16 h,
diluted with methylene chloride (30 mL) and washed with brine (2 × 10 mL). The organic layer
was dried over sodium sulfate, filtered and concentrated. Purification by chromatography (silica
1
gel, 0–45% ethyl acetate/hexanes) yielded 1 : (120 mg, 43%). H NMR (300 MHz, CDCl3) δ
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 43
1
2
3
4
5
6
7.31–7.41 (m, 5H); 7.25 (dd, J = 2.7, 9.3 Hz, 1H); 7.13 (d, J = 2.4 Hz, 1H); 6.76 (d, J = 9.3 Hz,
1H); 5.20 (s, 2H); 2.29 (s, 3H); 2.15 (s, 3H); ESI MS m/z 281 [M + H]+; HPLC purity: >99%.
7
8
9
5-(3,5-Dimethylisoxazol-4-yl)-1-(2-fluorobenzyl)pyridin-2(1H)-one 2 (ZEN-2906)
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
To a solution of 5ꢀbromoꢀ1,2ꢀdihydropyridinꢀ2ꢀone (150 mg, 0.86 mmol) in acetonitrile (3 mL),
1ꢀ(bromomethyl)ꢀ2ꢀfluorobenzene (1.03 mmol) and potassium carbonate (237 mg, 1.72 mmol)
was added. The reaction mixture was heated at 80 oC for 16 h. The reaction mixture was cooled
to room temperature, concentrated and purified by chromatography (silica gel, 0–20% ethyl aceꢀ
tate/hexanes) to give 5ꢀbromoꢀ1ꢀ[(2ꢀfluorophenyl)methyl]ꢀ1,2ꢀdihydropyridinꢀ2ꢀone (180 mg,
1
74%) as a white solid: H NMR (300 MHz, DMSO–d6) δ 8.10 (d, J = 2.8 Hz, 1H), 7.58 (dd, J =
9.9, 2.8 Hz, 1H), 7.42–7.30 (m, 1H), 7.28–7.08 (m, 3H), 6.42 (d, J = 9.7 Hz, 1H), 5.12 (s, 2H).
To a solution of 5ꢀbromoꢀ1ꢀ[(2ꢀfluorophenyl)methyl]ꢀ1,2ꢀdihydropyridinꢀ2ꢀone (0.64 mmol) in
1,4ꢀdioxane (4 mL), 3,5ꢀdimethylꢀ4ꢀ(4,4,5,5ꢀtetramethylꢀ1,3,2ꢀdioxaborolanꢀ2ꢀyl)isoxazole (170
mg, 0.76 mmol),
2
M
aq. sodium carbonate (0.48 mL, 0.95 mmol) and
tetrakis(triphenylphosphine)palladium(0) (37 mg, 0.032 mmol) were added. The reaction mixture
was purged with nitrogen and heated at 80 oC for 16 h. The mixture was diluted with methylene
chloride (30 mL) and washed with brine (2 × 10 mL). The organic layer was dried over sodium
sulfate, filtered and concentrated.
