Novel scaffold for cathepsin K inhibitors

Article abstract of DOI:10.1016/j.bmcl.2007.09.047

Pyrrolopyrimidine, a novel scaffold, allows to adjust interactions within the S3 subsite of cathepsin K. The core intermediate 10 facilitated the P3 optimization and identified highly potent and selective cathepsin K inhibitors 11-20.

Full text of DOI:10.1016/j.bmcl.2007.09.047

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 17 (2007) 6096–6100
Novel scaffold for cathepsin K inhibitors
a
a,
a
a
a
*
Naoki Teno, Takahiro Miyake, Takeru Ehara, Osamu Irie, Junichi Sakaki,
a
a
a
a,ꢀ
a
Osamu Ohmori, Hiroki Gunji, Naoko Matsuura, Keiichi Masuya, Yuko Hitomi,
a
a,ꢁ
a,§
a
Kazuhiko Nonomura, Miyuki Horiuchi, Keigo Gohda, Atsuko Iwasaki,
a
a,–
a
a
Ichiro Umemura, Sachiyo Tada, Motohiko Kometani, Genji Iwasaki,
b
b
b
a,ꢀ
Sandra W. Cowan-Jacob, Martin Missbach, Ren e´ Lattmann and Claudia Betschart
a
Novartis Institutes for BioMedical Research, Ohkubo 8, Tsukuba, Ibaraki 300-2611, Japan
b
Novartis Institutes for BioMedical Research, CH-4002 Basel, Switzerland
Received 20 August 2007; revised 10 September 2007; accepted 12 September 2007
Available online 15 September 2007
Abstract—Pyrrolopyrimidine, a novel scaffold, allows to adjust interactions within the S3 subsite of cathepsin K. The core interme-
diate 10 facilitated the P3 optimization and identified highly potent and selective cathepsin K inhibitors 11–20.
Ó 2007 Elsevier Ltd. All rights reserved.
A healthy bone metabolism depends on an appropriate
balance between osteoclast-mediated bone resorption
and osteoblast-mediated bone formation. If the bone
resorption is exceeding the bone formation, bone mass
is reduced and leads to disease conditions like post-men-
opausal osteoporosis and tumor-induced osteolysis. A
reduction of the bone formation rate, which is observed
in senile-osteoporosis, also reduces bone mass and
finally induces bone fractures. In the 1980s bisphospho-
nates were found to act as bone resorption inhibitors
lished, and its inhibition is now perceived as a promising
new principle for a treatment of the diseases caused by
increased bone resorption.
6
The sugar derivative 1 was identified in a high through-
put screening as a potent inhibitor of cathepsin K
7
[IC50 = 14 nM]. Surprisingly, removal of the hydro-
philic sugar moiety and also simplification of the ring
system to the pyrimidine 2 resulted with retained
8
potency [IC50 = 170 nM]. A molecular modeling with
(
likely through inhibition of farnesyl diphosphate
the nitrile group covalently bound to the active site cys-
teine suggests that the cyclohexyl group occupies the S2
subsite while the S3 is unoccupied. Therefore, the small
molecule 2 was a good starting point for further optimi-
zation. We expected to make further increase in the
potency by extending into the S3. In addition, due to
differences in the corresponding S3 subsites, higher
selectivity over related proteases, such as cathepsin L
and cathepsin S, was anticipated.
synthase in osteoclasts), and today, bisphosphonates
are widely used for the treatment of osteoporosis and
tumor-induced osteolysis.
1
,2
In the 1990s three independent groups have reported on
cathepsin K, cysteine protease predominantly
expressed in osteoclasts.
a
3
–5
In the meantime, the role
of cathepsin K in degradation of bone matrix was estab-
Three attachment sites for the P3 moiety were consid-
Keywords: Cathepsin K inhibitors; Pyrrolopyrimidine; Novel scaffold.
