Structure-Guided Design of Group i Selective p21-Activated Kinase Inhibitors

Article abstract of DOI:10.1021/acs.jmedchem.5b00572

The p21-activated kinases (PAKs) play important roles in cytoskeletal organization, cellular morphogenesis, and survival and have generated significant attention as potential therapeutic targets for cancer. Following a high-throughput screen, we identified an aminopyrazole scaffold-based series that was optimized to yield group I selective PAK inhibitors. A structure-based design effort aimed at targeting the ribose pocket for both potency and selectivity led to much-improved group I vs II selectivity. Early lead compounds contained a basic primary amine, which was found to be a major metabolic soft spot with in vivo clearance proceeding predominantly via N-acetylation. We succeeded in identifying replacements with improved metabolic stability, leading to compounds with lower in vivo rodent clearance and excellent group I PAK selectivity.

Full text of DOI:10.1021/acs.jmedchem.5b00572

Subscriber access provided by UNIV OF CALIFORNIA SAN DIEGO LIBRARIES
Article
Structure-Guided Design of Group I Selective p21-Activated Kinase Inhibitors
James J. Crawford, Wendy Lee, Ignacio Aliagas, Simon Mathieu, Klaus Hoeflich, Wei Zhou, Weiru
Wang, Lionel Rouge, Lesley Murray, Hank La, Ning Liu, Peter Wei-Jen Fan, Jonathan Cheong,
Christopher Heise, Sreemathy Ramaswamy, Robert Mintzer, Yanzhou Liu, Qi Chao, and Joachim Rudolph
J. Med. Chem., Just Accepted Manuscript • Publication Date (Web): 01 Jun 2015
Downloaded from http://pubs.acs.org on June 2, 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 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
StructureꢀGuided Design of Group I Selective p21ꢀ
Activated Kinase Inhibitors
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
James J. Crawford, †* Wendy Lee, † Ignacio Aliagas, † Simon Mathieu, † Klaus P. Hoeflich, ‡ Wei
Zhou, ‡ Weiru Wang, § Lionel Rouge, § Lesley Murray, ┴ Hank La, ┴ Ning Liu, ┴ Peter W. Fan, ┴
Jonathan Cheong, ┴ Christopher E. Heise, ◊ Sreemathy Ramaswamy, ◊ Robert Mintzer, ◊ Yanzhou
Liu, † Qi Chao, ǁǁ Joachim Rudolph †
Departments of †Discovery Chemistry, ‡Translational Oncology, §Structural Biology, ┴Drug
Metabolism and Pharmacokinetics, ◊Biochemical and Cellular Pharmacology, Genentech, Inc., 1
DNA Way, South San Francisco, CA 94080, USA
ǁǁShanghai Chempartner Inc., 998 Halei Road, Zhangjiang HiꢀTech Park, Pudong New Area,
Shanghai 201203, People’s Republic of China
p21ꢀactivated kinases, ATPꢀcompetitive kinase inhibitors, kinase selectivity, ligand efficiency,
oncology, Nꢀacetyltransferase, overexpression, genomic amplification, aminopyrazole
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 45
1
2
3
4
ABSTRACT
5
6
7
8
9
The p21ꢀactivated kinases (PAKs) play important roles in cytoskeletal organization, cellular
morphogenesis and survival and have generated significant attention as potential therapeutic
targets for cancer. Following a highꢀthroughput screen, we identified an aminopyrazole scaffoldꢀ
based series that was optimized to yield group I selective PAK inhibitors. A structureꢀbased
design effort aimed at targeting the ribose pocket for both potency and selectivity led to muchꢀ
improved group I vs. II selectivity. Early lead compounds contained a basic primary amine,
which was found to be a major metabolic soft spot with in vivo clearance proceeding
predominantly via Nꢀacetylation. We succeeded in identifying replacements with improved
metabolic stability, leading to compounds with lower in vivo rodent clearance and excellent
group I PAK selectivity.
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
Introduction
The p21ꢀactivated kinases (PAKs) are a family of Rac1 and Cdc42 effectors, consisting of six
members subdivided into two groups: PAK1ꢀ3 (group I) and PAK4ꢀ6 (group II). PAK1 is the
most widely studied of the family members and plays a fundamental role in controlling cell
motility by linking a variety of extracellular signals to changes in actin cytoskeleton
organization, cell shape and adhesion dynamics.1,2
Although PAK1 is widely expressed in a variety of normal tissues, mice deficient in PAK1
were found to be viable.3 PAK1 genomic amplification and significantly increased expression
occur in ovarian, breast, bladder and other cancers.4,5 A recent study found that breast cancer
cell lines were particularly sensitive to PAK group I inhibitors.6 Not only is PAK function
increased in many human cancers, there is a further correlation with advanced tumor grade and
ACS Paragon Plus Environment
2

Page 3 of 45
Journal of Medicinal Chemistry
1
2
patient survival.7,8 As a result, PAKs, and PAK1 in particular, have been the subject of
significant drug discovery efforts.9ꢀ11
3
4
5
6
7
8
9
To date, no PAK inhibitor has advanced to proofꢀofꢀconcept studies in human. One possible
contributing factor is that kinome selectivity, including selectivity within the PAK family, has
been difficult to achieve. As a result, it has not been possible to test the efficacy and safety
associated with inhibiting specific PAK isoforms. While some highly selective allosteric
inhibitors have been disclosed, the successful optimization of these to drugꢀlike compounds has
yet to be achieved.12,13 More recently, a series of highly selective PAK4 inhibitors was reported,
which contained a propargyl substituent that protruded into a back pocket,14 in addition to a pair
of group I vs II selective scaffolds (2ꢀamino pyrido[2,3ꢀd]pyrimidineꢀ7(8H)ꢀone15 and 7ꢀ
azaindole cores16).
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 most advanced PAK inhibitor reported thus far is the Pfizer pyrrolopyrazole PAK4
clinical candidate, PFꢀ3758309 (1, Figure 1),17 which was progressed into Phase 1 clinical trials
in patients with advanced/metastatic solid tumors. Although designed to be a PAK4 inhibitor, the
compound is, in fact, a panꢀPAK inhibitor. Unfortunately, PFꢀ3758309 did not advance beyond
Phase I due to insufficient oral bioavailability in human studies.18
Based on the preclinical data supporting a role for PAK1 in tumorigenesis, our program goal
was to discover a group I selective PAK inhibitor. Group I and II PAKs share high sequence
homology in the active site, indicating a significant hurdle to selectivity.1,11 Nonetheless,
progress has been described in recent literature reports. One tactic that has emerged is to engage
backꢀpocket interactions to differentiate between PAK1 and PAK4 (as representatives of groups
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 45
1
2
I and II).11,14,15 In this report, we present a complementary approach, targeting key differences in
the sugarꢀbinding pocket.
3
4
5
6
7
8
9
Our efforts began with a high throughput screening (HTS) campaign, which yielded few hits of
interest.19 We attributed the scarcity of HTS hits to the open and conspicuously flexible ATPꢀ
binding pocket of PAK1 noted in the literature.20 One hit that piqued our interest was the
pyrazolyl quinazoline amine 2, derived from an internal Aurora A inhibitor program (Figure 2).
With an encouraging PAK1 Ki of 100 nM and bearing a structural resemblance to known PAK
inhibitors, such as 3,21,22 we decided to develop this hit further and rapidly identified the
monocyclic 4ꢀaminopiperidine analog, 4, a compound with 4ꢀfold improved potency over 2
(PAK1 Ki = 22 nM). This compound showed a lower preference for PAK4 over PAK1, with a
similar ratio to that observed for the original hit, 2. Moreover, given potential toxicity concerns
with the azabicyclo[3.1.0]hexanꢀ6ꢀamine motif, and cyclopropyl amines in general,23 we
proceeded with this monocyclic group for our initial investigations.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
A coꢀcrystal structure of 4 with PAK1 showed that its hingeꢀbinding interactions were, as
expected, akin to those of PFꢀ3758309, as shown in Figure 3. In addition to the classic 3ꢀpoint
donorꢀacceptorꢀdonor hydrogen bond interactions with the kinase hinge residues Glu345 and
Leu347, another such interaction is observed from the piperidine NH2 group to a backbone
carbonyl in the sugarꢀbinding pocket, Asp393. This differs from the protonated dimethylamine
pendant motif in 1, which makes a crucial polar interaction with Asp458 in PAK4 (Asp407 in
PAK1) of the DFG loop (Figure 3A).
With this encouraging hit in hand, we first set out to investigate variation at the pyrazole 3ꢀ
position (Table 1). While a methyl group is present in both the Aurora inhibitor, MKꢀ0457 and
ACS Paragon Plus Environment
4