Purification by chromatography (silica gel, 0–45% ethyl acetate/hexanes) yielded 2 (48 mg,
25%) as a white solid: 1H NMR (300 MHz, DMSO–d6) δ 7.96 (d, J = 2.1 Hz, 1H), 7.52 (dd, J1 =
6.6 Hz, J2 = 2.7 Hz, 1H), 7.46–7.36 (m, 1H), 7.24–7.06 (m, 3H), 7.96 (d, J = 2.1 Hz, 1H), 5.14
(s, 2H), 2.36 (s, 3H), 2.19 (s, 3H); ESI m/z 299 [M + H]+; HPLC purity: >99%
1-Benzyl-6-(3,5-dimethylisoxazol-4-yl)-lH-benzo[d]imidazol-4-amine 3 (ZEN-3212)
22
ACS Paragon Plus Environment

Page 23 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
A mixture of 6ꢀbromoꢀ1Hꢀ1,3ꢀbenzodiazolꢀ4ꢀamine (1.0 g, 4.7 mmol), 3,5ꢀdimehylꢀ4ꢀ(4,4,5,5ꢀ
tetramethylꢀ1,3,2ꢀdioxaborolanꢀ2ꢀyl)ꢀ1,2ꢀoxazole (4.5 g, 6.6 mmol), K2CO3 (1.3 g, 9.4 mmol),
1,4ꢀdioxane (48 mL) and water (3.5 mL) was degassed with nitrogen for 20 minutes followed by
addition of tetrakis (triꢀphenylphosphine)palladium(0) (550 mg, 0.94 mmol). The mixture was
heated at 90 °C for 18 hours then cooled to room temperature. The crude reaction mixture was
purified by chromatography (silica gel, 0–10% methanol/dichloromethane) to obtain 6ꢀ(3,5ꢀ
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
1
dimethylisoxazolꢀ4ꢀyl)ꢀ1Hꢀbenzo[d] imidazolꢀ4ꢀamine (760 mg, 68%) as an offꢀwhite solid: H
NMR (500 MHz, CDCl3) δ 7.98 (s, 1H), 6.74 (s, 1H), 6.40 (s, 1H), 4.45 (s, 2H), 2.38 (s, 3H),
2.25 (s, 3H); ESI m/z 229 [M + H]+.
To a solution of 6ꢀ(3,5ꢀdimethylisoxazolꢀ4ꢀyl)ꢀ1Hꢀbenzo[d] imidꢀazolꢀ4ꢀamine (290 mg, 1.27
mmol) in CH3CN (15 mL), potassium carbonate (350 mg, 2.54 mmol) and benzyl chloride (200
mg, 1.59 mmol) were added. The reaction mixture was stirred at 60 °C for 16 h. The mixture was
diluted with methylene chloride (20 mL) and filtered through a layer of Celite. The filtrate was
concentrated and purified by chromatography (silica gel, 0ꢀ10% CH3OH/CH2Cl2) to obtain 3
1
(109 mg, 27%) as an offꢀwhite solid: H NMR (300 MHz, CDCl3) δ 7.95 (s, 1H), 7.37ꢀ7.34 (m,
3H), 7.23ꢀ7.20 (m, 2H), 6.46 (d, J = 1.2 Hz, 1H), 6.40 (d, J = 1.2 Hz, 1H), 5.34 (s, 2H), 2.31 (s,
3H), 2.16 (s, 3H); ESI MS m/z 319 [M + H]+; HPLC purity: >99%
5-(Dimethyl-1,2-oxazol-4-yl)-1-{[4-(oxiran-2-yl)phenyl]methyl}-1,2-dihydropyridin-2-one
1A (ZEN-3219)
A mixture of compound of 5ꢀ(3,5ꢀdimethylisoxazolꢀ4ꢀyl)pyridinꢀ2(1H)ꢀone (100 mg, 0.49
mmol), 2ꢀ(4ꢀ(chloromethyl)phenyl) oxirane (90 mg, 0.54 mmol) and K2CO3 (135 mg, 0.98
mmol) in CH3CN (6 mL) was stirred at 70 °C for 16 h. The reaction mixture was allowed to cool
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 43
1
2
3
4
to room temperature and was filtered through a layer of Celite. The filtrate was concentrated and
5
6
1
purified on silica gel to obtain the desired product 1A as an offꢀwhite solid (94 mg, 57%). H
7
8
9
NMR (300 MHz, CDCl3) δ 7.40–7.12 (m, 6H), 6.64 (d, J = 9.3 Hz, 1H), 5.17 (s, 2H), 3.85–3.84
(m, 1H), 3.16–3.13 (m, 1H), 2.78–2.74 (m, 1H), 2.30 (s, 3H), 2.15 (s, 3H); ESI MS m/z 323 [M +
H]+; HPLC purity: >99%.
