Discovery and SAR studies of a novel series of noncovalent cathepsin S inhibitors

Article abstract of DOI:10.1016/j.bmcl.2005.01.045

A novel series of competitive, reversible cathepsin S (CatS) inhibitors was discovered and optimized. The 4-(2-keto-1-benzimidazolinyl)-piperidin-1-yl moiety was found to be an effective replacement for the 4-arylpiperazin-1-yl group found in our earlier series of CatS inhibitors. This replacement imparted improved PK properties as well as decreased off-target activity. Optimization of the ketobenzimidazole moiety led to the discovery of the lead compound JNJ 10329670, which represents a novel class of selective, noncovalent, reversible, and orally bioavailable inhibitors of cathepsin S.

Full text of DOI:10.1016/j.bmcl.2005.01.045

Bioorganic & Medicinal Chemistry Letters 15 (2005) 1687–1691
Discovery and SAR studies of a novel series of
noncovalent cathepsin S inhibitors
Darin J. Gustin, Clark A. Sehon, Jianmei Wei, Hui Cai, Steven P. Meduna,
Haripada Khatuya, Siquan Sun, Yin Gu, Wen Jiang, Robin L. Thurmond,
Lars Karlsson and James P. Edwards*
Johnson and Johnson Pharmaceutical Research and Development, LLC, 3210 Merryfield Row, San Diego, CA 92121, USA
Received 25 August 2004; revised 17 January 2005; accepted 19 January 2005
Abstract—A novel series of competitive, reversible cathepsin S (CatS) inhibitors was discovered and optimized. The 4-(2-keto-1-
benzimidazolinyl)-piperidin-1-yl moiety was found to be an effective replacement for the 4-arylpiperazin-1-yl group found in our
earlier series of CatS inhibitors. This replacement imparted improved PK properties as well as decreased off-target activity. Opti-
mization of the ketobenzimidazole moiety led to the discovery of the lead compound JNJ 10329670, which represents a novel class
of selective, noncovalent, reversible, and orally bioavailable inhibitors of cathepsin S.
Ó 2005 Elsevier Ltd. All rights reserved.
Cathepsin S (CatS) is a cysteine protease of the papain
family that is involved in the presentation of antigens
to the cell surface of certain antigen-presenting cells
(APCs) for recognition by CD4+ T-cells. The main tar-
get of the proteolytic activity of CatS is the invariant
chain (Ii), a chaperone molecule for major histocompati-
bility complex class II molecules (MHCII).1–3 Inhibi-
tion of CatS activity slows the removal of Ii and
attenuates antigen presentation to CD4+ T-cells. The
only selective CatS inhibitors disclosed to date, such as
the dipeptidyl vinyl sulfone LHVS,4 rely upon the form-
ation of a covalent adduct at the active site cysteine of
CatS to achieve potent inhibition.5–8
F
N
Cl
N
N
N
OH
N
Me
O
F
1
N
I
CN
N
N
N
O
O
NH2
N
Our discovery and initial optimization of the screening
hit 1 leading to the potent (IC50 = 20 nM) and selective
noncovalent cathepsin S inhibitor 2 have recently been
disclosed.9
NH2
2
O
azine inhibitors are active in a secondary cellular assay
measuring Ii degradation in (human) JY cells, they
had limited oral bioavailability.10 We thus set out to
identify new CatS inhibitors with improved selectivity
and bioavailability compared to the arylpiperazines.
While compound 2 and analogues are completely selec-
tive against other proteases tested, including the related
cathepsins B, E, F, K, and L, several analogues were
found to have affinity for the a1a adrenergic receptor
(Table 1). In addition, although several of the arylpiper-
During a survey of commercially available aryl-substi-
tuted piperazines and piperidines, we discovered that a
4-(2-keto-1-benzimidazolinyl)-piperidine was a suitable
replacement for the 4-arylpiperazine moiety. Reported
Keywords: Cathepsin S; Cysteine protease inhibitor.
