Optimization of Highly Kinase Selective Bis-anilino Pyrimidine PAK1 Inhibitors

Article abstract of DOI:10.1021/acsmedchemlett.6b00322

Group I p21-activated kinase (PAK) inhibitors are indicated as important in cancer progression, but achieving high kinase selectivity has been challenging. A bis-anilino pyrimidine PAK1 inhibitor was identified and optimized through structure-based drug design to improve PAK1 potency and achieve high kinase selectivity, giving in vitro probe compound AZ13705339 (18). Reduction of lipophilicity to lower clearance afforded AZ13711265 (14) as an in vivo probe compound with oral exposure in mouse. Such probes will allow further investigation of PAK1 biology.

Full text of DOI:10.1021/acsmedchemlett.6b00322

Subscriber access provided by UNIV OF CALIFORNIA SAN DIEGO LIBRARIES
Letter
Optimization of highly kinase selective bis-anilino pyrimidine PAK1 inhibitors
William McCoull, Edward J Hennessy, Kevin Blades, Claudio Chuaqui, James E Dowling,
Andrew D. Ferguson, Frederick Woolf Goldberg, Nicholas Howe, Christopher R Jones,
Paul Kemmitt, Gillian Lamont, Jeffrey G. Varnes, Richard A Ward, and Bin Yang
ACS Med. Chem. Lett., Just Accepted Manuscript • DOI: 10.1021/acsmedchemlett.6b00322 • Publication Date (Web): 14 Sep 2016
Downloaded from http://pubs.acs.org on September 20, 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.
ACS Medicinal Chemistry Letters is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 6
ACS Medicinal Chemistry Letters
1
2
3
4
5
6
7
8
9
Optimization of highly kinase selective bis-anilino pyrimidine PAK1
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
William McCoull,*† Edward J. Hennessy,*‡ Kevin Blades,§,ǂ Claudio Chuaqui,‡ James Dowling,‡ Anꢀ
drew D. Ferguson,‡ Frederick W. Goldberg,† Nicholas Howe,§ Christopher R Jones,§ Paul Kemmitt,†
Gillian Lamont,† Jeffrey Varnes,‡ Richard A. Ward,† and Bin Yang.‡
†AstraZeneca, 310 Cambridge Science Park, Milton Road, Cambridge, CB4 0WG, UK
§AstraZeneca, Alderley Park, Macclesfield, Cheshire, SK10 4TG, UK.
‡AstraZeneca, Gatehouse Park, Waltham, Massachusetts 02451, USA.
KEYWORDS: PAK, LLE, kinase selectivity, probe, bis-anilino pyrimidine.
ABSTRACT: Group I p21ꢀactivated kinase (PAK) inhibitors are indicated as important in cancer progression but achieving high
kinase selectivity has been challenging. A bisꢀanilino pyrimidine PAK1 inhibitor was identified and optimized through structureꢀ
based drug design to improve PAK1 potency and achieve high kinase selectivity, giving in vitro probe compound AZ13705339
(18). Reduction of lipophilicity to lower clearance afforded AZ13711265 (14) as an in vivo probe compound with oral exposure in
mouse. Such probes will allow further investigation of PAK1 biology.
pyrimidine 1 was identified as a modestly potent PAK1 inhibiꢀ
tor (IC50 = 100 nM) with 11ꢀfold selectivity over PAK4 (Table
1). While selectivity within the group I PAKs was not exꢀ
pected, PAK4 was utilized to monitor selectivity against group
II PAKs. Selectivity against some key kinase antiꢀtargets,
KDR and FGFR1,13 was modest. Since 1 originated from an
EphB4 kinase project,14 a member of the Src kinase family,
high activity for Src kinase was expected. Src family kinase
activity was of less concern to us but was monitored as a
measure of overall kinase selectivity. A crystal structure of 1
bound to PAK1 was obtained and compared with a crystal
structure of a 7ꢀazaindole PAK1 inhibitor 2 (PAK1 IC50 = 260
nM) previously reported by us.10 Overlay of these different
chemotypes allowed us to visualize four areas of structure 1 to
be explored to optimize PAK1 binding.