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
6-(Dimethyl-1,2-oxazol-4-yl)-1-{[4-(oxiran-2-yl)phenyl]methyl}-1H-1,3-benzodiazol-4-
amine 3A (ZEN-3411)
A mixture of 6ꢀbromoꢀ1Hꢀ1,3 benzodiazolꢀ4ꢀamine (1.0 g, 4.7 mmol), 3,5ꢀdimehylꢀ4ꢀ(4,4,5,5ꢀ
tetramethylꢀ1,3,2ꢀdioxaborolanꢀ2ꢀyl)ꢀ1,2ꢀoxazole (4.5 g, 6.6 mmol), K2CO3 (1.3 g, 9.4 mmol),
1,4ꢀdioxane (48 mL) and water (3.5 mL) was degassed with nitrogen for 20 minutes followed by
addition of tetrakis (triꢀphenylphosphine) palladium(0) (550 mg, 0.94 mmol). The mixture was
heated at 90 °C for 18 hours then cooled to room temperature. The crude reaction mixture was
purified by chromatography (silica gel, 0–10% methanol/dichloromethane) to obtain 6ꢀbromoꢀ
1Hꢀ1,3 benzodiazolꢀ4ꢀamine 6ꢀ(3,5ꢀdimethylꢀ1,2 –oxazolꢀ4ꢀyl)ꢀ1Hꢀ1,3ꢀbenzodiazolꢀ4ꢀamine
1
(760 mg, 68%) as an offꢀwhite solid: H NMR (500 MHz, CDCl3) δ 7.98 (s, 1H), 6.74 (s, 1H),
6.40 (s, 1H), 4.45 (s, 2H), 2.38 (s, 3H), 2.25 (s, 3H); ESI m/z 229 [M + H]+. To a solution of 6ꢀ
(3,5ꢀdimethylisoxazolꢀ4ꢀyl)ꢀ1Hꢀbenzo[d] imidazolꢀ4ꢀamine (100 mg, 0.44 mmol) in CH3CN (5
mL) was added potassium carbonate (121 mg, 0.88 mmol) and 2ꢀ(4ꢀ(chloromethyl)phenyl)
oxirane (81 mg, 0.48 mmol). The reaction mixture was stirred at 60 °C for 16 h. The mixture
was diluted with methylene chloride (20 mL) and filtered through a layer of Celite. The filtrate
was concentrated and purified by chromatography (silica gel, 0–10% CH3OH/CH2Cl2) followed
1
by reverse phase comiflash to obtain 3A (70 mg, 44%) as an offꢀwhite solid: H NMR (300
MHz, CD3OD) δ 8.18 (s, 1H), 7.27 (s, 4H), 6.55 (d, J = 1.2 Hz, 1H), 6.45 (d, J = 1.5 Hz, 1H),
24
ACS Paragon Plus Environment

Page 25 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
5.44 (s, 2H), 3.84 (dd, J = 4.2, 2.7 Hz, 1H), 3.09 (dd, J = 5.4, 4.2 Hz, 1H), 2.74 (dd, J = 5.4, 2.7
Hz, 1H), 2.24 (s, 3H), 2.14 (s, 3H); ESI MS m/z 361 [M + H]+; HPLC purity: 97.3%.
7
8
9
5-(Dimethyl-1,2-oxazol-4-yl)-1-(3-fluoro-4-(oxiran-2-yl)benzyl)pyridin-2(1H)-one 2A
(ZEN-3862)
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
A reaction mixture of (4ꢀbromoꢀ3ꢀfluorophenyl)methanol (500 mg, 2.44 mmol), pyridine (578
mg, 7.32 mmol) and MsCl (559 mg, 4.88 mmol) in CH2Cl2 (10 mL) was stirred at rt for 16 h.
