Structure-driven HtL: Design and synthesis of novel aminoindazole inhibitors of c-Jun N-terminal kinase activity

Article abstract of DOI:10.1016/j.bmcl.2005.05.008

The design and synthesis of a new series of c-Jun N-terminal kinase inhibitors are reported. The novel series of substituted amino indazoles were designed based on a combination of hits from high-throughput screening and X-ray crystal structure informatio

Full text of DOI:10.1016/j.bmcl.2005.05.008

Bioorganic & Medicinal Chemistry Letters 15 (2005) 3459–3462
Structure-driven HtL: Design and synthesis of novel
aminoindazole inhibitors of c-Jun N-terminal kinase activity
Michael J. Stocks,a,* Simon Barber,a Rhonan Ford,a Frederic Leroux,a
Steve St-Gallay,b Simon Teaguea and Yafeng Xuec
aDepartment of Medicinal Chemistry, AstraZeneca R&D Charnwood, Bakewell Road, Loughborough LE11 5RH, UK
bDepartment of Physical and Metabolic Science, AstraZeneca R&D Charnwood, Bakewell Road, Loughborough LE11 5RH, UK
cStructural Chemistry Laboratory, AstraZeneca R&D Mo¨lndal, S-431 83, Mo¨lndal, Sweden
Received 4 January 2005; revised 25 April 2005; accepted 2 May 2005
Available online 9 June 2005
Abstract—The design and synthesis of a new series of c-Jun N-terminal kinase inhibitors are reported. The novel series of substituted
amino indazoles were designed based on a combination of hits from high-throughput screening and X-ray crystal structure infor-
mation of the compounds crystallised into the JNK-1 ATP binding site.
ꢀ 2005 Elsevier Ltd. All rights reserved.
Cell surface receptors play a fundamental role by which
cells receive information from extracellular signals.
These receptors transmit information into the cell where
it is propagated by activation or suppression of bio-
chemical pathways. Protein kinases and phosphatases
are important components of such intracellular signal-
ling pathways as they allow the information to be cas-
caded to numerous effector molecules as well as giving
amplification of the extracellular signal. The mitogen-
activated protein (MAP) kinase signalling pathways
are activated by engagement of a number of cell surface
receptors. One of these pathways, the JNK pathway
(named after one of the key enzymes in the pathway,
c-Jun N-terminal kinase or JNK), is activated specifi-
cally by stress or pro-inflammatory cytokines.1 Mem-
bers of the JNK family (JNK-1, 2 and 3) act as an
integration point for multiple intracellular biochemical
signals governing a wide variety of cellular processes
such as proliferation,2 apoptosis,3 migration4 and
transcriptional regulation.5 Activation of the JNK
pathway has been documented in a number of disease
settings, suggesting that specific inhibition of JNK
activity could provide an effective therapy for a variety
of conditions.6,7
Compounds 18 and 2 (Fig. 1) were identified from high-
throughput screening and possessed promising potency9
as well as attractive structural features amenable to opti-
misation by rapid parallel synthesis.
A series of synthesis/screening cycles based around com-
pound 2 resulted in new lead series of compounds 3–10
(Fig. 2), which demonstrated good potency against
JNK-1 (Table 1).
An X-ray crystal structure of 4 bound into the ATP
binding site of JNK-1 was obtained.10,11 In the crystal
structure (Fig. 3), the phenoxy group binds in the pocket
occupied by the ribose of the competitive binder ATP
(n.b. AMP-PNA, adenylyl imidodiphosphate, was used
in the crystallisation experiment, but only the ADP por-
tion was well defined in the electron density map). The
amide present in 4 is acting as a H-bond acceptor mak-
ing a key interaction with the hinge-region through
Keywords: Aminoindazole; c-Jun N-terminal kinase; Inhibitors;
Bioavailable.
*
Corresponding author. Fax: +44 1509 645571; e-mail: mike.stocks@
astrazeneca.com
Figure 1. Initial hits from high-throughput screening.
0960-894X/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.05.008