Page 5 of 45
Journal of Medicinal Chemistry
1
2
the JAK2 inhibitor, AZD1480,24,25 this change resulted in a significant loss in PAK1 potency.
PAK1 activity appeared to return when the size of the substituent was gradually increased.
Small carbocycles, such as cyclopropyl or cyclobutyl, seem to fit best in the binding pocket,
efficiently filling the space alongside the (Met) gatekeeper side chain. Overall, cyclopropyl was
regarded as the best compromise between potency and human microsomal stability, and we
proceeded with this group in further SAR exploration.
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
Next we turned our attention to investigating replacements for the thiophene moiety, which is
known to harbor the potential to form a reactive metabolite.26,27 Thiophene rings are susceptible
to activation via a number of mechanisms, such as hepatic oxidation (often epoxidation) by
cytochrome P450 enzymes, oxidation of the thiophene sulfur (activating the ring towards
nucleophilic addition of biomolecules), or peroxidase addition of chlorine to the thiophene sulfur
(also activating the ring towards nucleophilic addition), and any of the resultant species would
pose a potential reactive metabolite risk. Substituting the thiophene ring for a furan (11, Table 2)
resulted in a 6ꢀfold drop in potency, perhaps due to the loss of a favorable 3.3Ǻ sulfurꢀoxygen
interaction with the mainꢀchain carbonyl of Leu347.28 In addition to thiophene replacements, we
sought to improve physicochemical properties with a deletion strategy to reduce molecular
weight and logP. Accordingly, we next targeted removal of the second ring, leaving a simple diꢀ
substituted pyrimidine, 12. While a loss in PAK1 potency was observed, we were encouraged by
the unchanged ligand lipophilicity efficiency29 and improved metabolic stability (PAK1 Ki 454
nM, LLE 3.8, HLM CLhep 1.4 mL/min/kg), where CLhep corresponding to <30% of human liver
blood flow, i.e. less than 6.2 mL/min/kg, was desired. Subsequent stepwise removal of the
pyrimidine nitrogen atoms resulted in further loss of PAK1 potency particularly when the
nitrogen proximal to the aminopyrazole was removed. A nonꢀclassical hydrogen bond between
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 45
1
2
3
4
5
6
7
8
9
the pyrimidine central ring and the CꢀH at the pyrazole 4ꢀpostion imparts near coꢀplanarity of the
two ring systems. This allows an effective fit into the hinge binding area and we attribute this
loss in potency to the disruption of planarity. In addition, the altered central ring orientation
might compromise the interaction between amino piperidine tail and ribose pocket (Figure 3C).
Given the good overall profile that the monocyclic pyrimidine represented, we continued our
investigations with this system as our primary scaffold as it (a) did not represent an obvious
metabolic liability and (b) offered the greatest flexibility to make changes at this vector at a later
stage.
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
At the outset, our primary goal was to improve activity against PAK1. We postulated that it
might be possible to do this by optimizing hydrogenꢀbonding interactions in the sugarꢀbinding
pocket. Additionally, given our project goal of PAK 1 vs. 4 selectivity, in reꢀexamining the
crystal structures of 1 and 4 we became interested in the differing orientations of the DFGꢀ
aspartate (Asp407 in PAK1) between PAK1 and PAK4. While in the PAK1 structure of 4
residue Asn394 seems to lock Asp407 in an orientation pointing away from the ligand, the
corresponding residue in PAK4, Ser445, cannot perform the same function. As a result, the
DFGꢀaspartate side chain is pointing towards the ligand in 80% of the PAK4 crystal structures
known to us.30 Seeking to take advantage of the different spatial constraints in the ribose binding
pocket resulting from the Asn394/Ser445 residue difference, we investigated elongated tail
groups, capable of making hydrogen bond or ionic interactions within the ribose pocket (Table
3), with the joint goals of improved PAK1 enzyme affinity and PAK1/4 selectivity. In targeting
this residue difference we hoped to gain selectivity over group II PAKs, but not the broader
kinome since 98.6% of kinases have an Asn at this position.31
ACS Paragon Plus Environment
6

Page 7 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Transposing and elongating the tail group as in piperidine 15 did not yield a significant potency
improvement. We had more success when a primary amine was appended at the ligand
terminus. Transꢀcyclohexyl diamineꢀcontaining 16 was more potent against PAK1 than 15, and,
perhaps most interestingly, did not have higher affinity for PAK4. A significant potency
improvement came when this group was elongated with an additional methylene, adding some
flexibility to the ligand. The resultant transꢀ4ꢀ(aminomethyl)cyclohexanamine compound 17
displayed a 6ꢀfold improvement over the more rigid 16 (PAK1 Ki = 26 nM for 17), while also
displaying the first hint of selectivity vs. PAK4 (1.5ꢀfold). The cisꢀisomer 18 had a similarly
good profile, but lost any semblance of PAK1/4 selectivity.
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
Given that 17 represented one of our most potent compounds, we obtained another coꢀcrystal
structure in order to confirm the interactions (Figure 4). As anticipated, the primary amine tail
motif makes two polar contacts in the ribose pocket (plus one to a sulfate anion as a result of
crystallization conditions), but does not engage Asp407.
Compound 17 displayed activity in a phosphoꢀS298ꢀMEK (pMEK) EBCꢀ1 cellular assay, albeit
with a significant cellꢀshift (105ꢀfold), possibly caused by its relatively low permeability
(MDCK Papp AB/BA 0.5/2.0 10ꢀ6 cm/s). We next investigated its pharmacokinetic profile in
vivo. When dosed i.v. in rats (2.5 mg/kg bolus; CLp 158 mL/mg/kg) and mice (2 mg/kg bolus;
CLp 83 mL/mg/kg) much higher clearance was observed than with earlier, nonꢀbasic analogues
(see supporting information). In particular, the clearance in rats was underꢀpredicted by in vitro
measurements in liver microsomes or hepatocytes. Similarly, compound 4 (also featuring an
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 45
1
2
3
4
NH2ꢀcontaining tailꢀgroup), while displaying moderate clearance in vitro (RLM 29, RH 34
5
6
mL/min/kg), had unexpectedly high rat i.v. clearance (CLp 1034 mL/min/kg).
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
Further investigations were aimed at rationalizing the unexpectedly high clearance of 17. A
number of possible nonꢀCYP mediated mechanisms of metabolism were investigated and among
these was Nꢀacetyl transferase (NAT).32 The contribution of NAT was conceivable given the
presence of exposed primary amine moieties in both compounds. The ubiquitous expression of
NAT1 in rats could explain underprediction of in vivo total clearance by our hepatocyte stability
assay, which would not take into account any contribution of Nꢀacetyltransferase from other
tissues such as lung and kidney.33 To test for this, rat liver S9 incubation studies were conducted
for 4 and 17. Compounds were exposed to rat liver S9 both with acetylꢀCoA, NADPH, and also
without cofactors.34 For both compounds, the highest turnover of parent compound was
observed for incubations with acetylꢀCoA, with acetylated metabolite detected in the +acetylꢀ
CoA incubation samples (for more information, see supporting information). In contrast, parent
compound was stable over 60 minutes in the absence of NADPH.35 These data indicate that both
17 and 4 are NAT substrates, potentially contributing to very high (greater than hepatic blood
flow) in vivo clearance in rat.36 Given the concerns about the inclusion of a basic pendant amine
in our ligand,29 but cognizant of the advantages that it seemed to offer, we set out to explore
bioisosteres, seeking similar interactions in the ribose pocket and improved selectivity (Table 4).
Replacement of the cyclohexyl amine with a 2ꢀpyridyl amine, 19, yielded an inhibitor that
maintained some PAK1 potency, as well as improved PAK1/4 selectivity (PAK1 Ki 138 nM; 4ꢀ
fold selectivity). Unfortunately, when dosed i.v. to rats, this compound still displayed high
ACS Paragon Plus Environment
8

Page 9 of 45
Journal of Medicinal Chemistry
1
2
clearance well above hepatic blood flow (Rat i.v. CLp 122 mL/min/kg).37 Indeed, NAT and
NADPHꢀdependent enzymes are also contributing to the metabolism of 19, 50% of parent
remained after 80 minutes of incubation with AcetylꢀCoA, and the Nꢀacetylated metabolite
detected. Following incubation with NADPH, 60% of parent remained with 3 oxidative
metabolites observed. In contrast, very little turnover occurred following incubation with
UDPGA, with no glucuronide metabolites detected in this study.
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
Subsequent addition of methyl groups to the exocyclic amine caused a drop in activity,
demonstrating the importance of the riboseꢀpocket interactions. The breakthrough came when
we cyclized to a bicyclic indole tail, 22. This nonꢀbasic tail had PAK1 affinity of the same
magnitude as the highly basic 17, with improved PAK1/4 selectivity of 5ꢀfold. Computational
modelling suggested that an (S)ꢀMe might be tolerated at the benzylic position, with a potentially
beneficial conformational orientation. While this was, indeed, tolerated, it did not offer any clear
advantage over the unsubstituted analogue. In contrast, the (R)ꢀenantiomer showed little PAK1
activity (1.9 µM, compound not shown).
We further postulated that modulating the pKa of the donor NH might allow for a better
interaction with the protein carbonyl groups of Asp393 and Asn394, thus benzimidazole 24 was
prepared. While this imparted a loss in potency relative to the indole analogs, possibly due to the
desolvation costs associated with the additional imidazolyl N, we were encouraged by the
PAK1/4 selectivity of this analogue of almost 8ꢀfold. When the benzimidazole was Nꢀ
methylated this resulted in ~5ꢀfold loss of PAK1 potency, regardless of which nitrogen was
alkylated (examples 25 and 26, Table 4).
We were further heartened when we dosed analogues 22ꢀ24 in rats. Each of these compounds
showed moderate i.v. clearance, broadly in line with predictions from rat liver microsomes (in
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 45
1
2
3
4
5
6
7
8
9
vivo CL of 30.1, 30, and 24.1 mL/min/kg, respectively). To our delight, and in line with the in
vivo data, 24 was not found to be a substrate of NAT when tested in the rat S9 fraction in the
presence of AcetylꢀCoA.
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
Of this group, compound 23 demonstrated the strongest inhibition of downstream
phosphorylation of MEK (S298) in PAK1ꢀamplified EBCꢀ1 cells, with an IC50 of 120 nM.
Encouraged by this, and curious to see if more cellular mechanistic information could be gleaned
using 23, the compound was further tested in C6 glioma cells, which contain high endogenous
levels of activated PAK1, and several downstream markers were measured (Figure 6).
Encouragingly, a good correlation between inhibition of pꢀPAK1ꢀS144 (autophosphorylation)
and inhibition of pꢀMEK1ꢀS298 and pꢀstathminꢀS16 (substrates) were observed in the C6 cells,
and this inhibition was closely correlative with the inhibition of MEK (S298) phosphorylation
seen in the EBCꢀ1 cells. Consistent with the role of PAK family members in cytoskeletal
regulation, stathmin is a microtubule destabilizing protein, and phosphorylation at serine 16 by
PAK and other kinases alters stathminꢀtubulin binding.38 PAK4ꢀS474 phosphorylation was
unaffected by compound 23 treatment.
Between 22 and 24 we observed a 0.5 unit difference in measured logD7.4 (3.0 vs. 2.5) in
addition to greater stability in HLM (15.5 vs. 8.9 mL/min/kg). We hoped to exploit this lower
lipophilicity through modifications of the benzimidazole tail. To this end, we first made the
analogue incorporating the benzylic (S)ꢀMe, 27 (Table 5). This substituent synergized better
with the benzimidazole than with the indole group, yielding our most potent inhibitor at the time,
with a PAK1 Ki of 12 nM, and with improved selectivity at 8ꢀfold over PAK4, but with poor
permeability and signs of efflux (MDCK Papp AB:0.9, BA:6.8). We sought to mimic the possible
ACS Paragon Plus Environment
10