The reaction mixture was diluted with CH2Cl2, and it was filtered through a layer of Celite. The
filtrate was concentrated to give 1ꢀbromoꢀ4ꢀ(chloromethyl)ꢀ2ꢀfluorobenzene as an offꢀwhite
1
solid (4.2 g, 95%): H NMR (500 MHz, CDCl3) δ 7.63–7.43 (m, 3H), 6.31 (s, 2H). A mixture of
5ꢀ(3,5ꢀdimethylisoxazolꢀ4ꢀyl)pyridinꢀ2(1H)ꢀone
(285 mg, 1.5 mmol) which was prepared
according to literature procedure
(WO 2014096965, compound 1421), 1ꢀbromoꢀ4ꢀ
(chloromethyl)ꢀ2ꢀfluorobenzene (368 mg, 1.65 mmol) and K2CO3 (414 mg, 3.0 mmol) in
CH3CN (10 mL) was stirred at 70 °C for 16 h. The reaction mixture was allowed to cool to room
temperature and filtered through a layer of Celite. The filtrate was concentrated and purified on
silica gel to obtain 1ꢀ[(4ꢀbromoꢀ3ꢀfluorophenyl)methyl]ꢀ5ꢀ(3,5ꢀdimethylꢀ1,2ꢀoxazolꢀ4ꢀylꢀ1,2ꢀ
dihydropyridinꢀ2ꢀone as an offꢀwhite solid (180 mg, 32%). 1H NMR (300 MHz, CDCl3) δ
7.59–7.54 (m, 1H), 7.30–7.26 (m, 2H), 7.17–7.12 (m, 2H), 7.03 (d, J = 8.1 Hz, 1H), 5.14 (s, 2H),
2.36 (s, 3H), 2.22 (s, 3H); ESI MS m/z 378 [M + H]+. To a solution of 1ꢀ[(4ꢀbromoꢀ3ꢀ
fluorophenyl)methyl]ꢀ5ꢀ(3,5ꢀdimethylꢀ1,2ꢀoxaꢀzolꢀ4ꢀylꢀ1,2ꢀdihydropyridinꢀ2ꢀone (180 mg, 0.48
mmol) in DMF (5 mL) was added tributyl(vinyl)stannane (197 mg, 0.62 mmol) and
Pd(PPh3)2Cl2 (34 mg, 0.048 mmol). The reaction mixture was purged with nitrogen and heated at
85 °C for 16 h. The mixture was filtered through a layer of Celite. The filtrate was concentrated.
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 43
1
2
3
4
5
6
Purification by chromatography (silica gel, 0–40% ethyl acetate/dichloromethane) left 5ꢀ(3,5ꢀ
dimethylꢀ1,2ꢀoxazolꢀ4ꢀyl)ꢀ1ꢀ[(4ꢀethenylꢀ3ꢀfluorophenyl)methyl]ꢀ1,2ꢀdihydropyridinꢀ2ꢀone (83
7
8
1
mg, 53%) as an orange oil: H NMR (300 MHz, CDCl3) δ 7.51–7.46 (m, 1H), 7.26–7.25 (m,
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
1H), 7.14 (d, J = 1.8 Hz, 1H), 7.09–7.00 (m, 2H), 6.91–6.74 (m, 2H), 5.82 (d, J = 17.7 Hz, 1H),
5.40 (d, J = 11.1 Hz, 1H), 5.15 (s, 2H), 2.32 (s, 3H), 2.18 (s, 3H); ESI MS m/z 325 [M + H]+. 5ꢀ
(3,5ꢀdimethylꢀ1,2ꢀoxazolꢀ4ꢀyl)ꢀ1ꢀ[(4ꢀethenylꢀ3ꢀfluorophenyl)methyl]ꢀ1,2ꢀ dihydroꢀpyridinꢀ2ꢀone
was dissolved in acetone (3 mL) and water (1 mL) at 0 °C and combined with NBS (50 mg, 0.29
mmol). The reaction mixture was stirred at rt for 6 h. To the mixture above at 0 °C, NaOH (21
mg, 0.52 mmol) was added and the mixture was stirred at rt for 16 h. The reaction mixture was
filtered through a layer of Celite. The filtrate was concentrated. Purification by chromatography
(silica gel, 0–40% ethyl acetate/dichloromethane) left 5ꢀ(dimethylꢀ1,2ꢀoxazolꢀ4ꢀyl)ꢀ1ꢀ(3ꢀfluoroꢀ
4ꢀ(oxiranꢀ2ꢀyl)benzyl) pyridinꢀ2(1H)ꢀone (53 mg, 60%) as an offꢀwhite solid. 1H NMR (500
MHz, DMSO–d6) δ 7.94 (d, J = 2.5 Hz, 1H), 7.50 (dd, J = 9.0, 2.5 Hz, 1H), 7.24 (d, J = 11.5 Hz,
1H), 7.19–7.17 (m, 2H), 6.51 (d, J = 9.0 Hz, 1H), 5.12 (s, 2H), 4.08 (dd, J = 4.0, 2.5 Hz, 1H),
3.14 (dd, J = 5.5, 4.0 Hz, 1H), 2.87 (dd, J = 5.5, 2.5 Hz, 1H), 2.36 (s, 3H), 2.19 (s, 3H); ESI m/z
341 [M + H]+; HPLC purity: 98.1%.
