Primary amides as selective inhibitors of cathepsin K

Article abstract of DOI:10.1016/j.bmcl.2007.05.024

The nitrile warhead used in a series of cathepsin K inhibitors can be replaced by a less electrophilic primary amide. The accompanying loss of potency can be partially recovered by introducing a substituent α to the amide. The potency gain resulting from this addition is not achieved with the nitrile derivatives due to a different geometry of the cysteine adduct in the enzyme active site. This study led to the identification of the primary amide 2g, which is an inhibitory substrate, with an IC₅₀ of 10 nM against cathepsin K and excellent selectivity versus the other cathepsins.

Full text of DOI:10.1016/j.bmcl.2007.05.024

Bioorganic & Medicinal Chemistry Letters 17 (2007) 4328–4332
Primary amides as selective inhibitors of cathepsin K
*
´
Serge Leger, Christopher I. Bayly, W. Cameron Black, Sylvie Desmarais,
´ ´
´
Jean-Pierre Falgueyret, Frederic Masse, M. David Percival and Jean-Franc¸ois Truchon
Merck Frosst Centre for Therapeutic Research, PO Box 1005, Pointe-Claire-Dorval, Que., Canada H9R 4P8
Received 21 February 2007; revised 5 May 2007; accepted 9 May 2007
Available online 16 May 2007
Abstract—The nitrile warhead used in a series of cathepsin K inhibitors can be replaced by a less electrophilic primary amide. The
accompanying loss of potency can be partially recovered by introducing a substituent a to the amide. The potency gain resulting
from this addition is not achieved with the nitrile derivatives due to a different geometry of the cysteine adduct in the enzyme active
site. This study led to the identification of the primary amide 2g, which is an inhibitory substrate, with an IC50 of 10 nM against
cathepsin K and excellent selectivity versus the other cathepsins.
Ó 2007 Published by Elsevier Ltd.
Cathepsin K is a lysosomal cysteine protease that is in-
volved in the proteolysis of Type I collagen which is the
major organic constituent of bone. Cathepsin K is
highly and selectively expressed in osteoclasts, the cells
which degrade bone during the continuous cycle of bone
degradation and formation.1,2 Inhibition of cathepsin K
represents a potential therapeutic approach for diseases
characterized by excessive bone resorption such as
osteoporosis.3
CF3
H
R
N
N
H
O
MeO2S
Cat K IC50 = 0.2 nM
Cat K IC50 = 238 nM
1a R = CN
2a R = C(O)NH2
Figure 1. Potency of nitrile versus amide warhead.
Many of the reported inhibitors of cathepsin K rely on
an electrophilic functionality to interact with the nucle-
ophilic cysteine residue found in the catalytic site of this
class of enzymes.4 Our laboratories have recently re-
ported the preparation and SAR of a series of nitrile-
containing compounds that are potent inhibitors of
cathepsin K.5,6 The electrophilic nitrile warhead has
been shown by X-ray crystallography to form a covalent
adduct with the active site cysteine of the enzyme.6 This
covalent, but reversible, bond is proposed to account for
much of the inhibitor’s binding energy.
1a (L-873724) is replaced by the less electrophilic pri-
mary amide (compound 2a).
The addition of cysteine to a nitrile warhead results in
the formation of a planar sp2 adduct, a thioimidate
(Fig. 2a). Conversely, the cysteine addition to an sp2
electrophile such as amide gives rise to an sp3 tetrahedral
adduct (Fig. 2b). These hybridizations exert different
torsional constraints on the adjacent a-carbon, the
attachment point for P1 substitutions, differentially sta-
bilizing the complex for a given P1 moiety. This differ-
ence in conformation can lead to an improved enzyme
interaction of the P1 substituent of an sp2 electrophile
that is not observed with a nitrile. This is supported
by the observation that large effects of P1 substituents
have been observed with aldehyde electrophiles,7 while
only small effects have been reported in the case of ni-
trile inhibitors.8
Given the importance of the electrophile to the inhibi-
tion, the replacement of the nitrile with a less electro-
philic moiety such as an amide will inevitably result in
a loss of potency. This is demonstrated by the loss of
activity shown in Figure 1 when the nitrile in compound
Keywords: Cathepsin inhibitor; Primary amide; Selective; Substrate.
*
An additional source of binding may be achieved by
an amide warhead by the potential formation of an
Corresponding author. Tel.: +1 514 428 3077; fax: +1 514 428
4939; e-mail: leger@merck.com
0960-894X/$ - see front matter Ó 2007 Published by Elsevier Ltd.
doi:10.1016/j.bmcl.2007.05.024