Page 11 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
conformational effect of the methyl group, likely the preferential orientation of the tail motif, by
substituting the phenyl ring of the bicyclic system, which would have the beneficial effect of
eliminating a stereogenic center. While a 4ꢀF (28) offered no advantage over the unsubstituted
24, the 4ꢀCl, 29, set a new standard with 91ꢀfold PAK1/4 selectivity. Permeability also
improved (MDCK Papp A to B, 2.8 vs. 0.9 for 27), albeit still at the lowꢀmoderate level. We next
sought to mask the seemingly nonꢀessential polar hydrogen at the secondary amino linker to the
pyrimidine ring. While no permeability improvement was seen with the NꢀMe analog 32, the Nꢀ
Et counterpart 33 offered a measurable improvement (MDCK Papp A to B, 7.4). This came with
a concomitant improvement in cellular potency (pꢀMEK IC50 85 nM) and PAK1/4 selectivity
(91ꢀfold). Four analogs were dosed in vivo in rat (27, 29, 30 and 31). Of these, compound 29
showed the lowest i.v. clearance of any compound in this series. NMe analog 30 had
significantly higher clearance and relatively low bioavailability in rat (F% 12, following a 5
mg/kg p.o. dose), as did the NEt example, 31. This higher in vivo CL suggests that Nꢀ
dealkylation may be playing a role for the Nꢀalkyl analogs, despite relatively similar in vitro
predictions (rat hepatocyte CLhep 21, 23, 41, 44 mL/min/kg for 27, 29, 30, 31).
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
In order to better understand our excellent intraꢀPAK selectivity, we again sought to obtain
crystal structures, this time of 29 with both PAK1 and PAK4 (Figure 7). The reason for the
observed selectivity was immediately evident, and adhered to our initial hypothesis.
In PAK1, Asp407 (from the conserved AspꢀPheꢀGly [DFG] motif)39 is oriented away from the
benzimidazole motif, allowing the CH to have a nonꢀcanonical hydrogenꢀbond to Asn394, as
well as another hydrogenꢀbond from the NH to the carbonyl of Asp393. In contrast, in PAK4,
ACS Paragon Plus Environment
11

Journal of Medicinal Chemistry
Page 12 of 45
1
2
3
4
5
6
7
8
9
Asp458 (the corresponding Asp from the DFG motif) occupies a position that occludes the
ligand, preventing it from occupying the same location and causes it to adopt a more strained
conformation. To assess relative conformational strain and energies of the methylꢀamino linker
to the benzimidazole of 29 in the two bound conformations found in PAK1 and PAK4, a
simplified system was studied consisting of a 2ꢀaminoꢀpyrimidine with the amino group linked
to a Clꢀphenyl. The conformation observed in PAK4 was found to be higher in energy than that
found in PAK1.40 In addition, the Hꢀbond to the Asp backbone carbonyl is also expected to be
weaker in PAK4 since the angle is less optimal than in PAK1. The benzimidazole moiety in
both PAK1 and PAK4 shows close contacts between one nitrogen and the backbone carbonyl of
Asp393 (PAK1) and Asp444 (PAK4). The distance between the imidazole nitrogen and the
carbonyl oxygen, 2.8Å in PAK1, is more indicative of a favorable hydrogen bond interaction
than that found in PAK4, 3.1Å (for more details, see supplemental material). Collectively, as
observed biochemically, these conformational factors relating to the orientation of the
benzimidazole group contributed to a weaker interaction of the ligand with PAK4.
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
Given the much improved PAK1 vs. PAK4 selectivity, broader kinome selectivity of 31 was
assessed (Figure 8). A panel of 74 kinases, broadly representative of the kinome, was selected
for screening at 100 nM, a concentration 20ꢀfold over the PAK1 Ki. As expected, excellent
group I PAK selectivity was observed, with very little inhibition of group II PAKs (9, 6, 11%
inhibition for PAKs 4, 5, 6 respectively). In total, 22 offꢀtarget kinases showed >50% inhibition,
including Aurora A. (Figure 8). When tested in a doseꢀresponse assay, 31 had a Ki of 5.3 nM
against Aurora A. Further improvement of kinome selectivity was necessary, and our efforts to
this end are the subject of a following communication.41
ACS Paragon Plus Environment
12

Page 13 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
CHEMISTRY
7
8
9
The preparation of compounds 4 – 10 was performed as shown in Scheme 1. Commercially
available aminopyrazoles, such as 3ꢀcyclopropylꢀ1Hꢀpyrazolꢀ5ꢀamine (33a), were reacted with 2,
4ꢀdichlorothieno[3,2ꢀd]pyrimidine (32) via an SNAr reaction to yield the corresponding monoꢀ
functionalized pyrimidine, 34a. A second SNAr, this time with a piperidinyl amine, and
subsequent acidꢀmediated NꢀBoc deprotection gave the desired products.
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 central ring analogs 11-14 were prepared by similar means (Scheme 2). Employing
either an SNAr on the appropriate pyrimidine, or a BuchwaldꢀHartwig reaction on a halopyridine,
42ꢀ44
the desired monohalo intermediate was obtained. In turn, a second SNAr or CꢀN coupling
reaction and NꢀBoc deprotection afforded the products in short order. For 13, no deprotection
was needed.
The primary source of structural diversity and complexity came from the ‘tail’ fragments. A
number of amine tails, such as those to prepare compounds 15ꢀ26, were commercially available.
These amines were reacted with the key common intermediate 37b as shown in Scheme 2. For
the preparation of 27, a stereospecific approach to install the (S)ꢀalphaꢀmethyl group using
Ellman’s chiral auxiliary is shown in Scheme 3.45 Palladiumꢀmediated coupling of butyl vinyl
ether with THPꢀprotected 5ꢀbromoꢀ1Hꢀbenzo[d]imidazole 40 gave the corresponding ketone 41,
which was then condensed with (R)ꢀ2ꢀmethylpropaneꢀ2ꢀsulfinamide. Reduction of the
sulfinamide and acidꢀmediated removal of the chiral functionality afforded the desired (S)ꢀ2ꢀ
methylbenzimidazole, 43, as the HCl salt.
ACS Paragon Plus Environment
13

Journal of Medicinal Chemistry
Page 14 of 45
1
2
3
4
5
6
7
8
9
The approach for the fluoroꢀ and chloroꢀbenzimidazole compounds 28ꢀ31 is outlined in Scheme
4. Phillips cyclization, with formic acid as the methylene source, on the commercially available
tetraꢀfunctionalized diaminobenzene 44a and 44b gave the functionalized benzimidazoles, 45a
and 45b,46 which were then protected as the 2ꢀtetrahydropyran adducts 46a and 46b,
respectively. Two synthetic pathways to the amino methyl intermediates 49a and 49b were
accomplished by either condensation with hydroxylamine via the aldehyde to form oxime 47, or
by direct palladiumꢀcatalyzed cyanation to give 48. Subsequent reduction of 47 and 48 to the
primary amines gave the requisite intermediates for the preparation of 28 and 29. For Nꢀ
alkylated analogs 30 and 31, monoꢀNꢀBoc protection, followed by alkylation and Bocꢀremoval,
gave the corresponding intermediates. Completion of the syntheses following the established
route of baseꢀmediated SNAr reaction and protecting group removal gave the final products.
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
CONCLUSIONS
The p21ꢀactivated kinases play important roles in cytoskeletal organization, cellular
morphogenesis and survival, and members of this family have been implicated in a wide range of
diseases including cancer, infectious diseases, and neurological disorders. From a highꢀ
throughput screen we identified aminopyrazole 2, which was optimized to a group I selective
PAK inhibitor.47 Treatment of cancer cells with 23 resulted in signaling pathway responses that
were consistent with previous reports using group I PAK inhibition via genetic or dominantꢀ
negative approaches. Through the application of structureꢀbased design, we were able to
successfully execute on selectivity hypotheses around the sugarꢀbinding pocket to impart group I
PAK selectivity, however this was not replicated across the kinome, and further optimization
will be required. A significant potency improvement was observed following addition of an NH2
ACS Paragon Plus Environment
14

