Discovery of novel imidazo[1,2-a]pyrazin-8-amines as Brk/PTK6 inhibitors

Article abstract of DOI:10.1016/j.bmcl.2011.07.101

A series of substituted imidazo[1,2-a]pyrazin-8-amines were discovered as novel breast tumor kinase (Brk)/protein tyrosine kinase 6 (PTK6) inhibitors. Tool compounds with low-nanomolar Brk inhibition activity, high selectivity towards other kinases and desirable DMPK properties were achieved to enable the exploration of Brk as an oncology target.

Full text of DOI:10.1016/j.bmcl.2011.07.101

Bioorganic & Medicinal Chemistry Letters 21 (2011) 5870–5875
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of novel imidazo[1,2-a]pyrazin-8-amines as Brk/PTK6 inhibitors
Hongbo Zeng a, , David B. Belanger a, Patrick J. Curran a, Gerald W. Shipps Jr. a, Hua Miao a,
⇑
Jack B. Bracken a, M. Arshad Siddiqui a, Michael Malkowski b, Yan Wang b
a Department of Chemistry, Merck Research Laboratories, 320 Bent Street, Cambridge, MA 02141, United States
b Department of Oncology, Merck Research Laboratories, 2015 Galloping Hill Road, Kenilworth, NJ 07033, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of substituted imidazo[1,2-a]pyrazin-8-amines were discovered as novel breast tumor kinase
(Brk)/protein tyrosine kinase 6 (PTK6) inhibitors. Tool compounds with low-nanomolar Brk inhibition
activity, high selectivity towards other kinases and desirable DMPK properties were achieved to enable
the exploration of Brk as an oncology target.
Received 5 April 2011
Revised 25 July 2011
Accepted 26 July 2011
Available online 3 August 2011
Ó 2011 Elsevier Ltd. All rights reserved.
Keywords:
Brk/PTK6 kinase inhibitors
Imidazo[1,2-a]pyrazin-8-amine
Cancer treatment
Breast tumor kinase (Brk, also known as PTK6) has been cloned
from metastatic breast tumor samples and from cultured human
melanocytes.1,2 Brk is normally expressed in the differentiating epi-
thelial cells of the intestine, skin, prostate, and oral cavity3–6 where
it has been shown to promote cellular differentiation, apoptosis,
and more recently to mediate migration/wound healing.7 In tumors
which over express Brk this Src-family, nonreceptor protein–tyro-
sine kinase has been implicated as a mediator of cancer cell pheno-
types, including increased proliferation, survival, and migration.8
Potential Brk substrates include RNA-binding proteins: Sam68,9
SLM-1,10 SLM-2,11 and PSF12; transcription factors: STAT313 and
STAT5a/b14; and a variety of signaling molecules: p190RhoGAP,15
paxillin,16 Akt,17 IRS-4,18 BKS/STAP-2,19 and KAP3A.20
Studies have shown that Brk interacts with the ErbB family sig-
naling pathway. Early work by Kamalati et al. demonstrated an
interaction between Brk and the ErbB family member EGFR and
over expression of Brk enhanced EGF-induced proliferation.21
Additional ErbB family members such as ErbR2, ErbR3, and ErbR4
showed similar interactions with Brk.22–24 Brk was also found to
enhance ERK1/2 activation in ErbB over-expressing cells.22
Rac1 activate migration and invasion programs in skin (A431)
and breast (MDA-MB231) cancer cell lines.15 Lange et al. showed
that breast cancer cells are dependent upon Brk expression for
their migratory response to EGF and heregulin-b1.8 It was also re-
ported that the Brk/Rac1/p38 MAPK pathway is required for EGF or
heregulin-b1 induced breast cancer cell migration.25 Recently it
was reported that Brk protects breast cancer cells from autophagic
cell death induced by loss of anchorage.26
All these findings suggested that Brk plays a divergent role as a
regulatory proto-oncogene in selected tissues and oncogenic role
in malignant settings. The specific cellular mechanisms of the Brk
pathway are very important in the rational drug design of cancer
therapy. Specific inhibition of Brk kinase activity may provide a
potentially novel approach to inhibit the growth of selected tu-
mors, sensitize the response of the tumor cells to other chemother-
apeutics and prevent/inhibit metastasis of cancer with enhanced
therapeutic windows.
