Design of two new chemotypes for inhibiting the Janus kinase 2 by scaffold morphing

Article abstract of DOI:10.1016/j.bmcl.2010.01.151

JAK2 is a target of high interest in chronic myeloproliferative disorders drug research. Starting from a screening hit, two new JAK2 inhibitor chemotypes were designed by scaffold morphing. The prototype compounds of these new series showed nanomolar inhi

Full text of DOI:10.1016/j.bmcl.2010.01.151

Bioorganic & Medicinal Chemistry Letters 20 (2010) 1858–1860
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Design of two new chemotypes for inhibiting the Janus kinase 2
by scaffold morphing
*
Pascal Furet , Marc Gerspacher, Carole Pissot-Soldermann
Novartis Institutes for BioMedical Research, WKL-136.P.12, CH-4002 Basel, Switzerland
a r t i c l e i n f o
a b s t r a c t
Article history:
JAK2 is a target of high interest in chronic myeloproliferative disorders drug research. Starting from a
screening hit, two new JAK2 inhibitor chemotypes were designed by scaffold morphing. The prototype
compounds of these new series showed nanomolar inhibition of the kinase.
Ó 2010 Elsevier Ltd. All rights reserved.
Received 18 December 2009
Revised 26 January 2010
Accepted 29 January 2010
Available online 4 February 2010
Keywords:
Scaffold morphing
Kinase inhibitor design
Inhibition of the Janus kinase 2 (JAK2) is considered as a prom-
ising therapeutic approach in the treatment of chronic myeloprolif-
erative neoplasms.1 Consequently, medicinal chemistry efforts
aiming at the discovery of chemical inhibitors of this kinase are
currently pursued.2–4 We initiated our own efforts in this direction
by screening an internal collection of kinase inhibitors using a bio-
chemical JAK2 enzymatic assay in a search for chemical starting
points. The screening resulted in the identification of compound
1, belonging to a series of pyrrolopyrimidines synthesized in the
course of a focal adhesion kinase (FAK) inhibitor program,5 as a po-
tent inhibitor of JAK2. Here, we report the design of two new JAK2
inhibitor chemotypes by scaffold morphing starting from com-
pound 1 (see Fig. 1). The synthesis, structure–activity relationships,
selectivity profiles, and pharmacokinetic properties of these new
series of JAK2 inhibitors are reported elsewhere.6,7
Designing molecular mimics of known biologically active mole-
cules, a process currently termed scaffold morphing, can be an
effective way to generate new chemotypes in medicinal chemistry.
In this process, the chemical functions conferring the desired bio-
logical activity are introduced in new, different, molecular frame-
works that can present advantages over the reference molecule
in terms of synthetic access, profile of selectivity or ADME proper-
ties. However, this is a very difficult exercise requiring a deep
understanding of the structural determinants of binding affinity
for the protein target coupled with innovative thinking in the de-
sign since no standard procedure or technology exists to accom-
plish this task with guaranteed success.
To establish a solid basis to the scaffold morphing of 1, the com-
pound was docked in the ATP pocket of an available crystal struc-
ture of the kinase domain of JAK2.8 The resulting docking model is
shown in Figure 2.9 Inspection of the possible orientations of 1 in
the ATP pocket of JAK2 led to the conclusion that the compound
most probably binds to this kinase with the same mode of interac-
tion as that observed in its co-crystal structure with FAK.5 Thus, in
the JAK2 model, its pyrimidine N1 atom and 2-amino group form
bidentate hydrogen bonds with the backbone of the hinge residue
L932.10 These position the trimethoxyphenyl moiety of the inhibi-
3
N
1 N
7
N
7
N
O
HN
HN
O
2
2
O
N
O
O
O
O
2
1
4
N
3
2
N
O
O
N
HN
8
O
O
O
O
O
S
O
O
S O
NH
O
NH
4
3
* Corresponding author. Tel.: +41 61 696 79 90; fax: +41 61 696 48 70.
E-mail address: pascal.furet@novartis.com (P. Furet).
Figure 1. Chemical structures of the screening hit and designed compounds.
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.01.151