The p21ꢀactivated kinases (PAKs) are a family of six serꢀ
ine/threonineꢀspecific intracellular protein kinases which are
positioned at the intersection of multiple signaling pathways of
importance in cancer progression.1,2 The PAK family is comꢀ
prised of two subgroups based on sequence homology: group
I (PAKs 1ꢀ3) and group II (PAKs 4ꢀ6). Group I PAKs have
high sequence identity in the kinase domain and possess an
autoꢀinhibitory domain which is relieved by binding of the
GTPꢀbinding proteins Rac or Cdc42, whilst group II PAKs
have lower kinase domain homology and are not activated by
Rho GTPases.3 Group I PAKs are overexpressed in a wide
variety of cancers and PAK1 is commonly overexpressed in
breast tumours with poor prognosis.4 In ovarian cancer cells
characterized by PAK1 amplification, treatment with a PAK1
inhibitor decreased proliferation and migration.5 In tumours
characterised by neurofibromatosis type 2 (NF2) inactivation,
group I PAKs have been shown to be hyperactivated.6 Conseꢀ
quently, there is increasing interest in identifying potent and
selective small molecule inhibitors of PAK1 for therapeutic
use in tumours characterised by PAK activation.7
There are relatively few chemotypes described in the literꢀ
ature for PAK1 inhibitors8 and only panꢀPAK inhibitor PFꢀ
35783099 progressed into clinical trials but is now stopped.
Ourselves10 and others11, 12 have recently disclosed PAK1 inꢀ
hibitors but achieving high kinase selectivity with a chemoꢀ
type amenable to achieving oral exposure has been challengꢀ
ing. Without high kinase selectivity it is difficult to discern
whether PAK1 is driving efficacy thus new alternative chemoꢀ
types for selective PAK1 inhibition are required.
1
2
Figure 1. PAK1 inhibitors use to guide structure-based
drug design
Firstly, the R4 position of the pyrimidine was unsubstitutꢀ
ed but a small space was available as occupied by the 7ꢀ
azaindole of 2 but not the pyrimidine of 1. Consequently,
addition of a fluorine at R4 in 3 increased potency 7ꢀfold and
this potency gain more than offset the higher lipophilicity (inꢀ
Following a kinaseꢀfocused subset screen (120k comꢀ
pounds) of the AstraZeneca compound collection, bisꢀanilino
creased
LLE15).
ACS Paragon Plus Environment

ACS Medicinal Chemistry Letters
Page 2 of 6
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. X-ray crystallography of compounds in complex with PAK1: (a) 1 (PDB 5KBQ); (b) 2 (PDB 5KBR); (c) overlay of 1 & 2.
Table 1. “Head-group” SAR around initial screening hit
O
N
R3
R1
R4
N
R2
N
S
N
H
N
O
O
PAK1
pPAK1
PAK4
KDR
FGFR1
ratiob
Src
Cpd
R1
R2
R3
R4
logD7.4
LLEc
IC50 (nM)a
IC50 (ꢁM)a
ratiob
11
ratiob
2.5
ratiob
1
3
4
5
6
7
8
Me
Me
H
H
H
CH2OH
CH2OH
CH2OH
CH2OH
CH2OMe
CH2OH
CH2OH
H
F
F
F
F
F
F
100
14
3.1
0.86
14
7.0
81
0.0021
0.034
0.11
2.9
3.3
3
4.1
4.6
4.4
4.4
3
14
93
25
2.0
0.52
1.7
18
H
45
43
H
Me
H
19
0.42
3.3d
0.48
0.22
0.088
21
1.4
1.7
63
0.069
0.0093
0.14
3.3
3.7
3.7
3.8
>4
Me
Cl
189
9.6
3.0
2.8
26
H
9.5
24
4.3
4.7
<4.6
Me
Me
22
66
0.17
FRAX-597
260
1.2
1.7
6.2
aGeometric mean of at least two experiments unless otherwise stated. IC50 values were determined at ATP concentrations within 2ꢀfold of
the measured KM ATP for their respective kinases; bselectivity calculated from ratio of IC50 values; cLLE = pIC50 (PAK1) ꢀ logD7.4; dn=1
This was viewed as a significant improvement and F at R4
was utilized in all subsequent compounds (Table 1).