There are no specific Brk/PTK6 inhibitors reported to date.
Dasatinib (3, Fig. 1), a Src/Abl multi-kinase inhibitor that is ap-
proved for chronic myelogenous leukemia (CML), is a potent
PTK6/Brk (IC50 = 9 nM).27
Focused library screening using a HTRF-format biochemical as-
say28a as well as ALIS29 screening of our internal compound library
against the Brk enzyme yielded several classes of compounds,
including the imidazo[1,2-a]pyrazin-8-amine series (1, Fig. 1)
described herein. Compound 1 had a moderate inhibitory activity
(IC50 = 500 nM) to Brk.
A report from Kamalati et al. showed that over expression of Brk
potentially activates the PI3k/Akt pathway.23 Brk has also been
shown to play a role in the activation of STAT signaling path-
way.14,13 Several developing lines of evidence suggest that Brk is
involved directly in the processes of migration and invasion that
characterize metastatic breast malignancy. For example, Chen
et al. reported that Brk is upstream of paxcillin, and CrkII and
The imidazo[1,2-a]pyrazine series is known to have strong
affinity to Aurora kinases.30 In addition, Brk itself is highly homol-
ogous to other Src family kinases. Therefore our primary objectives
⇑
Corresponding author. Tel.: +1 617 992 3489; fax: +1 617 992 2406.
E-mail addresses: zenghb@hotmail.com, hongbo.zeng@merck.com (H. Zeng).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.07.101

H. Zeng et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5870–5875
5871
H
N
I
H
N
OH
R6
Br
Br
N
N
2
N
N
N
N
R2
a, b
c, d
N
N
R2
Cl
N
5
N
N
N
N
6
NH2
3
S
N
N
N
S
Br
S
7 N
N
N
N 1
N
HN
Cl
8
6
5
H
N
4
HN
HN
S
S
NH
N
R3
N
R3
5
O
R6
1
2
R6
3
N
Cl
O
R2
O
N
R2
i, j
e, f
g, h
N
N
2
N
N
1
8
Figure 1. Brk screening hits and Dasatinib (3, Sprycel).
HN
O S
O
OH
7
8
were to improve biochemical potency and mechanism-based cellu-
lar activity (inhibition of phosphorylation of SAM68) as well as
selectivity towards Aurora and Src family kinases (i.e., Lck). An
analysis of compounds in our internal file and subsequent fol-
low-up screening resulted in the identification of compound 2
( Fig. 1), which had relatively potent enzymatic activity (Brk
IC50 = 7 nM), moderate cellular activity (p-SAM68 IC50 = 156 nM)
and some selectivity towards both Aurora B (IC50 = 754 nM) and
Lck (IC50 = 420 nM). In comparison, Dasatinib inhibited Aurora B
and Lck with IC50 = 6485 and 3 nM, respectively.
Cl
Scheme 1. Reagents and conditions: (a) 50% chloraldehyde in water, IPA, reflux,
79%; (b) NaSMe, MeOH, rt, 93%; (c) trimethylboroxine, 5 mol % Pd(Ph3P)4, K2CO3,
DMF, 100 °C, 54%; (d) N-iodosuccinimide, DMF, 60 °C, 97%; (e) SEM-pyrazolobo-
ronic acid pinacol ester, 5 mol % Pdcl2(dppf), K3PO4, 1,4-dioxane/water (9:1), 100 °C,
92%; (f) m-CPMA, DCM, rt, 90%; (g) methyl 4-amino-2-chlorobenzoate, NaH, DMF,
rt, 83%; (h) LiOH, THF/water (1:1) rt; (i) Boc-piperazine, HATU, DIEA, DMF; (j) 4 N
HCl in dioxane, 40 °C.
A computational model was generated by docking compound 2
into a homology model of BRK (Fig. 2) using SRC as the template
and predicted that the series bound to BRK in the ATP pocket.