P. Furet et al. / Bioorg. Med. Chem. Lett. 20 (2010) 1858–1860
1859
Figure 2. Model of compound 1 docked in the ATP pocket of JAK2. Key hydrogen
Figure 4. Model of prototype compound 2. (yellow) docked in the ATP pocket of
bonds appear as dashed lines.
JAK2. The mimicked compound
represented as dashed lines.
1 appears in green. Key hydrogen bonds are
tor in the hydrophobic channel formed by residues G935 and L855
at the entrance of the cavity while placing its pyrrole ring in the
hydrophobic environment provided by V911, G993, and the gate
keeper residue M92910 at the bottom of the pocket. This orienta-
tion allows, in addition, a favorable hydrophobic interaction to
occur between its pyridine ring and residue V863 of the kinase
P-loop.
Examination of the above docking model, inspired the original
idea to invert the six-membered and five-membered rings in the
pyrrolopyrimidine hinge binding moiety of 1. This implied that
the hydrogen bond acceptor functionality interacting with the
backbone N–H group of residue L932 should be introduced in the
five-membered ring while the six-membered ring would mimic
the pyrrole ring of 1 in its hydrophobic interactions with residues
V911, G993, and M929. The 2-amino-aryl-7-aryl-benzoxazole scaf-
fold shown in Figure 3 was designed by interactive molecular mod-
eling following this concept. The binding model of the prototype
molecule of this series, compound 2, is represented in Figure 4.
As can be seen, the inversion of the pyrrolopyrimidine five- and
six-membered rings to give a benzoxazole bicycle provides an
excellent molecular mimic of 1. The N3 atom of the benzoxazole
core can form the expected hinge hydrogen bond interaction while
the attached 2-amino-aryl and 7-aryl moieties match exactly the
positions of those of the pyrrolopyrimidine inhibitor in the ATP
pocket.11 The prototype compound 2 was synthesized.6 Very
encouragingly, 2 turned out to potently inhibit JAK2 with an IC50
value of 15 nM comparing well with that of 1, 5.9 nM, in the bio-
chemical assay.12 Following up on this promising result, we under-
took a synthesis program around the new scaffold that has led to a
new class of potent, selective, and orally available JAK2 inhibitors.6
One of the most potent inhibitors obtained in the new benzox-
azole series was compound 3 bearing a sulfonamide group in para
position of the 7-phenyl moiety (12 nM). Looking for additional
new inhibitor chemotypes, we were inspired by the concept of
C–HÁÁÁO pseudo hydrogen bond.13,14 This type of interaction involv-
Figure 5. Model of prototype compound 4 (green) docked in the ATP pocket. The
mimicked compound 3 appears in yellow. Key hydrogen bonds are represented as
dashed lines.
ing an aromatic C–H group of the inhibitor polarized by an adjacent
nitrogen atom and the backbone carbonyl function of one of the
hinge residues has been observed in crystal structures of kinase-
inhibitor complexes.15 However, it is usually not considered in
the design of new inhibitors.
Thus, the 2,8-diaryl-quinoxaline scaffold shown in Figure 3 was
designed as a mimic of the benzoxazole one and a prototype, com-
pound 4, morphing the potent benzoxazole derivative 3 was envis-
aged for synthesis. As illustrated in Figure 5 with the docking
model of 4, the driving idea here was to mimic the hydrogen bond
between the 2-amino substituent of the benzoxazole inhibitor and
the backbone carbonyl group of hinge residue L932 by a non con-
ventional pseudo hydrogen bond involving the aromatic C-H group
in position 3 of the quinoxaline ring.16 This implied to attach a tri-
methoxyphenyl moiety in position 2 of the quinoxaline to match
that of the benzoxazole compound. The resulting molecule 4 was
able to form the same interactions with the ATP binding site as 3
as can be judged from the excellent overlap of the two molecules
displayed in Figure 5.
4
3
N
N
3
2
7
HN
N
2
O
8
To our satisfaction, taking the risk of replacing a standard
hydrogen bond interaction by a non conventional one paid off.
With an IC50 value of 42 nM in the biochemical assay, 4 showed po-
tent inhibition of JAK2. Thus, we had access to an additional new
R1
R2
R1
R2
Figure 3. Designed scaffolds.

