Scaffold oriented synthesis. Part 3: Design, synthesis and biological evaluation of novel 5-substituted indazoles as potent and selective kinase inhibitors employing [2+3] cycloadditions

Article abstract of DOI:10.1016/j.bmcl.2011.01.007

We report the synthesis and biological evaluation of 5-substituted indazoles and amino indazoles as kinase inhibitors. The compounds were synthesized in a parallel synthesis fashion from readily available starting materials employing [2+3] cycloaddition r

Full text of DOI:10.1016/j.bmcl.2011.01.007

Bioorganic & Medicinal Chemistry Letters 21 (2011) 1476–1479
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Scaffold oriented synthesis. Part 3: Design, synthesis and biological evaluation
of novel 5-substituted indazoles as potent and selective kinase inhibitors
employing [2+3] cycloadditions
⇑
Irini Akritopoulou-Zanze , Brian D. Wakefield, Alan Gasiecki, Douglas Kalvin , Eric F. Johnson,
Peter Kovar, Stevan W. Djuric
Global Pharmaceutical Research and Development, Abbott Laboratories, 100 Abbott Park Road, Abbott Park, IL 60044, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
We report the synthesis and biological evaluation of 5-substituted indazoles and amino indazoles as
kinase inhibitors. The compounds were synthesized in a parallel synthesis fashion from readily available
starting materials employing [2+3] cycloaddition reactions and were evaluated against a panel of kinase
assays. Potent inhibitors were identified for numerous kinases such as Rock2, Gsk3b, Aurora2 and Jak2.
Ó 2011 Elsevier Ltd. All rights reserved.
Received 24 November 2010
Revised 29 December 2010
Accepted 3 January 2011
Available online 11 January 2011
Keywords:
Kinase scaffolds
Kinase inhibitors
Rock2 inhibitors
Gsk3b inhibitors
Aurora2 inhibitors
Jak2 inhibitors
[2+3] Cycloadditions
Triazoles
Isoxazoles
As part of our efforts to develop patentable, high quality lead
molecules for kinase programs at Abbott,1 we reported the discov-
ery of underutilized unique heterocycles as hinge binders with po-
tent inhibitory activity against kinase targets.2 A second part of our
strategy has been the implementation of underutilized but robust
chemistries to engage known kinase hinge binding elements
(Fig. 1). The rational behind this approach was the ability to rapidly
explore the ATP binding site of numerous kinases, utilizing readily
available starting materials, without compromising the novelty of
the final molecules.
Br
TMS
N
N
N
N
a, b
H
H
1
2
R1
R1
c
e
R2I
c
d
R2
R3
X
X
N
FG
Het
2+3 Cycloadditions
N
N
N
N
N
N
N
N
O
N
N
N
N
N
H
H
N
H
N
N
H
H
X = H, NH2
Proposed new
scaffolds
3
4
5
Scheme 1. Reagents and conditions: (a) Boc2O, DMAP, CH2Cl2, rt, quant; (b) CuI,
Figure 1. Design of novel molecules based on known kinase hinges.
(Ph3)2Cl2Pd, DMF, Et3N, 95 °C, 72%, then KOH, MeOH, 95%; (c) DMSO, NaN3, L-
proline, Na2CO3, sodium ascorbate, CuSO4Á5H2O, 65 °C, 6–39%; (d) R2Cl, NaN3,
CuSO4Á5H2O, Cu(0), t-BuOH, H2O, CEM microwave, 100 W, 125 °C, 10 min, 3–26%;
(e) R3CHO, NH2OHÁHCl, 6 N NaOH, chloramine-TÁ3H2O, CuSO4, Cu wire, t-BuOH,
H2O, 50 °C, 4–5%.
⇑
Corresponding author. Tel.: +1 847 9375006; fax: +1 847 9350310.
E-mail address: irini.zanze@abbott.com (I. Akritopoulou-Zanze).
Present address: 1201 Lockwood Drive, Buffalo Grove, IL 60089, USA.
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.01.007