*
7
9
Corresponding author. Tel.: +81 29 865 2360; fax: +81 29 865
308; e-mail: naoki.teno@novartis.com
ered (A, B , and C in Fig. 1). For scaffold C, the mod-
eling suggested acetylenes as rigid linkers between the
pyrimidine and S3 (scaffold D in Fig. 2). Indeed, the
compounds with high potency and good selectivity
against related proteases could be identified in this series
2
ꢀ
Present address: Novartis Institutes for BioMedical Research,
CH-4002 Basel, Switzerland.
ꢁ
§
YASUTOMI & Associates, MT-2 BLDG, 5-36, Miyahara 3-chome,
Yodogawa-ku, Osaka 532-0003 Japan.
Computer-Aided Molecular Modeling Research Center Kansai,
9
as reported elsewhere.
1
-3-12, Honjyocho, Higashinada-ku, Kobe, Hyogo 658-0012, Japan.
2
2
–
Depending on the structure of R (in the case of R with
benzyl propargyl moiety) in Figure 2, the direct product
Central Research Laboratories, Sysmex Co., 4-4-4, Takatsuka-dai,
Nishi-ku, Kobe, Hyogo 651-2271, Japan.
0
960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.09.047

N. Teno et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6096–6100
6097
1
R
Table 1. Inhibition of human cathepsins K, L, and S by compounds
N
5
a–g
N
N
R
HN
N
N
OH
O
R
S3
HO
A
OH
N
N
N
N
N
N
R
N
N
N
N
N
R
HN
N
N
N
N
HN
N
N
B
2
1
a
Compound
R
IC50 (nM)
HTS-hit
Minimal scaffold
HN
N
N
Cat K
Cat L
Cat S
C
5
a
H
Cl
15
120
82
390
490
5
b
2.3
Figure 1. Design of new pyrimidine scaffolds, based on a HTS hit,
reaching the S3.
5
c
47
3.4
<1
4.7
1.5
>1000
>1000
410
>1000
980
O
2
R
5d
5e
N
2
N
R
N
N
N
HN
R1
N
N
N
N
R1
460
N
N
D
E
Figure 2. Design of new chemotypes based on the pyrimidine scaffold C.
N
5
f
260
>1000
730
of the coupling could not be obtained but the product
immediately rearranged to form a substituted pyrrolo-
pyrimidine E. These pyrrolopyrimidine compounds also
inhibited cathepsin K with comparable potency as to the
acetylene derivatives. In this paper we report on the syn-
thetic methods and the biological activity of cathepsin K
inhibitors.
5
g
110
N
a
Inhibition of recombinant human cathepsins K, L, and S in a fluo-
7
rescence assay.
was tested. The compounds 5f and 5g with a negative
charge in the P3 moiety showed inhibitory activity
for cathepsin K at single digit nM range, but not high
selectivity against both cathepsins S and L simulta-
neously, while 5c and 5d revealed the countertrend.
The methylpiperizine derivative 5e was found to show
sub nM potency and the selectivity against cathepsins
L and S was obtained in more than 400-folds.
The synthesis of pyrrolopyrimidine 5a is outlined in
Scheme 1. Regioselective substitution of the chloride
at the 4- position of 5-bromo-2,4-dichloropyrimidine
gave the intermediate 3. The second chloride was ex-
changed with cyanide to give the pyrimidine 4. Cas-
1
0,11
tro-Stephens type cyclization
immediate ring closure afforded the pyrrolopyrimidine
followed by
A co-crystal structure of the compound 5e with cathep-
sin K showed that the benzyl group is pointing toward
the S3 subsite (Fig. 3). The aromatic ring is involved
in an edge-to-face interaction with the side chain of
Tyr67. The co-crystal structure further indicated that
the para-substituents on the compound 5a are posi-
tioned between the Asp61 and Tyr67, as seen for 5e.
The cyclohexyl group of 5e occupies the lipophilic S2
subsite.