1860
P. Furet et al. / Bioorg. Med. Chem. Lett. 20 (2010) 1858–1860
6. Gerspacher, M.; et al. Bioorg. Med. Chem. Lett., in press, doi:10.1016/
j.bmcl.2010.01.069.
7. Pissot-Soldermann, C.; et al., submitted for publication.
8. Lucet, I. S.; Fantino, E.; Styles, M.; Bamert, R.; Patel, O.; Broughton, S. E.; Walter,
M.; Burns, C. J.; Treutlein, H.; Wilks, A. F.; Rossjohn, J. Blood 2006, 107, 176 (PDB
entry code 2B7A).
9. Modeling and docking was performed with a version of MacroModel enhanced
for graphics by A. Dietrich. MacroModel: Mohamadi, F.; Richards, N. G. J.;
Guida, W. C.; Liskamp, R.; Lipton, M.; Caufield, C.; Chang, G.; Hendrickson, T.;
Still, W. C. J. Comput. Chem. 1990, 11, 440.
10. The ‘hinge’ is the amino-acid stretch connecting the N-terminal and C-terminal
domains of a kinase in a widely used terminology. The ‘gate keeper’ is the main
residue determining the selectivity of ATP site directed kinase inhibitors. It is
named like this because it controls the access to the hydrophobic back pocket
of the ATP binding site. It is located at the beginning of the hinge segment. See
for instance: Furet, P.; Bold, G.; Meyer, Thomas; Roesel, J.; Guagnano, V. J. Med.
Chem. 2006, 49, 4451.
11. In the first prototype compound, a simple phenyl ring was envisaged at
position 7, the nitrogen atom of the pyridine ring of 1 showing no particular
interaction with the cavity in the docking model.
series of potent JAK2 inhibitors that was subsequently optimized
towards compounds suitable for in vivo efficacy evaluation.7
The two examples presented here illustrate the efficiency of
innovative scaffold morphing to generate new useful chemotypes
in the very competitive field of kinase inhibitor research. In partic-
ular, our work gives support to the notion that a pseudo hydrogen
bond formed between an inhibitor aromatic C–H group and one of
the backbone amide carbonyl groups of the hinge segment is a
favorable interaction to seek in kinase inhibitor design.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2010.01.151.
12. For a detailed description of the biochemical assays see: Gerspacher, M.; Furet,
P.; Vangrevelinghe, E.; Pissot Soldermann, C.; Gaul, C.; Holzer, P. PCT Int. Appl.
WO 2008148867, 2008; Chem. Abstr. 2008, 150, 56197.
13. Taylor, R.; Kennard, O. J. Am. Chem . Soc. 1982, 104, 5063.
14. Desiraju, G. R. Acc. Chem. Res. 1996, 29, 441.
References and notes
1. For recent reviews see: (a) Atallah, E.; Verstovsek, S. Exp. Rev. Antican. Ther.
2009, 9, 663; (b) Pardanani, A. Leukemia 2008, 22, 23; (c) Levine, R. L.;
Pardanani, A.; Tefferi, A.; Gilliland, D. G. Nat. Rev. Cancer 2007, 7, 673.
2. Antonysamy, S.; Hirst, G.; Park, F.; Sprengeler, P.; Stappenbeck, F.; Steensma, R.;
Wilson, M.; Wong, M. Bioorg. Med. Chem. Lett. 2009, 19, 279.
15. Pierce, A. C.; Sandretto, K. L.; Bemis, G. W. Protein: Struct. Funct. Genet. 2002, 49,
567.
16. In our model, the designed C–HÁÁÁO interaction had the attributes of a H-bond
(see Ref. 13): a distance between the proton and the acceptor oxygen atom of
2.3 Å which is significantly less than the sum of the van der Waals radii (2.7 Å),
an angle C–HÁÁÁO of 140° approaching the ideal value of 180° (linearity) and an
elevation angle of 30° approaching the ideal value of 0° indicating the tendency
of the HÁÁÁO contact to lie in the plane containing the oxygen lone pairs. A
crystal structure of compound 4 in complex with the kinase domain of JAK2
was subsequently solved (Ref. 7). It fully confirmed the validity of the design
concept. In this structure a C–HÁÁÁO pseudo hydrogen bond is observed between
the C–H group in position 3 of the quinoxaline ring and the backbone carbonyl
oxygen atom of residue L932. The geometric parameters of this hydrogen bond
are the following: distance proton-oxygen of 2.34 Å, angle C–HÁÁÁO of 132° and
elevation angle of 40.4°.
3. Burns, C. J.; Bourke, D. G.; Andrau, L.; Bu, X.; Charman, S. A.; Donohue, A. C.;
Fantino, E.; Farrugia, M.; Feutrill, J. T.; Joffe, M.; Kling, M. R.; Kurek, M.; Nero, T.
L.; Nguyen, T.; Palmer, J. T.; Phillips, I.; Shackleford, D. M.; Sikanyika, H.; Styles,
M.; Su, S.; Treutlein, H.; Zeng, J.; Wilks, A. F. Bioorg. Med. Chem. Lett. 2009, 19,
5887.
4. Kiss, R.; Polgar, T.; Kirabo, A.; Sayyah, J.; Figueroa, N. C.; List, Al. F.; Sokol, L.;
Zuckerman, K. S.; Gali, M.; Bisht, K. S.; Sayeski, P. P.; Keseru, G. M. Bioorg. Med.
Chem. Lett. 2009, 19, 3598.
5. Choi, H-S.; Wang, Z.; Richmond, W.; He, X.; Yang, K.; Jiang, T.; Karanewsky, D.;
Gu, X-J.; Zhou, V.; Liu, Y.; Che, J.; Lee, C. C.; Caldwell, J.; Kanazawa, T.;
Umemura, I.; Matsuura, N.; Ohmori, O.; Honda, T.; Gray, N.; He, Y. Bioorg. Med.
Chem. Lett. 2006, 16, 2689.