According to this model the pyrazole ring resides inside the en-
zyme while the C-2 nitrogen and the C-8 NH form hydrogen bonds
with the hinge region of Brk and is similar to that of published Aur-
ora A/B inhibitors.30 The 2-chlorobenzoic amide interacts within
the hydrophobic region in front of the ATP binding pocket and ex-
tends to the solvent front. A distinct hydrogen bond was also
expected between the amide carbonyl group and the side chain
of Arg-195. Aurora kinases prefer an amino-isothiazole ring in this
region to stabilize the bioactive conformation through a polar
interaction between the 7-position core nitrogen and the isothiaz-
ole sulfur atom.30 This difference suggested that the core orienta-
tion can be changed through substitutions on the benzene or
piperazine ring; or that the ring could be replaced with other het-
erocycles to improve selectivity toward Aurora kinases.
The general synthetic route for compound 2 and its analogs are
shown in Scheme 1. Condensation of 3,5-dibromopyrazin-2-amine
and 2-chloroacetaldehyde in i-PrOH at reflux afforded the imi-
dazo[1,2-a]pyrazine core. An 8-Br moiety was substituted with a
methylthiol group to give intermediate 5 (for 2, R2 = H, 5a). Iodiza-
tion at the C-3 position provided the key intermediate 6 (for 2,
R2 = H, R6 = Me 6a), which facilitated installation of different
functional groups at C-3 by coupling with readily available boronic
acids. The methylthiol group was then converted to methylsulfone
by using m-CPBA to give key intermediate 7 (for 2, R2 = H, R3 = 3-
pyrazol, R6 = Me, 7a), which could be replaced with a variety of
nucleophilic amines. Reaction of 7a with methyl 4-amino-2-chlo-
robenzoate followed by hydrolysis gave key intermediate 8 (for
2, R2 = H, R3 = 3-pyrazol, R6 = Me, 8a). Coupling of acid 8 with var-
ious amines generated amide-based analogs and global deprotec-
tion of acid-labile protecting groups with HCl gave the final
product 2.
Syntheses of the C-2 substituted analogs 11a–d (Scheme 2.)
started with 4 and corresponding 2-chloro-aldehydes followed by
subsequent steps to provide the key intermediate sulfones 7
(R2 = Me, Et, c-Pr and i-Pr and R3 = 3-pyrazol, R6 = Me). Coupling
of 2-chloro-4-aminobenzoic acid with 4-Boc-piperazine afforded
intermediate 10 and replacement of the sulfone group with 10 fol-
lowed by global deprotection resulted in compounds 11a–d.
Early data (Table 1, compounds 11a–d) revealed that the space
at the C-2 position of the imidazo[1,2-a]pyrazine core was very
limited. Only a hydrogen was tolerated (compound 2) and all other
larger groups caused substantial loss of inhibitory activity toward
Brk. In some cases compounds found to be relatively weak in the
enzyme assay were not tested further in the cellular assay
(denoted as not determined or N.D.).
Variation of the C-3 group (Table 2) indicated that the 3-N-H-
pyrazole was an optimal group for BRK inhibitory activity; and
other more basic heterocycles (12d–f) resulted in a loss of Brk
activity. Methylation of the pyrazole (12a) decreased the potency
(Brk IC50 = 256 nM). Analog 12b containing a 3-thiophene group
gave a moderate Brk inhibitory potency (Brk IC50 = 51 nM).
Starting with intermediate 4, Pd-mediated coupling reactions
were employed to install different functional groups at C-6. The
resulting products were then carried through similar steps as
shown in Scheme 1 to give final compounds 13a–d in Table 3.
The hydrogen analog 13a only gave moderate potency to Brk.
The SAR showed that small alkyl groups, such as ethyl (13b) and
cylcopropyl (13c) were tolerated and preferred. These functional
groups not only improved the cell-based activity, but also in-
creased the selectivity window toward both kinases Aurora B and
Lck. Inhibitor 13c demonstrated a favorable cell-based profile with
a p-SAM68 IC50 of 22 nM and more than a 300-fold selectivity win-
dow to both kinases Aurora B and Lck.
Figure 2. Docking of compound 2 into the homology models of BRK using SRC as
the template (2src.pdb).

5872
H. Zeng et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5870–5875
N
NH
N
Boc
H2N
R2
H2N
N
N
a
b, c
N
OH
N
HN
Cl
O
NH
Cl
10
O
N
9
11
Cl
O
Scheme 2. Reagents and conditions: (a) 4-Boc-piperazine, HATU, DIEA, DMF, rt, 96%; (b) sulfone 7, NaH, DMF, rt, 81%; (c) 4 N HCl in dioxane, 40 °C.