5
a. The new inhibitor 5a exhibited an IC50 value of
5 nM against cathepsin K but with poor selectivity
1
against cathepsins S and L (Table 1). Compounds
5
5
b–5g were prepared by an analogous procedure to
a, and their inhibitory activity against cathepsins
Br
Br
a
b
Br
Cl
N
N
N
HN
N
3
Cl
HN
N
N
Cl
N
On the basis of the results from a quenched fluorogenic
1
2
4
substrate library for cathepsin K , the neopentyl moiety
having a branched alkyl group was introduced instead
of the cyclohexyl group of 5a. Introduction of the neo-
pentyl P2 group provided a potent cathepsin K inhibitor
5h [IC50 (cathepsin K) = 1.2 nM] (Table 2), which was
prepared in the same way as for 5a shown in Scheme
c
N
N
N
5
N
a
1
. The compound, however, still had moderate inhibi-
tory activity toward cathepsins S and L. [IC50 (cathepsin
S) = 100 nM, IC50 (cathepsin L) = 380 nM]. It should be
noted that an increase in inhibitory activity for cathep-
sin K is caused by a suitable interaction between the
Scheme 1. Reagents and conditions: (a) Cyclohexylamine, NaHCO
MeOH, rt, 6 h, 54%; (b) KCN, 1,4-diazabicyclo[2.2.2]octane, DMSO-
O, 60 °C, 6 h, 84%; (c) prop-2-ynyl-benzene, Pd(PPh Cl , CuI,
N, DMF, 80 °C, 12 h, 15%.
3
,
H
2
3
)
2
2
Et
3

6098
N. Teno et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6096–6100
Table 2. Inhibition of human cathepsins K, L, and S by compounds
h–w
5
Asp61
R
N
N
N
N
S3
a
IC50 (nM)
Covalent
bond with
Cys25
Compound
R
Tyr67
Cat K
Cat L
Cat S
S2
5
5
5
h
H
1.2
<1
380
170
96
100
930
460
i
CH
Cl
3
j
<1
Figure 3. Co-crystal structure of compound 5e with cathepsin K
inhibitor is extending into S3 between Asp61 and Tyr67 (PDB code:
5
5
k
l
O
O
<1
1.1
540
110
820
2R6N).
>1000
O
neopentyl group and the S2 subsite. The observation is
8
consistent with that of the previous report.
5m
N
<1
<1
970
>1000
320
H
N
5
5
5
5
5
5
n
o
p
q
r
>1000
Another promising result was observed for compounds
O
O
1
3
5
2
5
i–5w. As the preparation of 5q depicted in Scheme
, the P3 with benzyl propargyl group was coupled with
-bromo-4-(2,2-dimethyl-propylamino)-pyrimidine-2-
<1
<1
<1
<1
1.1
>1000
140
>1000
>1000
160
N
carbonitrile (6), which was prepared in the same way as
for 4 shown in Scheme 1. As listed in Table 2, the com-
pounds 5m, 5o, 5t, and 5v were considered satisfactory
in terms of the potency for the target enzyme and the
selectivity against other cysteine protease, cathepsins S
and L.
N
O
S N
O
N
H
>1000
>1000
>1000
O
H
As shown in Scheme 3, we designed and prepared a use-
ful intermediate 10 which can be used to introduce a
wide variety of P3 moieties at the late stage of the com-
pound preparation. Bromide 6 was coupled with 2-prop-
H N S N
220
2
O
F
F
F
O
S N
O
s
H
440
1
4
2
and CuI as catalysts to afford 7. Formation of the pyr-
-ynyloxy-tetrahydro-pyran
by using Pd(PPh ) Cl
3 2 2
1
5
N
rolopyrimidine ring was carried out by DBU , followed
by tetrahydropyran removal using TsOH-H O. Alcohol
5
t
<1
<1
>1000
430
>1000
>1000
2
9
was converted to the bromine 10 and the P3 optimiza-
N
5
u
tion using a parallel synthesis became feasible as
depicted in Figure 4.