Page 15 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
group, and recognizing the importance of a hydrogen bond, rather than ionic, interaction from
this, a number of bioisosteres were designed and tested. Ultimately, the incorporation of a
substituted benzimidazole allowed the identification of analogs with excellent PAKꢀgroup
selectivity, which also circumvented observed in vivo acetylation in rat by Nꢀacetyltransferase.
In this series, the observation of differing conformations of a conserved aspartate residue
between PAKs 1 and 4, and understanding the possible reasons for this through the examination
of coꢀcrystal structures, was crucial to obtaining PAK group selectivity.
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
EXPERIMENTAL SECTION
All chemicals were purchased from commercial suppliers and used as received. Flash
chromatography was carried out with prepacked silica cartridges from either ISCO or SiliCycle
on an ISCO Companion chromatography system using gradient elution. NMR spectra were
recorded on a Bruker AV III 400 NMR spectrometer at 300K or 350K and referenced to
tetramethylsilane. Preparative HPLC was performed on a Polaris C18 5 ꢁm column (50 mm × 21
mm), eluting with mixtures of water/acetonitrile. All final compounds were purified to >95%
chemical and optical purity, as assayed by LC/MS. This analysis was performed on an Agilent
1100 HPLC coupled with Agilent MSD mass spectrometer using ESI as the ionization source.
The LC separation was performed using an Agilent ZORBAX SBꢀC18, 1.8 mm, 30 × 2.1 mm
column with a flow rate of 0.4 mL/min. The gradient consisted of 3−97% acetonitrile in water,
each containing 0.05% TFA, over 7 min; 97% acetonitrile was then maintained for 1.5 min,
followed by reꢀequilibration at 3% acetonitrile for 1.5 min. The LC column temperature was
maintained at 40 °C. UV absorbance was measured at 220 and 254 nm along with a full scan
mass spectrum. For select compounds, a longer gradient LC was used (method B). In these cases,
ACS Paragon Plus Environment
15

Journal of Medicinal Chemistry
Page 16 of 45
1
2
3
4
5
6
7
8
9
the gradient consisted of 2−98% acetonitrile in water, each containing 0.05% TFA, over 25.5
min; 98% acetonitrile was then maintained for 2.5 min, followed by reꢀequilibration at 2%
acetonitrile for 1.5 min. UV and MS detection was performed as before.
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
General Procedure 1. Compounds 34a–34g were synthesized by treating 2,4ꢀ
dichloropyrimidine 1 with 5ꢀ(alkyl or aryl)ꢀ1Hꢀpyrazoles 33a–33g (1 equiv) and triethylamine
(1.5 equiv) in DMA at 120 °C for 16h. Solvent was evaporated followed by addition of water
(40 mL). The resultant solid was filtered, and the solid was then suspended in diethyl ether (30
mL) and stirred at room temperature for 16h. The suspended solid was collected, vacuum dried
and used in the next step without further purification.
Analysis of Representative Compound. 2-chloro-N-(5-cyclopropyl-1H-pyrazol-3-
yl)thieno[3,2-d]pyrimidin-4-amine (34a). Yield: 75.6%. [ESꢀMS] (ESI+): m/z calcd for
C12H10ClN5S [M + H]+, 292.0; found, 292.1.
General Procedure 2. The final compounds 4–10 were prepared by treating intermediates 34a–
34g with 4ꢀ(NꢀBocꢀamino)ꢀpiperidine (2.0 equiv) and triethylamine (2.6 equiv) in EtOH
microwaved at 170 °C for 40 minutes. The reaction mixture was cooled, and 4.0 M HCl in 1,4ꢀ
dioxane (10 equiv) was added. After stirring at RT for 24, the reaction mixture was concentrated
in vacuo. Final compounds 4–10 were all purified by preparative HPLC.
Compounds 4–10. Compounds 4–10 were synthesized via general procedure 1 followed by
general procedure 2.
2-(4-Aminopiperidin-1-yl)-N-(5-cyclopropyl-1H-pyrazol-3-yl)thieno[3,2-d]pyrimidin-4-
1
amine (4). Yield: 51%. H NMR (400 MHz, DMSOꢀd6, 350 K) δ 11.97 (s, 1H), 9.25 (s, 1H),
7.86 (d, J = 5.4 Hz, 1H), 7.04 (d, J = 5.4 Hz, 1H), 6.16 (s, 1H), 4.57 – 4.41 (m, 2H), 3.09 – 2.96
ACS Paragon Plus Environment
16

Page 17 of 45
Journal of Medicinal Chemistry
1
2
3
4
(m, 4H), 2.88 – 2.79 (m, 1H), 1.95 – 1.86 (m, 1H), 1.82 – 1.71 (m, 2H), 1.27 – 1.14 (m, 2H),
5
6
13
1.02 – 0.90 (m, 2H), 0.74 – 0.62 (m, 2H). C NMR (101 MHz, DMSOꢀd6, 350 K) δ 163.28,
7
8
9
160.69, 155.64, 147.05, 133.69, 123.59, 105.66, 95.29, 49.04, 43.41, 35.67, 8.04, 7.54. HRMS
(ESI+): m/z calcd for C17H22N7S [M + H]+, 356.1652; found, 356.1654.
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-(4-Aminopiperidin-1-yl)-N-(5-methyl-1H-pyrazol-3-yl)thieno[3,2-d]pyrimidin-4-amine
(5). Yield: 10%. 1H NMR (400 MHz, DMSOꢀd6, 350 K) δ 12.00 (s, 1H), 9.25 (s, 1H), 7.86 (d, J
= 5.3 Hz, 1H), 7.04 (d, J = 5.3 Hz, 1H), 6.29 (s, 1H), 4.57 – 4.46 (m, 2H), 3.06 – 2.96 (m, 4H),
13
2.91 – 2.80 (m, 1H), 2.25 (s, 3H), 1.82 – 1.72 (m, 2H), 1.30 – 1.15 (m, 2H). C NMR (101
MHz, DMSOꢀd6, 350 K) δ 163.25, 160.73, 155.72, 147.10, 139.64, 133.70, 123.56, 105.68,
101.5, 98.34, 49.02, 43.40, 35.36, 15.99, 11.53. HRMS (ESI+): m/z calcd for C15H20N7S [M +
H]+, 330.1495; found, 330.1496.
2-(4-Aminopiperidin-1-yl)-N-(5-ethyl-1H-pyrazol-3-yl)thieno[3,2-d]pyrimidin-4-amine (6).
Yield: 29%. 1H NMR (400 MHz, DMSOꢀd6, 350 K) δ 12.00 (s, 1H), 9.34 (s, 1H), 7.85 (d, J = 5.4
Hz, 1H), 7.04 (d, J = 5.3 Hz, 1H), 6.30 (s, 1H), 4.58 – 4.45 (m, 2H), 3.07 – 2.95 (m, 4H), 2.88 –
13
2.78 (m, 1H), 2.67 – 2.59 (m, 2H), 1.80 – 1.71 (m, 2H), 1.29 – 1.14 (m, 5H). C NMR (101
MHz, DMSOꢀd6, 350 K) δ 162.43, 159.89, 154.87, 146.10, 145.08, 134.09, 132.82, 122.77,
104.84, 96.16, 53.49, 48.21, 42.60, 34.82, 18.54, 12.90. HRMS (ESI+): m/z calcd for C16H22N7S
[M + H]+, 344.1652; found, 344.1654.
2-(4-Aminopiperidin-1-yl)-N-(5-propyl-1H-pyrazol-3-yl)thieno[3,2-d]pyrimidin-4-amine
(7). Yield: 67%. 1H NMR (400 MHz, DMSOꢀd6, 350 K) δ 12.0 (s, 1H), 9.34 (s, 1H), 7.86 (d, J =
3.6 Hz, 1H), 7.04 (d, J = 5.3 Hz, 1H), 6.29 (s, 1H), 4.62 – 4.47 (m, 2H), 3.06 – 2.86 (m, 5H),
2.59 (t, J = 7.5 Hz, 2H), 1.86 – 1.74 (m, 2H), 1.71 – 1.58 (m, 2H), 1.36 – 1.29 (m 2H), 1.03 –
ACS Paragon Plus Environment
17

Journal of Medicinal Chemistry
Page 18 of 45
1
2
0.90 (m, 3H). 13C NMR (101 MHz, DMSOꢀd6, 350 K) δ 163.23, 160.66, 155.75, 146.93, 144.42,
133.71, 123.58, 105.78, 97.66, 48.94, 43.30, 34.41, 28.04, 22.34, 13.85. HRMS (ESI+): m/z
calcd for C17H24N7S [M + H]+, 358.1808; found, 358.1812.
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
2-(4-Aminopiperidin-1-yl)-N-(5-isopropyl-1H-pyrazol-3-yl)thieno[3,2-d]pyrimidin-4-amine
(8). Yield: 47%. 1H NMR (400 MHz, DMSOꢀd6, 350 K) δ 12.03 (s, 1H), 9.35 (s, 1H), 7.86 (d, J
= 5.4 Hz, 1H), 7.08 – 7.03 (m, 1H), 6.33 (s, 1H), 4.61 – 4.42 (m, 2H), 3.06 – 2.93 (m, 6H), 2.93
13
– 2.81 (m, 1H), 1.82 – 1.72 (m, 2H), 1.32 – 1.16 (m, 7H). C NMR (101 MHz, DMSOꢀd6, 350
K) δ 163.25, 160.68, 155.65, 150.73, 146.80, 133.65, 123.61, 105.72, 95.76, 49.01, 43.38, 35.24,
26.10, 22.72. HRMS (ESI+): m/z calcd for C17H24N7S [M + H]+, 358.1818; found, 358.1810.
2-(4-Aminopiperidin-1-yl)-N-(5-cyclobutyl-1H-pyrazol-3-yl)thieno[3,2-d]pyrimidin-4-amine
(9). Yield: 55%. 1H NMR (400 MHz, DMSOꢀd6, 350 K) δ 11.94 (s, 1H), 9.36 (s, 1H), 7.87 (d, J
= 5.4 Hz, 1H), 7.05 (d, J = 5.4 Hz, 1H), 6.39 (s, 1H), 4.59 – 4.47 (m, 2H), 3.58 – 3.46 (m, 1H),
3.08 – 2.97 (m, 4H), 2.92 – 2.81 (m, 1H), 2.41 – 2.30 (m, 2H), 2.21 – 2.07 (m, 2H), 2.06 – 1.86
13
(m, 2H), 1.82 – 1.73 (m, 2H), 1.32 – 1.16 (m, 2H). C NMR (101 MHz, DMSOꢀd6, 350 K) δ
163.27, 160.69, 155.66, 148.87, 147.03, 133.69, 123.61, 105.72, 96.31, 49.02, 43.40, 35.33,
32.04, 29.39, 18.68. HRMS (ESI+): m/z calcd for C18H24N7S [M + H]+, 370.1808; found,
370.1811.
2-(4-Aminopiperidin-1-yl)-N-(5-phenyl-1H-pyrazol-3-yl)thieno[3,2-d]pyrimidin-4-amine
(10). Yield: 72%. 1H NMR (400 MHz, DMSOꢀd6, 350 K) δ 12.44 (s, 1H), 9,43 (s, 1H), 7.89 (d, J
= 5.4 Hz, 1H), 7.77 – 7.71 (m, 2H), 7.49 – 7.44 (m, 2H), 7.38 – 7.32 (m, 1H), 7.08 (d, J = 5.4 Hz,
1H), 6.88 (s, 1H), 4.61 – 4.49 (m, 2H), 3.20 – 2.96 (m, 4H), 2.94 – 2.86 (m, 1H), 1.87 – 1.74 (m,
13
2H), 1.34 – 1.19 (m, 2H). C NMR (101 MHz, DMSOꢀd6, 350 K) δ 163.42, 160.69, 155.55,
ACS Paragon Plus Environment
18