proved PAK1 potency to 3 nM with a similar selectivity proꢀ
file to monoꢀMe 3. It is notable that using LLE as a measure
of compound quality that 3 and 8 contain the optimal head
groups from this SAR exploration. Furthermore, 8 already
compared favorably in terms of PAK1 potency and selectivity
with FRAXꢀ597,16 the best external benchmarking PAK1 inꢀ
hibitor available at the time (Table 1). Also, 8 already has
improved selectivity against our key kinase antiꢀtargets KDR
and FGFR1 although FRAXꢀ597 was 10ꢀfold more selective
against PAK4.
Secondly, the benzylic alcohol “head group” formed some
beneficial interactions in the kinase selectivity pocket with Me
at R1 occupying a lipophilic cavity and methanol at R3 realizꢀ
ing a polar interaction with Glu315 Based on the overlay, the
unsubstituted R2 position appears most likely to tolerate furꢀ
ther modification into the same region as the chlorine atom in
2. Consequently we explored some destructive SAR plus R2
exploration and the key findings are shown in Table 1. Reꢀ
moval of the Me at R1 resulted in 4 with 3ꢀfold reduced PAK1
potency and reduced KDR selectivity. Moving the Me group
from R1 to R2 in 5 maintained PAK1 potency but eroded all
KDR and FGFR1 selectivity, indicating the need for an R1
substituent for improved selectivity. Methylation of the R3
alcohol in 6 reduced PAK1 potency 14ꢀfold confirming the
importance of the alcohol to Glu315 interaction. Switching
Me for Cl at R1 in 7 maintained PAK1 potency and selectivity
against the other kinases indicating that either of these groups
was beneficial. The diꢀMe combination compound 8 imꢀ
The third area suggested from the overlay was in the solꢀ
vent accessible channel region of PAK1. The piperidine of 2
bisected the two aryl substituent of 1, suggesting that a 6
membered cyclic base would be tolerated para to the aniline of
1. Thus paraꢀpiperazine 9 was prepared which exhibited enꢀ
couraging PAK1 potency of 33 nM and increased selectivity
of 150ꢀfold over PAK4 (Table 2). In our 7ꢀazaindole series
we had achieved kinase selectivity improvements in the solꢀ
vent channel area10 thus we expected similar selectivity gains
could be achieved by further optimization. The methanesulꢀ
ACS Paragon Plus Environment

Page 3 of 6
ACS Medicinal Chemistry Letters
fonyl group was tolerated in combination with piperazine and again gave similar LLE in line with a lipophilicity driven 4ꢀ
1
2
3
4
5
6
10 exhibited significantly improved selectivity against KDR
and FGFR1 while maintaining potency relative to matched
pair morpholine compound 3. Increasing the bulk of the sulꢀ
fone to ethyl in 11 gave a 4ꢀfold PAK1 potency increase and
in excess of 100ꢀfold selectivity against the three main kinase
antiꢀtargets. The alternative diꢀMe head group compound 12
fold PAK1 potency increase. The matched pair morpholine
compound 13 was informative in that it was isolipophilic with
12 but it was less potent against PAK1 and has lower selectiviꢀ
ty against all the other kinases measured. Lipophilicity could
be lowered without loss of PAK1 potency or kinase selectivity
by utilizing a bridged piperazine as in compound 14.