3460
M. J. Stocks et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3459–3462
the glutamic acid residue in the hinge-region leading to
more highly bound inhibitors.
A series of aminoindazoles was prepared and their JNK-
1 activity is shown in Table 2.
An analysis of the X-ray crystal structure of 23 in JNK-1
(Fig. 7) shows the formation of an internal hydrogen
bond between the benzylic nitrogen and the phenyl ether
Figure 2. Result of synthesis/screening cycles—identification of a new
lead series 3–10, the aryloxy pyridines.
˚
of the aminoindazole (2.9 A) as well as giving evidence
of strong H-bond donor/acceptor interactions with
˚
˚
˚
Met111 (2.8 A). Interestingly, unlike many other kinase
Glu109 (2.9 A) and Met111 (3.1 A). Further analysis
of the structural requirements for binding showed that
changing the template from indazole to azaindazole
(11 vs 22; 14 vs 23) resulted in a ꢀ10-fold increase in po-
tency against the isolated enzyme. This increase might,
in part, be accounted for by considering the X-ray crys-
tal structure, which shows that the 5-position nitrogen
of the azaindazole is hydrogen bonding to a water
molecule. This water is not observed in all JNK-1 crystal
structures, but is positioned close to the well-conserved
Lys55 along with a small cluster of other water mole-
cules, suggesting that the local environment in the plane
of the aromatic ring is hydrophilic. The azaindazole
template can make a good hydrogen bond to the water
molecule, an interaction which is absent in the indazole
template. A second rational for the observed increase in
potency might be due to differences in the electronics of
inhibitors mimicking the adenine of ATP, compound 4
does not possess a H-bond donor to interact with the
hinge-region through Glu109 (Fig. 4).12
Unfortunately, structurally similar compounds of the
class 3–10 were known and exemplified in the patent lit-
erature14 against other biological targets, so a new
chemical lead series was required. The X-ray crystal
structure of 1 crystallised into JNK-1 was solved and
in this case compound 1 makes a donor/acceptor inter-
action with both the methionine and glutamic acid in
the ATP binding site. An overlay of 1 with the X-ray
crystal structure of 4 (Fig. 5) suggested that amino
indazoles of the type shown in Figure 6 would represent
a series of compounds to explore, as the H-bond donat-
ing group within the amino indazole could interact with
Table 1. Inhibition results for compounds 3–10 against JNK-1
Compound
X
A
B
D
E
Activitya pIC50 ( 0.2)
3
4
C
H
H
H
H
Me
H
H
H
H
H
H
H
H
Me
H
H
H
2-Cl
2-Cl
H
5.4
5.5
5.4
i.ab
5.2
7.0
5.8
5.4
N
N
N
N
N
N
N
H
5
H
6
H
H
7
H
H
8
H
2,5-diOMe
2-Cl
2-Cl
9
10
2-F
4-OMe
a Activity reported against the isolated JNK-1 enzyme.
b <40% inhibition at 100 lM.
Figure 4. Schematic representation of overlay of 4 with ADP showing
key H-bond interactions.
Figure 3. X-ray crystal structure of 4 in the JNK-1 active site overlaid
with ADP.13

M. J. Stocks et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3459–3462
3461
Figure 5. Overlay of X-ray crystal structures of 1 and 4 in the JNK-1
ATP binding site.15
Figure 7. Cutaway of X-ray crystal structure of 23 in the JNK-1 ATP
binding site showing the key interactions with Glu109 and Met111.16
Compound 23 was found to be metabolically unstable in
vitro in both rat and human (rat heps Clint = 43 lL/min/
1 · 106 cells; human mics. Clint = 74 lL/min/mg). How-
ever, compound 22 displayed a good level of in vitro met-
abolic stability (rat heps Clint = 10 lL/min/1 · 106 cells;
human mics. Clint = 12 lL/min/mg) and was chosen as
the candidate for in vivo PK dosing. The compound dem-
onstrated good in vivo characteristics with good bioavail-
ability (Cl = 34 mL/min/kg; t1/2 = 0.7 h; F = 50%).
The synthesis of the aminoindazoles was achieved by
known chemistry17 from commercially available 2-phen-
oxy-6-fluorobenzonitrile (Scheme 1).
The synthesis of compounds 22 and 23 required the
preparation of 4-methoxy-3-cyano-2-chloropyridine18
(Scheme 2).
Figure 6. Rational design of the aminoindazole series of compounds.
In summary, a new series of aminoindazole inhibitors of
c-Jun N-terminal kinase activity has been developed
from a structure-based drug design programme initiated
from a high-throughput screening/synthesis campaign.
The initial hit series proved to be reasonably active
inhibitors, and X-ray structural information obtained
the pyrazole hinge region between the azaindazole and
the indazole templates. This is enforced by calculations
showing the N–H at the 1-position of the azaindazole
ring system is a better H-bond donor to Glu109 than
the corresponding indazole.
Table 2. Inhibition results for compounds 11–23
Compound
X
A
B
D
Activitya pIC50 ( 0.2)
11
12
13
14
15
16
17
18
19
20
21
22
23
C
C
C
C
C
C
C
C
C
C
C
N
N
H
H
4-NHAc
2,5-diCl
3-Cl
6.1
6.1
5.9
6.9
6.3
6.7
6.1
6.4
6.1
5.8
5.3
6.8
7.8
H
H
H
H
H
H
4-F
2,5-diOMe
4-NHAc
2,5-diOMe
4-NHAc
2,5-diOMe
2,5-diOMe
2-Cl
H
H
4-F
4-SO2Me
4-SO2Me
H
H
NO2
NO2
NH2
H
H
H
H
H
2-Cl
4-NHAc
2,5-diOMe
a Activity reported against the isolated JNK-1 enzyme.