Page 19 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
146.80, 144.37, 133.94, 131.25, 129.34, 128.31, 125.51, 125.39, 123.67, 105.78, 96.23, 48.99,
43.45, 35.21. HRMS (ESI+): m/z calcd for C20H22N7S [M + H]+, 392.1652; found, 392.1654.
7
8
9
2-chloro-N-(5-cyclopropyl-1H-pyrazol-3-yl)furo[3,2-d]pyrimidin-4-amine (37a). The title
compound was prepared via general procedure 1, using 3ꢀcyclopropylꢀ1Hꢀpyrazolꢀ5ꢀamine
(0.30 g, 2.4 mmol) and 2,4ꢀdichlorofuro[3,2ꢀd]pyrimidine (0.45 g, 2.4 mmol) as starting
materials to give 304 mg of 37a as yellow solid. Yield: 46%. [ESꢀMS] (ESI+): m/z calcd for
C12H10ClN5O [M + H]+, 275.1; found, 276.1.
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-(4-Aminopiperidin-1-yl)-N-(5-cyclopropyl-1H-pyrazol-3-yl)furo[3,2-d]pyrimidin-4-amine
(11). The title compound was synthesized via general procedure 2, replacing 34a with 2ꢀchloroꢀ
Nꢀ(5ꢀcyclopropylꢀ1Hꢀpyrazolꢀ3ꢀyl)furo[3,2ꢀd]pyrimidinꢀ4ꢀamine 37a. Yield: 41%. 1H NMR
(400 MHz, DMSOꢀd6, 350 K) δ 9.33 (s, 1H), 7.95 (d, J= 5.4 Hz, 1H), 6.67 (d, J = 5.4 Hz, 1H),
6.20 (s, 1H), 4.54 – 4.36 (m, 2H), 3.06 – 2.93 (m, 4H), 2.87 – 2.77 (m, 1H), 1.94 – 1.85 (m, 1H),
1.81 – 1.69 (m, 2H), 1.30 – 1.13 (m, 2H), 0.99 – 0.88 (m, 2H), 0.71 – 0.58 (m, 2H). 13C NMR
(101 MHz, DMSOꢀd6, 350 K) δ 159.48, 152.63, 149.82, 147.09, 145.55, 129.17, 107.30, 93.77,
49.00, 43.81, 35.56, 8.00, 7.58. HRMS (ESI+): m/z calcd for C17H22N7O [M + H]+, 340.1880;
found, 340.1884.
2-Chloro-N-(5-cyclopropyl-1H-pyrazol-3-yl)pyrimidin-4-amine (37b). A mixture of 2,4ꢀ
dichloropyrimidine (2.70 g, 18.1 mmol), 5ꢀcyclopropylꢀ1Hꢀpyrazolꢀ3ꢀamine (2.45 g, 19.9
mmol), and DIPEA (4.70 mL, 27.2 mmol) in DMSO (25 mL) was stirred at 60 °C for 4 days.
The reaction was allowed to cool, then was poured into water (~ 200 mL) and sonicated for 15
minutes until a white solid precipitated out. The precipitate was filtered and dried under vacuum
to give 3.5 g of desired product. Yield: 82%. 1H NMR (400 MHz, DMSOꢀd6) δ 12.16 (s, 1H),
ACS Paragon Plus Environment
19

Journal of Medicinal Chemistry
Page 20 of 45
1
2
3
4
5
6
10.23 (s, 1H), 8.15 (s, 1H), 7.19 (s, 1H), 5.98 (s, 1H), 1.93 ꢀ 1.84 (m, 1H), 0.96 ꢀ 0.89 (m, 2H),
0.71 – 0.65 (m, 2H). [ESꢀMS] (ESI+): m/z calcd for C10H11ClN5 [M + H]+, 236.1; found, 237.1.
7
8
9
2-(4-Aminopiperidin-1-yl)-N-(5-cyclopropyl-1H-pyrazol-3-yl)pyrimidin-4-amine (12). In a
microwave vial was placed 37b (50 mg, 0.21 mmol), tertꢀbutyl Nꢀ(4ꢀpiperidyl)carbamate (130
mg, 0.64 mmol), nꢀbutanol (1.5 mL), and triethylamine (49 mg, 0.48 mmol). The vial was
capped, and the reaction mixture was stirred in a microwave, heating at 180 °C, for 15 min. The
reaction solvent was evaporated under reduced pressure. The crude material was then dissolved
in 4M HCl in dioxane (1 mL), and the reaction mixture was stirred at room temperature for 30
min. The solvent was again removed under reduced pressure, and the crude residue was purified
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
by preparative HPLC to give 24 mg of compound 12. Yield: 38%. H NMR (400 MHz, DMSOꢀ
d6, 350 K) δ 11.76 (s, 1H), 8.97 (s, 1H), 7.86 (d, J = 5.7 Hz, 1H), 6.24 (d, J = 5.7 Hz, 1H), 6.01
(s, 1H), 4.53 – 4.38 (m, 2H), 3.03 – 2.92 (m, 4H), 2.87 – 2.79 (m, 1H), 1.91 – 1.83 (m, 1H), 1.78
– 1.69 (m, 2H), 1.23 – 1.10 (m, 2H), 0.95 – 0.88 (m, 2H), 0.69 – 0.62 (m, 2H). 13C NMR (101
MHz, DMSOꢀd6, 350 K) δ 161.80, 160.50, 156.61, 148.57, 146.44, 95.93, 92.89, 48.95, 42.81,
35.57, 7.95, 7.35. HRMS (ESI+): m/z calcd for C15H22N7 [M + H]+, 300.1931; found, 300.1927.
6-Chloro-N-(5-cyclopropyl-1H-pyrazol-3-yl)pyridin-2-amine (37c). 2,6ꢀDichloropyridine
(0.300 g, 2.0 mmol), 5ꢀcyclopropylꢀ1Hꢀpyrazolꢀ3ꢀamine (0.26 g, 2.1 mmol),
tri(dibenzylideneacetone)dipalladium(0) (0.04 g, 0.04 mmol), DavePhos (0.03 g, 0.08 mmol),
lithium hexamethyldisilazide (0.81 g, 4.9 mmol), and anhydrous THF (11 mL) were combined in
a microwave vial. The reaction mixture was purged with N2 and then heated, with stirring, at 130
°C in the microwave for 15 min. The cooled reaction mixture was poured into ethyl acetate. The
organic layer was extracted with water and brine, dried over Na2SO4, filtered, and evaporated
under reduced pressure. The crude mixture was purified by column chromatography (SiO2:
ACS Paragon Plus Environment
20

Page 21 of 45
Journal of Medicinal Chemistry
1
2
heptane / ethyl acetate gradient) to obtain 240 mg of 37c. Yield: 50%. 1H NMR (400 MHz,
DMSOꢀd6) δ 11.87 (s, 1H), 9.41 (s, 1H), 7.55 (t, J = 7.9 Hz, 1H), 7.24 (d, J = 8.3 Hz, 1H), 6.70
(d, J = 7.5 Hz, 1H), 5.91 (d, J = 2.2 Hz, 1H), 1.91 – 1.82 (m, 1H), 0.96 – 0.87 (m, 2H), 0.70 –
0.60 (m, 2H).
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
6-(4-Aminopiperidin-1-yl)-N-(5-cyclopropyl-1H-pyrazol-3-yl)pyridin-2-amine (13).
Compound 37c (0.030 g, 0.13 mmol) and tertꢀbutyl Nꢀ(4ꢀpiperidyl)carbamate (0.21 mL) were
combined in a vial. The vial was capped, and the reaction mixture was stirred at 150 °C for 16
hours. The reaction mixture was concentrated under reduced pressure, and the crude mixture was
purified by preparative HPLC to give 13 mg of compound 13. Yield: 34%. 1H NMR (400 MHz,
DMSOꢀd6, 350 K) δ 8.33 (s, 1H), 7.26 (t, J = 8.0 Hz, 1H), 6.32 (d, J = 7.9 Hz, 1H), 6.08 (d, J =
8.2 Hz, 1H), 5.90 (s, 1H), 4.15 – 4.03 (m, 2H), 2.99 – 2.75 (m, 5H), 1.90 – 1.80 (m, 1H), 1.80 –
1.72 (m, 2H), 1.33 – 1.19 (m, 2H), 0.92 – 0.86 (m, 2H), 0.66 – 0.61 (m, 2H). 13C NMR (101
MHz, DMSOꢀd6, 350 K) δ 165.38, 158.51, 154.50, 149.12, 146.94, 139.04, 97.77, 96.72, 91.41,
48.88, 44.30, 34.99, 7.89, 7.66. HRMS (ESI+): m/z calcd for C16H23N6 [M + H]+, 299.1979;
found, 299.1982.
2-Chloro-N-(5-cyclopropyl-1H-pyrazol-3-yl)pyridin-4-amine (37d). tertꢀButyl 3ꢀaminoꢀ5ꢀ
cyclopropylꢀ1Hꢀpyrazoleꢀ1ꢀcarboxylate (0.40 g, 1.6 mmol), 2ꢀchloroꢀ4ꢀiodopyridine (0.41 g, 1.8
mmol), tri(dibenzylideneacetone)dipalladium(0) (0.06 g, 0.07 mmol), 4,5ꢀ
bis(diphenylphosphino)ꢀ9,9ꢀdimethylxanthene (0.08 g, 0.13 mmol), cesium carbonate (0.82 g,
2.5 mmol), and degassed toluene (11 mL) were combined in a microwave vial. The reaction
mixture was purged with N2. The vial was capped, and the mixture was stirred at 145 °C in the
microwave for 15 min. The cooled reaction mixture was filtered through a bed of Celite® and
rinsed with ethyl acetate. The filtrate was concentrated under reduced pressure, and the crude
ACS Paragon Plus Environment
21