I. Akritopoulou-Zanze et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1476–1479
1477
oxides.6 Similarly, 5-substituted-3-amino indazoles could be
prepared by Sonogashira coupling of trimethylsilylacetylene with
commercially available 2-fluoro-5-iodobenzonitrile 6 (Scheme 2).
Deprotected intermediate 7 was subjected to cycloaddition reac-
tions, followed by treatment with hydrazine to afford the final
products 9.
Following these procedures numerous analogs were prepared in
a short period of time in a parallel synthesis fashion. The com-
pounds were tested in a panel of kinase assays7 covering all of
the branches of the kinome tree.8
We were pleased to find that the majority of the compounds
showed inhibitory activity in at least one of the kinase assays. In
the case of triazoloindazoles 3 (Table 1) most of the compounds
were potent inhibitors of Rock2 but they also inhibited Gsk3b,
Aurora2 and Jak2. Depending on the substitution pattern the activ-
ity against the various kinases could be modulated. For example,
although compounds 3g and 3h had similar potencies against
Rock2, 3h was more potent than 3g against Aurora2 and Jak2. Fur-
thermore, a cyclohexylmethyl substituent in compound 3f greatly
improved the Jak2 activity, while tethered phenyl groups in com-
pounds 3g–3i improved Gsk3b activity.
In the case of triazoloindazoles 4 (Table 2) we observed a differ-
ent potency profile. A direct comparison of 3g to 4b showed that in
contrast to 3g, 4b was more potent against Gsk3b and Aurora2
than Rock2 while the 3-amino indazole 4c lost some of its Aurora2
and Gsk3b potency and gained Rock2 and Jak2 activity. Simple
substitutions on the benzyl ring also have a dramatic effect on
the potency and selectivity of the analogs. For example, although
meta substituted benzyl ring derivatives improved potency against
Aurora2, Gsk3b and Rock2 the methyl group appears to be the best
substituent for Aurora2 activity, the fluoro for Gsk3b and the
chloro for Rock2. For the dichloro analogs 4l, 4m and 4n the
Br
CN
F
CN
F
c
TMS
a, b
6
7
R3
R3
N
NH2
N
N
CN
F
d
O
O
N
H
8
9
Scheme 2. Reagents and conditions: (a) CuI, (Ph3)2Cl2Pd, DMF, Et3N, 95 °C, 24 h,
93%; (b) TBAF, THF, 55%; (c) R3CHO, NH2OHÁHCl, 6 N NaOH, chloramine-TÁ3H2O,
CuSO4, Cu wire, t-BuOH, H2O, 50 °C, 33–69%; (d) NH2NH2, EtOH, 70 °C, 24 h, 53–78%.
In this report, we describe our efforts in applying the above
strategy to indazoles and aminoindazoles, which are well known
kinase hinge binding motifs.3 We were pleased to discover that
application of [2+3] cycloaddition reactions to suitably substituted
indazoles rendered the compounds novel.
Starting with 5-bromo indazole 1 (Scheme 1), 5-(4-substituted-
1,2,3-triazol-1-yl)-indazoles 3 could be obtained directly by cyclo-
addition reactions of in situ generated 5-azido-indazole with
commercially available acetylenes following literature procedures.4
Alternatively, Sonogashira coupling of Boc protected 1 with tri-
methylsilylacetylene and subsequent base induced desylilation
and Boc deprotection provided 5-ethynyl-indazole 2 which upon
cycloaddition reactions with azides4,5 yielded 5-(1-substituted-
1,2,3-triazol-4-yl)-indazoles 4. Intermediate 2 could also be utilized
for the synthesis of isoxazoles 5 via the in situ formation of nitrile
Table 1
Kinase inhibitory activity of 5-(4-substituted-1,2,3-triazol-1-yl)-indazoles 3a
R1
N
N
N
N
N
H
Compound R1
Aurora2 Ki (lM) Egfr Ki (
l
M) Gsk3b Ki (
l
M) Jak2 Ki (
l
M) Kdr Ki (
l
M) Pak4 Ki (
l
M) Pim1 Ki (
l
M) Rock2 Ki (lM)
3a
3b
>4.900
>4.900
>1.800
>1.800
2.026
0.756
>8.880
>8.880
>3.750
>3.750
8.359
0.108
0.635
>5.450
>1.450
>8.570
O
3c
>4.900
>4.900
>1.800
>1.800
5.216
3.466
1.453
1.453
>8.880
>8.880
>3.750
>3.750
>8.570
>8.570
0.230
0.356
HO
3d
3e
>4.900
>4.900
>1.800
>1.800
1.298
0.891
0.791
0.281
>8.880
>8.880
>3.750
>3.750
>8.570
>8.570
0.044
0.100
3f
3g
3h
3i
1.577
0.524
1.930
>1.800
>1.800
>1.800
0.542
0.446
0.528
1.282b
0.898
0.644
>8.880
>8.880
>8.880
>3.750
>3.750
>3.750
>8.570
>8.570
>8.570
0.018
0.019
0.062
a
Ki values are based on six point curves unless otherwise noted.
Ki value is based on an eleven point curve done in triplicate.
b