Table 2
Table 1
Brk and p-SAM68 inhibition data for 12a–f
Brk and p-SAM68 inhibition data for 11a–d
R3
H
N
N
N
N
N
5
3
1
HN
N
7
R2
NH
11
N
N
N
HN 8
12
Cl
O
NH
Compds
R3
IC50 (nM)
Brk
N
Cl
O
p-SAM68
Compds
R2
IC50 (nM)
Brk
N
N
12a
12b
12c
256
51
>3000
p-SAM68
Me
Et
S
11a
11b
11c
11d
1524
>10,000
>10,000
>10,000
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
c-Pr
i-Pr
S
177
N
Assay conditions listed in Ref. 28.
12d
12e
12f
355
319
N.D.
N.D.
N.D.
N
The clear SAR at C-2, C-3, and C-6 helped us to focus on the
potential at the C-8 amide region as we mentioned early in the
proposed binding mode. Starting with intermediate 6, replacement
of the sulfone with amines followed by hydrolysis, amide forma-
tion and global deprotection provided final compounds 14a–e in
Table 4. Initial results (Table 4) showed that the para-orientation
of the amide to the 8-amino group on the benzene ring was re-
quired to maintain the potency. Free acid 14a and meta amide
14b gave moderate inhibitory activity to Brk (IC50 = 94 and
104 nM, respectively). Saturation of the middle ring (14c) caused
loss of activity, possibly due to an unfavorable angle change of
the amide carbonyl group relative to the piperazine or phenyl ring.
Replacement of the benzene ring of the amide with other
five-member rings, such as thiophene (14d) and thiazole (14e)
maintained the potency (Brk IC50 = 10 and 12 nM, respectively).
However, this class of compounds possessed potent activity
against Aurora B (Aur B IC50 less than 13 nM, the limit of the
assay); consistent with results reported in publications from the
Aurora kinase program.30
As shown in Table 5, replacement of the benzoic amide of 2 with
substituted pyridines (15a–c) gave moderate to good inhibitory
activity in both biochemical and cell-based assays. This is consis-
tent with the proposed binding mode wherein a hydrophobic ring
was anticipated to be preferred in this region. Replacement of the
benzene ring with a non-amide substituted thiazole (15d) and thi-
ophene (15e) showed very good potency to Brk (IC50 = 13 and
10 nM) as well as to Aurora B (IC50 <13 nM).30 Therefore we shifted
our efforts towards replacing the original chlorine on the benzene
ring.
N
N
2643
Assay conditions listed in Ref. 28.
Table 3
Brk and p-SAM68 inhibition and counti-screening data for 13a–d
N
NH
R6
N
N
N
HN
NH
N
13
Cl
IC50 (nM)
Brk
O
Compds
R6
p-SAM68
Aur B
Lck
Me
H
2
13a
13b
7
391
10
157
N.D.
38
754
N.D.
653
420
N.D.
722
Et
13c
13d
6
22
2541
N.D.
1830
N.D.
477
N.D.
Assay conditions listed in Ref. 28.
Data in Table 6 demonstrated that a variety of functional groups
can be tolerated at the ortho position adjacent to the amide car-
bonyl group. Hydrogen (16a) and fluorine (16b) substitutions gave
similar results as chlorine substitution. Substitutions with CF3
(16c), methoxy (16d), cyclopropyl (16e), and isopropyl (16f) all

H. Zeng et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5870–5875
5873
Table 4
Table 6
Brk and p-SAM68 inhibition and counti-screening data for 14a–e
Brk and p-SAM68 inhibition and counti-screening data for 16a–e
N
N
NH
NH
N
N
N
N
N
14
N
HN
HN
NH
R8
N
R8
IC50 (nM)
Brk
16
Compds
X
IC50 (nM)
Brk
O
p-SAM68
Aur B
N.D.
Lck
Compds
X
p-SAM68
Aur B
Lck
OH
N
14a
14b
97
N.D.
N.D.
16a
16b
16c
16d
16e
15
32
7
167
153
35
232
143
84
603
N.D.
722
N.D.
169
Cl
O
H
O
F
104
N.D.
N.D.
N.D.
NH
NH
CF3
OMe
c-Pr
Cl
23
38
22
837
128
N
14c
14d
799
10
N.D.
N.D.
N.D.