Journal of Medicinal Chemistry
Page 22 of 45
1
2
3
4
5
6
7
8
9
mixture was purified by column chromatography (SiO2: heptane / ethyl acetate gradient) to
obtain 58 mg of compound 37d. Yield: 15%. [ESꢀMS] (ESI+): m/z calcd for C11H12ClN4 [M +
H]+, 236.1; found, 236.1.
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-(4-Aminopiperidin-1-yl)-N-(5-cyclopropyl-1H-pyrazol-3-yl)pyridin-4-amine (14). The
title compound was prepared via general procedure 2, using 37d in place of 34a to give 13 mg of
compound 14.Yield: 18%. 1H NMR (400 MHz, DMSOꢀd6 + 1% TFA, 350 K) δ 8.34 (s, 1H),
7.76 (d, J = 5.6 Hz, 1H), 6.74 (d, J = 5.3 Hz, 1H), 6.46 – 6.43 (m, 1H), 5.55 (s, 1H), 4.17 – 4.08
(m, 2H), 4.06 – 3.97 (m, 1H), 3.93 – 3.80 (m, 1H), 3.16 – 3.02 (m, 1H), 2.89 – 2.79 (m, 2H),
2.51.91 – 1.93 (m, 2H), 1.84 – 1.75 (m, 1H), 1.49 – 1.35 (m, 2H), 0.93 – 0.87 (m, 2H), 0.70 –
0.65 (m, 2H). HRMS (ESI+): m/z calcd for C16H23N6 [M + H]+, 299.1979; found, 299.2018.
N4-(5-Cyclopropyl-1H-pyrazol-3-yl)-N2-(piperidin-4-ylmethyl)pyrimidine-2,4-diamine (15).
Compound 37b (50 mg, 0.21 mmol), tertꢀbutyl 4ꢀ(aminomethyl)piperidineꢀ1ꢀcarboxylate (81
mg, 0.27 mmol), nꢀbutanol (2 mL), and DIPEA (0.10 mL, 0.63 mmol) were combined in a vial.
The vial was capped, and the reaction mixture was stirred at 120 °C for 18 hours. The reaction
was cooled and concentrated under reduced pressure. To the crude residue was added 4M HCl in
dioxane (2 mL), and the mixture was stirred at room temperature for 16 hours. Solvent was
removed under reduced pressure, and the crude residue was purified by preparative HPLC to
give 34 mg of compound 15. Yield: 52%. 1H NMR (400 MHz, DMSOꢀd6 + 1% TFA, 350 K) δ
11.77 (s, 1H), 9.06 (s, 1H), 7.79 (d, J = 5.7 Hz, 1H), 6.51 (s, 1H), 6.14 (s, 1H), 6.09 (s, 1H), 3.23
– 3.11 (m, 5H), 2.75 – 2.64 (m, 2H), 1.90 – 1.73 (m, 4H), 1.33 ꢀ 116 (m, 2H), 0.96 – 0.82 (m,
2H), 0.72 – 0.61 (m, 2H) 13C NMR (101 MHz, DMSOꢀd6 +1% TFA, 350 K) δ 162.80, 160.32,
158.44, 158.13, 156.57, 145.00, 96.28, 93.07, 49.06, 46.60, 44.77, 35.13, 28.80. HRMS (ESI+):
m/z calcd for C16H24N7 [M + H]+, 314.2088; found, 314.2091.
ACS Paragon Plus Environment
22

Page 23 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
N2-(Trans-4-aminocyclohexyl)-N4-(5-cyclopropyl-1H-pyrazol-3-yl)pyrimidine-2,4-diamine
(16). The title compound was prepared in the same manner as 15, using tertꢀbutyl Nꢀ[4ꢀ
aminocyclohexyl]carbamate in place of tertꢀbutyl 4ꢀ(aminomethyl)piperidineꢀ1ꢀcarboxylate to
give 42 mg of compound 16. Yield: 64%. 1H NMR (400 MHz, DMSOꢀd6 +1% TFA, 350 K) δ
11.90 (s, 1H), 9.09 (s, 1H), 7.79 (d, J = 5.7 Hz, 1H), 6.13 (s, 1H), 6.04 (s, 2H), 3.72 – 3.57 (m,
3H), 2.86 – 2.74 (m, 1H), 2.07 – 1.95 (m, 2H), 1.94 – 1.80 (m, 3H), 1.37 – 1.21 (m, 2H), 1.21 –
1.07 (m, 2H), 0.95 – 0.82 (m, 2H), 0.74 – 0.61 (m, 2H). 13C NMR (101 MHz, DMSOꢀd6 + 1%
TFA, 350 K) δ 161.99, 160.31, 158.39, 158.10, 156.57, 147.32, 96.23, 92.72, 50.35, 49.99,
49.22, 49.06, 32.82, 31.31, 7.88. HRMS (ESI+): m/z calcd for C16H24N7 [M + H]+, 314.2088;
found, 314.2091.
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
N2-((Trans-4-aminocyclohexyl)methyl)-N4-(5-cyclopropyl-1H-pyrazol-3-yl)pyrimidine-2,4-
diamine (17). The title compound was prepared in the same manner as 15, using tertꢀbutyl transꢀ
[4ꢀ(aminomethyl)cyclohexyl]carbamate in place of tertꢀbutyl 4ꢀ(aminomethyl)piperidineꢀ1ꢀ
carboxylate to give 130 mg of compound 17. Yield: 37%. 1H NMR (400 MHz, DMSOꢀd6 +1%
TFA, 350 K) δ 11.93 (s, 1H), 9.04 (s, 1H), 7.78 (d, J = 5.6 Hz, 1H), 6.70 (s, 1H), 6.42 – 5.75 (m,
2H), 3.13 (t, J = 6.4 Hz, 2H), 2.71 – 2.54 (m, 1H), 2.40 – 2.50 (m, 2H), 1.90 – 1.74 (m, 5H), 1.58
– 1.43 (m, 1H), 1.16 – 0.81 (m, 6H), 0.72 – 0.57 (m, 2H). 13C NMR (101 MHz, DMSOꢀd6 + 1%
TFA, 350 K) δ 162.63, 160.05, 158.41, 158.11, 156.44, 148.83, 145.78, 96.07, 92.71, 50.72,
47.15, 37.21, 34.39, 29.63, 8.12, 7.43. HRMS (ESI+): m/z calcd for C17H26N7 [M + H]+,
328.2244; found, 328.2249.
N2-((Cis)-4-aminocyclohexyl)methyl)-N4-(5-cyclopropyl-1H-pyrazol-3-yl)pyrimidine-2,4-
diamine (18). The title compound was prepared in the same manner as 15, using tertꢀbutyl cisꢀ
[4ꢀ(aminomethyl)cyclohexyl]carbamate in place of tertꢀbutyl 4ꢀ(aminomethyl)piperidineꢀ1ꢀ
ACS Paragon Plus Environment
23