1478
I. Akritopoulou-Zanze et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1476–1479
Table 2
Kinase inhibitory activity of 5-(1-substituted-1,2,3-triazol-4-yl)-indazoles 4a
R2
N
X
N
N
N
N
H
Compound R2
X
Aurora2 Ki
M)
Egfr Ki
M)
Gsk3b Ki
M)
Jak2 Ki
M)
Kdr Ki
Pak4 Ki
M)
Pim1 Ki
Rock2 Ki
M)
(
l
(
l
(
l
(
l
(lM)
(
l
(l
M)
(l
4a
4b
H
H
H
0.493
ND
0.448
1.453b
3.554
69.911
2.219
0.209
0.073
>1.800
0.039
0.284b
>8.880c
3.978
>5.450
0.923
0.097
4c
NH2 0.252
>1.800
ND
0.265
0.045
0.091b
>3.750
ND
>8.570
4.153
0.064
0.097
Cl
4d
H
0.553
>1.453b
ND
Cl
4e
4f
H
H
0.160
0.186
ND
ND
0.023
0.197
1.450
ND
13.142
ND
ND
ND
0.039
0.068
2.710
37.846
Cl
F
F
4g
H
0.016
ND
0.091
ND
>0.915b
13.090
10.153
0.285
4h
4j
H
H
H
0.042
ND
ND
ND
ND
0.026
0.070
0.070
ND
ND
ND
ND
ND
ND
ND
12.000
165.230
ND
0.180
0.285
0.375
ND
F
4k
0.015
>0.915b
Cl
Cl
Cl
4l
H
H
0.066
ND
ND
ND
0.048
0.016
ND
ND
ND
67.636
ND
20.307
36.000
0.015
0.018
4m
126.500
Cl
Cl
4n
H
0.240
ND
0.028
ND
>0.915b
ND
41.538
0.041
Cl
a
Ki values are based on six point curves unless otherwise noted.
Ki value is based on an eleven point curve done in triplicate.
b
position of the second chloro group modulates the Gsk3b and
Rock2 activity with 4l being more potent in Rock2, 4m being equi-
potent against Gsk3b and Rock2 and 4n being more potent against
Gsk3b.
When the heterocyclic ring at the 5-position of the indazole ring
is an isoxazole (Table 3) the compound 5a maintained its Rock2
and Gsk3b potency but it was also more potent against Aurora2
and Jak2 as compared to 3g. Addition of an amino group at the
3-position of the indazole improved significantly its Aurora2 and
Jak2 activity. Aliphatic substituents on the isoxazole ring main-
tained moderate Rock2 activities.
In conclusion, we have discovered multiple series of easily
accessible 5-substituted indazoles as potent kinase inhibitors.
Although the compounds showed high potency against multiple
kinases we have already observed selectivity trends based on sub-
stitution patterns of very close analogs. In addition to the panel of
kinase assays reported here, numerous compounds were also
tested in a large panel of more than 100 kinases. Many interesting