3462
M. J. Stocks et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3459–3462
26, 957; Vincenti, M. P.; Brinckerhoff, C. E. J. Clin. Invest.
2001, 108, 181.
7. Selected chemical patents relating to JNK inhibitors:
Bhagwat, S. S.; Satoh, Y.; Sakata, S. T.; Buhr, C. A.;
Albers, R.; Sapienza, J.; Plantevin, V.; Chao, Q.; Saha-
srabudhe, K.; Ferri, R. US2004127536, 2004; Chem.
Abstr. 2004, 141, 89085; Satoh, Y.; Bhagwat, S. S. U.S.
Pat. Appl. Publ. (2004), US2004106634, 2004; Chem.
Abstr. 2004, 141, 23544; Cao, J.; Green, J.; Moon, Y.-C.;
Wang, J.; Ledeboer, M.; Harrington, E.; Gao, H.
WO2002083667, 2002; Chem. Abstr. 2002, 137, 325327;
Bhagwat, S. S.; Satoh, Y.; Sakata, Steven T.
WO2002010137, 2002; Chem. Abstr. 2002 136, 151163.
8. Grayshan, R.; French, A.; McKinnon, A.-K.; Hamad; De
Boos, G. A. WO198910924, 1989; Chem. Abstr. 1990, 112,
216936.
Scheme 1. Reagents: (a) hydrazine hydrate, ethanol, (b) aldehyde
R*CHO, sodium triacetoxyborohydride, methanol, acetic acid.
9. As part of a JNK-1 inhibitor programme, the AstraZeneca
compound collection was screened in scintillation prox-
imity assay-based HTS that employed human recombi-
nant JNK-1a1 and a biotin-tagged ATF-2. The activities
and potencies of hits were subsequently confirmed in a
conventional filter wash assay. For specific conditions see:
King, S.; Teague, S.; Xue, Y.; Swahn, B.-M.
WO2003051277, 2003; Chem. Abstr. 2003, 139, 53019.
10. The expression and crystallisation of JNK-1 will be
published elsewhere. Co-crystallisation was performed
using hanging-drop method. Complete data were collected
using in-house X-ray source (CuKa radiation; wavelength
Scheme 2. Reagents: (a) dimethylformamide dimethyl acetal, metha-
nol, (b) dry HCl, methanol, (c) phenol, cesium carbonate, (d)
hydrazine hydrate, ethanol, (e) aldehyde R*CHO, sodium triacetoxy-
borohydride, methanol, acetic acid.
˚
˚
k 1.54 A) to a resolution of 2.6 A on a Mar-Research 345-
mm image-plate detector system. X-ray radiation was
generated by a Rigaku RU300HB rotating anode operated
at 50 kV and 100 mA. The crystal belongs to orthorhom-
bic space group P212121, with cell parameters of
of these compounds in the active site of JNK-1 eventu-
ally afforded the aminoindazole series. These were
shown to be potent inhibitors of JNK-1, demonstrating
good in vitro pharmacokinetics and good bioavailability
when dosed in rats.
˚
˚
˚
a = 49.30 A, b = 79.75 A and c = 107.41 A.
11. Xie, X.; Gu, Y.; Fox, T.; Coll, J. T.; Fleming, M. A.;
Markland, W.; Caron, P. R.; Wilson, K. P.; Su, M. S.
Structure 1998, 6, 983.
12. Cherry, M.; Williams, D. H. Curr. Med. Chem. 2004, 11,
663; Williams, D. H.; Mitchell, T. Curr. Opin. Pharmacol.
2002, 2, 567.
