Discovery of a Selective and Potent Inhibitor of Mitogen-Activated Protein Kinase-Interacting Kinases 1 and 2 (MNK1/2) Utilizing Structure-Based Drug Design

Article abstract of DOI:10.1021/acs.jmedchem.5b01657

The discovery of a highly potent and selective small molecule inhibitor 9 for in vitro target validation of MNK1/2 kinases is described. The aminopyrazine benzimidazole series was derived from an HTS hit and optimized by utilization of a docking model, conformation analysis, and binding pocket comparison against antitargets.

Full text of DOI:10.1021/acs.jmedchem.5b01657

Subscriber access provided by GAZI UNIV
Article
Discovery of a Selective and Potent Inhibitor of Mitogen-
Activated Protein Kinase-Interacting Kinases 1 and
2 (MNK1/2) Utilizing Structure-Based Drug Design
Wooseok Han, Yu Ding, Yongjin Xu, Keith Pfister, Shejin Zhu, Robert Warne, Mike Doyle, Mina
Aikawa, Payman Amiri, Brent Appleton, Stuart D. Darrin, Abdallah Fanidi, and Cynthia M. Shafer
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.5b01657 • Publication Date (Web): 22 Mar 2016
Downloaded from http://pubs.acs.org on March 24, 2016
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 44
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Discovery of a Selective and Potent Inhibitor of
MitogenꢀActivated Protein KinaseꢀInteracting
Kinases 1 and 2 (MNK1/2) Utilizing Structureꢀ
Based Drug Design
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
Wooseok Hanχ*, Yu Dingχ, Yongjin Xuχ, Keith Pfisterχ, Shejin Zhuχ, Bob Warneξ, Mike Doyleξ,
Mina Aikawaξ, Payman Amiriξ, Brent Appletonχ, Darrin D. Stuartξ, Abdallah Fanidiξ, Cynthia
M. Shaferχ
χGlobal Discovery Chemistry, Novartis Institutes for BioMedical Research, 4560 Horton Street,
Emeryville, CA 94608
ξOncology, Novartis Institutes for BioMedical Research, 4560 Horton Street, Emeryville, CA
94608
KEYWORDS. mitogenꢀactivated protein kinaseꢀinteracting kinase, MNK, EIF4E, protein
kinase, cancer, oncology, small molecule inhibitor, structureꢀbased drug design.
ABSTRACT. The discovery of a highly potent and selective small molecule inhibitor 9 for in
vitro target validation of MNK1/2 kinases is described. The aminopyrazine benzimidazole series
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 44
1
2
3
4
5
6
was derived from an HTS hit and optimized by utilization of a docking model, conformation
analysis, and binding pocket comparison against antiꢀtargets.
7
8
9
INTRODUCTION
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
Mitogenꢀactivated protein (MAP) kinaseꢀinteracting kinases 1 and 2 (MNK1/2) are
serine/threonine protein kinases that phylogenetically belong to the group of Ca2+/calmodulinꢀ
dependent kinases (CaMK). Each MNK gene, MKNK1 and MKNK2, produce two alternatively
spliced isoforms, encoding MNK1a, MNK1b, MNK2a and MNK2b proteins. MNK1a/2a contain
a MAPKꢀbinding site in the Cꢀterminal regions, which can be activated through phosphorylation
by ERK and p38 MAPK, while the ‘b’ isoforms are constitutively active and lack this MAPKꢀ
binding site.1,2
In tumor cells, aberrant regulation of MAPK signaling pathways3 leads to uncontrolled cell
division, invasiveness, survival and underlying these cancer phenotypes is dysregulated cell
metabolism and protein synthesis.4 Protein synthesis is tightly controlled through regulation of
mRNA translation and a key player in this process is the eukaryotic initiation factor 4F (eIF4F)
which is a capꢀbinding complex that binds the 5′ 7ꢀmethyl guanosine cap found on all cellular
mRNAs. It is comprised of eukaryotic initiation factor E (EIF4E), the DEADꢀbox helicase,
eukaryotic initiation factor 4A (EIF4A), and the scaffold protein eukaryotic initiation factor 4G
(EIF4G).5 Among them, the capꢀbinding protein EIF4E is the limiting factor for the assembly of
the eIF4F complex6,7 and acts as a convergence point for RASꢀRAFꢀMEKꢀERK and PI3KꢀAKTꢀ
mTOR pathways.8 Phosphorylation of EIF4E occurs at a single site, S209, which is mediated by
MNK1/2 kinases.9 High levels of phosphoꢀEIF4E are found in various types of cancers and
several studies demonstrate that MNK1/2ꢀmediated phosphorylation is essential for EIF4E’s role
in oncogenic transformation.10–12
ACS Paragon Plus Environment

Page 3 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
For example, genetic studies using mouse models clearly demonstrate an oncogenic role for
Mnk1/2 mediated phosphorylation of Eif4e in an Eµ-Myc mouse lymphoma model.13 In this
model, wildꢀtype Eif4e and a phosphoꢀmimetic mutant, Eif4e (S209D) cooperated with Myc in
accelerating lymphomagenesis, whereas a phosphoꢀdefective mutant Eif4e (S209A) did not. A
direct role for Mnk1ꢀmediated phosphorylation was demonstrated by transfecting hematopoietic
stem cells from these mice with a kinaseꢀdefective Mnk1 mutant (T2A2) or a constitutively
active mutant (T332D). While the active mutant readily formed tumors, the kinaseꢀdefective
mutant did not. Additional evidence for Mnk1/2 in mouse tumorigenesis comes from Ras and
Pten−/− genetically engineered mouse models.14,15
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
There is also evidence that MNK1/2 could play a role in human tumor cells.11 Feedback
activation of MNK1/2 in response to targeted therapies such as imatinib16, rapamycin and other
rapalogs17–20 or chemotherapies such as cytarabine21 results in increased phosphorylation of
EIF4E which potentially diminishes the therapeutic activity. Inhibition of MNK1/2 in
combination with these therapies enhanced the antitumor activity. Simultaneous inhibition of
MNK1/2 and Bruton’s tyrosine kinase (BTK) or androgen receptor (AR) as a dual inhibitor
enhanced antiproliferative activity against activated diffuse large Bꢀcell lymphoma (DLBCL)
cell lines or AR sensitive prostate cancer cell lines compared to a single agent treatment.22–24
Taken together, these data support a role for MNK1/2ꢀmediated phosphorylation of EIF4E in
oncogenesis and cancer therapeutic resistance.11,12
While other MNK1/2 substrates such as SPRY2,25,26 PSF,27 hnRNP A1,28 and cPLA229 could
play a role in tumorigenesis,12 the most compelling evidence supports a critical role of EIF4E.30
In contrast to their role in tumorigenesis, MNK1/2 appear to be dispensable for normal
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 44
1
2
development in mouse knockꢀout studies31 and this potential difference may provide a
therapeutic avenue for targeting translational requirements in cancer.11,32,33
3
4
5
6
7
8
9
Several MNK inhibitors such as N3ꢀ(4ꢀfluorophenyl)ꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidineꢀ3,4ꢀ
diamine (1, CGP57380) and cercosporamide have been widely described in literature.11,34–36 1,
which was first identified by Novartis,37 is a weak inhibitor of MNK1/2 with low micromolar
IC50 values and exhibits poor selectivity against other kinases.12,38,39 While cercosporamide is
more potent in in vitro assays and shows antitumor efficacy in mouse xenograft models,40,41 it
inhibits JAK3 and PKCβ with submicromolar activites.42 Therefore, we undertook to develop
more potent and selective MNK1/2 inhibitors that could be used to better understand the
consequences of MNK1/2 inhibition in cancer. Herein, we describe the discovery of a selective
and potent inhibitor of MNK1/2 kinases via a structureꢀbased drug design approach.
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. (a) Coꢀcyrstal structure of staurosporine in a mutated MNK2 (2HW7); (b) Overlay
model of compound 2 on staurosporine in a mutated MNK2.
ACS Paragon Plus Environment