I. Akritopoulou-Zanze et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1476–1479
1479
Table 3
Kinase inhibitory activity of 5-(3-substituted-isoxazol-5-yl)-indazoles 5a
R3
N
X
O
N
N
H
Compound R3
X
Aurora2 Ki (lM) Egfr Ki (
l
M) Gsk3b Ki (
l
M) Jak2 Ki (
l
M) Kdr Ki (
l
M) Pak4 Ki (
l
M) Pim1 Ki (
l
M) Rock2 Ki (lM)
5a
H
0.323
>1.800
0.572
0.351
>8.880
>3.750
>8.570
0.028
5b
5c
NH2
H
0.026
2.366
ND
0.450
2.530
0.016b
0.482b
3.291
3.588
5.155
0.016
0.113
>1.800
>8.880
>3.750
>8.570
O
N
5d
5e
H
H
>4.900
>4.900
ND
4.272
3.810
ND
>8.880
>8.880
>3.750
>3.750
3.685
0.402
0.315
N
H
>1.800
>1.453
>8.570
a
Ki values are based on six point curves unless otherwise noted.
Ki value is based on an eleven point curve done in triplicate.
b
500
concentration, then
controls were prequenched with EDTA. ATP, enzyme and biotinylated peptide
l
M. The plate was diluted 17-fold in assay buffer to reduce the DMSO
activity profiles emerged out of these activities and the results will
be reported in due course.
8 ll
was transferred to black HTRF assay plates. Low
were added in sequence for a total volume of 24
at room temperature.
The receptor tyrosine kinase (RTK) reaction buffer consisted of 50 mM HEPES
ll and the reaction run for 1 h
Acknowledgments
pH 7.5, 1 mM DTT, 10 mM MgCl2, 2 mM MnCl2, 0.01% BSA, and 100
Na3VO4.
lM
The authors thank the Abbott High Throughput Synthesis, High
Throughput Purification and Structural Chemistry groups for their
support with library synthesis, purification and analytical services,
respectively.
The serine/threonine (SK) kinase reaction buffer consisted of 20 mM HEPES
pH 7.5, 1 mM DTT, 10 mM MgCl2, 100 M Na3VO4, and 0.0075% Triton X 100.
Jak2 was run at 5 M ATP, Gsk3b, Pak4, Aurora2, Rock2, Kdr and Pim1 were
run at 10 M ATP and Egfr at 50 M ATP.
Peptide substrates used were: pGS (Gsk3b), L15 (Pak4), kemptide
PDKtide (Jak2), 0.5 FGFR
l
l
l
l
2 lM
References and notes
(Aurora2), longS (Rock2), AL1 (Pim1), 0.5
(Egfr, Kdr).
lM
lM
The final compound concentration run from 10
DMSO concentration was 2%.
lM to 3.2 nM and the final
1. Akritopoulou-Zanze, I.; Hajduk, P. J. Drug Discovery Today 2009, 14, 291.
2. (a) Akritopoulou-Zanze, I.; Darczak, D.; Sarris, K.; Phelan, K. M.; Huth, J. R.; Song,
D.; Johnson, E. F.; Jia, Y.; Djuric, S. W. Bioorg. Med. Chem. Lett. 2006, 16, 96; (b)
Gracias, V.; Ji, Z.; Akritopoulou-Zanze, I.; Abad-Zapatero, C.; Huth, J. R.; Song, D.;
Hajduk, P. J.; Johnson, E. F.; Glaser, K. B.; Marcotte, P. A.; Pease, L.; Soni, N. B.;
Stewart, K. D.; Davidsen, S. K.; Michaelides, M. R.; Djuric, S. W. Bioorg. Med.
Chem. Lett. 2008, 18, 2691.
The RTK enzyme reactions were stopped with an equal volume of Revelation
Quench buffer containing 0.04 M Hepes 7.4, 0.48 M KF, 0.01% Tween20, 0.1%
BSA, 0.06 M EDTA, Cisbio anti-phospho serine/threonine antibody labeled
with Eu 3+-Cryptate and Streptavidin–Allophycocyanin. After equilibration for
1 h the plates were read on an Envision plate reader optimized for HTRF:
320 nm excitation and 665/615 nm emission.
3. Ghose, A. K.; Herbertz, T.; Pippin, D. A.; Salvino, J. M.; Mallamo, J. P. J. Med. Chem.
2008, 51, 5149.
The S/T enzyme reactions were stopped with 30
lM final EDTA in PBS
transferred to PerkinElmer Streptavidin labeled Flashplate, washed three
a
4. Feldman, A. K.; Colasson, B.; Fokin, V. V. Org. Lett. 2004, 6, 3897.
5. Appukkuttan, P.; Dehaen, W.; Fokin, V. V.; Van der Eycken, E. Org. Lett. 2004, 6,
4223.
6. Hansen, T. V.; Wu, P.; Fokin, V. J. Org. Chem. 2005, 70, 7761.
7. Using a Tecan Gemini robot, compounds were diluted in 100% DMSO into six
points and five-fold dilutions into a 384 well polypropylene plate, starting at
times with PBS/0.05% Tween20 and counted on a Packard Topcount.
Results were analyzed utilizing an in house MS Excel macro to calculate %
inhibition, IC50 and Ki values.
8. Manning, G.; Whyte, D. B.; Martinez, R.; Hunter, T.; Sudarsanam, S. Science 2002,
298, 1912.