7
8
Table 2. “Solvent tail” & p-loop SAR
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
PAK1
pPAK1
PAK4 / KDR /
Cpd
R2
R5
H
R6
R7
H
R8
logD7.4
LLEc
4.5
IC50 (nM)a
IC50 (ꢁM)a
FGFR1 / Src ratiob
N
N
N
N
N
N
N
N
9
Me
Me
33
8.7
2.1
150 / 8.9 / 13 / 0.72
80 / 500/ 570 / 1.5
3
10
11
12
H
H
MeSO2
EtSO2
EtSO2
H
H
H
Me
Me
Me
0.81
0.49
0.20
2.6
3
5.5
5.7
5.6
2.1
100 / 760 / 1400 / 2.2
160 / 330 / 1300 / 3.3
Me
0.45
3.8
Me
EtSO2
H
O
N
Me
2.8
0.27
17 / 8.2 / 41 / 0.12
3.8
4.7
13
0.58d
2.0
0.11
0.53
0.23
0.13
90 / 2100 / 880 / 0.96
200 / 24 / 620 / 2.3
2.5
3.5
2.8
3.3
6.7
5.2
6.0
6.2
N
N
N
N
N
N
N
N
14
15
16
17
Me
H
EtSO2
EtSO2
EtSO2
EtSO2
H
H
H
H
Me
Et
H
Bn
1.5
1500 / 420 / 1600 / 11
3800 / 950 / 1700 / 9.0
H
0.35d
F
N
N
18
H
EtSO2
H
0.33
0.059
2600 / 4100 / 470 / 14
3.7
5.7
CN
aGeometric mean of at least two experiments unless otherwise stated. IC50 values were determined at ATP concentrations within 2ꢀfold of
the measured KM ATP for their respective kinases; bselectivity calculated from ratio of IC50 values; cLLE = pIC50 (PAK1) ꢀ logD7.4; dn=1.
ACS Paragon Plus Environment

ACS Medicinal Chemistry Letters
Page 4 of 6
Scheme 1. Synthesis of 14 and 18a
1
2
3
4
5
N
F
S
N
(i), (ii)
NO2
S
NH2
O
O
O
O
19
20
6
7
8
9
OH
OH
OH
R
OH
N
F
N
S
F
F
R
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
N
N
N
(iii)
(iv)
(v)
N
N
H
N
Cl
N
N
R = Me
H2
N
Cl
N
N
H
O
O
21
R = H, Me
22
23
14
OH
OH
N
F
N
S
F
N
N
(v)
(iv)
Cl
N
N
N
H
N
N
R = H
O
O
CN
CN
24
18
aReagents and conditions: (i) K2CO3, DMSO, 55 oC, 42%; (ii) Pd/C, H2, EtOAc, EtOH, 93%; (iii) nBuOH, 2,4ꢀdiCl,5ꢀFꢀpyrimidine,
(iPr)2NEt, 100 °C, R=Me 77%; R=H, 67%; (iv) R=Me, MeI, DMF 100%, R=H, 2ꢀCNꢀBnCl, DMF 79%; (v) HCl, TsOH, iPrOH,
R=Me; 20 30%; R=H, 3ꢀEtSO2ꢀ4ꢀ(1ꢀmethylpiperazine)ꢀaniline 47%.
The fourth and final area for optimization suggested from
crystal structure analysis was increasing the size of substituent
attached to the aniline beyond the current methyl substitution.
The NMe of 1 has a similar directionality to the ketone of 2,
with both being directed into an area of space not occupied by
the protein. We reasoned that increasing bulk could reach
towards the pꢀloop and potentially further improve potency
and selectivity. While ethyl substitution (compare 15 with 11)
gave no advantage, benzyl compound 16 improved PAK4
selectivity 15ꢀfold. Addition of F (17) or CN (18) to the ortho
position of the benzyl group gave a 4ꢀfold PAK1 potency inꢀ
crease with improved selectivity against all antiꢀtarget kinases
measured, including Src.
AZ13705339 (18) had 59 nM cellular pPAK1 potency,
high selectivity against PAK4, KDR and FGFR1, and even 14ꢀ
fold selectivity against Src so was considered to be an excelꢀ
lent in vitro probe compound. To further evaluate 18, screenꢀ
ing was conducted against a panel of 125 kinases (screened at
100 nM concentration at Millipore, equal to 303ꢀfold greater
than PAK1 IC50). High kinase selectivity was observed for 18,
with only 8 kinases giving greater than 80% inhibition (Figure
3). This included PAK1 and PAK2 and predominantly a few
Srcꢀfamily kinases. The IC50 values for these kinases were
determined as shown in Table 3. PAK1 was measured as out
of range which is consistent with our internal potency assay
and selectivity clearly exists for PAK1 over all kinases tested.