´
S. Leger et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4328–4332
4329
Table 1. Potency of a-substituted amides in purified enzyme assay
O
CF3
H
N
NH2
N
H
R
O
MeO2S
Compound
hrab Cat Ka IC50 (nM)
b
R
2a
2b
2c
2d
2e
2f
H
Me
238 37
174 31
66 11
CH2SMe
CH2SO2Me
CH2CH2SMe
109 26
13 2.0
151 26
10 1.4
273 75
CH2CH2SO2Me
CH2Ph
CH2CH2Ph
2g
2h
a Humanized rabbit enzyme.
b IC50s are an average of at least four independent titrations, SEM.
See Ref. 10 for assay conditions.
resulted in a 27-fold loss of potency. The methylsulfide
functionality, being smaller than a phenyl group, re-
quired a two-carbon chain to access similar interactions
with the S1 pocket of the enzyme’s catalytic site. This ex-
tra methylene of compound 2e resulted in a 5-fold
improvement of potency over its shorter analog 2c.
Polarity is not well-tolerated in this area of the catalytic
site as illustrated by comparing compounds 2d versus 2c
and compounds 2f versus 2e.
To confirm the required stereochemistry introduced at
P1, two benzyl analogs were prepared. As seen in Figure
3, the S configuration at P1 is optimal, whereas the R
diastereomer 3 is 600-fold less potent than 2g. This con-
firms the expected correlation with the S stereochemistry
of the natural aminoacids.
Figure 2. (a) Planar sp2 adduct resulting from the cysteine addition
onto the nitrile warhead of the cathepsin K inhibitor 1a.9 (b)
Tetrahedral sp3 adduct resulting from the cysteine addition onto the
amide warhead of the cathepsin K inhibitor 2a.9
Improvement of the intrinsic potency of these amide
derivatives resulting from P1 substitution was very
important when compared to the corresponding nitrile
analogs. As illustrated in Table 2, P1 substitution in
the nitrile series (compounds 1a–d) showed no improve-
ment in the ability of these compounds to inhibit the
additional hydrogen bond between the amide nitrogen
and the active site histidine of the enzyme (Fig. 2b).
The aim of the current study was to determine if the po-
tency loss observed with the replacement of a nitrile with
an amide warhead could be regained by forming a tetra-
hedral cysteine adduct with beneficial P1 interactions.
The effect of P1 substitution in this amide series was
evaluated for potency versus cathepsin K and selectivity
versus the other cathepsins.
O
CF3
H
N
NH2
N
H
O
The biological results obtained (Table 1) indicate that
P1 substitution in an amide series does indeed increase
potency. The smallest possible substitution showed a
marginal improvement in potency (compound 2b vs
2a). The chain length of the P1 substituent required
adjustment according to the appended functionality
(i.e., aromatic ring vs heteroatom). When a phenyl ring
is attached, the optimal linker was found to be the single
methylene of the phenylalaninamide derivative (com-
pound 2g). Lengthening the chain by preparation of
the homophenylalaninamide analog (compound 2h)
2g "S"
MeO2S
Cat K IC50 10 nM
O
CF3
H
N
N
NH2
H
O
3 "R"
MeO2S
Cat K IC50 6070 nM
Figure 3. Stereochemistry a to the amide warhead.