<13
N.D.
330
40
O
O
S
Assay conditions listed in Ref. 28.
N
O
S
N
Table 7
14e
12
N.D.
N.D.
N.D.
Brk and p-SAM68 inhibition and counti-screening data for 17a–j
N
O
N
NH
Assay conditions listed in Ref. 28.
N
N
N
HN
Table 5
R
17
Brk and p-SAM68 inhibition and counti-screening data for 15a–e
Cl
O
N
NH
Compds
R
IC50 (nM)
Brk
p-SAM68
>600
Aur B
N.D.
Lck
N
N
N
N
15
17a
17b
8
3
N.D.
N.D.
N
N
HN
R8
NH
O
>3000
N.D.
Compds
R8
IC50 (nM)
Brk
Ac
N
17c
17d
16
68
>3000
132
N.D.
N.D.
N.D.
p-SAM68
Aur B
561
Lck
N
NH
NH
OH
N
N
N
1200
15a
15b
11
75
109
3710
N.D.
N
N
N
N
17e
17f
93
8
608
254
N.D.
43
N.D.
244
N.D.
O
O
2000
N
N
17g
17h
17i
379
72
N.D.
120
137
113
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
10
15c
15d
15e
495
13
N.D.
94
N.D.
<13
<13
N.D.
N.D.
N.D.
N
N
N
O
OH
O
S
N
N
60
N
N
S
10
N.D.
OH
NH
N
17j
34
70
Assay conditions listed in Ref. 28.
Assay conditions listed in Ref. 28.
increased cell-based activity to Brk. However, the selectivity win-
dows to Aurora B and Lck both dropped to within 10-fold.
ment activity in the cell. Attachment of a linear alcohol (17d),
replacement with piperidine (17e) or alcohol (17h) or methoxy
(17i) substituted piperidines afforded moderate potency in the
p-SAM68 assay. A morpholine-based replacement (17f) gave mod-
erate cell-based activity (p-SAM68 IC50 = 254 nM) but poor selectiv-
ity toward Aurora B (IC50 = 43 nM). Interestingly, 17f also showed
Docking experiments suggested the piperazine ring of 2 was ori-
ented toward the solvent front (Fig. 2). Replacement with various
functional groups gave diverse results (Table 7, 17a–j). Methylation
(17a), lactamation (17b), and amidation (17c) of the piperazine
amine maintained the biochemical activity, but lost target-engage-
good PK in rats (PO, 10 mpk, AUC = 12.6 lM h, C6h = 1 lM). Surpris-

5874
H. Zeng et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5870–5875
Table 8
Table 10
Brk and p-SAM68 inhibition and counti-screening data for 18a–f
Data for selected 2-Cl substituted analogs 20a–d
N
N
NH
NH
R6
N
N
N
N
N
N
HN
HN
Cl
R8
R
18
Cl
IC50 (nM)
Brk
O
20
IC50 (nM)
O
Compds
R
Compds
R6
R8
p-SAM68
Aur B
1227
Lck
Brk
p-SAM68
Aur B
Lck
NH
NH
18a
18b
10
16
23
2262
Me
N
N
NH
NH
O
20a
20b
20c
20d
64
N.D.
>20,000
>20,000
>20,000
>20,000
>20,000
>20,000
>20,000
>20,000
N
42
194
4648
63
N.D.
N
N
N
NH
NH
NH
NH
Me
18c
18d
18e
18f
18
19
7
19
62
305
11,000
339
3417
539
N
N
N
N
156
156
N.D.
O
N.D.
47
1501
3208
Assay conditions listed in Ref. 28.
11
177
242
Assay conditions listed in Ref. 28.
Table 11
Data for selected 2-F substituted analogs 21a–d
N
NH
Table 9
Data for combination compounds 19a–g
R6
N
N
NH
N
N
HN
N
R
N
F
N
O
21
HN
Compds
R6
R
IC50 (nM)
R
19
Brk
10
3
p-SAM68
Aur B
Lck
Cl
O
Me
NH
Compds
R
IC50 (nM)
Brk
21a
21b
21c
54
7
>20,000
>20,000
>20,000
>20,000
>20,000
>20,000
N
p-SAM68
Aur B
2252
Lck
NH
O
N
N
NH
NH
19a
19b
2
5
180
597
Me
Me
N
N
70
N.D.