2. AlphaScreen
Single and dual bromodomain constructs with an Nꢀterminal Hisꢀtag were cloned, expressed, and
purified by nickel affinity and sizeꢀexclusion chromatography by Genscript or Xtal
BioStructures. Fractions representing monomeric protein were pooled and frozen at ꢀ80 °C for
use in subsequent experiments. The AlphaScreen® Histidine (Nickel Chelate) Detection Kit
(Perkin Elmer, 6760619M) containing 1 mL of Streptavidin Donor Beads, 1 mL of Nickel
Chelate Acceptor Beads, and 1.5 mL of 10X Buffer was used according to the manufacturer’s
26
ACS Paragon Plus Environment

Page 27 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
instructions. Briefly, Nꢀterminally Hisꢀtagged single and dual bromodomains (20ꢀ200 nM) and
biotinylated tetraꢀacetylated histone H4 peptide (3ꢀ75 nM) (Millipore 12ꢀ379) were mixed in 50
mM HEPES, 100 mM NaCl, and 0.1% bovine serum albumin buffer, pH 7.4, under green light.
Nickel chelate acceptor beads and streptavidin donor beads were added to a final concentration
of 2 µg/mL. Serially diluted compounds were added to the reaction mixture in a white 96 well
plate (Greiner). After 30 min incubation at room temperature, luminescence was measured at 570
nm by a SynergyH4 plate reader (Biotek). IC50 values were determined from a dose response
curve.
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
3. Protein Thermal Denaturation Assay
Binding of purified Nꢀterminally Hisꢀtagged recombinant BRD4(BD1) to ZEN inhibitors was
investigated using the protein thermal shift assay (PTS™, Life Technologies). 5 µM of bromoꢀ
domains were incubated with 100 µM of the irreversible inhibitors or the control compounds in
the presence of 5X SYPRO orange protein stain (Life Technologies Sꢀ6650) for 30 min or 4
hours in 50 mM NaCl/ 50 mM HEPES buffer, pH 7.4. Melting temperatures (Tm) were
calculated using Applied Biosystems Protein Thermal Shift™ Software v1.1.
4. MALDI-TOF characterization
Analyses were performed on a Shimadzu Biotech Axima TOF2 (Shimadzu Instruments) matrixꢀ
assistedꢀlaser desorption/ ionization TimeꢀofꢀFlight (MALDIꢀTOF) mass spectrometer. Proteins
were analyzed in positive ion Linear mode. For intact protein mass measurement the instrument
was set with a mass range extending to 50 kd using a pulsed extraction setting of 16952.
Apomyoglobin MH+ of 16952 and MH2+ of 8476 were used as the standard to calibrate the
instrument. A 0.5 µL aliquot was applied to the MALDI target followed by the addition of 0.5
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 43
1
2
3
4
5
6
µL of Sinapinic acid as the desorption matrix (10 mg/mL in 0.1% TFA: Acetonitrile 50:50)
followed by air drying prior to insertion into the vacuum source.