´
S. Leger et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4328–4332
4330
Table 2. Effect of P1 substitution on the amides versus the nitriles
the biphenyl-trifluoroethylamine backbone of this class
of compounds for the S2 and S3 pockets of the catalytic
site of cathepsin K.
O
CF3
H
H
N
CN
N
NH2
N
H
On closer examination, it was found that these amides
are in fact slowly processed substrates for the enzyme.
Incubation of amide 2a with cathepsin K resulted in
the formation of the carboxylic acid 12 as observed by
reverse phase HPLC analysis of the incubation media
(Fig. 4).
VS.
R
O
R
2
1
MeO2S
R
Amides hrab Cat
Nitriles hrab Cat
Ka IC50 (nM)
Ka IC50 (nM)
b
b
H
Me
2a
2b
238 37
174 31
13 2.0
10 1.4
1a
1b
1c
1d
0.2 0.04
0.9 0.05
0.2 0.01
0.6 0.07
Kinetic constants were determined for three amide
inhibitors in competition experiment with the synthetic
substrate Z-Leu-Arg-AMC (Table 4).11
CH2CH2SMe 2e
CH2Ph 2g
a Humanized rabbit enzyme.
b IC50s are an average of at least four independent titrations, SEM
(except compound 1c where n = 2). See Ref. 10 for assay conditions.
O
O
H
H
[hrab Cat K] = 4 nM
N
N
enzyme, while the same substitution in the amide series
gave rise to more than 20-fold improvement in potency.
NH2
OH
O
O
2a
12
Other electrophilic functionalities, which would also
form a tetrahedral cysteine adduct with the enzyme,
were evaluated. As seen in Table 3, the primary amide
2e remains the most active analog despite the fact that
it is not the most electrophilic. For example the methyl
ketone 6 and the ester 7, which are more electrophilic
than the amide 2e, do not display improved potency.
This is consistent with the possibility for the amide to
form hydrogen bonding interaction with the catalytic
histidine as mentioned earlier (Fig. 2a).
160
140
120
100
80
160
T= 0 min
T= 120
140
min
Compond 2a
R.T. = 7.75 min
120
100
80
60
40
20
0
Compound 2a
R.T. = 7.75 min
Compound 12
R.T. = 8.25 min
60
40
20
0
-20
-20
10
0
2
4
6
8
10
12
2
4
6
8
0
12
Time (min)
Time (min)
Any substitution on the amide nitrogen resulted in a de-
crease in potency (compounds 4, 5, 9, and 10). The acid
analog (compound 8), which is potentially the hydrolysis
product of the amide, remains relatively potent
(IC50 = 120 nM) considering the poor electrophilic char-
acter of this warhead. This reflects the strong affinity of
Figure 4. Enzymatic hydrolysis of the amide 2a as observed by reverse
phase HPLC.
Table 4. Kinetics of amide compounds as enzyme substrates
30
2e
2b
2a
25
Z-LR-AMC
Table 3. Effect of modifying the electrophilic warhead
20
15
10
5
CF
3
H
Electrophile
SMe
N
N
H
O
0
MeO2S
Compound
0
200
400
600
800
1000
1200
Time (min)
b
Electrophile
hrab Cat Ka IC50 (nM)
a
b
Compound
R
H
kcat
(sÀ1
Km
(nM)
kcat/Km
(sÀ1 nMÀ1
)
2e
4
CONH2
CONHMe
CONMe2
COMe
13 2.0
67 15
)
2a
2b
2e
23.2
23.2
119
87
0.19
0.26
0.54
0.04
5
1045 160
41 8.3
27 7.7
120 22
87 16
Me
CH2CH2SMe
Z-Leu-Arg-AMC
6
3.25
79
6
7
CO2Me
CO2H
2000c
8
9
CONHBn
CONHCH2CF3
CONHSO2Me
a All compounds (25 lM) were added to 4 nM of humanized rabbit
cathepsin K and incubated at room temperature.
b Km of the compound was estimated from its IC50 value with the
following relationship: IC50 = Km (1 + S/Ks) using Z-Leu-Arg as
substrate.
10
11
277 44
238 18
a Humanized rabbit enzyme.
b IC50s are an average of at least four independent titrations, SEM
(except compound 6 where n = 2). See Ref. 10 for assay conditions.
c Km was measured in a conventional kinetic experiment.