O
10
14
2987
21d
21e
30
9
52
46
>20,000
>20,000
>20,000
>20,000
N
N
N
NH
NH
NH
NH
O
19c
19d
19e
19f
3
13
5
47
33
4588
2501
2739
2517
8484
476
945
NH
NH
N
N
N
N
N
21f
3
7
>20,000
14,114
45
284
Me
NH
NH
21g
21h
10
6
38
<7
>20,000
>20,000
>20,000
>20,000
N
N
10
17
365
19g
124
>600
10,400
Assay conditions listed in Ref. 28.
Assay conditions listed in Ref. 28.
Experience suggested that combination compounds with these
substituted piperazines (from Table 7) and the cyclopropyl group
we discovered at C-6 of the core (from Table 3) might further
improve the potency and selectivity window. As shown in Table
9, compounds 19a–f all demonstrated single-digit nM enzymatic
potency and low double-digit nM cell-based activity. Compound
19a even demonstrated single-digit nM cell-based activity, so far
one of the best on the program. However, the selectivity window
to Lck unexpectedly dropped to 50- to 100-fold. Surprisingly, the
morpholine analog 19g showed only moderate enzymatic potency
ingly, homopiperazine analog 17j showed only moderate potency
to Brk (IC50 = 34 nM) and picked up Lck activity (IC50 = 70 nM).
As shown in Table 8, compounds 18a–f with direct substitution
and bridge-linking on the piperazine ring dramatically improved
cell-based activity to the low double-digit nM range. Geminal di-
methyl substitution on the piperazine ring (18a) not only im-
proved the cell-based potency, but also lifted the selectivity
window to more than 100-fold for both Aurora B and Lck.

H. Zeng et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5870–5875
5875
Table 12
Mouse PK data for compound 21a
4. Vasioukhin, V.; Serfas, M. S.; Siyanova, E. Y.; Polonskaia, M.; Costigan, V. J.; Liu,
B.; Thomason, A.; Tyner, A. L. Oncogene 1995, 10, 349.
5. Derry, J. J.; Prins, G. S.; Ray, V.; Tyner, A. L. Oncogene 2003, 22, 4212.
6. Petro, B. J.; Tan, R. C.; Tyner, A. L.; Lingen, M. W.; Watanabe, K. Oral. Oncol. 2004,
40, 1040.
7. Castro, N. E.; Lange, C. A. Breast Cancer Res. 2010, 12, R60.
8. Ostrander, J. H.; Daniel, A. R.; Lange, C. A. Curr. Opin. Pharmacol. 2010, 10, 662.
9. Lukong, K. E.; Larocque, D.; Tyner, A. L.; Richard, S. J. Biol. Chem. 2005, 280,
38639.
Parameter (units)
IP
PO
PO
PO
Number of animals (N)
Dose (mg/kg)
AUC (0–6 h) (ng h/mL)
15
30
15
10
42
0.1
18
15
30
15
100
3388
8.1
675
1.60
4.0
7205
17.1
7513
17.9
0.5
606
1.4
176
0.42
2.0
AUC (0–6 h) (lM h)
Cmax (ng/ml)
10. Haegebarth, A.; Heap, D.; Bie, W.; Derry, J. J.; Richard, S.; Tyner, A. L. J. Biol.
Chem. 2004, 279, 54398.
Cmax
(l
M)
0.04
6.0
Tmax (h)
11. Coyle, J. H.; Guzik, B. W.; Bor, Y. C.; Jin, L.; Eisner-Smerage, L.; Taylor, S. J.;
Rekosh, D.; Hammarskjold, M. L. Mol. Cell. Biol. 2003, 2, 92.
12. Lukong, K. E.; Huot, M. E.; Richard, S. Cell. Signal. 2009, 21, 1415.
13. Weaver, A. M.; Silva, C. M. Breast Cancer Res. 2007, 9, R79.
14. Liu, L.; Gao, Y.; Qiu, H.; Miller, W. T.; Poli, V.; Reich, N. C. Oncogene 2006, 25,
4904.
(IC50 = 124 nM) and greater than 600 nM cell-based activity,
although the selectivity windows toward Aurora and Lck were
maintained.
15. Chen, H. Y.; Shen, C. H.; Tsai, Y. T.; Lin, F. C.; Huang, Y. P.; Chen, R. H. Mol. Cell.
Biol. 2004, 24, 10558.
Moving the chlorine on the benzene ring to the next position
caused the enzymatic potency to drop (Table 10, compounds
20a–d). Surprisingly, fluorine substitution at this position provided
the final top compounds (Table 11, compounds 21a–h). Almost all
of these compounds showed single-digit to low double-digit nM
cell-based activity. The enhanced activity might be due to the im-
proved hydrogen bond between the amide carbonyl group and
Arg-195 as we proposed in the binding mode in Figure 2. The selec-
tivity windows to Aurora B and Lck were both significantly im-
proved to greater than 300-fold. For compound 21a, it had clean
16. Ikeda, O.; Miyasaka, Y.; Sekine, Y.; Mizushima, A.; Muromoto, R.; Nanbo, A.;
Yoshimura, A.; Matsuda, T. Biochem. Biophys. Res. Commun. 2009, 384, 71.
17. Lukong, K. E.; Richard, S. Cell. Signal. 2008, 20, 432.
18. Shen, C. H.; Chen, H. Y.; Lin, M. S.; Li, F. Y.; Chang, C. C.; Kuo, M. L.; Settleman, J.;
Chen, R. H. Cancer Res. 2008, 68, 7779.
19. Zhang, P.; Ostrander, J. H.; Faivre, E. J.; Olsen, A.; Fitzsimmons, D.; Lange, C. A. J.
Biol. Chem. 2005, 280, 1982.
20. Qiu, H.; Zappacosta, F.; Su, W.; Annan, R. S.; Miller, W. T. Oncogene 2005, 24,
5656.
21. Kamalati, T.; Jolin, H. E.; Mitchell, P. J.; Barker, K. T.; Jackson, L. E.; Dean, C. J.;
Page, M. J.; Gusterson, B. A.; Crompton, M. R. J. Biol. Chem. 1996, 271,
30956.
22. Xiang, K.; Chatti, H.; Qiu, B.; Lakshmi, A.; Krasnitz, J.; Hicks, M.; Yu, W.; Miller,
T.; Muthuswamy, S. K. Proc. Natl. Acad. Sci. 2008, 105, 12463.
23. Kamalati, T.; Jolin, H. E.; Fry, M. J.; Crompton, M. R. Oncogene 2000, 19, 5471.
24. Aubele, M.; Walch, A. K.; Ludyga, N.; Braselmann, H.; Atkinson, M. J.; Luber, B.;
Auer, G.; Tapio, S.; Cooke, T.; Bartlett, J. M. Br. J. Cancer 2008, 99, 1089.
25. Ostrander, J. H.; Daniel, A. R.; Lofgren, K.; Kleer, C. G.; Lange, C. A. Cancer Res.
2007, 67, 4199.
CYP inhibition profile (all >20
hepatocyte-based clearance assays (human = 2.0
and rat = 5.3 L/min/mill cells). The mouse PK results are shown
l
M) and low intrinsic clearance in
lL/min/mill cells
l
in Table 12. The bioavailability of 21a was only about 5.8% and is
likely attributed to poor permeability (CACO-2: 0 nm/s). Com-
pound 21d showed a similar potency and selectivity profile as
21a. Surprisingly, 21d had much better permeability (CACO-2:
26. Harvey, A. J.; Pennington, C. J.; Porter, S.; Burmi, R. S.; Edwards, D. R.; Court, W.;
Eccles, S. A.; Crompton, M. R. Am. J. Pathol. 2009, 175, 1226.
27. Hochhaus, A. Exp. Rev. Anticancer Ther. 2007, 7, 1529.
28. (a) Brk/PTK6 kinase assay was in HTRF format and the CisBio HTRF kinease TK
(62TK0PEJ) kit was used. Assay reagents are solubilized as per kit protocol.
Compounds are diluted and pre-incubated with PTK6 enzyme (Invitrogen) in
assay plate for 30 min at RT while shaking. Substrate is added followed by ATP
to start reaction in 384 well Corning assay plates. Reaction is stopped after one
hour by addition of kit supplied XL665/antibody Cryptate solution. Plate is read
after 1 h using Pherastar Microplate reader. (b) p-Sam68 alpha-screen assay.
293 WT-PTK6 cells are plated overnight in 96 well TC plates containing 10%
DMEM with supplements and G418. Media is flicked off and complete media
containing compound is added and incubated for 3 h at 37 °C. Media is
314 nm/s) and PK profile (rat, po, 10 mpk, AUC0–6h = 31.1
lM h,
C
6h = 3.5 M). These results suggested that compounds 21a and d
l
may be appropriate tool compounds to evaluate the in vivo activity
of Brk inhibitors in xenograft breast tumor models to further vali-
date the potential of Brk/PTK6 as an oncology target.
In summary, we have discovered imidazo[1,2-a]pyrazin-8-
amines as potent Brk kinase inhibitors. The overall SAR suggested
that this type of inhibitors probably bind to the enzyme as shown
in Figure 2. Several inhibitors, with single-digit nanomolar target-
engagement cell-based activity and an appealing overall DMPK
profile, could be used as tool compounds to further validate Brk/
PTK6 as a potential target for cancer treatment.
aspirated and cells are rinsed with PBS. Cells are lysed in 50 lL lysis buffer
(50 mM Tris, pH 7.4; 250 mM NaCl; 5 mM EDTA; 50 mM NaF; 1 mM Na3VO4;
1% Nonidet P40 (NP40); 0.02% NaN3; 2 mM Na3VO4) is added and placed on a
plate shaker for 5 min. Lysates are now ready for the assay. Five micro-liter of
lysate and 5
20) were added per well (Perkin Elmer). Ten microliter of 3X PT66 acceptor
beads were then added to each well (Final = 20 g/ml, Perkin Elmer), and the
plat was put on a shaker for 90 min. Subsequently, 5 L of 6x Biotinylated-
SAM68. (Final = 0.2 g/mL, Santa Cruz) was added and the plate was out on a
shaker for 60 min before addition of 5 l of 6x Donor beads (Final: 20 g/mL).
ll dilution buffer (25 mM HEPES, pH 7.5; 100 mM NaCl; 0.01% T-
l
l
Acknowledgments
l
l
l
We would like to thank Drs. John Piwinski, Neng-Yang Shih, and
William Windsor for support of this work. We also like to thank Dr.
Li Xiao for modeling studies and Michael Starks, Jason Hill, and
Mark Pietrafitta for purification support.
The plate was put on a shaker for 30 min before reading by Envision Plate
Reader (Perkin Elmer).
29. Annis, D. A.; Nickbarg, E.; Yang, X.; Ziebell, M. R.; Whitehurst, C. E. Curr. Opin.
Chem. Biol. 2007, 11, 518.
30. (a) Belanger, D. B.; Curran, P. J.; Hruza, A.; Voigt, J.; Meng, Z.; Mandal, A. K.;
Siddiqui, M. A.; Basso, A. D.; Gray, K. Bioorg. Med. Chem. Lett. 2010, 20, 5170; (b)
Belanger, D. B.; Williams, M. J.; Curran, P. J.; Mandal, A. K.; Meng, Z.; Rainka, M.
P.; Yu, T.; Shih, N.-Y.; Siddiqui, M. A.; Liu, M.; Tevar, S.; Lee, S.; Liang, L.; Gray,
K.; Yaremko, B.; Jones, J.; Smith, E. B.; Prelusky, D. B.; Basso, A. D. Bioorg. Med.
Chem. Lett. 2010, 20, 6739; (c) Meng, Z.; Kulkarni, B. A.; Kerekes, A. D.; Mandal,
A. K.; Esposite, S. J.; Belanger, D. B.; Reddy, P. A.; Basso, A. D.; Tevar, S.; Gray, K.;
Jones, J.; Smith, E. B.; Doll, R. J.; Siddiqui, M. A. Bioorg. Med. Chem. Lett. 2011, 21,
592.
References and notes
1. Mitchell, P. J.; Barker, K. T.; Martindale, J. E.; Kamalati, T.; Lowe, P. N.; Page, M.
J.; Gusterson, B. A.; Crompton, M. R. Oncogene 1994, 9, 2383.
2. Lee, S. T.; Strunk, K. M.; Spritz, R. A. Oncogene 1993, 8, 3403.
3. Siyanova, E. Y.; Serfas, M. S.; Mazo, I. A.; Tyner, A. L. Oncogene 1994, 9, 2053.