*
Corresponding author. Tel.: +1 8584502024; fax: +1 8587843116;
e-mail: jedward7@prdus.jnj.com
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.01.045

1688
D. J. Gustin et al. / Bioorg. Med. Chem. Lett. 15 (2005) 1687–1691
Table 1. Arylpiperazine cathepsin S inhibitorsa
N
R1
N
N
I
N
O
R2
N
P
Compound
R1
R2
P
CatS IC50, nM
a1a Ki, nM
JY Cell (Ii deg.) IC50, lM
2
3
4
CN
CN
CN
CONH2
CONH2
Ac
CONH2
20
120
50
12
0.88
3.9
H
H
ND
36
1.3
a1a Ki values were determined by MDS PanLabs.
a CatS IC50 and JY Ii degradation assay IC50 values are the mean of n P 3 runs and determined as described previously.9,11
NH
O
R1
N
N
R3
R3
N
N
HN
N
O
a
R2
b or c
N
N
P
5
P
6
R3
N
N
N
O
OH
R1
N
N
N
P
R2
7-12, 18-37
Scheme 1. Reagents and conditions: (a) epichlorohydrin (4–6 equiv), Cs2CO3 (2 equiv), DMF, rt; (b) amine, 3–10% Yb(OTf)3, CH2Cl2, rt, 0 °C; (c)
amine, EtOH, reflux.
Table 2. Ketobenzimidazole CatS inhibitorsa
herein is an SAR study of ketobenzimidazole-containing
CatS inhibitors that led to the preparation of potent,
orally bioavailable inhibitors of CatS.11
R3
N
N
N
O
4
OH
N
3
HN
2
The epoxide intermediates 6, prepared as described pre-
viously,9 provided the ability to quickly survey Ôhead
groupÕ piperidine inputs and to assess the corresponding
structure–activity relationships (Scheme 1). The addi-
tion of the piperidines to the epoxides was promoted
either thermally (EtOH, 80 °C) or by 3–10% Yb(OTf)3
(CH2Cl2, rt) to afford the secondary alcohols 7–12 and
18–37 in 50–90% yields.
N
Me
O
Compound
R3
CatS IC50, nM
7
4-Cl
460
130
130
120
80
8
3-Me, 4-Cl
4-CF3O
4-I
9
10
11
12
As shown in Table 2, the parent ketobenzimidazole
afforded analogues with excellent CatS inhibition when
paired with a variety of substituted pyrazoles. The 4-
chlorophenyl pyrazole derivative 7 inhibited CatS with
an IC50 of 0.46 lM, comparable to the level of activity
observed previously for 1 (1.0 lM). The addition of a
meta-substituent to the phenyl pyrazole (8) or replace-
ment of the chlorine atom with more lipophilic groups
(9–12) improved inhibitory activity by threefold or bet-
ter. These new CatS inhibitors also had sub-micromolar
affinity for a1a.
4-Br
4-CF3
100
a See Table 1 for details.
on the ketobenzimidazole aromatic ring and the distal
nitrogen atom with the aim of reducing a1 affinity and
improving bioavailability. The requisite piperidine inter-
mediates were prepared by the route depicted in Scheme
2.12 Condensation of a 2-fluoronitrobenzene 13 with the
4-aminopiperidine 14 afforded the nitroanilines 15.
Reduction of the nitro group by transfer hydrogenation
and subsequent treatment with triphosgene afforded
the ketobenzimidazoles 16 in good yields, providing
Given the encouraging CatS activity of these analogues,
we next set out to prepare derivatives with substitutions

D. J. Gustin et al. / Bioorg. Med. Chem. Lett. 15 (2005) 1687–1691
1689
CO2Et
N
F
O2N
CO2Et
HN
O2N
N
+
b, c
a
H2N
R2
13
R2
14
15
CO2Et
NH
N
O
O
d (R1 = H)
or
R1
N
N
N
HN
e, d (R1 =/ H)
R2
16
R2
17
Scheme 2. Reagents and conditions: (a) DMF, 60 °C; (b) Pt/C, NH4HCO2, EtOH, rt; (c) triphosgene, Et3N, CH2Cl2, rt; (d) 1 N NaOH, EtOH,
reflux; (e) KHMDS, R1X, THF.
prolonged handling of the unstable phenylene diamines
was avoided. Basic hydrolysis afforded the piperidines 17
(R1 = H). Alternatively, N-substitution of the ketobenz-
imidazole nitrogen could be effected prior to removal of
the ethoxycarbonyl protecting group. Basic hydrolysis
then afforded the N-substituted ketobenzimidazoles 17.