Journal of Medicinal Chemistry
Page 24 of 45
1
2
carboxylate to give 50 mg of compound 18. Yield: 60%. 1H NMR (400 MHz, DMSOꢀd6 +1%
TFA, 350 K) δ 11.65 (s, 1H), 9.06 (s, 1H), 7.79 (d, J = 5.7 Hz, 1H), 6.47 (s, 1H), 6.13 (d, J = 5.7
Hz, 1H), 6.03 (s, 1H), 3.27 – 3.16 (m, 2H), 3.07 – 3.98 (m, 1H), 1.89 – 1.80 (m, 1H), 1.80 – 1.68
(m, 1H), 1.65 – 1.33 (m, 10H), 0.93 – 0.83 (m, 2H), 0.71 – 0.62 (m, 2H). 13C NMR (101 MHz,
DMSOꢀd6 + 1% TFA, 350 K) δ 162.81, 160.27, 156.55, 147.46, 96.11, 92.48, 48.08, 45.12,
34.93, 30.13, 29.84, 25.28, 24.76, 7.82. HRMS (ESI+): m/z calcd for C17H26N7 [M + H]+,
328.2244; found, 328.2249.
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
N2-((6-Aminopyridin-3-yl)methyl)-N4-(5-cyclopropyl-1H-pyrazol-3-yl)pyrimidine-2,4-
diamine (19). Compound 37b (0.97 g, 4.1 mmol), 5ꢀ(aminomethyl)pyridinꢀ2ꢀamine (0.66 g, 5.3
mmol), DIPEA (2.1 mL, 12 mmol), and ethanol (10 mL) were combined in a vial. The vial was
capped, and the reaction mixture was stirred at 120 °C for 18 hours. The solvent was removed
under reduced pressure, and the crude residue was purified by preparative HPLC to give 0.12 g
of compound 19 as a white solid. Yield: 30%. 1H NMR (400 MHz, DMSOꢀd6, 350 K) δ 11.74 (s,
1H), 8.96 (s, 1H), 7.89 (d, J = 2.4 Hz, 1H), 7.82 (d, J = 5.9 Hz, 1H), 7.36 (dd, J = 8.4, 2.4 Hz,
1H), 6.68 (s, 1H), 6.41 (d, J = 8.6 Hz, 1H), 6.17 (s, 2H), 5.50 (s, 2H), 4.30 (d, J = 6.1 Hz, 2H),
1.88 – 1.79 (m, 1H), 0.93 – 0.81 (m, 2H), 0.69 – 0.59 (m, 2H). 13C NMR (101 MHz, DMSOꢀd6,
350 K) δ 162.50, 160.41, 159.12, 156.61, 148.90, 147.24, 145.75, 137.22, 124.39, 108.05, 96.51,
93.47, 42.08, 7.85, 7.29. HRMS (ESI+): m/z calcd for C16H19N8 [M + H]+, 323.1727; found,
323.1734.
N4-(5-Cyclopropyl-1H-pyrazol-3-yl)-N2-((6-(methylamino)pyridin-3-yl)methyl)pyrimidine-
2,4-diamine (20). The title compound was prepared in the same manner as 19, using 5ꢀ
(aminomethyl)ꢀNꢀmethylpyridinꢀ2ꢀamine in place of 5ꢀ(aminomethyl)pyridinꢀ2ꢀamine to give 51
mg of compound 20 as a white solid. Yield: 21%. 1H NMR (400 MHz, DMSOꢀd6, 350 K) δ
ACS Paragon Plus Environment
24

Page 25 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
11.76 (s, 1H), 8.98 (s, 1H), 7.97 (d, J = 2.4 Hz, 1H), 7.82 (d, J = 5.8 Hz, 1H), 7.38 (dd, J = 8.4,
2.4Hz, 1H), 6.70 (s, 1H), 6.38 (d, J = 8.4 Hz, 1H), 6.30 – 5.90 (m, 3H), 4.31 (d, J = 6.0 Hz, 2H),
2.76 (d, J = 4.8 Hz, 3H), 1.87 – 1.79 (m, 1H), 0.98 – 0.78 (m, 2H), 0.72 – 0.57 (m, 2H). 13C
NMR (101 MHz, DMSOꢀd6, 350 K) δ 162.49, 160.41, 159.25, 156.59, 148.96, 147.19, 145.75,
136.92, 123.67, 107.44, 96.50, 93.48, 42.10, 28.58, 7.86, 7.25. HRMS (ESI+): m/z calcd for
C17H21N8 [M + H]+, 337.1884; found, 337.1883.
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
N4-(5-Cyclopropyl-1H-pyrazol-3-yl)-N2-((6-(dimethylamino)pyridin-3-
yl)methyl)pyrimidine-2,4-diamine (21). The title compound was prepared in the same manner
as 19, using 5ꢀ(aminomethyl)ꢀN,Nꢀdimethylpyridinꢀ2ꢀamine in place of 5ꢀ(aminomethyl)pyridinꢀ
2ꢀamine to give 42 mg of compound 21 as a white solid. Yield: 4%. 1H NMR (400 MHz,
DMSOꢀd6 + 1% TFA, 350 K) δ 12.00 (s, 1H), 9.88 (s, 1H), 8.08 (d, J = 2.2 Hz, 1H), 7.82 (d, J =
6.4 Hz, 1H), 7.66 – 7.53 (s, 1H), 7.50 (dd, J = 8.7, 2.4 Hz, 1H), 6.60 (d, J = 8.5 Hz, 1H), 6.32 (d,
J = 6.3 Hz, 1H), 6.09 (s, 1H), 4.40 (d, J = 5.4 Hz, 2H), 3.00 (s, 6H), 1.91 – 1.81 (m, 1H), 0.93 –
0.86 (m, 2H), 0.68 – 0.61 (m, 2H). 13C NMR (101 MHz, DMSOꢀd6 + 1% TFA, 350 K) δ 162.54,
160.52, 158.94, 156.61, 148.97, 147.17, 145.80, 137.34, 123.79, 105.82, 96.53, 93.45, 41.96,
38.22, 7.86, 7.20. HRMS (ESI+): m/z calcd for C18H23N8 [M + H]+, 351.2040; found, 351.2048.
N2-((1H-indol-5-yl)methyl)-N4-(5-cyclopropyl-1H-pyrazol-3-yl)pyrimidine-2,4-diamine (22).
The title compound was prepared in the same manner as 19, using (1Hꢀindolꢀ5ꢀyl)methanamine
in place of 5ꢀ(aminomethyl)pyridinꢀ2ꢀamine to give 99 mg of compound 22 as a white solid.
Yield: 30%. 1H NMR (400 MHz, DMSOꢀd6 + 1% TFA, 350 K) δ 11.96 (s, 1H), 10.83 (s, 1H),
9.75 (s, 1H), 7.82 (d, J = 6.2 Hz, 1H), 7.55 – 7.43 (m, 2H), 7.35 (d, J = 8.5 Hz, 1H), 7.28 (t, J =
2.8 Hz, 1H), 7.14 – 7.06 (m, 1H), 6.42 – 6.36 (m, 1H), 6.35 – 6.20 (s, 1H), 6.20 – 5.94 (s, 1H),
4.62 (d, J = 5.6 Hz, 2H), 1.90 – 1.77 (m, 1H), 0.93 – 0.82 (m, 2H), 0.67 – 0.59 (m, 2H). 13C
ACS Paragon Plus Environment
25

Journal of Medicinal Chemistry
Page 26 of 45
1
2
3
4
5
6
7
8
9
NMR (101 MHz, DMSOꢀd6 + 1% TFA, 350 K) δ 160.45, 159.63, 151.41, 146.92, 135.71,
130.28, 128.21, 125.88, 121.30, 119.01, 111.66, 101.39, 97.11, 93.51, 45.24, 7.90, 7.53. HRMS
(ESI+): m/z calcd for C19H20N7 [M + H]+, 346.1775; found, 346.1778.
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)-N2-(1-(1H-indol-5-yl)ethyl)-N4-(5-cyclopropyl-1H-pyrazol-3-yl)pyrimidine-2,4-diamine
(23). The title compound was prepared in the same manner as 19, using (S)ꢀ1ꢀ(1Hꢀindolꢀ5ꢀ
yl)ethanamine in place of 5ꢀ(aminomethyl)pyridinꢀ2ꢀamine and nꢀbutanol in place of ethanol to
give 190 mg (35%) of compound 23. Yield: 35%. 1H NMR (400 MHz, DMSOꢀd6, 350 K) δ
11.91 (s, 1H), 10.79 (s, 1H), 9.54 (s, 1H), 7.79 (d, J = 6.0 Hz, 1H), 7.55 (d, J = 1.6 Hz, 1H), 7.34
(d, J = 8.5 Hz, 1H), 7.29 – 7.25 (m, 1H), 7.25 – 7.19 (s, 1H), 7.15 (dd, J = 8.6, 1.9 Hz, 1H), 6.39
– 6.35 (m, 1H), 6.28 – 6.12 (s, 1H), 6.12 – 5.86 (s, 1H), 5.27 – 5.16 (m, 1H), 1.92 – 1.82 (m,
1H), 1.53 (d, J = 6.9 Hz, 3H), 0.97 – 0.86 (m, 2H), 0.77 – 0.58 (m, 2H). 13C NMR (101 MHz,
DMSOꢀd6, 350 K) δ 160.30, 159.66, 152.63, 146.85, 135.94, 135.54, 128.09, 125.82, 120.14,
117.41, 111.59, 101.47, 96.82, 93.22, 50.70, 25.90, 23.88, 7.91. HRMS (ESI+): m/z calcd for
C20H22N7 [M + H]+, 360.1931; found, 360.1932.
N2-((1H-benzo[d]imidazol-5-yl)methyl)-N4-(5-cyclopropyl-1H-pyrazol-3-yl)pyrimidine-2,4-
diamine (24). Compound 37b (0.58 g, 2.5 mmol), (1ꢀ(tetrahydroꢀ2Hꢀpyranꢀ2ꢀyl)ꢀ1Hꢀ
benzo[d]imidazolꢀ5ꢀyl)methanamine (0.74 g, 3.2 mmol), DIPEA (9.5 mL, 7.4 mmol), and
ethanol (10 mL) were combined in a vial. The vial was capped, and the reaction mixture was
stirred at 120 °C for 18 hours. The cooled reaction mixture was poured into ethyl acetate, and the
organic layer was washed with water and brine, dried over Na2SO4, filtered, and concentrated
under reduced pressure. The crude residue was purified by column chromatography (SiO2:
heptane / ethyl acetate gradient) to obtain 1.1 g (81%) of N4ꢀ(5ꢀcyclopropylꢀ1Hꢀpyrazolꢀ3ꢀyl)ꢀN2ꢀ
((1ꢀ(tetrahydroꢀ2Hꢀpyranꢀ2ꢀyl)ꢀ1Hꢀbenzo[d]imidazolꢀ5ꢀyl)methyl)pyrimidineꢀ2,4ꢀdiamine. To a
ACS Paragon Plus Environment
26