Page 5 of 44
Journal of Medicinal Chemistry
1
2
3
4
RESULTS AND DISCUSSION
5
6
7
8
Since dual inhibition of MNK1 and MNK2 kinases was desired to maximize antitumor
activity, the homology of the two MNK homologs was evaluated. The amino acid sequence
identity of the catalytic domain of the two isoforms was found to be 72−78%, with the ATP
binding sites being even higher (90%) with only two residues out of 21 being different.9,11 The
two residues, M162L and R97K, point away from the pocket and were not expected to affect the
shape of the pocket. Therefore, it was anticipated that an inhibitor in the active site could inhibit
both MNK1 and NMK2 kinases with similar potencies. The catalytic domain was also compared
with that of 17 other closelyꢀrelated kinases43 mostly in the Ca2+/calmodulinꢀdependent kinase
(CaMK) group. This effort indicated that MNK2 had significant sequence identity (>30%) with
the catalytic domain of each of these kinases which was of concern for our ability to design
selective MNK inhibitors.
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
With the goal of dual inhibition of MNK1 and MNK2 kinases, an HTS campaign was carried
out to identify starting points for our medicinal chemistry efforts. A number of chemotypes with
single digit micromolar and submicromolar MNK1/2 activity and good ligand efficiency (LE)44
of 0.33−0.41 were identified. The hits were then docked in a coꢀcrystal structure of staurosporine
in MNK2 (Figure 1a, 2HW7) where the DFD motif on the activation loop was mutated to DFG
to relieve autoꢀinhibition.45 Beside the unique DFD motif, MNK kinases have a unique
combination of F159 gatekeeper and C225 secondary gatekeeper, which results in a small
hydrophobic pocket near the gatekeeper F159. Additionally, due to a hydrogen bond between
M162 and G165, a Cꢀshaped loop is formed near lower hinge region producing a larger cavity
than in many kinases.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 44
1
2
3
4
5
6
7
8
9
One of the hits identified from the HTS, a fragmentꢀlike aminopyridine benzimidazole 2,
showed submicromolar potency in biochemical assays of MNK1 and MNK2 and a high ligand
efficiency (LE) value of 0.38 for both MNK1 and MNK2 (Table 1). When 2 was docked in the
aforementioned staurosporine/MNK2 coꢀcrystal structure, two hydrogen bond interactions of the
aminopyridine moiety (Figure 1b) overlapped well with the donorꢀacceptor motif of the lactam
of staurosporine. The aryl ring of the benzimidazole is pointing toward the small hydrophobic
area under the F159 gatekeeper, suggesting a Tꢀshaped edgeꢀtoꢀface πꢀπ stacking interaction.46
The overlay model of 2 on staurosporine suggested that we could identify key interactions in the
lower hinge region and the acidic region around E209 for gaining potency and selectivity (Figure
1b). Based on its ligand efficiency and proposed binding mode, compound 2 was considered as a
good starting point to explore the SAR of the lower hinge and the acidic patch regions.
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. SAR of benzimidazole aminopyridine series at lower hinge region.
ACS Paragon Plus Environment

Page 7 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
The homology model indicated that an aryl or heteroaryl ring at the C5ꢀposition of 2 could
make contact with lower hinge residues and be used as a handle to explore the large Cꢀshaped
lower hinge cavity (Table 1). Compound 3 exhibited improved biochemical MNK1 and MNK2
potencies, but the selectivity against two of the identified offꢀtargets, CDK1 and CDK2,
deteriorated significantly. CDK1 and CDK2 were among the CaMKII kinases identified in the
sequence alignment analysis previously discussed which had significant sequence identity to
MNK2. However, the docking pose of 3 in MNK2 revealed that a polar interaction with S16647
could be made by employing a polar substituent at the meta position of the phenyl ring. It was
anticipated that the selectivity against CDK1 and CDK2 would be increased because MNK2 has
the unique Cꢀshaped lower hinge conformation due to G164 insertion compared to CDK1 and
CDK2. CDK2 has a D86 in the place of the S166 in the MNK2 and CDK2 binding pocket
overlay (Figure 2). Due to the Cꢀshaped conformation, the D86 in CDK2 is slightly further away
from the ligand than S166 in MNK2. Also, the S166 in MNK2 should be attractive to a hydrogen
bond accepting group whereas the D86 in CDK2 is likely to be repulsive due to a loneꢀpair
repulsion. The examination of SAR on the meta position with hydrogen bond acceptorꢀbearing
groups led to compound 4 which further improved the biochemical potency against MNK1 and
MNK2 (0.036 and 0.034 ꢀM, respectively) and improved the selectivity against CDK1 and
CDK2. The docking model of 4 (Figure 2) clearly suggested that the carbonyl oxygen of the
amide functionality has a hydrogen bond interaction with the backbone NH of S166. In addition,
one of the amide methyl groups is anticipated to have a hydrophobic interaction with Pꢀloop
residues. The optimization in the lower hinge region not only increased potency and selectivity,
but also the lipophilic efficiency (LipE)48 by 2.0 and 2.8 units for MNK1 and MNK2. However,
the solubility was not improved, as reflected in high logD value.
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 8 of 44
1
2
3
4
Figure 2. Analysis of the Cꢀshaped lower hinge region in MNK2 and CDK2 docked with 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
The next round of SAR examined the acidic patch region. The crystal structure and docking
model suggested that the acidic patch could be reached from the 7’ꢀposition of the benzimidazole
of 4 (Figure 1b). We reasoned that a basic amineꢀcontaining group would improve not only the
solubility of 4 but also the MNK1 and MNK2 potencies by making hydrogen bond interactions
with the glutamic acid E209 in the acidic patch. Several basic amines such as piperidine,
piperazine, and aminocyclohexane were examined at the 7’ꢀposition of the benzimidazole of 4.
Compound 5 which incorporated a piperazine ring led to >10ꢀfold increase in potency against
MNK1 and MNK2 (0.004 and 0.002 ꢀM, respectively) without loss of selectivity against CDK1
and CDK2. As anticipated, the solubility improved significantly (>100 ꢀM), and the logD value
lowered by 2.2 units. This polar interaction further improved LipE to 6.1 and 6.4 for MNK1 and
MNK2, indicating that the optimization effort was progressing well.49
ACS Paragon Plus Environment