Acknowledgments
13. Figures 3 and 5 were generated using QUANTA and
Figure 7 used SYBYL [SYBYLꢁ 6.8, Tripos Inc., 1699
South Hanley Rd., St. Louis, MO 63144, USA].
14. Marfat, A.; Chambers, R. J.; Watson, J. W.; Cheng, J. B.;
Duplantier, A. J.; Kleinman, E. F. WO9845268, 1998;
Chem. Abstr. 1998, 129, 682365.
15. All structures were overlaid by the a carbon atoms of
the protein backbone in corresponding residues within
QUANTA [Accelrys Inc., Quanta, release 98, San
Diego: Accelrys Inc., 1998]. Hydrogen atoms were
added using CHARMm [Brooks, B. R.; Bruccoleri, R.
E.; Olafson, B. D.; States, D. K.; Swaminathan, S.;
Karplus, M. Comput. Chem. 1983, 4, 187] v26.2,
parameters version 22.0.
The authors thank John Unitt and Paul Hemsley,
Department of Discovery Bioscience, AstraZeneca
R&D Charnwood, for biological studies. Stefan Gesch-
˚
windner and Pernilla Lo¨fas, Structural Chemistry Labo-
ratory, AstraZeneca R&D Mo¨lndal, for support in
protein production and initial crystallisation studies.
References and notes
1. Waetzig, G. H.; Schreiber, S. Aliment. Pharmacol. Ther.
2003, 18, 17; Johnson, G. L.; Lapadat, R. Science 2002,
298, 1911.
2. Du, L.; Lyle, C. S.; Obey, T. B.; Gaarde, W. A.; Muir, J.
A.; Bennett, B. L.; Chambers, T. C. J. Biol. Chem. 2004,
279, 11957; Cheng, Y.; Zhizhin, I.; Perlman, R. L.;
Mangoura, D. J. Biol. Chem. 2000, 275, 23326.
16. The Connolly solvent accessible surface [Connolly, M. L.
Science 1983, 221, 709] was coloured by lipophilicity
[Ghose, A. K.; Crippen, G. M. J. Comput. Chem. 1986, 7,
565], with brown representing more lipophilic regions and
blue representing more hydrophilic regions.
17. Kaltenbach, R. F.; Patel, M.; Waltermire, R. E.; Harris,
G. D.; Stone, B. R. P.; Klabe, R. M.; Garber, S.; Bacheler,
L. T.; Cordova, B. C.; Logue, K.; Wright, M. R.;
Erickson-Viitanen, S.; Trainor, G. L. Bioorg. Med. Chem.
Lett. 2003, 13, 605.
18. Bierer, D.; McClure, A.; Fu, W.; Achebe, F.; Ladouceur,
G. H.; Burke, M. J.; Bi, C.; Hart, B.; Dumas, J.; Sibley,
R.; Scott, W. J.; Johnson, J.; Asgari, D. WO2003027085,
2003; Chem. Abstr. 2003, 138, 287663. Mittelbach, M.;
Kastner, G.; Junek, H. Arch. Pharmacol. 1985, 318, 481.
3. Ammendrup, A.; Maillard, A.; Nielsen, K.; Andersen, N.
A.; Serup, P.; Madsen, O. D.; Mandrup-Poulsen, T.;
Bonny, C. Diabetes 2000, 49, 1468; Vivo, C.; Liu, W.;
Broaddus, V. C. J. Biol. Chem. 2003, 278, 25461.
4. Javelaud, D.; Laboureau, J.; Gabison, E.; Verrecchia, F.;
Mauviel, A. J. Biol. Chem. 2003, 278, 24624.
5. Higuchi, H.; Grambihler, A.; Canbay, A.; Bronk, S. F.;
Gores, G. J. J. Biol. Chem. 2004, 279, 51.
6. Orchard, S. Curr. Opin. Drug Discovery Dev. 2002, 5, 713–
717; Harper, S. J.; LoGrasso, P. Drugs of the Future 2001,