Interestingly, this includes PAK2, albeit PAK2 potency is still
high at 6 nM. Selectivity between PAK1 and PAK2 was not
expected but this can be attributed to differences in KM for
ATP (71 ꢁM for PAK2 and 17 ꢁM for PAK1). PAK2 is
known to lead to embryo lethality thus a compound with
PAK1 over PAK2 selectivity could be of value.17 18 was
screened in an alternative binding assay format at DiscoveRx
and the PAK1 and PAK2 affinities were measured at Kd of
0.28 and 0.32 nM respectively, indicating that 18 does equally
bind to PAK1 and PAK2.
A representative synthesis of this series of compounds is
shown in Scheme 1. Reaction of commercially available 3ꢀ
(ethanesulfonyl)ꢀ4ꢀfluoroaniline
methylpiperazine or
diazabicyclo[2.2.1]heptane followed by hydrogenation gave
the desired anilines for solvent tail exploration. 2,4ꢀDichloroꢀ
5ꢀfluoropyrimidine underwent an SNAr reaction with either 3ꢀ
aminoꢀ4ꢀmethylbenzyl alcohol 21 (R=H) or (5ꢀaminoꢀ2,4ꢀ
dimethylphenyl)methanol 21 (R=Me) in 67% and 61% yields
respectively. NꢀAlkylation followed by a final SNAr reaction
with the appropriate aniline provides the desired compounds
14 and 18.
19
with
either
1ꢀ
(1S,4S)ꢀ2ꢀmethylꢀ2,5ꢀ
Throughout the optimization process, cellular pPAK1 inꢀ
hibition was measured in an MCF10A cell line.10
This
cell potency was found to correlate well with the PAK1 enꢀ
zyme activity, with a shift as expected based on different ATP
concentrations in the two assays (see Supporting Information).
A cross check with a similar assay measuring pPAK1 inhibiꢀ
tion in the PAK1 amplified cell line OVCAR3 for 11 showed
comparable inhibition (IC50 = 0.54 vs 0.49 ꢁM for OVCAR3
and MCF10A respectively).
Figure 3. Kinase inhibition for 18 (125 kinases @ 100 nM)
ACS Paragon Plus Environment

Page 5 of 6
ACS Medicinal Chemistry Letters
Table 3. Kinase inhibition for 18 (>80% @ 100 nM)a
tive, cellular active in vitro probe 18. Compound quality was
1
2
3
4
5
6
7
8
monitored throughout the process and this enabled in vivo
probe 14 to also be identified from this series. Such probes
will be valuable tools for those looking to understand PAK1
biology and our own further investigations will be the pubꢀ
lished in due course.
kinase
% inhibition @ 100 nM
IC50 (nM)
PAK1
100
98
95
91
91
87
86
83
< 1
4
Yes
PAK2
Lyn
6
31
7
ASSOCIATED CONTENT
Supporting Information
PKCtheta
Fyn
45
93
36
9
Procedures for synthesis and characterization of compounds; kiꢀ
nase selectivity data for compound 18; correlation of pPAK1 cell
assay with PAK1 enzyme assay; effect of 18 in in vitro radioligꢀ
and binding, enzyme and functional assays; protein expression,
purification, crystallization and structure determination for 1 and
2 in PAK1 (PDF). The Supporting Information is available free of
charge on the ACS Publications website.