Page 9 of 44
Journal of Medicinal Chemistry
1
2
3
4
Table 2. Multiꢀparameter optimization of the aminopyrazine series.
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
Having identified a potent and selective inhibitor of MNK1 and MNK2, a mechanismꢀbased
cellular assay measuring the phosphorylation of EIF4E, a direct downstream target, of MNK1
and MNK2 was used to determine whether the biochemical inhibition of MNK1 and MNK2
kinases had an effect in a cellular context. The EC50 for inhibition of EIF4E phosphorylation in
KMS11ꢀluc myeloma by compound 5 was 0.13 ꢀM.50 While subsequent SAR examination in the
aminopyridine series failed to improve cellular activity further,51 we turned our attention to the
isosteric aminopyrazine series. Although the aminopyrazines were less potent in the biochemical
assays than the corresponding aminopyridine compounds, they exhibited a narrower ratio
between the enzymatic and cellular readouts. For example, the aminopyrazine 6 (Table 2) was 5ꢀ
to 10ꢀfold less potent against MNK1 and MNK2 than the aminopyridine 5 (Table 1), but the
cellular target modulation tracked with the biochemical potency. Because of this, the decision
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 44
1
2
3
4
5
6
was made to focus on the optimization of the aminopyrazine series with the hope of improving
the biochemical potency.
7
8
9
To understand what was driving the difference between the enzymatic potencies of 5 and 6, the
compounds were docked in MNK2. The docking model of the more potent aminopyridine 5
indicated that the dihedral angle between the aminopyridine and benzimidazole rings was 20
degrees. The torsional energy barrier from the ground state conformation to a bioactive
conformation was anticipated to be higher in the aminopyrazine 6 which has a flat conformation
in the ground state due to two internal hydrogen bonds. A quantum mechanics (QM) study
disclosed a ~1.4 kcal/mol of the torsional energy difference between aminopyridine and
aminopyrazine benzimidazole, corresponding to 10ꢀfold difference in potency.52 We thought that
the introduction of a methyl group on the nitrogen on the benzimidazole in 6 would disrupt the
planarity, enabling 6 to attain the bioactive conformation of 5. This change53 did improve the
biochemical potency 20ꢀ to 40ꢀfold (0.002 and 0.001 ꢀM for MNK1 and MNK2, respectively) as
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
shown in 7 (Table 2). The target modulation potency of 7 also increased 10ꢀfold to 0.005 ꢀM. In
this case, the “magic methyl” effect54 not only influenced the biochemical potency but also the
cell permeability and physicochemical properties. The removal of a hydrogen bond donor led to
an increase in cell permeability (Cacoꢀ2 PappAꢀB: 0.7 × 10ꢀ6 cm/s for 6 vs 9.0 × 10ꢀ6 cm/s for 7),
even with a decrease of logD by 0.8 units. The disruption of molecular planarity enhanced
solubility 3ꢀfold (254 ꢀM) and drugꢀlikeness as demonstrated in the increase of LipE value by
1.6 units.
ACS Paragon Plus Environment

Page 11 of 44
Journal of Medicinal Chemistry
1
2
3
4
Figure 3. Overlaying View of Van Der Waals (VDW) Surface of MNK2 and CDK2 with 7.
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
Incorporating the methyl on the benzimidazole caused the piperazine ring to become
perpendicular to the plane of the benzimidazole ring. This presented the ethylene bridge of the
piperazine ring to the Pꢀloop enabling a hydrophobic interaction. The data suggest that there are
differences in the Pꢀloop regions of MNK2 and CDK2 (Table 2).55 We therefore examined the Pꢀ
loop areas in MNK2 and CDK2 by overlaying the binding pockets in order to find a difference
that can be utilized to improve selectivity further. In addition to the lower hinge region as
mentioned before, we found differences in both Pꢀloop and acidic patch areas (Figure 3), where
MNK2 has a different space arrangement than CDK2. It appeared from the overlay study that
CDK2 might have less space in the Pꢀloop area than MNK2. Further, the Pꢀloop area is primarily
hydrophobic. Thus, a methyl group was incorporated on the piperazine ring of 7 to provide 8
which had similar properties and biochemical potencies as 7 and improved selectivity against
CDK1 and CDK2.
Finally, the lower hinge region was reꢀexamined. Although the docking model suggested that a
plausible position for the amide group should be the rightꢀhand side of the phenyl ring to make a
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 44
1
2
3
4
5
6
7
8
9
hydrogen bond interaction with S166, analysis of the model suggested that the Cꢀshaped loop
between M162 and G165 at the lower hinge area is spacious enough to accommodate the amide
group of 8 if it were presented on the left hand side of the phenyl ring. Thus, if the phenyl bound
in a flipped binding mode which placed the amide group away from the S166 residue, the
potency would be reduced due to the loss of hydrogen bond interaction. To restrict the position
of the amide to the S166 residue, the phenyl ring was replaced with an ortho pyridine (9). Similar
to a 2,2’ꢀbipyridyl system, the rotational barrier of these two rings should be >1.5 kcal/mol based
on the torsion distribution from small molecule crystal structures with the same feature.56 This
predisposes the amide group to be predominantly towards the S166 residue. While the resulting
compound 9 maintained the desired potencies of MNK1 and MNK2, the activities of CDK1 and
CDK2 were completely dialed out (IC50 >25 ꢀM).57
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. Overlaying View of MNK2 and CDK2 with 9.
ACS Paragon Plus Environment

Page 13 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
By closely analyzing the docking model of 9 in CDK2 (Figure 4), the high selectivity was
found to be due to the fact that an oxygen atom of the carbonyl group on a lower hinge L83
residue of CDK2 is in closer proximity to the pyridyl nitrogen of compound 9 (3.2 Å) than that
on the corresponding residue of MNK2 (3.5 Å). More importantly, the lone pair trajectories on
the pyridyl nitrogen of 9 and the oxygen atom at the residue L83 in CDK2 resulted in repulsion
and the high selectivity observed for CDK2. Further, compound 9 showed excellent selectivity
against 53 other kinases from an internal protein kinase panel, where only four kinases were
inhibited with IC50’s <1 ꢀM (CaMK2D 0.69 ꢀM, FLT3 0.28 ꢀM, PIM2 0.73 ꢀM, ROCK2 0.37
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
ꢀM).
While the optimization of the aminopyrazine series led to potent compounds in the pꢀEIF4E
modulation assay, we did not collect cell proliferation data for many of the compounds in the
series because the data were not informative on early compounds due to the offꢀtarget activity,
including CDK1/2. We now have potent, selective compounds which exhibit moderate inhibition
of cell proliferation in the KMS11ꢀluc human multiple myeloma tumor cell line (e.g. EC50 for 9
= 1.7 ꢀM)58 and will begin exploring MNK1/2 dependency across different tumor types.
Compound 9 also exhibited good cell permeability, modest efflux in Cacoꢀ2 assay, good
physicochemical properties such as solubility and logD, and the highest LipE in the
aminopyrazine series. Furthermore, compound 9 did not inhibit any CYP450 isoforms (IC50 >40
ꢀM) and exhibited moderate in vitro clearance in rat liver microsomes (CLint 32 ꢀL/min/mg; ERh
0.54) suggesting that it could be used as an in vivo tool compound, although we did not
investigate its behavior in vivo.
Overall, our structureꢀbased design resulted in an enhanced kinase selectivity profile as well as
>80 fold improvement in pꢀEIF4E EC50 for compound 9 compared to compound 6 (Table 2).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 44
1
2
3
4
5
6
7
8
9
Based on the in vitro biochemical and target modulation potency, ADME and selectivity profile,
compound 9 was selected as a lead to probe the dependence on MNK1/2 in cancer, and this is the
focus of ongoing efforts.
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
Compound 2 was synthesized via Suzuki crossꢀcoupling reaction of commercially available 3ꢀ
(4,4,5,5ꢀtetramethylꢀ1,3,2ꢀdioxaborolanꢀ2ꢀyl)pyridinꢀ2ꢀamine
and
2ꢀbromoꢀ1Hꢀ
benzo[d]imidazole (Scheme 1). Treatment of 2 with Nꢀbromosuccinimide (NBS) provided
brominated intermediate 10, which was then coupled with 11 or 12 to yield 3 or 4. The synthesis
of piperazineꢀcontaining compounds 5 and 6 started from commerically available 3ꢀchloroꢀ2ꢀ
nitroaniline and tertꢀbutyl piperazineꢀ1ꢀcarboxylate, which were subjected to SNAr amination
reaction to provide nitro intermediate 13 (Scheme 2). Subsequent palladiumꢀcatalyzed
hydrogenation of 13 afforded bisꢀanilino intermediate 14, which was coupled with 15 or 16 to
form the corresponding amide and then cyclized to give the benzimidazole intermediate of
aminopyridine and aminopyrazine, 17 or 18. Compounds 5 and 6 were obtained via Suzuki
coupling reaction of boronic acid 12 with the corresponding bromo intermediate 17 or 18
followed by deprotection of Bocꢀgroup.
Scheme 1. Synthesis of compounds 2−4.
ACS Paragon Plus Environment