Conversion to the target molecules proceeded as shown
in Scheme 1.
We next turned our attention to inhibitors with nonhy-
drogen substituents on the aromatic ring of the keto-
benzimidazole, and the CatS inhibitory activity for
these derivatives is tabulated in Table 4, with the data
from the parent inhibitors 8, 10, 11, and 12 shown for
comparison.
Substitutions were generally well tolerated, particularly
at the 50 and 60 positions of the aromatic ring, affording
no more than 2-fold changes in activity versus the parent
structures. For the 4-bromophenylpyrazole 11, for
example, chloro (26), methoxy (28), or methyl (30)
Several N-alkylated derivatives of 8 and 12 were pre-
pared and tested for their ability to inhibit CatS activity
(Table 3).
For the 3-methyl-4-chlorophenyl pyrazole derivative 8,
addition of an N-methyl group resulted in a half-log loss
in activity (18) that could be regained by more lipophilic
N-substitutions (19–22). On the other hand, N-methyla-
tion of the 4-trifluoromethyl phenylpyrazole 12 afforded
an inhibitor (23) with equivalent potency. Although N-
substitution on the ketobenzimidazole moiety had only
a moderate effect on CatS activity, qualitatively the solu-
bility of these analogues was greatly improved, making
them easier to handle in many of the biological assays.
Unfortunately, screening against a1a showed little
improvement in selectivity.
Table 4. Substituted ketobenzimidazolesa
R3
N
4
N
N
O
OH
3
R1
N
2
N
N
P
7'
4'
R2
6'
5'
Compound R1 R2
R3
P
CatS IC50
nM
,
Table 3. Effect of N-substitution on CatS activitya
8
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
3-Me, 4-Cl Ac
130
120
80
R3
10
11
12
24
25
26
27
28
29
30
31
32
33
34
35
36
37
H
4-I
4-Br
4-CF3
Ac
Ac
Ac
N
N
N
4
O
H
OH
H
100
100
230
120
110
140
210
90
R1
3
N
2
50-Me
60-F
60-Cl
3-Me, 4-Cl Ac
N
4-I
4-Br
Ac
Ac
Ac
Ac
Ac
Ac
N
Me
50,60-diCl 4-Br
O
50-OMe
70-Me
50-Me
H
4-Br
4-Br
4-Br
Compound
R1
R3
CatS IC50, nM
8
H
3-Me-4-Cl
4-CF3
130
100
500
310
140
100
90
3-Me, 4-Cl MeSO2
3-Me, 4-Cl MeSO2
60
50
12
18
19
20
21
22
23
H
60-Cl
Me
Et
3-Me-4-Cl
3-Me-4-Cl
3-Me-4-Cl
3-Me-4-Cl
3-Me-4-Cl
4-CF3
H
4-CF3
4-CF3
4-Br
MeSO2
MeSO2
MeSO2
MeSO2
MeSO2
50
60
60-Cl
60-Cl
n-Bu
i-Pr
50
30
Me 60-Me
Me 60-Cl
4-CF3
4-CF3
CF3CH2
Me
80
110
a See Table 1 for details.
a See Table 1 for details.

1690
D. J. Gustin et al. / Bioorg. Med. Chem. Lett. 15 (2005) 1687–1691
substituents at 50 or 60 afforded inhibitors with activity
comparable to 11. Dichloro substitution (27) was tolera-
ted equally well.
In order to understand the effect of the linker hydroxyl
on activity, two pairs of enantiomers containing the 6-
chloroketobenzimidazole were synthesized (Scheme 3).
The pyrazole 5 was condensed with t-butyldimethylsil-
yl-protected glycidols (R)- or (S)-38 (>98% ee), fol-
lowed by cleavage of the TBS ether with CSA. The diols
(S)- and (R)-39 were converted13 to the epoxides (S)- or
(R)-6 and then to the desired analogues 40–43.
As shown by the CatS inhibitory activity (Scheme 3),
the absolute stereochemistry of the inhibitors had little
effect on CatS potency.