N
O
5
5
v
<1
<1
>1000
520
>1000
>1000
The parallel synthesis approach allowed us to access
compounds with structural diversity by using an almost
unlimited choice of commercially available nucleophiles
with the functional groups (amine, hydroxyl, thiol). By
this approach, structurally diverse P3 moieties can be
introduced using the common intermediate 10. In addi-
tion, we will be able to control the lipophilicity of the
inhibitor by modifying P3 moiety, leading to improving
physicochemical properties.
N
w
a
Inhibition of recombinant human cathepsins K, L, and S in a fluo-
7
rescence assay.
with significant selectivity for cathepsin K over cathep-
sins L and S.
Various P3 parts (R in Table 3) were coupled with 10 in
polar solvent (e.g., DMF, DMSO, or acetone) in the
presence of a base (K CO , NaH, or diisopropylethy-
In summary, a novel pyrrolopyrimidine scaffold E
derived from the pyrimidine derivative contributed to
the discovery of new and specific cathepsin K inhibitors.
The use of a key intermediate 10 has facilitated the P3
optimization and identified novel cathepsin K inhibitors
2
3
1
3,16
amine).
listed in Table 3. Unlike the purine derivatives and
Representative inhibitors in this series are
7
8
2
-cyano-pyrimidine inhibitors, the inhibitors 11–20 dis-
played highly potent cathepsin K inhibition combined

N. Teno et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6096–6100
6099
Table 3. Inhibitory activity of new chemotype with azole, heterocyclic
Si
H
N
O
or spirohydantoin group against cathepsins K, L, and S
2
H N
c
N
a, b
Br
S
Si
O
R
N
N
N
N
O
S N
O
H
d
a
Compound
R
IC50 (nM)
Br
HN
N
N
Cat K
6
Cat L
>1000
Cat S
>1000
N
N
N
N
N
N
6
5
q
1
1
N
Scheme 2. Reagents and conditions: (a) Mg (powder), I
(b) 1-methoxy-propa-1,2-diene, CuBr, THF, rt, 24 h, 61%; (c) MsCl,
pyridine, CH
2
, THF;
N
Cl
Cl
12
4
5
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
2
Cl
2
, rt, 1 h, quant.; (d) Pd(PPh
3
)
2
Cl
2
, CuI, Et
3
N, DMF,
N
70 °C, 2 h, 61%.
N
N
1
3
N
O
O
O
O
a
c
b
d
N
N
N
6
14
3
N
N
N
O
HN
N
N
N
O
O
7
8
1
5
4
N
Br
O
HO
N
N
N
N
N
N
N
16
17
F
N
N
N
1
9
10
O
N
O
Scheme 3. Reagents and conditions: (a) 2-prop-2-ynyloxy-tetrahydro-
pyran, Pd(PPh Cl , CuI, Et
N, DMF, 80 °C, 2 h, 92%; (b) DBU,
DMF, 100 °C, 2 h, 93%; (c) TsOHÆH O, THF, MeOH, rt, 3 h, 87%;
<1
N
)
2
2
3
H
3
2
(
d) CBr
4
, Ph
3
2
P, CH Cl
2
, 0 °C, 0.5 h, rt, 3 h, 78%.
H
N
N
O
1
1
8
9
<1
<1
770
>1000
>1000
N
X
O
R
X
X
N
N
HN
HN
O
>1000
N
N
N
N
O
F
R2
R
1
N
S
N
N
2
0
1.7
>1000
930
R
Br
N
N
N
N
N
N
N
N
N
N
O
N
a
1
0
Inhibition of recombinant human cathepsins K, L, and S in a fluo-
7
rescence assay.
R
H
R N
N
O
N
N
N
to Dr. Robert Hof and Dr. Muneto Mogi for helpful
critical reading of the manuscript.