7
8
9
5. X-ray Crystallography
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
The complex with 1 (ZEN-2759) was prepared by adding 2 mM 1 to 1 mg/mL
BRD4(BD1)(BRD4 amino acids 44ꢀ170). The mixture was incubated for 30 minutes at room
temperature; concentrated to 7 mg/mL; and setup with broad formulation screens using the
sitting drop vapor diffusion method. Coꢀcrystals grew to usable size in three days using 0.2 M
potassium fluoride and 8% (W/V) PEG 3350 as reservoir formulation. To generate BRD4ꢀ
(BD1)/1A coꢀcrystals, 1A (100 mM stock in 100% DMSO) was added to 1 mg/mL BRD4(BD1)
with gentle mixing to achieve a final concentration of 0.33 mM 1A, resulting in a
compound:protein ratio of 5:1. The BRD4(BD1)/inhibitor complex was incubated at room
temperature for 4 hours to form an covalent adduct and then concentrated up to 3.0ꢀ4.0 mg/mL
using the 3 kDa cutꢀoff Centricon concentrator. This complex was crystallized using the sitting
drop vapor diffusion method at room temperature. The screening process yielded several
crystallization conditions that produced crystals of sufficient size and grew to a maximum size
after three days. The crystallization condition that resulted in the crystal selected for Xꢀray
diffraction data collection was obtained from a well solution of 0.2 M Ammonium Sulfate, 0.1 M
BisꢀTris pH 6.5, 25% PEG 3350. Coꢀcrystals of both complexes were harvested, treated with the
crystallization solution supplemented with 20ꢀ27% ethylene glycol and flash frozen in liquid
nitrogen. Xꢀray diffraction data were measured at Beamline X29A, NSLS synchrotron facility,
Brookhaven National Laboratory. The Xꢀray diffraction data were reduced using HKL2000
(Otwinowski and Minor, 1997) (Supplementary Table S2). The crystal structures were solved by
molecular replacement with Molrep. Model building and refinement were pursued using COOT
28
ACS Paragon Plus Environment

Page 29 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(Emsley, 2004) and REFMAC (Vagin, 2004), respectively. Inspection of the initial electron
density maps showed unambiguous compound density in each of the two structures. 1 was
modeled in with the protein and refined to a resolution of 1.88 Å with Rfactor and Rfree of
16.2% and 22.6%, respectively and good stereochemistry (Supplementary data Table S2). In the
other structure initial electron density map, 1A epoxide was observed as a linearized, covalent
adduct to Met149. Initial model building and refinement were done with neither of Met149 nor
this compound in the structure. After placement of the solvent molecules, Met149 and 1A
compound were modeled and refined. A residual element of density linked to the sulfur linker
atom was modeled as a methyl, but this group could not be unambiguously assigned. The coꢀ
structure was further refined to a resolution of 1.47 Å with Rfactor and Rfree of 14.5% and
19.4%, respectively and good stereochemistry (Supplementary Table S3).
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
6. Proliferation assay
The MV4ꢀ11 cell line was obtained from ATCC and propagated according to manufacturer’s
instructions. Cells were counted using a Countess automated cell counter (Life Technologies)
and seeded in 96 well plates between 25,000 and 50,000 cells/well. Twoꢀfold dilutions of
covalent inhibitors vs. nonꢀcovalent controls were added in triplicate wells and compounds were
added at the time of seeding. DMSO at a final concentration of 0.1% was used as a negative
control. Compounds and cells were incubated for 72 hours at 37 °C, 5% CO2. Twenty microliters
of the CellTiter 96 Aqueous One Solution Reagent (Promega) were then added to each well, and
the cells were incubated at 37 °C, 5% CO2 for three hours. Absorbance at 490 nm was read on a
MultiSkan GO plate reader (Thermo Scientific) using the precision measurement mode. The
percentage of proliferation was calculated as follows: [(ODSampleꢀODblank)/OD DMSO]*100.
Determination of the IC50 values was done with the GraphPad Prism software. In the washout
29
ACS Paragon Plus Environment