An interesting effect regarding the nature of tetrahydro-
pyridine N-substituent on CatS activity was also
observed. Replacement of the acetamide with
a
methylsulfonamide resulted in a modest 2-fold improve-
ment in potency for both the 3-methyl-4-chloro-
phenylpyrazole (compound 8 vs compound 31); and 4-
trifluoromethylphenylpyrazole (12 vs 33). This trend
was maintained with a 60-chloroketobenzimidazole (26
vs 35). In addition to the modest improvements in CatS
potency, a1a selectivity of certain analogues was greatly
improved. For example, compound 37 had a Ki of 3 lM
against the a1a adrenergic receptor.
Given the modest effect of hydroxyl stereochemistry on
CatS inhibition, achiral analogues that did not bear the
linker hydroxyl group were targeted and prepared as
shown in Scheme 4.
R3
R3
*
N
N
HO
N
HN
a, b
OH
*
O
+
TBSO
N
N
38
Me
39
c, d, e
Me
O
O
5
R3
R3
*
N
*
N
N
N
4
O
N
O
OH
f
N
3
2
HN
N
N
Me
Me
O
O
Cl
6
(S)-40: R3 = 3-Me, 4-Cl CatS IC50 = 190 nM
(R)-41: R3 = 3-Me, 4-Cl CatS IC50 = 350 nM
(S)-42: R3 = 4-CF3 CatS IC50 = 200 nM
(R)-43: R3 = 4-CF3 CatS IC50 = 180 nM
Scheme 3. Reagents and conditions: (a) Cs2CO3, DMF, rt; (b) CSA, MeOH; (c) MeC(OMe)3, PPTS; (d) AcBr, MeOH, CH2Cl2; (e) KOEt, MeOH;
(f) 6-chloro-1-piperidin-4-yl-1,3-dihydro-benzoimidazol-2-one, EtOH, reflux.
NH
O
R1
N
O
N
R3
a, b
R3
N
N
HN
H
N
R2
c
N
N
SO2Me
SO2Me
5
44
R3
N
N
N
O
R1
N
N
N
SO2Me
R2
45-53
Scheme 4. Reagents and conditions: (a) 3-bromopropanol, Cs2CO3, DMF; (b) Dess–Martin periodinane; (c) Na(OAc)3BH, AcOH, CH2Cl2.

D. J. Gustin et al. / Bioorg. Med. Chem. Lett. 15 (2005) 1687–1691
1691
Table 5. CatS inhibition of achiral noncovalent inhibitorsa
macological characterization of JNJ 10329670 has
recently been described.11
R3
N
4
N
N
O
In conclusion, replacement of the head-group N-arylpip-
erazine in our original series of CatS inhibitors with a
ketobenzimidazole piperidine resulted in a new series
of potent, noncovalent CatS inhibitors. The incorpora-
tion of substituents on the ketobenzimidazole aryl moi-
ety proved useful for improving selectivity over the a1
adrenergic receptors and attaining inhibitors with im-
proved oral bioavailability. In addition, a selective non-
covalent reversible inhibitor, JNJ 10329670, was selected
for further investigation in order to elucidate the phar-
macology of CatS and its potential as a target for immu-
nosuppressive therapies.
3
R1
N
2
N
N
7'
4'
SO2Me
R2
6'
5'
Compound
R1
R2
R3
CatS IC50, nM
45
46
47
48
49
50
51
52
53
H
H
60-Cl
3,4-diCl
4-CF3
4-CF3
3,4-diCl
4-Br
10
30
H
H
H
50
50
Me
H
H
H
60-Cl
50
Me
4-CF3
4-CF3
4-Br
100
100
140
150
NCCH2
Me
Me
H
H
60-Cl
References and notes
3,4-diCl
a See Table 1 for details.
1. Villandangos, J. A.; Bryant, R. A. R.; Deussing, J.;
Driessen, C.; Lennon-Dumenil, A.-M.; Riese, R. J.; Roth,
W.; Saftig, P.; Shi, G.-P.; Chapman, H. A.; Peters, C.;
Ploegh, H. L. Immunol. Rev. 1999, 172, 109.
2. Nakagawa, T. Y.; Rudensky, A. Y. Immunol. Rev. 1999,
172, 121.