Page 27 of 45
Journal of Medicinal Chemistry
1
2
solution of N4ꢀ(5ꢀcyclopropylꢀ1Hꢀpyrazolꢀ3ꢀyl)ꢀN2ꢀ((1ꢀ(tetrahydroꢀ2Hꢀpyranꢀ2ꢀyl)ꢀ1Hꢀ
benzo[d]imidazolꢀ5ꢀyl)methyl)pyrimidineꢀ2,4ꢀdiamine (0.40 g, 0.93 mmol) in MeOH (10 mL)
was added pꢀtoluenesulfonic acid (32 mg, 0.19 mmol), and the reaction mixture was stirred at
reflux for 3h. The solvent was removed under reduced pressure, and the crude residue was
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
1
purified by preparative HPLC to give 61 mg of compound 24. Yield: 19%. H NMR (400 MHz,
DMSOꢀd6) δ 12.46 (s, 1H), 11.94 (s, 1H), 9.39 (s, 1H), 8.41 (s, 1H), 7.81 (s, 1H), 7.17 – 7.52 (m,
5H), 6.09 – 6.24 (s, 1H), 4.60 (d, 2H), 1.80 (s, 1H), 0.84 – 0.85 (m, 2H), 0.47 – 0.59 (m, 2H).
[ESꢀMS] (ESI+): m/z calcd for C18H19N8 [M + H]+, 347.2; found, 347.2.
N4-(5-Cyclopropyl-1H-pyrazol-3-yl)-N2-((1-methyl-1H-benzo[d]imidazol-5-
yl)methyl)pyrimidine-2,4-diamine (25). The title compound was prepared in the same manner
as 19, using (1ꢀmethylꢀ1Hꢀbenzo[d]imidazolꢀ5ꢀyl)methanamine in place of 5ꢀ
(aminomethyl)pyridinꢀ2ꢀamine and isopropanol in place of ethanol to give 75 mg of compound
25. Yield: 15%. 1H NMR (400 MHz, DMSOꢀd6 + 1% TFA, 350 K) δ 12.02 (s, 1H), 9.92 (s, 1H),
8.11 (s, 1H), 7.94 – 7.74 (m, 2H), 7.63 (d, J = 1.5 Hz, 1H), 7.49 (d, J = 8.3 Hz, 1H), 7.30 (dd, J =
8.4, 1.6 Hz, 1H), 6.41 – 6.23 (s, 1H), 6.20 – 5.96 (s, 1H), 4.68 (d, J = 5.9 Hz, 2H), 3.82 (s, 3H),
1.88 – 1.78 (m, 1H), 0.94 – 0.80 (m, 2H), 0.69 – 0.54 (m, 2H). 13C NMR (101 MHz, DMSOꢀd6 +
1% TFA, 350 K) δ 162.22, 160.10, 156.05, 148.47, 145.21, 144.46, 143.59, 134.30, 133.58,
121.91, 117.70, 109.42, 96.03, 92.93, 44.36, 30.52, 7.33, 6.64, 0.01.HRMS (ESI+): m/z calcd for
C19H21N8 [M + H]+, 361.1884; found, 361.1888.
N4-(5-cyclopropyl-1H-pyrazol-3-yl)-N2-((1-methyl-1H-benzo[d]imidazol-6-
yl)methyl)pyrimidine-2,4-diamine (26). The title compound was prepared in the same manner
as 25, using (1ꢀmethylꢀ1Hꢀbenzo[d]imidazolꢀ6ꢀyl)methanamine in place of (1ꢀmethylꢀ1Hꢀ
benzo[d]imidazolꢀ5ꢀyl)methanamine to give 48 mg of compound 26. Yield: 14%. 1H NMR (400
ACS Paragon Plus Environment
27

Journal of Medicinal Chemistry
Page 28 of 45
1
2
3
4
5
6
7
8
9
MHz, DMSOꢀd6 + 1% TFA, 350 K) δ 12.05 (s, 1H), 9.98 (s, 1H), 8.11 (d, J = 2.5 Hz, 1H), 8.01 –
7.76 (s, 1H), 7.85 (d, J = 6.5 Hz, 1H), 7.59 (d, J = 8.1 Hz, 1H), 7.54 (s, 1H), 7.24 (dd, J = 8.2, 1.6
Hz, 1H), 6.44 – 6.24 (s, 1H), 6.16 – 5.97 (s, 1H), 4.71 (d, J = 4.1 Hz, 2H), 3.80 (s, 3H), 1.88 –
1.78 (m, 1H), 0.93 – 0.80 (m, 2H), 0.66 – 0.54 (m, 2H). 13C NMR (101 MHz, DMSOꢀd6 + 1%
TFA, 350 K) δ 162.19, 160.11, 156.06, 148.44, 145.21, 144.26, 142.44, 135.38, 134.67, 121.11,
118.82, 108.26, 96.08, 92.96, 44.55, 30.43, 7.34, 6.64, 0.00. HRMS (ESI+): m/z calcd for
C19H21N8 [M + H]+, 361.1884; found, 361.1888.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5-Bromo-1-(tetrahydro-2H-pyran-2-yl)-1H-benzo[d]imidazole (40). A mixture of 5ꢀbromoꢀ
1Hꢀbenzo[d]imidazole (4.5 g, 23 mmol), 3,4ꢀdihydroꢀ2Hꢀpyran (9.2 mL, 0.11 mol), pꢀ
.
TsOH H2O (1.0 g, 3.4 mmol) in THF (200 mL) was stirred under nitrogen at 80 °C for 18 hours.
The solvent was evaporated under reduced pressure, and the crude residue was purified by
column chromatography (SiO2: MeOH / DCM gradient) to afford 6.0 g of 5ꢀbromoꢀ1ꢀ
(tetrahydroꢀ2Hꢀpyranꢀ2ꢀyl)ꢀ1Hꢀbenzo[d]imidazole as yellow oil. Yield: 100%. [ESꢀMS] (ESI+):
m/z calcd for C12H14BrN2O [M + H]+, 281.1; found, 281.1.
1-(1-(Tetrahydro-2H-pyran-2-yl)-1H-benzo[d]imidazol-5-yl)ethanone (41). A mixture of
27b (2.8 g, 0.01 mol), butyl vinyl ether (1.5 g, 0.02 mol), PdCl2 (26 mg, 0.14 mmol), tri(oꢀ
tolyl)phosphine (88 mg, 0.29 mmol), potassium phosphate (1.5 g, 0.02 mol), and
isopropylacetate (20 mL) was heated at reflux under nitrogen for 18 hours. The reaction mixture
was cooled to room temperature and the solvent was removed under reduced pressure. To the
residue was added 6M HCl (60 mL) and the resulting solution was stirred for 15 min. The
mixture was adjusted to pH 8 with aqueous NH4OH (35%) and was extracted with EtOAc (300
mL). The organic layer was washed with brine (50 mL), dried (Na2SO4), filtered and
concentrated under reduced pressure. The residue was purified by column chromatography
ACS Paragon Plus Environment
28

Page 29 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(SiO2: MeOH/ DCM gradient) to afford 1.2 g of 1ꢀ(1ꢀ(tetrahydroꢀ2Hꢀpyranꢀ2ꢀyl)ꢀ1Hꢀ
benzo[d]imidazolꢀ5ꢀyl)ethanone as yellow oil. Yield: 86%. [ESꢀMS] (ESI+): m/z calcd for
C14H17N2O2 [M + H]+, 245.1; found, 245.3.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(S)-2-Methyl-N-((1S)-1-(1-(tetrahydro-2H-pyran-2-yl)-1H-benzo[d]imidazol-5-
yl)ethyl)propane-2-sulfinamide (42). To a solution of (R)ꢀ2ꢀmethylpropaneꢀ2ꢀsulfinamide (6.9
g, 57 mmol) and titanium ethoxide (26 g, 0.11 mol) in THF (200 mL) at room temperature was
added 41 (14 g, 57 mmol), and the reaction mixture was heated to 75 °C for 18 hours. The
mixture was then cooled to ꢀ48 oC, and LꢀSelectride (170 mL, 1M solution in THF) was added
dropwise and stirred at room temperature for an hour. The reaction mixture was cooled to 0 oC
and MeOH was added dropwise until gas evolution was no longer observed. The crude reaction
mixture was poured into an equal volume of rapidly stirred brine. The resulting suspension was
filtered through a plug of Celite® and the filter cake was washed with EtOAc. The filtrate was
washed with brine, and the brine layer was extracted with EtOAc, dried (Na2SO4) and evaporated
under reduced pressure. The residue was purified by chromatography (SiO2: MeOH / DCM
gradient) to afford 10 g of (S)ꢀ2ꢀmethylꢀNꢀ((1S)ꢀ1ꢀ(1ꢀ(tetrahydroꢀ2Hꢀpyranꢀ2ꢀyl)ꢀ1Hꢀ
benzo[d]imidazolꢀ5ꢀyl)ethyl)propaneꢀ2ꢀsulfinamide as yellow oil. Yield: 50%. [ESꢀMS] (ESI+):
m/z calcd for C18H28N3O2S [M + H]+, 350.2; found, 350.1.
(1S)-1-(1-(tetrahydro-2H-pyran-2-yl)-1H-benzo[d]imidazol-5-yl)ethanamine
hydrochloride (43). To a solution of 42 (3.10 g, 8.6 mmol) in ethyl acetate (10 mL) was added
dropwise 2M HCl in ethyl acetate (25 mL), and the reaction mixture was stirred at room
temperature for 18 hours. The resultant precipitate was collected and washed with ethyl acetate
to afford 2.1 g of (1S)ꢀ1ꢀ(1ꢀ(tetrahydroꢀ2Hꢀpyranꢀ2ꢀyl)ꢀ1Hꢀbenzo[d]imidazolꢀ5ꢀyl)ethanamine as
ACS Paragon Plus Environment
29