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
Src
Lck
aData was obtained using Millipore KinaseProfiler service
Furthermore, in vitro pharmacological profiling18 of 18
against a panel of over 80 diverse targets was conducted and
exhibited good selectivity (see Supporting Information). No
targets showed activity below 100 nM and 10 targets showed
activity below 1 ꢁM (M1, M2, M5, D3, 5HT2B, 5HT2C, DT,
CaVꢀL and ARa1B, ARa2C)
AUTHOR INFORMATION
Corresponding Author
*(W.M.) Tel: +44 (0) 1625 236807; Eꢀmail:
william.mccoull@astrazeneca.com
While 18 is an excellent in vitro tool compound for group I
PAK inhibition, it does not have optimal properties for in vivo
exposure, mainly due to moderate clearance (CLint = 30
ꢁL/min/106 in rat hepatocytes) likely driven by high lipophilicꢀ
ity. The lower lipophilicity analogue AZ13711265 (14) disꢀ
played lower clearance (CLint = 11 ꢁL/min/106) and was proꢀ
filed in mouse pharmacokinetics (Table 4). Oral exposure was
achieved for 14 with moderate clearance in vivo and an oral
Cmax of 7.7 ꢁM from a 100 mg/Kg dose. Correcting for plasma
protein binding this translates into 2ꢀfold excess free cover at
Cmax of the cellular IC50 for pPAK1. A recent publication19 has
linked free cover above a cellular PAK1 endpoint with acute
toxicity, thus it is worth noting that we did not observe any
visible toxicity in this experiment. However, this was not a
maximum tolerated dose experiment and 14 achieved free
cover for about 4 h above the pPAK1 IC50 which may not be
exactly comparable with the cell assay used to calculate acute
toxicity in the publication,19 thus we cannot rule out the possiꢀ
bility of toxicity at higher doses.
Present Addresses
ǂRedx Antiꢀinfectives, Alderley Park, Macclesfield, Cheshire,
SK10 4TG, UK
Author Contributions
The manuscript was written through contributions of all authors.
All authors have given approval to the final version of the manuꢀ
script. *These authors contributed equally.
Notes
The authors declare no competing financial interest
ACKNOWLEDGMENT
Claire McWhirter, Martina Fitzek, Renee GarciaꢀArenas and
Brenda McDermott are acknowledged for generating kinase inhiꢀ
bition data. Robert Godin, Mark Roth and Paul Clarkson are
acknowledged for expert technical assistance in interpretation of
biological data. Caroline Rivard is acknowledged for expert
DMPK input
Table 4. Mouse pharmacokinetic parameters for 14a
ABBREVIATIONS
ATP, adenosine triphosphate; PAK, p21ꢀactivated kinase; LLE,
ligand lipophilicity efficiency; PPB, plasma protein binding.
parameter
14
IV CL (mL/min/Kg)
Vdss (L/Kg)
22
1.4
1.0
7.7
2.9
REFERENCES
1. Radu, M.; Semenova, G.; Kosoff, R.; Chernoff, J. PAK signalling
during the development and progression of cancer. Nat Rev Cancer
2014, 14, 13ꢀ25.
2. King, H.; Nicholas, N. S.; Wells, C. M. Chapter Seven ꢀ Role of pꢀ
21ꢀActivated Kinases in Cancer Progression. In International Review
of Cell and Molecular Biology; Kwang W. Jeon, Ed.; Academic
Press: 2014; Vol. 309, pp 347ꢀ387.
3. Kumar, R.; Gururaj, A. E.; Barnes, C. J. p21ꢀActivated Kinases in
Cancer. Nat Rev Cancer 2006, 6, 459ꢀ471.
4. Ong, C. C.; Jubb, A. M.; Haverty, P. M.; Zhou, W.; Tran, V.; Truꢀ
ong, T.; Turley, H.; O'Brien, T.; Vucic, D.; Harris, A. L.; Belvin, M.;
Friedman, L. S.; Blackwood, E. M.; Koeppen, H.; Hoeflich, K. P.
Targeting p21ꢀactivated kinase 1 (PAK1) to induce apoptosis of tuꢀ
mor cells. Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 7177ꢀ7182,
S7177/1ꢀS7177/16.
IV halfꢀlife (h)
po Cmax (ꢁM)
PPB (%free)
a18 was dosed IV at 0.5 mg/Kg (n=2) in 5% DMSO: 95%
SBEꢀβꢀCD (30% w/v) in water, and po at 100 mg/Kg (n=2) in
10% DMSO: 90% kleptose (45% w/v) in water.
In conclusion, bisꢀanilino pyrimidine inhibitors of PAK1
were obtained from a kinase subset screen. Using an overlay
of two chemotypes bound to PAK1, we were able to optimize
potency against PAK1 by maximizing the efficiency of interꢀ
actions at four positions around the molecule. Increasing kiꢀ
nase selectivity was incorporated into this optimization, parꢀ
ticularly against key antiꢀtargets, resulting in a highly selecꢀ
ACS Paragon Plus Environment

ACS Medicinal Chemistry Letters
Page 6 of 6
5. Prudnikova, T. Y.; VillamarꢀCruz, O.; Rawat, S. J.; Cai, K. Q.;
Gꢀ5555: Improving Properties by Employing an Unorthodox Lowꢀ
pKa Polar Moiety. ACS Med. Chem. Lett. 2015, 6, 1241ꢀ1246.
13. Norman, R. A.; Schott, A.; Andrews, D. M.; Breed, J.; Foote, K.
M.; Garner, A. P.; Ogg, D.; Orme, J. P.; Pink, J. H.; Roberts, K.;
Rudge, D. A.; Thomas, A. P.; Leach, A. G. ProteinꢀLigand Crystal
Structures Can Guide the Design of Selective Inhibitors of the FGFR
Tyrosine Kinase. J. Med. Chem. 2012, 55, 5003ꢀ5012.
14. Barlaam, B.; Ducray, R.; Lambertꢀvan, d. B.; Ple, P.; Bardelle, C.;
Brooks, N.; Coleman, T.; Cross, D.; Kettle, J. G.; Read, J. Inhibitors
of the tyrosine kinase EphB4. Part 4: Discovery and optimization of a
benzylic alcohol series. Bioorg. Med. Chem. Lett. 2011, 21, 2207ꢀ
2211.
15. Tarcsay, A.; Nyiri, K.; Keseru, G. M. Impact of Lipophilic Effiꢀ
ciency on Compound Quality. J. Med. Chem. 2012, 55, 1252ꢀ1260.
16. Licciulli, S.; Maksimoska, J.; Zhou, C.; Troutman, S.; Kota, S.;
Liu, Q.; Duron, S.; Campbell, D.; Chernoff, J.; Field, J.; Marmorstein,
R.; Kissil, J. L. FRAX597, a Small Molecule Inhibitor of the p21ꢀ
activated Kinases, Inhibits Tumorigenesis of Neurofibromatosis Type
2 (NF2)ꢀassociated Schwannomas. Journal of Biological Chemistry
2013, 288, 29105ꢀ29114.
17. Radu, M.; Lyle, K.; Hoeflich, K. P.; VillamarꢀCruz, O.; Koeppen,
H.; Chernoff, J. p21ꢀactivated kinase 2 regulates endothelial develꢀ
opment and function through the Bmk1/Erk5 pathway. Mol. Cell.
Biol. 2015, 35, 3990ꢀ4005.
18. Bowes, J.; Brown, A. J.; Hamon, J.; Jarolimek, W.; Sridhar, A.;
Waldron, G.; Whitebread, S. Reducing safetyꢀrelated drug attrition:
the use of in vitro pharmacological profiling. Nat. Rev. Drug Discov-
ery 2012, 11, 909ꢀ922.
19. Rudolph, J.; Murray, L. J.; Ndubaku, C. O.; O'Brien, T.; Blackꢀ
wood, E.; Wang, W.; Aliagas, I.; Gazzard, L.; Crawford, J. J.; Drobꢀ
nick, J.; Lee, W.; Zhao, X.; Hoeflich, K. P.; Favor, D. A.; Dong, P.;
Zhang, H.; Heise, C. E.; Oh, A.; Ong, C. C.; La, H.; Chakravarty, P.;
Chan, C.; Jakubiak, D.; Epler, J.; Ramaswamy, S.; Vega, R.; Cain, G.;
Diaz, D.; Zhong, Y. Chemically Diverse Group I p21ꢀActivated Kiꢀ
nase (PAK) Inhibitors Impart Acute Cardiovascular Toxicity with a
Narrow Therapeutic Window. J. Med. Chem. 2016, 59, 5520ꢀ5541.
Chernoff, J. Effects of p21ꢀactivated kinase 1 inhibition on 11q13ꢀ
amplified ovarian cancer cells. Oncogene 2016, 35, 2178ꢀ2185.
6. Menges, C. W.; Sementino, E.; Talarchek, J.; Xu, J.; Chernoff, J.;
Peterson, J. R.; Testa, J. R. Group I p21ꢀActivated Kinases (PAKs)
Promote Tumor Cell Proliferation and Survival through the AKT1
and RafꢀMAPK Pathways. Mol. Cancer Res. 2012, 10, 1178ꢀ1188.
7. Senapedis, W.; Crochiere, M.; Baloglu, E.; Landesman, Y. Theraꢀ
peutic Potential of Targeting PAK Signaling. Anticancer Agents Med
Chem 2015, 16, 75ꢀ88.
8. Rudolph, J.; Crawford, J. J.; Hoeflich, K. P.; Wang, W. Inhibitors
of p21ꢀActivated Kinases (PAKs). J. Med. Chem. 2015, 58, 111ꢀ129.
9. Murray, B. W.; Guo, C.; Piraino, J.; Westwick, J. K.; Zhang, C.;
Lamerdin, J.; Dagostino, E.; Knighton, D.; Loi, C.; Zager, M.;
Kraynov, E.; Popoff, I.; Christensen, J. G.; Martinez, R.; Kephart, S.
E.; Marakovits, J.; Karlicek, S.; Bergqvist, S.; Smeal, T. Smallꢀ
molecule p21ꢀactivated kinase inhibitor PFꢀ3758309 is a potent inhibꢀ
itor of oncogenic signaling and tumor growth. Proc. Natl. Acad. Sci.
U. S. A. 2010, 107, 9446ꢀ9451
10. McCoull, W.; Hennessy, E. J.; Blades, K.; Box, M. R.; Chuaqui,
C.; Dowling, J. E.; Davies, C. D.; Ferguson, A. D.; Goldberg, F. W.;
Howe, N. J.; Kemmitt, P. D.; Lamont, G. M.; Madden, K.; McWhirtꢀ
er, C.; Varnes, J. G.; Ward, R. A.; Williams, J. D.; Yang, B. Identifiꢀ
cation and optimisation of 7ꢀazaindole PAK1 inhibitors with imꢀ
proved potency and kinase selectivity. MedChemComm 2014, 5,
1533ꢀ1539.
11. Rudolph, J.; Aliagas, I.; Crawford, J. J.; Mathieu, S.; Lee, W.;
Chao, Q.; Dong, P.; Rouge, L.; Wang, W.; Heise, C.; Murray, L. J.;
La, H.; Liu, Y.; Manning, G.; Diederich, F.; Hoeflich, K. P. Leveragꢀ
ing the PreꢀDFG Residue Thrꢀ406 To Obtain High Kinase Selectivity
in an AminopyrazoleꢀType PAK1 Inhibitor Series. ACS Med. Chem.
Lett. 2015, 6, 711ꢀ715.
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
12. Ndubaku, C. O.; Crawford, J. J.; Drobnick, J.; Aliagas, I.; Campꢀ
bell, D.; Dong, P.; Dornan, L. M.; Duron, S.; Epler, J.; Gazzard, L.;
Heise, C. E.; Hoeflich, K. P.; Jakubiak, D.; La, H.; Lee, W.; Lin, B.;
Lyssikatos, J. P.; Maksimoska, J.; Marmorstein, R.; Murray, L. J.;
O'Brien, T.; Oh, A.; Ramaswamy, S.; Wang, W.; Zhao, X.; Zhong,
Y.; Blackwood, E.; Rudolph, J. Design of Selective PAK1 Inhibitor
Insert Table of Contents artwork here
ACS Paragon Plus Environment