3. Shi, G.-P.; Villadangos, J. A.; Dranoff, G.; Small, C.; Gu,
L.; Haley, K. J.; Riese, R.; Ploegh, H. L.; Chapman, H. A.
Immunity 1999, 10, 197.
4. Palmer, J. T.; Rasnick, D.; Klaus, J. L.; Bromme, D. J.
Med. Chem. 1995, 38, 3193.
Regioselective alkylation of pyrazole 5 with 3-bromo-
propanol in the presence of cesium carbonate followed
by oxidation of the primary alcohol with Dess–Martin
periodinane afforded the aldehydes 44. Installation of
the piperidines via reductive amination afforded the
desired achiral deshydroxy analogs 45–53 (Table 5).
Interestingly, removal of the hydroxyl group resulted in
compounds with equal or improved activity compared
to the hydroxy analogues. For example, the hydroxy
compound 34 (Table 4) had an IC50 of 60 nM, while
the deshydroxy analogue 46 had an IC50 of 30 nM
(Table 5). Likewise, the deshydroxy version of 11 (CatS
IC50 = 80 nM) was also an excellent CatS inhibitor (49,
CatS IC50 = 50 nM).
5. Katunuma, N.; Murata, E.; Kakegawa, H.; Matsui, A.;
Tsuzuki, H.; Tsuge, H.; Turk, D.; Turk, V.; Fukushima,
M.; Tada, Y.; Asao, T. FEBS Lett. 1999, 458, 6.
6. Ward, Y. D.; Thomson, D. S.; Frye, L. L.; Cywin, C. L.;
Morwick, T.; Emmanuel, M. J.; Zindell, R.; McNeil, D.;
Bekkali, Y.; Giradot, M.; Hrapchak, M.; DeTuri, M.;
Crane, K.; White, D.; Pav, S.; Wang, Y.; Hao, M.-H.;
Grygon, C. A.; Labadia, M. E.; Freeman, D. M.;
Davidson, W.; Hopkins, J. L.; Brown, M. L.; Spero, D.
M. J. Med. Chem. 2002, 45, 5471.
Among the analogues prepared and tested, JNJ
10329670 (50, Table 5) was chosen for further pharma-
cological characterization for several reasons. First, it
was selective against the adrenergic receptors a1a, a1b,
and a1d (1.1–1.9 lM). This is a dramatic improvement
in selectivity (50–100x) when compared to the unsubsti-
tuted ketobenzimidazoles and the arylpiperazines from
the original lead series. In addition, JNJ 10329670 has
encouraging activity in the secondary cellular assay
measuring inhibition of cleavage of the invariant chain
in JY cells9,11 (IC50 = 0.60–0.80 lM). Finally, 50 also
has good oral bioavailability in rats, dogs, and monkeys
(40–75%) and an excellent in vivo half-life (4–14 h).
Compound 50 was also tested against cathepsins B, C,
D, E, F, K, L, L2, Z, and legumain: IC50Õs of greater
than 50 lM were observed for these enzymes. The phar-
7. Bromme, D.; Smith, R. A.; Coles, P. J.; Kirschke, H.;
Storer, A. C.; Krantz, A. Biol. Chem. Hoppe–Seyler 1994,
375, 343.
8. For a recent review, see: Leroy, V.; Thurairatnam, S.
Expert Opin. Ther. Pat. 2004, 14, 301.
9. Thurmond, R. L.; Beavers, M. P.; Cai, H.; Meduna, S. P.;
Gustin, D. J.; Sun, S.; Almond, H. J.; Karlsson, L.;
Edwards, J. P. J. Med. Chem. 2004, 47, 4799.
10. Unpublished results from these laboratories.
11. For the pharmacological characterization of a cathepsin S
inhibitor from this series, see: Thurmond, R. L.; Sun, S.;
Sehon, C. A.; Baker, S. M.; Cai, H.; Gu, Y.; Jiang, W.;
Riley, J. P.; Williams, K. N.; Edwards, J. P.; Karlsson, L.
J. Pharmcol. Exp. Ther. 2004, 308, 268.
12. Henning, R.; Lattrell, R.; Gerhards, H. J.; Leven, M. J.
Med. Chem. 1987, 30, 814.
13. Kolb, H. C.; Sharpless, K. B. Tetrahedron 1992, 48, 10515.