N
N
N
N
N
Figure 4. Exploring the P3 moiety by a parallel synthesis.
References and notes
with nM potency with excellent selectivity against
cathepsins L and S. The structure activity relationships
and the pharmacokinetic properties of this series
reported here will be covered in future publications.
1. Van Beek, E.; Pieterman, E.; Cohen, L.; Lowik, C.;
Papapoulos, S. Biochme. Biophys. Res. Commun. 1999,
264, 108.
2
3
4
. Dunford, J. E.; Thompson, K.; Coxon, F. P.; Luckman, S.
P.; Hahn, F. M.; Poulter, C. D.; Ebetino, F. H.; Rogers,
M. J. J. Pharmacol. Exp. Ther. 2001, 296, 235.
. Tezuka, K.; Tezuka, Y.; Maejima, A.; Sato, T.; Nemoto,
K.; Kamioka, H.; Hakeda, Y.; Kumegawa, M. J. Biol.
Chem. 1994, 269, 1106.
Acknowledgments
We thank Mr. Junichi Yamanaka and Ms. Ibuki Misawa
for excellent technical assistance. The authors are grateful
. Br o¨ mme, D.; Okamoto, K.; Wang, B. B.; Biroc, S. J. Biol.
Chem. 1996, 271, 2126.

6100
N. Teno et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6096–6100
5
. Draket, F. H.; Dodds, R. A.; James, I. E.; Connor, J. R.; R.; Gasser, J.A.; Lattmann, R.; Missbach, M.
Debouck, C.; Richardson, S.; Lee-Rykaczewski, E.; Cole-
man, L.; Rieman, D.; Barthlow, R.; Hastings, G.; Gowen,
M. J. Biol. Chem. 1996, 271, 12511.
. Grabowska, U. B.; Chambers, T. J.; Shiroo, M. Curr.
Opin. Drug Dis. Dev. 2005, 8, 619.
Abstracts of Papers, 12th Annual Meeting of Division
of Medicinal Chemistry, Tsukuba, Japan, Nov 24–26,
2004.; The Pharmaceutical Society of Japan, 2004;
P1-28.
6
7
8
13. Betschart, C.; Hayakawa, K.; Irie, O.; Sakaki, J.; Iwasaki,
G.; Lattmann, R.; Missbach, M.; Teno, N. WO
2003020721, 2003.
. Altmann, E.; Cowan-Jacob, S. W.; Missbach, M. J. Med.
Chem. 2004, 47, 5833.
. Altmann, E.; Aichholz, R.; Betschart, C.; Buhl, T.; Green,
J.; Irie, O.; Teno, N.; Lattmann, R.; Tintelnot-Blomley,
M.; Missbach, M. J. Med. Chem. 2007, 50, 691.
. Paper in preparation.
14. Trost, B. M.; Jonasson, C.; Wuchrer, M. J. Am. Chem.
Soc. 2001, 123, 12736.
15. Unlike the synthesis of 5a (Scheme 1), the coupling of 6
and 2-prop-2-ynyloxy-tetrahydro-pyran did not give 8
directly. The intermediate 8 was prepared from 7 by using
the reaction condition analogous to the synthesis of
indoles Kondo, Y.; Kojima, S.; Sakamoto, T. J. Org.
Chem. 1997, 62, 6507.
9
1
1
1
0. Stephens, R. D.; Castro, C. E. J. Org. Chem. 1963, 28,
3313.
1. Kondo, Y.; Watanabe, R.; Sakamoto, T.; Yamanaka, H.
Chem. Pharm. Bull. 1989, 37, 2933.
2. Teno, N.; Kometani, M.; Betschart, C.; Toriyama, K.;
Niwa, S.; Sakaki, J.; Altmann, E.; Buhl, T.; Gamse,
16. Irie, O.; Iwasaki, G.; Masuya, K.; Miyake, T.; Teno,
N. WO 2004076455, 2004.