Page 15 of 44
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
Reagents and conditions: (a) PdCl2(dppf)·CH2Cl2, DME, 2 M Na2CO3 (aq), 120 °C, microwave,
32%; (b) NBS, DMF, 0 °C, 20 min, 57%; (c) PdCl2(dppf)·CH2Cl2, DME, 2 M Na2CO3 (aq), 130
°C, microwave, 15 min, 16% for 3 and 21% for 4.
Scheme 2. Synthesis of compounds 5 and 6.
H2N
O2N
H2N
H2N
NH2
O
H2N
O2N
a
b
N
OH
+
N
N
Y
Cl
N
N
Br
13
14
15
16
Boc
Boc
Y = CH
Y = N
NH2
Y
N
N
NH2
N
N
N
H
c, d
e, f
N
H
N
N
NH
Y
N
Br
N
Boc
17
18
5
6
Y = CH
Y = N
Y = CH
Y = N
O
Reagents and conditions: (a) tertꢀButyl piperazineꢀ1ꢀcarboxylate, K2CO3, DMF, 130 °C, 28 h,
48%; (b) H2, Pd/C, MeOH, 1 h, 96%; (c) 15 or 16, HATU, DIEA, MeCN; (d) AcOH, 120 °C, 1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 44
1
2
3
4
5
6
7
8
h, 41% for 17 (over 2 steps) and 35% for 18 (over 2 steps); (e) 12, PdCl2(dppf)·CH2Cl2, DME, 2
M Na2CO3 (aq), 120 °C, microwave; (f) HCl in doxane, 32% for 5 (over 2 steps); TFA, CH2Cl2,
20% for 6 (over 2 steps).
9
Scheme 3. Synthesis of compounds 7−9.
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
O2N
F
O2N
H
N
O2N
F
N
H
a
c
b
+
N
N
N
Z
Br
Boc
N
Z
N
Z
Boc
Boc
19 Z = H
20 Z = (S)-CH3
21 Z = H
22 Z = (S)-CH3
H2N
NH2
N
NH2
N
N
N
H
N
N
d, e
N
f
N
O
N
N
N
Z
N
N
N
Br
B
Boc
Boc
N
Z
O
25 Z = H
26 Z = (S)-CH3
Boc
27
23 Z = H
24 Z = (S)-CH3
g, i
h, i
NH2
N
NH2
N
N
N
N
N
N
N
N
Z
N
NH
NH
N
N
N
7 Z = H
8 Z = (S)-CH3
9
O
O
Reagents and conditions: (a) tertꢀButyl piperazineꢀ1ꢀcarboxylate or tertꢀbutyl (S)ꢀ2ꢀ
methylpiperazineꢀ1ꢀcarboxylate, Pd2(dba)3, Xantphos, Cs2CO3, dioxane, 110 °C, 16 h, 68% for
19 or 32% for 20; (b) Methylamine, K2CO3, DMF, 130 °C, 16 h, 54% for 21 and 48% for 22; (c)
H2, Pd/C, MeOH, 2 h, 78% for 23 and 85% for 24; (d) 16, EDC, HOAt, DMF; (e) AcOH, 130
ACS Paragon Plus Environment

Page 17 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
6
°C, 0.5 h, 37% for 25 (over 2 steps) and 50% for 26 (over 2 steps); (f) Bis(pinacolato)diboron,
PdCl2(dppf)·CH2Cl2, KOAc, dioxane, 120 °C, microwave; (g) 12, PdCl2(dppf)·CH2Cl2, DME, 2
7
8
9
M
Na2CO3 (aq), 120 °C, microwave; (h) 2ꢀBromoꢀN,Nꢀdimethylisonicotinamide,
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
PdCl2(dppf)·CH2Cl2, DME, 2 M Na2CO3 (aq), 120 °C, microwave; (i) TFA, CH2Cl2, 15% for 7
(over 2 steps) and 20% for 8 (over 2 steps), and 36% for 9 (over 3 steps).
For the synthesis of compounds 7−9 containing Nꢀmethyl and piperidinyl groups on
benzimidazole ring, 1ꢀbromoꢀ2ꢀfluoroꢀ3ꢀnitrobenzene was reacted with tertꢀbutyl piperazineꢀ1ꢀ
carboxylate or tertꢀbutyl (S)ꢀ2ꢀmethylpiperazineꢀ1ꢀcarboxylate under a palladiumꢀcatalyzed
amination reaction condition59 (Scheme 3). The SNAr amination reaction of 19 or 20 with
methylamine followed by the reduction of the nitro group of 21 or 22 yielded bisꢀanilino
intermediate 23 or 24. The following amide bond formation/cyclization reactions with 16
provided the benzimidazole derivative 25 or 26, which further underwent Suzuki coupling
reaction with 12 to afford 7 or 8 after Bocꢀdeprotection. Compound 26 was converted to the
corresponding boronic ester 27, and subsequent palladiumꢀcatalyzed cross coupling reaction with
2ꢀbromoꢀN,Nꢀdimethylisonicotinamide and the removal of Boc group led to compound 9.
CONCLUSION
In summary, structureꢀbased rational drug design utilizing a published MNK crystal structure
allowed for the rapid development of a fragment hit to a potent and selective inhibitor of MNK1
and MNK2 kinases (9) which had good solubility and permeability. This compound exhibited
superior potency and selectivity compared to MNK inhibitors described in the literature and can
be used as a lead to further evaluate and prosecute this pathway and as a starting point for future
drug discovery efforts.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 44
1
2
3
4
EXPERIMENTAL SECTION
5
6
7
8
Chemistry. The compounds and/or intermediates were characterized by high performance liquid
chromatography (HPLC) using a Waters Millenium chromatography system with a 2695
Separation Module (Milford, MA). The analytical columns were Alltima C18 reverse phase,
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
4.6 × 50 mm, flow 2.5 mL/min, from Alltech (Deerfield, IL). A gradient elution was used (flow
2.5 mL/min), typically starting with 5% acetonitrile/95% water and progressing to 100%
acetonitrile over a period of 10 min. All solvents contained 0.1% TFA. Compounds were
detected by ultraviolet light (UV) absorption at either 220 or 254 nm. HPLC solvents were from
Burdick and Jackson (Muskegan, MI) or Fisher Scientific (Pittsburgh, PA). All compounds
where biological data are presented have >95% purity as determined by HPLC. In some
instances, purity was assessed by thin layer chromatography (TLC) using glass or plastic backed
silica gel plates, such as, for example, BakerꢀFlex Silica Gel 1B2ꢀF flexible sheets. TLC results
were readily detected visually under ultraviolet light, or by employing well known iodine vapor
and other various staining techniques.
Mass spectrometric analysis was performed on one of two LCMS instruments: a Waters
System (Alliance HT HPLC and a Micromass ZQ mass spectrometer; Column: Eclipse XDBꢀ
C18, 2.1 × 50 mm; solvent system: 5ꢀ95% (or 35ꢀ95%, or 65ꢀ95% or 95ꢀ95%) acetonitrile in
water with 0.05% TFA; flow rate 0.8 mL/min; molecular weight range 200ꢀ1500; cone Voltage
20 V; column temperature 40 °C) or a Hewlett Packard System (Series 1100 HPLC; Column:
Eclipse XDBꢀC18, 2.1 × 50 mm; solvent system: 1ꢀ95% acetonitrile in water with 0.05% TFA;
flow rate 0.8 mL/min; molecular weight range 150ꢀ850; cone Voltage 50 V; column temperature
30 °C). All masses were reported as those of the protonated parent ions.
ACS Paragon Plus Environment

Page 19 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
HRMS ESI‐MS data were recorded using a Synapt G2 HDMS (TOF mass spectrometer,
Waters) with electrospray ionization source. The resolution of the MS system was approximately
15000. Leucine Enkephalin was used as lock mass (internal standard) infused from lockspray
probe. The compound was infused into the mass spectrometer by UPLC (Acquity, Waters) from
sample probe. The separation was performed on Acquity UPLC BEHꢀC18 1 × 50 mm column at
0.2 mL/min flow rate with the gradient from 5% to 95% in 3 min. Solvent A was water with
0.1% formic acid and solvent B was acetonitrile with 0.1% formic acid. The mass accuracy of
the system has been found to be <5 ppm with lock mass.
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
1HꢀNuclear magnetic resonance (NMR) analysis was performed on some of the compounds
with a Varian 300 or 400 MHz NMR and a Bruker 500 MHz NMR. The spectral reference was
13
either TMS or the known chemical shift of the solvent. C‐NMR spectra were recorded using a
Bruker AVANCE‐500 NMR spectrometer operating at a frequency of 125.77 MHz for 13C
equipped with a 5 mm BBO probe.
Preparative separations were carried out using a Teledyne ISCO chromatography system, by
flash column chromatography using silica gel (230ꢀ400 mesh) packing material, or by HPLC
using a Waters 2767 Sample Manager, C18 reversed phase column, 30 × 50 mm, flow 75
mL/min. Typical solvents employed for the Teledyne ISCO chromatography system and flash
column chromatography were dichloromethane, methanol, ethyl acetate, hexane, acetone,
aqueous ammonia (or ammonium hydroxide), and triethylamine. Typical solvents employed for
the reverse phase HPLC were varying concentrations of acetonitrile and water with 0.1% TFA.
3-(1H-Benzo[d]imidazol-2-yl)pyridin-2-amine (2).60,61 To a solution of 3ꢀ(4,4,5,5ꢀtetramethylꢀ
1,3,2ꢀdioxaborolanꢀ2ꢀyl)pyridinꢀ2ꢀamine (633 mg, 2.88 mmol) in DME (9 mL) and 2 M Na2CO3
(aq, 3.3 mL) in a microwave vial was added 2ꢀbromoꢀ1Hꢀbenzo[d]imidazole (436 mg, 2.2 mmol)
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 44
1
2
3
4
5
6
7
8
9
and PdCl2(dppf)·CH2Cl2 (126 mg, 160 µmol). The reaction was heated at 130 °C for 20 min
under microwave irradiation. The reaction mixture was partitioned between EtOAc (30 mL) and
water (30 mL). After the two layers were separated, the organic layer was washed with water and
brine, dried over anhydrous Na2SO4, filtered, and concentrated in vacuo. The crude product was
purified by automated flash chromatography (30ꢀ100% EtOAc in heptane) to yield 2 (144 mg,
31%). LCMS (m/z): 211 (MH+), 0.46 min. HRMS (m/z) calcd for C12H11N4 (MH+) 211.0984,
found 211.0981. HPLC: 2.35 min, >98% purity. 1H NMR (400 MHz, CD3OD) δ ppm 8.13 (dd, J
= 7.83, 1.57 Hz, 1H), 8.03 (dd, J = 5.09, 1.57 Hz, 1H), 7.59 (br. s, 2H), 7.24 (dd, J = 5.87, 3.13
Hz, 2H), 6.77 (dd, J = 7.43, 5.09 Hz, 1H).
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-(1H-Benzo[d]imidazol-2-yl)-5-bromopyridin-2-amine (10). To a solution of 2 (132 mg, 0.63
mmol) in DMF (4 mL) was added NBS (112 mg, 0.63 mmol) at 0 °C. The reaction mixture was
stirred at 0 °C for 20 min. After water (0.5 mL) was added, a precipitate formed and was filtered
off. The residue was purified by preparative reverse phase HPLC. After the removal of
acetonitrile from the combined pure fractions followed by addition of saturated Na2CO3 (aq), the
product was extracted with EtOAc (200 mL). The organic layer was washed with water, filtered,
and concentrated in vacuo to yield 10 (104 mg, 57%). LCMS (m/z): 288.9, 290.9 (MH+), 0.55
1
min. H NMR (400 MHz, CD3OD) δ ppm 8.29 (d, J = 2.35 Hz, 1H), 8.08 (d, J = 1.96 Hz, 1H),
7.60 (br. s, 2H), 7.32ꢀ7.18 (m, 2H).
3-(1H-Benzo[d]imidazol-2-yl)-5-phenylpyridin-2-amine (3).62 To 10 (20 mg, 0.069 mmol) in a
microwave vial was added 11 (17.1 mg, 0.14 mmol), PdCl2(dppf)·CH2Cl2 (8.5 mg, 10.4 µmol),
DME (800 µL), and 2 M Na2CO3 (aq, 140 µL). The reaction was heated at 130 °C for 15 min
under microwave irradiation. The reaction mixture was partitioned between EtOAc (3 mL) and
water (3 mL). After the two layers were separated, the organic layer was washed with water and
ACS Paragon Plus Environment

Page 21 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
brine, dried over anhydrous Na2SO4, filtered, and concentrated in vacuo. The crude product was
purified by preparative reverse phase HPLC. The pure fractions were combined and lyophilized
to yield 3 as TFA salt (4.4 mg, 16%). LCMS (m/z): 287 (MH+), 0.67 min. HRMS (m/z) calcd for
C18H15N4 (MH+) 287.1297, found 287.1295. HPLC: 2.57 min, >99% purity. 1H NMR (400 MHz,
CD3OD) δ ppm 8.98 (d, J = 1.96 Hz, 1H), 8.30 (d, J = 1.96 Hz, 1H), 7.79ꢀ7.74 (m, 2H), 7.70 (dd,
J = 5.48, 3.52 Hz, 2H), 7.56 (t, J = 7.63 Hz, 2H), 7.51ꢀ7.42 (m, 1H), 7.40ꢀ7.32 (m, 2H).
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-(6-Amino-5-(1H-benzo[d]imidazol-2-yl)pyridin-3-yl)-N,N-dimethylbenzamide (4). To 10
(20 mg, 0.069 mmol) in a microwave vial was added 12 (26.7 mg, 0.14 mmol),
PdCl2(dppf)·CH2Cl2 (8.5 mg, 10.4 µmol), DME (800 µL), and 2 M Na2CO3 (aq, 140 µL). The
reaction was heated at 130 °C for 15 min under microwave irradiation. The reaction mixture was
partitioned between EtOAc (3 mL) and water (3 mL). After the two layers were separated, the
organic layer was washed with water and brine, dried over anhydrous Na2SO4, filtered, and
concentrated in vacuo. The crude product was purified by preparative reverse phase HPLC. The
pure fractions were combined and lyophilized to yield 4 as TFA salt (7 mg, 21%). LCMS (m/z):
358.1 (MH+), 0.62 min. HRMS (m/z) calcd for C21H20N5O (MH+) 358.1668, found 358.1667.
1
HPLC: 1.17 min, >99% purity. H NMR (400 MHz, CD3OD) δ ppm 8.98 (d, J = 1.96 Hz, 1H),
8.36 (d, J = 1.96 Hz, 1H), 7.87 (d, J = 7.83 Hz, 1H), 7.82 (s, 1H), 7.71 (dd, J = 5.87, 3.52 Hz,
2H), 7.65 (t, J = 7.63 Hz, 1H), 7.53 (d, J = 7.83 Hz, 1H), 7.40ꢀ7.32 (m, 2H), 3.16 (s, 3H), 3.07
(s, 3H).
tert-Butyl 4-(3-amino-2-nitrophenyl)piperazine-1-carboxylate (13). To 3ꢀchloroꢀ2ꢀ
nitroaniline (1 g, 5.8 mmol) in a microwave vial was added K2CO3 (2 g, 14.5 mmol), tertꢀbutyl
piperazineꢀ1ꢀcarboxylate (2.7 g, 14.5 mmol), and DMF (18 mL) at room temperature. The
reaction mixture was heated at 130 °C in oil bath for 28 h. After diluted with water, the reaction
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 44
1
2
3
4
5
6
7
8
9
mixture was extracted with EtOAc (100 mL) three times. The combined organic layers were
washed with water and brine, dried over anhydrous Na2SO4, filtered, and concentrated in vacuo.
The crude product was purified by automated flash chromatography (0ꢀ50% EtOAc in heptane)
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
to yield 13 (890 mg, 48%). LCMS (m/z): 323 (MH+), 0.87 min. H NMR (400MHz, CDCl3) δ
ppm 7.15 (t, J = 8.0 Hz, 1H), 6.41 (d, J = 8.2 Hz, 1H), 6.45 (d, J = 8.2 Hz, 1H), 4.78 (br. s, 2 H),
3.58ꢀ3.48 (m, 4 H), 2.95 (br. s, 4 H), 1.48 (s, 9 H).
tert-Butyl 4-(2,3-diaminophenyl)piperazine-1-carboxylate (14). A solution of 13 (800 mg, 2.5
mmol) in MeOH (8.3 mL) was flushed with nitrogen gas for 10 min followed by addition of
Pd/C (26 mg, 0.25 mmol). After charged with hydrogen gas, the reaction mixture was equipped
with hydrogen gas balloon and stirred at room temperature for 1 h. The reaction mixture was
filterd through diatomaceous earth and washed with MeOH (10 mL) and EtOAc (20 mL). The
filtrate was concentrated in vacuo to yield 14 (700 mg, 96%) which was used in the next step
1
without further purification. LCMS (m/z): 293 (MH+), 0.6 min. H NMR (400MHz, CDCl3) δ
ppm 6.79ꢀ6.67 (m, 1H), 6.63ꢀ6.55 (m, 2H), 3.72ꢀ3.42 (m, 4H), 3.22 (br. s, 2 H), 2.84 (br. s, 4H),
1.55ꢀ1.42 (m, 9H).
tert-Butyl
4-(2-(2-amino-5-bromopyridin-3-yl)-1H-benzo[d]imidazol-7-yl)piperazine-1-
carboxylate (17). To a solution of 2ꢀaminoꢀ5ꢀbromonicotinic acid (15) (278 mg, 1.28 mmol) in
acetonitrile (6 mL) was added 14 (375 mg, 1.28 mmol), HATU (683 mg, 1.8 mmol), and DIEA
(0.43 mL, 2.4 mmol). The reaction mixture was stirred at room temperature for 6 h. LCMS
indicated that the coupling reaction was completed. LCMS (m/z): 491, 493 (MH+), 0.83 min.
After partitioning between EtOAc (150 mL) and saturated NaHCO3 (aq, 150 mL), the organic
layer was washed with water and brine, dried over anhydrous Na2SO4, filtered off, and
concentrated in vacuo. The crude product was dissolved in acetic acid (4 mL, 70 mmol). The
ACS Paragon Plus Environment

Page 23 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
reaction mixture was heated at 120 °C in oil bath for 1 h. After neutralizing with 1 N NaOH (aq),
the reaction mixture was extracted by EtOAc. The combined organic layers were washed with
water and brine, dried over anhydrous Na2SO4, filtered, and concentrated in vacuo. The crude
product was purified by automated flash chromatography to yield 17 (248 mg, 41%). LCMS
(m/z): 473, 475 (MH+), 0.81 min. 1H NMR (400MHz, CD3OD) δ ppm 8.29 (m, 1 H), 8.07 (d, J =
1.96 Hz, 1 H), 7.09ꢀ7.23 (m, 2 H), 6.70 (m, 1 H), 3.69 (m, 4 H), 3.45 (m, 4 H), 3.69 (s, 9 H).
3-(6-Amino-5-(7-(piperazin-1-yl)-1H-benzo[d]imidazol-2-yl)pyridin-3-yl)-N,N-
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
dimethylbenzamide (5). To 17 (32 mg, 0.068 mmol) in a microwave vial was added 12 (26.1
mg, 0.135 mmol), DME (600 µL), 2 M Na2CO3 (aq, 135 µL), and PdCl2(dppf)·CH2Cl2 (8.29 mg,
10.1 µmol). The reaction mixture was heated at 120 °C for 15 min under microwave irradiation.
The reaction mixture was partitioned between EtOAc (3 mL) and water (3 mL). After the two
layers were separated, the organic layer was washed with water and brine, dried over anhydrous
Na2SO4, filtered, and concentrated in vacuo. The crude product was dissolved in 2 mL of 4 M
HCl in dioxane. The reaction mixture was stirred at room temperature for 1 h. After the reaction
mixture was concentrated to dryness, the crude product was purified by preparative reverse phase
HPLC. The pure fractions were combined and lyophilized to yield 5 as TFA salt (12.1 mg, 32%).
LCMS (m/z): 457.1 (MH+), 0.56 min. HRMS (m/z) calcd for C25H28N7O (MH+) 442.2355, found
442.2349. HPLC: 1.92 min, >99% purity. 1H NMR (400 MHz, CD3OD) δ ppm 8.89 (d, J = 1.96
Hz, 1H), 8.36 (d, J = 1.96 Hz, 1H), 7.83 (d, J = 7.83 Hz, 1H), 7.78 (s, 1H), 7.62 (t, J = 7.63 Hz,
1H), 7.49 (d, J = 7.43 Hz, 1H), 7.26ꢀ7.30 (m, 2H), 6.77ꢀ6.84 (m, 1H), 3.71 (d, J = 5.09 Hz, 4H),
3.46ꢀ3.56 (m, 4H), 3.14 (s, 3H), 3.04 (s, 3H).
3-(5-Amino-6-(7-(piperazin-1-yl)-1H-benzo[d]imidazol-2-yl)pyrazin-2-yl)-N,N-
dimethylbenzamide (6). Following the steps c, d, e, and f in Scheme 2, compounds 12, 14 and
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 44
1
2
16 yielded 6 as TFA salt (8.2 mg, 20%). LCMS (m/z): 443.1 (MH+), 0.62 min. HRMS (m/z)
calcd for C24H27N8O (MH+) 443.2308, found 443.2310. HPLC: 2.56 min, >99% purity. 1H NMR
(300 MHz, DMSOꢀd6) δ ppm 8.83 (s, 1 H), 8.32 (d, J = 7.9 Hz, 1 H), 8.26 (s, 1 H), 7.55 (t, J =
7.6 Hz, 1 H), 7.37 (d, J = 7.3 Hz, 1 H), 7.29ꢀ7.17 (m, 2 H), 6.69 (d, J = 6.7 Hz, 1 H), 3.69 (br. s.,
4 H), 3.39 (m , 4H) 3.03 (s, 3 H), 2.94 (s, 3 H).
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
tert-Butyl 4-(2-fluoro-3-nitrophenyl)piperazine-1-carboxylate (19). A solution of tertꢀbutyl
piperazineꢀ1ꢀcarboxylate (1.9 g, 10.2 mmol), 1ꢀbromoꢀ2ꢀfluoroꢀ3ꢀnitrobenzene (1.5 g, 6.8
mmol), cesium carbonate (4.4 g, 13.6 mmol), Xantphos (237 mg, 0.409 mmol), Pd2(dba)3 (312
mg, 0.34 mmol), and dioxane (23 mL) in a sealed glass reaction vessel was flushed with nitrogen
gas for 10 min. Then, the reaction mixture was heated at 110 °C in oil bath overnight. After
diluted with water, the reaction mixture was extracted with EtOAc (100 mL) three times. The
combined organic layers were washed with water, saturated NaHCO3 (aq), and brine. The
organic layer was dried over anhydrous Na2SO4, filtered, and concentrated in vacuo. The crude
product was purified by automated flash chromatography (0ꢀ40% EtOAc/heptane) to yield 19
1
(2.2 g, 68%). LCMS (m/z): 270 (MH+ꢀBoc), 1.15 min. H NMR (400MHz, CDCl3) δ ppm 7.59
(d, J = 3.5 Hz, 1H), 7.21ꢀ7.15 (m, 2H), 3.68ꢀ3.55 (m, 4H), 3.14ꢀ 3.05 (m, 4 H), 1.54ꢀ1.43 (m,
9H).
tert-Butyl 4-(2-(methylamino)-3-nitrophenyl)piperazine-1-carboxylate (21). To a solution of
19 (500 mg, 1.54 mmol) in DMF (3 mL) was added methylamine (2 M solution in THF; 1.5 mL,
3.07 mmol). The reaction mixture was stirred at 130 °C overnight. After quenched with saturated
NaHCO3 (aq), the reaction mixture was extracted with EtOAc (10 mL) three times. The
combined organic layers were washed with water and brine, dried over anhydrous Na2SO4,
filtered, and concentrated in vacuo. The crude product was purified by automated flash
ACS Paragon Plus Environment

Page 25 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
chromatography (gradient EtOAc in heptane) to yield 21 (280 mg, 54%). Rf = 0.3 (30% EtOAc
in heptane). LCMS (m/z): 337 (MH+), 1.06 min. 1H NMR (400MHz, CDCl3) δ ppm 7.73 (dd, J =
1.2, 8.6 Hz, 1H), 7.15 (dd, J = 1.4, 7.6 Hz, 1H), 6.72 (t, J = 8.0 Hz, 1H), 3.69ꢀ3.52 (m, 4H), 3.00
(d, J = 3.9 Hz, 3H), 2.88 (br. s, 4H), 1.52ꢀ1.46 (m, 9H).
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
tert-Butyl 4-(3-amino-2-(methylamino)phenyl)piperazine-1-carboxylate (23). A solution of
21 (280 mg, 0.83 mmol) in MeOH (2.8 mL) was flushed with nitrogen gas for 10 min followed
by addition of Pd/C (89 mg). After charged with hydrogen gas, the reaction mixture was
equipped with hydrogen gas balloon and stirred at room temperature for 2 h. The reaction
mixture was filterd through diatomaceous earth and washed with MeOH (5 mL) and EtOAc (10
mL). The filtrate was concentrated in vacuo to yield 23 (200 mg, 78%), which was used in the
next step without further purification. LCMS (m/z): 307.1 (MH+), 0.67 min.
tert-Butyl
4-(2-(3-amino-6-bromopyrazin-2-yl)-1-methyl-1H-benzo[d]imidazol-7-
yl)piperazine-1-carboxylate (25). To a solution of 23 (253 mg, 0.826 mmol) in DMF (2.8 mL)
was added 16 (180 mg, 0.11 mmol), HOAt (135 mg, 0.99 mmol), and EDC (206 mg, 1.07
mmol). The reaction mixture was stirred at room temperature overnight. LCMS indicated that the
coupling reaction was completed. LCMS (m/z): 506, 508 (MH+), 0.82 min and 0.86 min. The
reaction mixture was partitioned between EtOAc (10 mL) and saturated NaHCO3 (aq, 10 mL).
After the two layers were separated, the organic layer was washed with water and brine, dried
over anhydrous Na2SO4, filtered off, and concentrated in vacuo. The crude product was dissolved
in acetic acid (2 mL). The reaction mixture was heated at 130 °C in oil bath for 20 min. After
neutralizing with 1 N NaOH (aq), the reaction mixture was extracted by EtOAc. The combined
organic layers were washed with water and brine, dried over anhydrous Na2SO4, filtered, and
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 44
1
2
3
4
5
6
concentrated in vacuo to yield 25 (150 mg, 37%) which was used in the next step without further
purification. LCMS (m/z): 488, 490 (MH+), 1.12 min.
7
8
9
3-(5-Amino-6-(1-methyl-7-(piperazin-1-yl)-1H-benzo[d]imidazol-2-yl)pyrazin-2-yl)-N,N-
dimethylbenzamide (7). To 25 (75 mg, 0.154 mmol) in a microwave vial was added 12 (59.3
mg, 0.31 mmol), DME (384 µL), 2 M Na2CO3 (aq, 128 µL), and PdCl2(dppf)·CH2Cl2 (18.1 mg,
23 µmol). The reaction mixture was heated at 120 °C for 15 min under microwave irradiation.
The reaction mixture was partitioned between EtOAc (3 mL) and water (3 mL). After the two
layers were separated, the organic layer was washed with water and brine, dried over anhydrous
Na2SO4, filtered, and concentrated in vacuo. The crude product was dissolved in DCM (1 mL)
and TFA (0.3 mL). The reaction mixture was stirred at room temperature for 1 h. After the
reaction mixture was concentrated to dryness, the crude product was purified by preparative
reverse phase HPLC. The pure fractions were combined and lyophilized to yield 7 as TFA salt
(35 mg, 40%). LCMS (m/z): 457.1 (MH+), 0.56 min. HRMS (m/z) calcd for C25H29N8O (MH+)
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
457.2464, found 457.2465. HPLC: 2.19 min, >99% purity. H NMR (400 MHz, DMSOꢀd6) δ
ppm 8.79 (s, 1H), 8.11 (d, J = 7.8 Hz, 1H), 8.03 (s, 1H), 7.60ꢀ7.49 (m, 2H), 7.38 (d, J = 7.4 Hz,
1H), 7.24 (t, J = 7.8 Hz, 1H), 7.07 (d, J = 7.8 Hz, 1H), 4.58 (s, 3H), 3.00 (s, 3H), 2.95 (s, 3H).
(S)-tert-Butyl 4-(2-fluoro-3-nitrophenyl)-2-methylpiperazine-1-carboxylate (20). A solution
of (S)ꢀtertꢀbutyl 2ꢀmethylpiperazineꢀ1ꢀcarboxylate (546 mg, 2.73 mmol), 1ꢀbromoꢀ2ꢀfluoroꢀ3ꢀ
nitrobenzene (400 mg, 1.82 mmol), cesium carbonate (1.18 g, 3.64 mmol), Xantphos (63.1 mg,
0.109 mmol), Pd2(dba)3 (83 mg, 0.091 mmol), and dioxane (6 mL) in a sealed glass reaction
vessel was flushed with nitrogen gas for 10 min. Then, the reaction mixture was heated at 110 °C
in oil bath overnight. After diluted with water, the reaction mixture was extracted with EtOAc
(15 mL) three times. The combined organic layers were washed with water, saturated NaHCO3
ACS Paragon Plus Environment