´
S. Leger et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4328–4332
4331
Table 5. Selectivity of the amide compounds versus the other
cathepsins
aminoamide 14 (Scheme 1). Subsequently, these amides
can be dehydrated using trifluoroacetic anhydride in
pyridine to give access to the nitrile analogs (compounds
1a–d). However, this approach gave low yields of nitriles
and many degradation products were observed, proba-
bly due to the presence of more than one amide bond.
An alternative route to the nitrile inhibitors involved
the preparation of substituted aminoacetonitrile precur-
sors (15a–d) and performing the peptide coupling reac-
tion at the end of the sequence as in the amide series.
Analysis of the resulting products by HPLC confirmed
the stereochemical integrity of the substituent a to the
nitrile.
Compound R
hrab Cat hCat Bb hCat
hCat
Ka IC50 IC50
Lb IC50 Sb IC50
c
c
c
c
(nM)
(nM)
(nM) (nM)
2a
2e
2g
H
238
>100,000 63,060 41,082
>100,000 55,728 76,938
>100,000 17,043 66,656
CH2CH2SMe 13
CH2Ph 10
a Humanized rabbit enzyme.
b Human enzyme.
c IC50s are an average of at least two independent titrations. See Ref.
10 for assay conditions.
In this novel amide series of cathepsin K inhibitors,
introduction of the appropriate P1 substituent
resulted in a partial recovery of the potency loss
encountered when a nitrile warhead is replaced by a
primary amide. These compounds were shown to be
competitive substrates for cathepsin K, although they
are turned over at such a slow rate that they behave
like inhibitors. This series of compounds also showed
excellent selectivity against the cathepsins B, L,
and S.
The value of the catalytic constant kcat of the amide
compound toward the enzyme appears to be related to
the size of the substitution a to the amide. At the same
time, the affinity of these compounds (estimated Km) for
the active site is improved by the P1 substitution. These
combined effects result in the compounds docking
tightly in the active site; their hydrolysis rates are very
slow when compared to the synthetic substrate, leading
these compounds to behave like enzyme inhibitors.
Introducing P1 substitution also has the potential to im-
prove selectivity over other cathepsins. Selected exam-
ples of primary amide compounds are listed in Table 5
with inhibition activity against cathepsins B, L, and
S. The unsubstituted amide 2a displays a 170-fold selec-
tivity for cathepsin K versus cathepsin S. The most po-
tent cathepsin K inhibitors 2e and 2g both display
excellent selectivity with 2e being slightly superior with
regard to its cathepsin L selectivity (4300-fold vs 1700-
fold for 2g). The substitution in P1 improves potency
against cathepsin K while having little effect in cathepsin
B, L, and S.
References and notes
1. Troen, B. R. Drug News Perspect. 2004, 17, 19.
2. Lecaille, F.; Kaleta, J.; Bro¨mme, D. Chem. Rev. 2002, 102,
4459.
3. Yasuda, Y.; Kaleta, J.; Bro¨mme, D. Adv. Drug Deliv. Rev.
2005, 57, 973.
4. Grabowska, U. B.; Chambers, T. J.; Shiroo, M. Curr.
Opin. Drug Discov. Devel. 2005, 8(5), 619.
5. Black, W. C.; Bayly, C. I.; Davis, D. E.; Desmarais, S.;
´
´
Falgueyret, J. P.; Leger, S.; Li, C. S.; Masse, F.; McKay,
D. J.; Palmer, J. T.; Percival, M. D.; Robichaud, J.; Tsou,
N.; Zamboni, R. Bioorg. Med. Chem. Lett. 2005, 15, 4741.
These amide inhibitors can be prepared in one step via a
peptide coupling of the acid precursor 1312,13 and an
ˆ
6. Li, C. S.; Deschesnes, D.; Desmarais, S.; Falgueyret, J. P.;
Gauthier, J. Y.; Kimmel, D. B.; Leger, S.; Masse, F.;
´
´
McGrath, M.; McKay, D. J.; Percival, M. D.; Riendeau,
´
D.; Rodan, S. B.; Therien, M.; Truong, V. L.; Wesolow-
ski, G.; Zamboni, R.; Black, W. C. Bioorg. Med. Chem.
Lett. 2006, 16, 1985.
CF3
H2N CONH2
CO2H
N
7. