Mitigation of Acetylcholine Esterase Activity in the 1,7-Diazacarbazole Series of Inhibitors of Checkpoint Kinase 1

Article abstract of DOI:10.1021/acs.jmedchem.5b00464

Checkpoint kinase 1 (ChK1) plays a key role in the DNA damage response, facilitating cell-cycle arrest to provide sufficient time for lesion repair. This leads to the hypothesis that inhibition of ChK1 might enhance the effectiveness of DNA-damaging therapies in the treatment of cancer. Lead compound 1 (GNE-783), the prototype of the 1,7-diazacarbazole class of ChK1 inhibitors, was found to be a highly potent inhibitor of acetylcholine esterase (AChE) and unsuitable for development. A campaign of analogue synthesis established SAR delineating ChK1 and AChE activities and allowing identification of new leads with improved profiles. In silico docking using a model of AChE permitted rationalization of the observed SAR. Compounds 19 (GNE-900) and 30 (GNE-145) were identified as selective, orally bioavailable ChK1 inhibitors offering excellent in vitro potency with significantly reduced AChE activity. In combination with gemcitabine, these compounds demonstrate an in vivo pharmacodynamic effect and are efficacious in a mouse p53 mutant xenograft model.

Full text of DOI:10.1021/acs.jmedchem.5b00464

Subscriber access provided by NEW YORK UNIV
Article
Mitigation of Acetylcholine Esterase Activity in the 1,7-
Diazacarbazole Series of Inhibitors of Checkpoint Kinase 1
Lewis Gazzard, Karen Williams, Huifen Chen, Lorraine Axford, Elizabeth Blackwood, Brenda Burton, Kerry
Chapman, Peter Crackett, Joy Drobnick, Charles Ellwood, Jennifer Epler, Michael Flagella, Emanuela
Gancia, Matthew Gill, Simon Goodacre, Jason Halladay, Joanne Hewitt, Hazel Hunt, Samuel Kintz,
Joseph Lyssikatos, Calum Macleod, Sarah Major, Guillaume Médard, Raman Narukulla, Judi Ramiscal,
Stephen Schmidt, Eileen Seward, Christian Wiesmann, Ping Wu, Sharon Yee, Ivana Yen, and Shiva Malek
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.5b00464 • Publication Date (Web): 19 May 2015
Downloaded from http://pubs.acs.org on May 21, 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.
Journal of Medicinal Chemistry 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 79
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
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Mitigation of Acetylcholine Esterase Activity in the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
,7ꢀDiazacarbazole Series of Inhibitors of
Checkpoint Kinase 1
a,*
g
a
g
b
Lewis Gazzard, Karen Williams, Huifen Chen, Lorraine Axford, Elizabeth Blackwood,
g
g
g
a
g
Brenda Burton, Kerry Chapman, Peter Crackett, Joy Drobnick, Charles Ellwood, Jennifer
b
c
g
g
g
Epler, Michael Flagella, Emanuela Gancia, Matthew Gill, Simon Goodacre, Jason
d
g
g
a
a
g
Halladay, Joanne Hewitt, Hazel Hunt, Samuel Kintz, Joseph Lyssikatos, Calum Macleod,
g
g
g
b
e
Sarah Major, Guillaume Médard, Raman Narukulla, Judi Ramiscal, Stephen Schmidt, Eileen
g
f
f
b
e
e
Seward, Christian Wiesmann, Ping Wu, Sharon Yee, Ivana Yen and Shiva Malek
a
b
c
d
Departments of Discovery Chemistry, Translational Oncology, Safety Assessment, Drug
e
f
Metabolism and Pharmacokinetics, Biochemical and Cellular Pharmacology, Structural
Biology, Genentech, Inc., 1 DNA Way, South San Francisco, California 94080, United States.
g
Argenta, A Charles River Company, 8ꢀ9 Spire Green Centre, Harlow, Essex, CM19 5TR,
United Kingdom.
1
ACS Paragon Plus Environment

Page 3 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Abstract
Checkpoint kinsase 1 (ChK1) plays a key role in the DNA damage response, facilitating cellꢀ
cycle arrest to provide sufficient time for lesion repair. This leads to the hypothesis that
inhibition of ChK1 might enhance the effectiveness of DNAꢀdamaging therapies in the treatment
of cancer. Lead compound 1 (GNEꢀ783), the prototype of the 1,7ꢀdiazacarbazole class of ChK1
inhibitors, was found to be a highly potent inhibitor of acetylcholine esterase (AChE) and
unsuitable for development. A campaign of analog synthesis established SAR delineating ChK1
and AChE activities, and allowing identification of new leads with improved profiles. In silico
docking using a model of AChE permitted rationalization of the observed SAR. Compound 19
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
(GNEꢀ900) and compound 30 (GNEꢀ145) were identified as selective, orally bioavailable ChK1
inhibitors offering excellent in vitro potency with significantly reduced AChE activity. In
combination with gemcitabine, these compounds demonstrate an in vivo pharmacodynamic
effect and are efficacious in a mouse p53ꢀmutant xenograft model.
Introduction
The eukaryotic cell cycle describes the sequential phases of growth, DNA replication,
organization and mitosis that a proliferating cell must successfully navigate. Progression through
the phases of the cell cycle is controlled by cell cycle checkpoints, biological safety mechanisms
1
that monitor and verify the fidelity of the cell and readiness for the next phase. Activation of
checkpoints can occur in response to the detection of DNA damage, resulting in cell cycle arrest
to facilitate repair of the damage. As such, the proper function of cell cycle checkpoints enable a
mechanism for tumor cells to recover from DNA damage that occurs as a result of radioꢀ or
2
chemotherapeutic intervention, and evade cell death resulting from these therapies.
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Checkpoint kinase 1 (ChK1) is a serine/threonine kinase that plays a central role in mediating
G2/M cell cycle arrest in response to DNA damage. Detection of damage results in activation of
2
ChK1 through the ATR checkpoint pathway and the initiation of repair mechanisms, or in the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
event of irreparable damage, induction of apoptosis. Successful repair results in the inactivation
of ChK1, allowing resumption of the cell cycle and progression into the mitotic phase.
Tumor cells suffering certain genetic deficiencies, notably mutation in the p53 gene, lack
3
functional G1/S checkpoint control early in the cell cycle. As a result, cells with damaged DNA
are able to progress through the G1 phase unabated, and are disproportionately reliant on ChK1ꢀ
mediated checkpoint control later in the cycle. Failure of ChK1 signaling to halt progression into
mitosis with damaged DNA would result in mitotic catastrophe and cell death. This opens the
possibility of preferentially sensitizing these p53ꢀdisfunctional tumor cells to the effects of DNA
damage by disabling ChK1. The potential therapeutic benefit of inhibiting ChK1 has attracted
4
significant interest and a number of preclinical programs have been reported. Several molecules
are undergoing clinical evaluation in combination with chemotherapies, and three programs
5
having progressed into Phase II.
We embarked on a program to discover a small molecule inhibitor of ChK1, with the goal of
improving the therapeutic benefit of existing DNAꢀdamaging chemotherapeutic agents used in
the treatment of cancer. We envisaged the ChK1 inhibitor would be coꢀadministered and
sensitize tumor cells to the cytotoxic effects of the agent. In order to maximize the flexibility of
treatment regimen, in combination with both current and future standard of care chemotherapies,
we focused our efforts on the discovery of an inhibitor that could be dosed orally. We recognized
that a short period of efficient ChK1 inhibition was likely to be sufficient to release damaged
cells from checkpoint arrest, and elicit a maximal effect. Our target profile allowed for little or
3
ACS Paragon Plus Environment

Page 5 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
no cytotoxic activity as a single agent, and minimal toxicity that might be additive to or
exacerbate the doseꢀlimiting toxicities associated with chemotherapy.
6
We previously described the discovery of the 1,7ꢀdiazacarbazole series of inhibitors and the
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
identification of 1 (GNEꢀ783) (Figure 1a), an attractive ATPꢀcompetitive lead compound
suitable for further optimization. Compound 1 demonstrated high biochemical and cellular
potency, synergy with chemotherapeutics by isobologram analysis, a pharmacokinetic profile
suitable for oral dosing and good kinase selectivity. This molecule has been successfully utilized
8
as a biological tool compound for further investigation of ChK1 as a therapeutic target.
The binding mode of 1 was revealed through Xꢀray crystallographic analysis of the complex
6
with the ChK1 kinase domain (Figure 1b), indicating the involvement of all nitrogen atoms in
the tricyclic core in polar interactions with either the hinge residues (Glu85 & Cys87) or the
highly conserved triad of waters in the kinase backꢀpocket. In addition, the proximity of C2ꢀH to
a hinge carbonyl (Cys87) suggests a productive, nonꢀcanonical interaction. The water network is
such that the ligand (N7) must accept a hydrogen bond from the nearest of the three waters
(
Figure 1c). The 6ꢀcyano function accepts a hydrogen bond from the protonated catalytic lysine
(Lys38). Further binding affinity is gained from hydrophobic interactions with either face of the
9
core tricycle (Leu15, Val23, Leu137) and the 3ꢀphenyl substituent (Gly90) in the ‘H2 pocket’.
The cationic piperazine moiety makes no direct contact with the protein, but protrudes from the
ATPꢀbinding pocket to allow maximal solvation.
4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
a)
b)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
(c)
Figure 1. (a) Structure of compound 1; (b) High resolution (1.9 Å) coꢀcrystal structure of 1 with
the kinase domain of ChK1 (PDB code: 4QYH); (c) Key water interactions in the back pocket.
Here we describe the structureꢀactivity relationships we developed through lead optimization,
following the identification of undesirable, nonꢀkinase activity in 1 and related analogs. Further,
an explanation is provided with the aid of computational modeling, allowing mitigation of this
activity and selection of candidates for in vivo efficacy evaluation.
5
ACS Paragon Plus Environment

Page 7 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Results and Discussion
With potent cellular activity, good pharmacokinetic properties and a good kinase selectivity
profile, compound 1 has been successfully employed as an in vitro and in vivo tool compound for
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
8
the investigation of ChK1 biology. Compound 1 also represented an excellent lead compound
for establishing SAR around this novel kinase scaffold. However, during expanded
pharmacokinetic evaluation, a significant undesired activity was discovered. Upon doseꢀ
escalation in rodents, muscle fasciculations were observed in the test subjects in a doseꢀ
dependent manner, implicating the engagement of neurological targets. Evaluation in a panel of
10
ion channel and receptor targets, designed to reveal pharmacological activities of significant
clinical concern, identified potent binding to acetylcholine esterase (AChE) as the key finding.
This unexpected activity was considered to represent an unacceptable risk and halted further
development of this compound as a therapeutic agent. We therefore began investigation of the
SAR of the 1,7ꢀdiazacarbazole scaffold, based on our lead compound 1, with the goal of
reducing AChE activity while maintaining the otherwise promising profile.
Inhibition of ChK1 kinase activity in vitro (ChK1 IC ) was determined using an AlphaScreen
5
0
1
1
peptide phosphorylation assay. In parallel, we monitored cellular activity in HTꢀ29 cells, a p53ꢀ
1
2
mutant human colorectal adenocarcinoma cell line. Following induction of mitotic arrest by
1
3
treatment with SNꢀ38, release of G2/M checkpoint control and progression into mitosis was
14
evidenced by phosphorylation of histone H3 (pHH3 EC ). We targeted a cellular potency
5
0
superior to that of compound 1 (EC 98 nM). In vitro inhibition of human AChE was also
50
1
5
monitored, with the goal of significantly improving the margin observed for lead compound 1,
which was found to be highly potent in this assay (IC 18 nM).
5
0
6
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Inspection of the chemical structure of 1 clearly implicated the presence of the exposed basic
amine as responsible for the observed AChE activity, ionization of which could mimic the
cationic moiety of acetylcholine (ACh). Our initial investigations therefore centered on
establishing the SAR related to the 1,7ꢀdiazacarbazole core 3ꢀaryl substituent, and the
importance of the basic amine function.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Initially a series of analogs was prepared including both phenyl replacements and nonꢀbasic
phenyl substitutions (Table 1). Significant ChK1 biochemical potency could be achieved with
simple nonꢀaromatic substituents (compounds 3–5), or even no substituent (compound 2),
16
translating to exceptionally high ligand efficiencies. These data indicate the tricyclic 1,7ꢀ
diazacarbazole core to be highly optimized ligand for the ATPꢀbinding pocket of ChK1, and
responsible for much of the affinity of the diverse set of elaborated analogs.
The synthesis of a library of 3ꢀaryl analogs was undertaken, including a comprehensive screen
of phenyl substitutions, however no analogs exhibited potency comparable to 1. The morpholine
analog 6, maintaining comparable hydrophobic contacts to compound 1 but lacking the
protonated amine, lost biochemical and cellular potency. The most potent analogs were those
containing a phenyl substituent bearing a hydrogenꢀbond donor (7–10), suggesting productive
polar contacts with residues surrounding the H2 pocket. Examination of Xꢀray coꢀcrystal
structures of the ChK1 kinase domain with these ligands confirmed a common binding mode
similar to compound 1. The coꢀcrystal structure with compound 7 revealed the phenolic hydroxyl
was engaging the carboxylate sidechain of Glu91 (Figure 2a). The coꢀcrystal with compound 8
indicated a hydrogen bond with hinge residue Tyr86, but with the acetamide carbonyl acting in
the role of acceptor (Figure 2b). Otherwise polar contacts were not observed, and in all cases
cellular potency was poor. Several heteroaromatic 3ꢀsubstituents offered promising biochemical
7
ACS Paragon Plus Environment

Page 9 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
potency, but in all cases lacked sufficient cellular activity to warrant further progression (11–17).
In general, the poor aqueous solubility of these nonꢀbasic analogs was assumed to be limiting
both activity in the cellular assay and the correlation with biochemical potency.
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 likely role of the basic amine in promoting AChE binding was supported by the lack of
AChE inhibitory activity observed in several nonꢀbasic 1,7ꢀdiazacarbazoles. Compound 5 and 3ꢀ
aryl analogs 11 and 12 were assayed for AChE inhibition and found to elicit no activity up to 25
ꢁ
M.
(a)
(b)
Figure 2. High resolution coꢀcrystal structures of (a) compound 7 (1.85 Å) and (b) compound 8
1.85 Å) with ChK1 kinase domain indicating polar and hydrophobic contacts (PDB codes:
RVL & 4RVK).
(
4
8
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 1. A basic 3ꢀsubstituent is required for optimal ChK1 potency
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ChK1
pHH3
ChK1
LE
cLogD
Compound
X
IC (nM)
EC (nM)
(pH 7.4)
50
50
N
N
1
1.3
97.7
0.44
2.8
2
3
4
5
H
Br
414
140
ꢀ
ꢀ
0.60
0.60
0.65
0.56
1.4
2.4
1.0
2.1
OH
NHEt
38.1
56.6
>5000
1792
O
N
6
7
78.7
219
0.37
3.2
41.7
13.5
>5000
425
0.47
0.44
3.0
2.6
OH
O
8
9
N
H
OH
OH
11.4
27.1
938
0.48
0.46
2.5
2.5
1
0
>5000
O
1
1
1
2
16.6
20.2
180
478
0.54
0.54
2.4
3.1
S
N
N
13
14
33.8
14.7
2495
0.49
0.55
2.2
1.5
N
>5000
N
H
1
1
5
6
127
76.5
42.7
>5000
>5000
>5000
0.46
0.47
0.49
2.3
2.2
2.1
N
N
N
17
9
ACS Paragon Plus Environment

Page 11 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
These data suggested that an ionized basic amine substituent was needed to achieve optimal
ChK1 potency. We therefore investigated alternative amine substitutions hoping to establish
SAR delineating ChK1 potency from AChE inhibition, and to maintain desirable drugꢀlike
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
17
properties we targeted amines with a calculated pK below 9.
a
This goal was quickly realized with our initial analogs. In fact, evaluation of a variety of basic
analogs demonstrated consistently high ChK1 potency, while dramatically modulating inhibition
of AChE (Table 2). These observations are consistent with the amine function being necessary
for optimal affinity for ChK1, but having a more critical and intimate role in interaction with
AChE.
Compared to 1, isomer 18 demonstrates a more than 250ꢀfold drop in potency in the AChE
assay with only a small loss in ChK1 activity. However this represented the highest potency
achievable among metaꢀsubstituted analogs (data not shown). Benzylic amine 19 demonstrated
comparable ChK1 potency to 1 and excellent cellular activity, and with AChE binding potency
reduced by 19ꢀfold. This modification additionally eliminated the undesirable aniline moiety.
Methyl substitutions around the benzylic position and piperidine moiety of compound 19 proved
effective in modulating AChE activity, while maintaining ChK1 potency (compounds 20–24).
The effect on AChE was unpredictable. For example, isomeric dimethylꢀderivatives 20 and 21
demonstrate similar ChK1 potency but a 30ꢀfold difference in AChE inhibition. Geminally
disubstituted isomers 22 and 23 demonstrate a more than 300ꢀfold difference in AChE inhibition.
Electron withdrawing substitutents on the piperidine of 19 reduced the basicity of the amine
and tended to further reduce AChE inhibition with limited effect on cellular potency (compounds
2
5–29). Weakly basic morpholine 29 demonstrated the greatest attenuation of AChE activity,
10
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
being more than 700ꢀfold less potent than 1 and 40ꢀfold less potent than the piperidine analog,
compound 19.
Differences in AChE activity not readily explained by either steric or electronic considerations
were also observed. Dimethylpiperidine compound 23 demonstrated the most potent AChE
activity observed (AChE IC 11 nM). In contrast, the spirocyclic oxetane 30, which differed by
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
the addition of a single oxygen atom, demonstrated a 220ꢀfold reduction in activity (AChE IC50
2.42 ꢁM) while maintaining good potency in the pHH3 cellular assay.
Heterocyclic and bicyclic analogs (compounds 31–34) were investigated but offered no further
improvements versus compound 19. In agreement with the unsubstituted heteroaryl derivatives,
pyridine and thiazole analogs bearing a basic amine substituent (compounds 31 and 32) exhibited
reduced biochemical and cellular potency. Bicyclic thiazole 34 was the most effective
heterocyclic analog, being comparable to 19 in cellular potency, but with greater activity versus
AChE.
11
ACS Paragon Plus Environment

Page 13 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 2. Basic amine structure acutely modulates AChE activity
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
ChK1
pHH3
AChE
cLogD
Compound
X
cpKa
IC (nM)
EC (nM)
IC (ꢁM)
50
(pH 7.4)
5
0
50
N
N
N
1
1.3
97.7
0.018
2.8
7.89
1
1
2
8
9
0
N
1.9
1.0
2.1
133
53.2
99.1
4.77
0.333
0.135
2.8
3.0
3.6
7.89
8.49
8.90
N
N
N
2
1
4.2
209
4.10
3.7
8.38
N
N
2
2
2
3
9.4
4.4
128
172
3.33
3.7
3.4
8.38
8.71
0.011
N
2
2
2
4
5
6
10.2
1.8
228
144
118
0.507
0.378
0.708
4.2
2.2
2.6
8.71
8.21
7.72
N
OH
OH
N
1.9
N
2
2
2
3
3
7
8
9
0
1
5.0
11.6
13.9
2.5
95.8
339
216
72.8
540
0.890
2.69
3.4
2.9
2.6
2.4
2.0
7.75
7.23
6.73
8.26
8.28
F
N
CN
O
N
13.3
N
2.42
O
N
5.3
0.187
N
12
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
N
3
2
13.6
2.3
658
152
50.9
0.018
0.160
0.259
2.9
2.4
2.1
7.84
8.52
8.62
N
S
N
33
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
N
3
4
0.78
S
We next explored replacement of the 6ꢀcyano function as a potential means to reducing AChE
potency (Table 3). With a promising improvement in AChE activity compound 19 was chosen as
the parent compound for this series. Unlike the 3ꢀaryl substituent, which does not form a discreet
contact with the protein, the 6ꢀcyano is involved in a polar interaction with the charged sidechain
of Lys38 and is likely to be less amenable to modification if ChK1 potency is to be maintained.
Removal of the cyano entirely yielded compound 35 that demonstrated a relatively modest 19ꢀ
fold drop in potency. This further supports the optimal nature of the tricyclic 1,7ꢀdiazacarbazole
core, providing much of the affinity through multiple polar contacts to the hinge and conserved
back pocket waters.
13
ACS Paragon Plus Environment

Page 15 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 3. Effects of 6ꢀsubtitution on AChE activity.
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
ChK1
pHH3
AChE
cLogD
Compound
X
IC (nM)
EC (nM) IC (ꢁM)
(pH 7.4)
50
50
50
N
19
35
1.0
53.2
0.333
3.0
3.2
H
18.6
4.3
1922
292
ꢀ
O
36
37
0.217
2.5
3.4
H N
2
O
102
2234
ꢀ
N
H
O
3
8
19.8
18.3
863
986
447
ꢀ
ꢀ
ꢀ
ꢀ
3.9
2.3
0.7
39
HO
HO
4
4
0
1
N
O
N
S
N
0.46
1.3
23.9
62.7
103
145
9.0
0.323
0.496
0.450
0.659
0.874
0.493
0.777
1.213
1.037
3.4
4.1
4
2
N
2
.6
43
0.95
1.1
O
N
4
4
4
5
O
3.6
3.7
4.1
4.1
3.2
3.1
N
N
0.31
0.95
9.1
46
75.1
1573
168
656
N
N
4
4
4
7
8
9
N
2.3
N
N
N
2.9
N
5
0
5.1
163
0.266
3.4
N
14
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
6
As previously observed in the discovery of 1, the 6ꢀcarboxamide analog 36 offered good
biochemical potency but weak cellular activity, and was not amenable to substitution (compound
3
7). The 6ꢀmethoxy and 6ꢀhydroxymethyl analogs (38 and 39) offered no great improvement
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
over 6ꢀH (35). The far lower potency of the 6ꢀhydroxy (40) versus 6ꢀmethoxy (38) provided an
interesting insight, as we may infer that the pyridone tautomer predominates. While the carbonyl
offers a good acceptor at the 6ꢀposition, this necessarily generates a hydrogen bond donor (N7ꢀ
H), which cannot satisfy the preference of the rigidly held back pocket water network (Figure
1b).
Replacement of 6ꢀcyano with heteroaromatic rings bearing an appropriately placed acceptor
proved highly effective (Table 3). Azole analogs were especially potent, however modulation of
AChE activity was limited and the improvement over 19 was modest at best (compounds 41–45).
Pyrazole 45 was the most potent analog in biochemical and cellular assays discovered in this
series and provided the greatest improvement in AChE inhibition, however reduced selectivity in
1
8
a kinase screening panel relative to 19 prevented further advancement. Azine derivatives
(
compounds 46–50) offered a greater improvement, with 3ꢀpyridyl most effective (46), but with
reduced cellular potency in all cases.
In an effort to rationalize the observed AChE activities we examined correlations with various
physical properties. Comparison with either calculated pK of the basic amine substituent or the
a
basicityꢀdependent calculated logD showed no significant correlation with potency and no
predictive potential. We therefore sought to use computational modeling of the AChE binding
site to rationalize the relative activities of the diazacarbazoles.
An in silico docking model of AChE was prepared utilizing an available high resolution
19
murine protein Xꢀray crystal structure (PDB code 2HA3, 2.25 Å). This provided the most
15
ACS Paragon Plus Environment

Page 17 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
suitable basis structure in which the ACh binding pocket was not occluded, due to the presence
of bound choline. The high sequence homology of the murine protein to human in the vicinity of
2
0
the ACh binding site was further improved by mutation of Ile294 to Val294 in our model.
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
Docking of 1 using the AChE model revealed a striking prediction (Figure 3a). The kinase
hingeꢀbinding 1,7ꢀdiazacarbazole core is predicted to form two hydrogenꢀbonded polar contacts
with the Ser293 and Phe295 residues near the mouth of the AChꢀbinding pocket. These kinase
hingeꢀlike contacts ideally orient the 3ꢀaryl substituent into the narrow, hydrophobic channel so
as to place the basic amine in a position very similar to that occupied by the ACh quaternary
ammonium cation. The close proximity of the cation to an electronꢀrich indole (Trp86) results in
2
1
stabilization through a πꢀcation interaction. This binding mode would be predicted to be of high
affinity and highly sensitive to variations in substitution around the basic amine.
Contrasting with 1, compound 19 is predicted to bind similarly and is accommodated in the
cationꢀbinding pocket in two conformations (Figure 3a). However due to the lesser proximity to
Trp86, neither conformation can benefit from πꢀcation stabilization, and experimentally a 19ꢀfold
loss in binding affinity is observed.
16
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
a)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(b)
(c)
17
ACS Paragon Plus Environment

Page 19 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 3. Pairwise comparisons of binding modes predicted by docking to AChE can be used to
rationalize differences in AChE inhibitory activity: (a) compound 1 (18 nM) vs. compound 19
(333 nM); (b) compound 20 (135 nM) vs. compound 21 (4.10 ꢁM); (c) compound 23 (11 nM)
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
vs. compound 30 (2.42 ꢁM). Binding surface colors are atomꢀbased (carbon grey, oxygen red,
nitrogen blue).
18
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Consideration of other pairs of similar analogs with contrasting potencies proved enlightening.
Compound 20 docked efficiently, placing the transꢀ2,6ꢀdimethylpiperidine moiety in a lipophilic
environment and permitting a low energy chair conformation (Figure 3b). In contrast, isomer 21
is unable to achieve a lowꢀenergy conformation and remain within the bounds of the pocket.
Compound 21 requires either a higher energy conformation or deformation of the protein, and
suffers a 30ꢀfold drop in affinity versus 20.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Compound 23 proved to be an exceptionally potent inhibitor of AChE, and is predicted by
docking to bind efficiently and fill the hydrophobic space (Figure 3c). Oxetane 30 docks in a
very similar conformation, but now suffers from the placement of oxygen in close proximity to
the carboxylate (Glu202) and a subsequent destabilization of this mode. Experimentally a 220ꢀ
fold reduction in activity is observed. These docking results allowed us to rationalize relative
AChE binding potencies and qualitatively predict effects of basic amine substitution.
Having established SAR around the 1,7ꢀdiazacarbazole core and identified a number of
analogs with an improved profile with respect to AChE binding, we utilized a screening cascade
to select candidates for evaluation in our mouse pharmacodynamic (PD) model. Selection was
2
2
made primarily using the results of isobologram analysis
to assess synergy with
chemotherapeutics in vitro, mouse oral pharmacokinetics and kinase profiling. PD results
provided a further selection criterion for evaluation in a mouse tumor xenograft model. Using
these criteria, compounds 19 and 30 were considered to have the best overall profile and were
selected for dosing in this efficacy model (Figure 4).
19
ACS Paragon Plus Environment

Page 21 of 79
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 4. In vitro activities of efficacy candidates 19 and 30.
Isobologram analysis allows the demonstration of synergy in inducing cytotoxicity in HTꢀ29
8b
cells upon treatment with gemcitabine in combination with a ChK1 inhibitor. Dose ranging
with both gemcitabine and ChK1 inhibitor allows generation of a twoꢀdimensional dose response
surface and determination of the concentration of sensitizer (ChK1 inhibitor) required to reduce
2
2
by tenꢀfold the IC of gemcitabineꢀinduced cytotoxicity (EC ). Both efficacy candidates 19
5
0
10fs
and 30 demonstrated excellent cellular chemopotentiation of gemcitabine, with EC10fs of 10 nM
and 16 nM respectively. In the absence of gemcitabine, compounds 19 and 30 demonstrated
single agent cytotoxicity (GI ) of 8.7 ꢁM and >2.5 ꢁM respectively. These data indicate a 870ꢀ
5
0
fold, and greater than 156ꢀfold margin for chemopotentiation over single agent cytotoxicity, and
indicate that combination treatment is synergistic rather than additive.
Selectivity for ChK1 relative to other kinases was assessed in an extensive kinase profiling
1
8
panel. Compounds 19 and 30 were assayed at 100 nM concentration, representing 100ꢀfold and
40ꢀfold excess respectively over the ChK1 biochemical IC . Compound 19 exhibited greater
5
0
than 80% inhibition of 15 of the 246 kinases assayed, and 50–80% inhibition of a further 17
kinases. Compound 30 exhibited greater than 80% inhibition of 21 of the 258 kinases assayed,
20
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
and 50–80% inhibition of a further 24 kinases. Cyclinꢀdependent kinases, important in regulating
the cell cycle, were not significantly inhibited. These results indicate the selected compounds
have good kinase selectivity profiles, with 19 modestly superior to 30.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 4. Crossꢀspecies pharmacokinetic parameters for efficacy candidates 19 & 30
Compound Species
IV (solution, pH 4.5)
PO (suspension, pH 6.5)
PPB%
Dose
Cl
Vss
Dose
AUC
Cmax
F
mg/kg mL/min/kg
L/kg
mg/kg
hr.uM
uM
%
*
1
3
9
0
Mouse
Rat
1
1
1
1
1
1
1
1
55.8
24.7
118
6.57
6.35
23.4
25.7
7.33
7.55
9.63
17.1
5
5
2
2
5
5
2
2
2.68
0.386
0.520
0.021
0.044
0.191
1.10
86.5
67.4
13.1
43.8
48.5
116
95.0
95.3
86.8
88.3
96.2
97.5
94.8
95.1
6.19
0.101
0.869
Dog
Cyno
Mouse
Rat
45.8
36.8
27.5
10.9
30.8
*
1.33
8.63
3.51
0.91
Dog
0.200
46.7
34.0
Cyno
0.086
* Due to the significant proportional increase when extrapolated after 9h, AUCs for mouse are to the 9h time point.
Consideration of our treatment hypothesis and dosing strategy—a single, appropriately timed
dose following chemotherapeutic treatment—required that a ChK1 inhibitor achieve a tumor
concentration above a threshold defined by cellular potency for a relatively short period. Crossꢀ
species pharmacokinetic parameters for compounds 19 and 30 are given in Table 4. Mouse PK
indicated both compounds provided sufficient oral exposure to warrant dosing in our in vivo
models. For compound 19, rodents demonstrated moderate clearance with good oral
bioavailability. However dog and cynomologous monkey demonstrated higher clearance (greater
than liver blood flow) and high volumes of distribution. Compound 30 demonstrated similarly
good oral exposures with improved consistency across species. Importantly, no sign of muscle
21
ACS Paragon Plus Environment

Page 23 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
fasciculations was detected with either compound in any species, or upon dose escalation in
rodents, consistent with the increased margins versus inhibition of AChE.
In our pharmacodynamics model, mice bearing HTꢀ29 xenografts were treated with
gemcitabine (60 mg/kg) followed after 24 h by a single oral dose of ChK1 inhibitor. Activation
of checkpoint control in response to gemcitabine, and subsequent abrogation of G2/M
checkpoint arrest in response to ChK1 inhibition, is evidenced by the level of activation of the
histone H2AX. Phosphorylation of H2AX (to form γH2AX) occurs in response to DNA doubleꢀ
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
3
strand breaks (DSB) resulting from mitotic entry with unrepaired DNA damage. The level of
phosphorylation was monitored by Western blot analysis of tumor tissues harvested 16 h after
ChK1 inhibitor administration. Compound 19 dosed orally at 10–40 mg/kg (Figure 5a) and
compound 30 dosed orally at 12.5–50 mg/kg (Figure 5b) demonstrated an increase in γH2AX in
a doseꢀdependent manner. In contrast, control arms indicated no increase in DSB upon treatment
with either gemcitabine or ChK1 inhibitor alone.
Compounds 19 and 30 were next evaluated in a multiꢀdose mouse efficacy study. NCR nude
2
4
mice bearing HTꢀ29 xenografts were treated with gemcitabine, followed after 24 h by ChK1
2
5
inhibitor dosed orally, a time interval that has been shown to be optimal in this model. For
compound 19, mice received five treatment cycles at five day intervals (Figure 5c). For
compound 30, mice received four treatment cycles at four day intervals (Figure 5d). Efficacy was
determined by monitoring tumor volume. Both compounds demonstrated improved reduction of
tumor growth at all doses compared to gemcitabine alone, and in a doseꢀdependent manner. At
the 50 mg/kg dose level, compound 30 demonstrated complete inhibition of tumor growth (111%
inhibition, p value 0.0002). Body weights were unaffected in all treatment groups, indicating the
2
6
test compounds did not reduce the tolerability relative to gemcitabine alone.
22
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
a)
(b)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
c)
(d)
mut
Figure 5. In vivo activity in NCR nude mice bearing HT29 p53 xenografts of compounds 19
and 30 dosed orally in MCT suspension 24 h following gemcitabine. Pharmacodynamics:
increase in proportional γH2AX 16 h after treatment with (a) compound 19 or (b) compound 30.
Efficacy: dose dependentꢀreduction in tumor growth following (c) five cycles of gemcitabine
23
ACS Paragon Plus Environment

Page 25 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
120 mg/kg) in combination with compound 19, or (d) four cycles of gemcitabine (60 mg/kg) in
combination with compound 30. Gemcitabine doses are indicated by arrows; the higher dose of
gemcitabine and frequency of treatment used with compound 19 limits the value of direct
comparison. Tumor growth inhibition and statistical analyses are provided in the supporting
information.
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
24
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Chemistry
The synthesis of the key 3ꢀbromoꢀ6ꢀcyanoꢀ1,7ꢀdiazacarbazole intermediate (3) was achieved in
six steps from commercially available 7ꢀazatryptophan (Scheme 1). Esterification and Pictetꢀ
Spengler reaction with formalin afforded an intermediate tetrahydroꢀ1,7ꢀdiazacarbazole (51), and
aromatization with selenium dioxide followed by regioselective bromination yielded methyl 9Hꢀ
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
6
pyrrolo[2,3ꢀb:5,4ꢀc']dipyridineꢀ6ꢀcarboxylate (52). Conversion to the 6ꢀcarboxamide derivative
was achieved by treatment with ammonia, and dehydration afforded intermediate 3. Protection as
the SEM derivative provided compound 53, and either of intermediates 3 or 53 could serve as the
biaryl coupling precursor. SuzukiꢀMiyaura couplings with various boronic acid or boronate
esters and subsequent deprotection afforded the corresponding 3ꢀarylꢀ6ꢀcyanoꢀ1,7ꢀ
diazacarbazoles (1, 6–10, 12, 18–20, 23 and 29) (Scheme 1). Similarly, Stille couplings with the
appropriate aryl stannane and subsequent deprotection afforded compounds 11 and 15–17.
Scheme 1. Synthesis of 6ꢀcyanoꢀ1,7ꢀdiazacarbazoles from 7ꢀazatryptophan. Reagents and conditions: (a) SOCl2,
MeOH, 0 °C–reflux; (b) CH O(aq), Py, 100 °C; (c) SeO , Py, dioxane, 100 °C; (d) Br , NaOAc, AcOH, 100 °C; (e)
2
2
2
NH , MeOH, 140 °C, autoclave; (f) TFAA, NEt , THF, 0 °C–rt; (g) SEMCl, NaH, DMF; (h) Boronic acid or
3
3
boronate ester, Pd(dppf)Cl , Cs CO (aq), dioxane, reflux; (i) Aryl stannane, Pd(PPh ) , LiCl, dioxane, ꢁW; (j)
2
2
3
3 4
TBAF, THF, 50 °C.
25
ACS Paragon Plus Environment

Page 27 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Compound 53 could be converted to the corresponding boronate pinacol ester 54 (Scheme 2).
SuzukiꢀMiyaura couplings of 54 with aryl halides, and subsequent removal of the SEM
protecting group, provided an alternate route to 3ꢀaryl derivatives (13, 14, 22, 24, 26–28 and 30–
34).
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
SuzukiꢀMiyaura coupling of 3 with 4ꢀ(bromomethyl)phenylboronic acid at elevated
temperature resulted in concomitant bromide hydrolysis, and subsequent Bocꢀprotection afforded
benzylic alcohol intermediate 55 (Scheme 3). Conversion to the mesylate and treatment with the
appropriate secondary amines resulted in mesylate displacement and deprotection to provide
benzylic amines 21 and 25.
Scheme 2. 3ꢀAryl diazacarbazoles via reversed SuzukiꢀMiyaura coupling route. Reagents and conditions: (a)
Bis(pinacolato)diboron, Pd(dppf)Cl , KOAc, dioxane, DMSO, 120 °C; (b) Aryl bromide, Pd(PPh )Cl , Cs CO (aq),
2
3
2
2
3
DMEꢀIMSꢀwater, ꢁW, 140 °C; (c) TBAF, THF, 55 °C.
Scheme 3. (4ꢀAminomethyl)phenyl derivatives via mesylate displacement route. Reagents and conditions: (a) 4ꢀ
bromomethyl)phenylboronic acid, Pd(dppf)Cl , KF(aq), MeCN, ꢁW, 175 °C; (b) Boc O, MeCN, 50 °C; (c) MsCl,
(
2
2
1
2
NEt , DCM, 0 °C–rt; (d) HNR R , NEt , MeCN, 50 °C.
3
3
26
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Compound 2 was prepared by reduction of 53 with zinc and ammonium formate, and SEM
deprotection. Compound 4 was prepared from boronate ester 54 by oxidation with Nꢀ
methylmorpholine Nꢀoxide and deprotection. Compound 5 was prepared from 53 by palladiumꢀ
catalyzed arylation of tertꢀbutyl carbamate, alkylation of the intermediate with sodium hydride
and ethyl iodide, and double deprotection.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Synthesis of the 1,7ꢀdiazacarbaole core with various 6ꢀsubstituents could be achieved in three
steps from 6ꢀsubstituted 3ꢀaminoꢀ4ꢀiodopyridines 56 (Scheme 4). SuzukiꢀMiyaura coupling of
56 with 5ꢀbromoꢀ2ꢀfluoropyridineꢀ3ꢀboronic acid affords the corresponding bipyridyls 57, and
subsequent coupling with the appropriate aryl boronate provides 58 (X=H, OMe). Alternatively,
boronic acid 59 can be prepared in two steps from 5ꢀbromoꢀ2ꢀfluoropyridine. SuzukiꢀMiyaura
coupling of 59 with 6ꢀbromoꢀ3ꢀaminoꢀ4ꢀiodopyridine provided a more convergent route to 58
(X=Br). Baseꢀmediated cyclization of the bipyridyl intermediates 58 yielded the 1,7ꢀ
diazacarbazoles (35, X=H; 38, X=OMe; 60, X=Br). Compounds 41, 42, 44–47 and 50 were
synthesized from intermediate 60 (X=Br) by SuzukiꢀMiyaura couplings with the appropriate
heteroaryl boronic acids. Compound 48 was synthesized from 60 by Stille coupling with 2ꢀ
(
tributylstannyl)pyrazine. Compound 49 was prepared by Stille coupling of 4ꢀ
tributylstannyl)pyridazine with 58 (X=Br) and subsequent baseꢀmediated cyclization.
(
27
ACS Paragon Plus Environment

Page 29 of 79
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
Scheme 4. Generalized biaryl synthesis of the 1,7ꢀdiazacarbazole core. Reagents and conditions: (a) Pd(dppf)Cl2,
KF(aq), MeCN, 95 °C; (b) 4ꢀ(Piperidinꢀ1ꢀylmethyl)phenylboronic acid hydrobromide, Pd(PPh )Cl , KF(aq), MeCN,
3
2
i
ꢁ
W, 100 °C; (c) LDA, B(O Pr) , THF, –10 °C; (d) Pd(dppf)Cl , KF(aq), MeCN, 90 °C; (e) NaHMDS, THF, rt; (f)
3 2
Pd(dppf)Cl , Na CO (aq), MeCN, ꢁW; (g) Aryl stannane, Pd(PPh ) , LiCl, dioxane, ꢁW.
2
2
3
3 4
6ꢀCarboxamides 36 and 37 were synthesized from the corresponding 6ꢀcarboxylic acid methyl
ester by thermal reaction with ammonia and ethylamine respectively. Reduction of the same ester
with lithium aluminum hydride yielded compound 39. Conversion of the methyl ester to 2,4,5ꢀ
oxadiazole 43 was accomplished via the acid hydrazide, by treatment with hydrazine and
subsequent condensation with ethyl formate. 6ꢀHydroxy compound 40 was prepared from the 6ꢀ
methoxy analog 38 by demethylation with hydrogen bromide.
28
ACS Paragon Plus Environment