Page 27 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
6
(aq), and brine. The organic layer was dried over anhydrous Na2SO4, filtered, and concentrated
in vacuo. The crude product was purified by automated flash chromatography (0ꢀ40% EtOAc in
7
8
9
1
heptane) to yield 20 (200 mg, 32%). LCMS (m/z): 240 (MH+ꢀBoc), 1.09 min. H NMR
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
(400MHz, CDCl3) δ ppm 7.63ꢀ7.54 (m, 1H), 7.20ꢀ7.13 (m, 2H), 4.35 (br. s, 1H), 3.98 (d, J =
13.3 Hz, 1H), 3.35ꢀ3.22 (m, 4H), 2.93 (dd, J = 3.5, 11.7 Hz, 1H), 2.87ꢀ2.78 (m, 1H), 1.49 (s,
9H), 1.36 (d, J = 6.7 Hz, 3H).
(S)-tert-Butyl 2-methyl-4-(2-(methylamino)-3-nitrophenyl)-piperazine-1-carboxylate (22).
To a solution of 20 (200 mg, 0.59 mmol) in DMF (2 mL) was added methylamine (2 M solution
in THF; 0.59 mL, 1.8 mmol). The reaction mixture was stirred at 130 °C overnight. After
quenched with saturated NaHCO3 (aq), the reaction mixture was extracted with EtOAc (10 mL)
three times. The combined organic layers were washed with water and brine, dried over
anhydrous Na2SO4, filtered, and concentrated in vacuo. The crude product was purified by
automated flash chromatography (gradient EtOAc in heptane) to yield 22 (99 mg, 48%). Rf = 0.3
1
(30% EtOAc in heptane). LCMS (m/z): 351 (MH+), 1.11 min. H NMR (400MHz, CDCl3) δ
ppm 7.69 (d, J = 8.6 Hz, 1H), 7.12 (d, J = 6.7 Hz, 1H), 6.72 (t, J = 8.2 Hz, 1H), 6.58 (br. s., 1H),
4.39 (m, 1H), 3.96 (d, J = 12.9 Hz, 1H), 3.37ꢀ3.23 (m, 1H), 3.13 (d, J = 10.2 Hz, 1H), 2.96 (d, J
= 5.5 Hz, 3H), 2.92ꢀ2.85 (m, 2H), 2.65ꢀ2.47 (m, 1H), 1.49 (s, 9H), 1.36 (d, J = 6.7 Hz, 3H).
(S)-tert-Butyl 4-(3-amino-2-(methylamino)phenyl)-2-methylpiperazine-1-carboxylate (24).
A solution of 22 (180 mg, 0.51 mmol) in MeOH (17 mL) was flushed with nitrogen gas for 10
min followed by addition of Pd/C (55 mg, 0.51 mmol). After charged with hydrogen gas, the
reaction mixture was equipped with hydrogen gas balloon and stirred at room temperature for 1
h. The reaction mixture was filtered through diatomaceous earth and washed with MeOH (20
mL) and EtOAc (50 mL). The filtrate was concentrated in vacuo to yield 24 (138 mg, 85%)
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 44
1
2
which was used in the next step without further purification. LCMS (m/z): 321.1 (MH+), 0.69
min.
3
4
5
6
7
8
9
(S)-tert-Butyl 4-(2-(3-amino-6-bromopyrazin-2-yl)-1-methyl-1H-benzo[d]imidazol-7-yl)-2-
methylpiperazine-1-carboxylate (26). To a solution of 16 (23.8 mg, 0.11 mmol) in DMF (364
µL) was added 24 (35 mg, 0.11 mmol), HOAt (17.8 mg, 0.13 mmol), and EDC (27.2 mg, 0.14
mmol). The reaction mixture was stirred at room temperature overnight. LCMS indicated that the
coupling reaction was completed. LCMS (m/z): 520, 522 (MH+), 0.90 min. After partitioning
between EtOAc (10 mL) and saturated NaHCO3 (aq, 10 mL), a solid was filtered off which was
the desired product. The filtrate was washed with water and brine, dried over anhydrous Na2SO4,
filtered off, and concentrated in vacuo to yield the crude product. The combined products were
dissolved in acetic acid (2 mL). The reaction mixture was heated at 110 °C in oil bath for 20 min.
After neutralizing with 1 N NaOH (aq), the reaction mixture was extracted by EtOAc. The
combined organic layers were washed with water and brine, dried over anhydrous Na2SO4,
filtered, and concentrated in vacuo to yield 26 (27 mg, 50%) which was used in the next step
without further purification. LCMS (m/z): 502, 504 (MH+), 1.16 min.
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
(S)-3-(5-amino-6-(1-methyl-7-(3-methylpiperazin-1-yl)-1H-benzo[d]imidazol-2-yl)pyrazin-
2-yl)-N,N-dimethylbenzamide (8). Following the steps g and i in Scheme 3, compound 12 and
26 yielded compound 8 as TFA salt (14.6 mg, 20% over 2 steps). LCMS (m/z): 471.2 (MH+),
0.59 min. HRMS (m/z) calcd for C26H31N8O (MH+) 471.2621, found 471.2616. HPLC: 2.26 min,
1
>99% purity. H NMR (400 MHz, DMSOꢀd6) δ ppm 9.05 (br. s, 1H), 8.83 (s, 1H), 8.15 (d, J =
7.8 Hz, 1H), 8.08 (s, 1H), 7.63ꢀ7.53 (m, 2H), 7.42 (d, J = 7.4 Hz, 1H), 7.27 (t, J = 7.8 Hz, 1H),
7.12 (d, J = 7.8 Hz, 1H), 4.61 (s, 3H), 3.52ꢀ3.37 (m, 6H), 2.99 (s, 3H), 2.95 (s, 3H), 2.87ꢀ2.77
(m, 1H), 1.30 (d, J = 5.9 Hz, 3H).
ACS Paragon Plus Environment

Page 29 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
tert-Butyl (S)-4-(2-(3-amino-6-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)pyrazin-2-yl)-1-
methyl-1H-benzo[d]imidazol-7-yl)-2-methylpiperazine-1-carboxylate (27). To a solution of
26 (44 mg, 0.088 mmol) in dioxane (146 µL) in a microwave vial was added 4,4,4',4',5,5,5',5'ꢀ
octamethylꢀ2,2'ꢀbi(1,3,2ꢀdioxaborolane) (44.5 mg, 0.175 mmol), PdCl2(dppf)·CH2Cl2, (7.15 mg,
8.76 µmol), and KOAc (25.8 mg, 0.263 mmol). After flushed with nitrogen gas for 10 min, the
reaction mixture was heated at 120 °C for 10 min under microwave irradiation. The reaction
mixture was filtered through diatomaceous earth. After washed with EtOAc, the filtrate was
concentrated in vacuo to yield 27 which was used in the next step without further purification.
LCMS (m/z): 468 (MH+), 0.8 min for the boronic acid of the desired product.
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
(S)-2-(5-Amino-6-(1-methyl-7-(3-methylpiperazin-1-yl)-1H-benzo[d]imidazol-2-yl)pyrazin-
2-yl)-N,N-dimethylisonicotinamide (9). To a solution of 27 (40 mg, 0.073 mmol) in DME (182
µL) and 2 M Na2CO3 (aq, 61 µL) in a microwave vial was added 2ꢀbromoꢀN,Nꢀ
dimethylisonicotinamide (25.0 mg, 0.11 mmol) and PdCl2(dppf)·CH2Cl2 (8.9 mg, 10.9 µmol).
The reaction mixture was heated at 120 °C for 10 min under microwave irradiation. The reaction
mixture was partitioned between EtOAc (3 mL) and water (3 mL). After the two layers were
separated, the organic layer was washed with water and brine, dried over anhydrous Na2SO4,
filtered, and concentrated in vacuo. The crude product was dissolved in CH2Cl2 (1 mL) and TFA
(0.3 mL). The reaction mixture was stirred at room temperature for 1 h. After diluted with
toluene (2 mL), the reaction mixture was concentrated to dryness. The crude product was
purified by preparative reverse phase HPLC. The pure fractions were combined and lyophilized
to yield 9 as TFA salt (16.4 mg, 36% over 3 steps). LCMS (m/z): 472.2 (MH+), 0.54 min. HRMS
(m/z) calcd for C25H30N9O (MH+) 472.2573, found 472.2570. HPLC: 1.84 min, >99% purity. 1H
NMR (500 MHz, DMSOꢀd6) δ ppm 9.28 (br. s, 1H), 9.11 (s, 1H), 8.75 (dd, J = 4.9, 0.7 Hz, 1H),
ACS Paragon Plus Environment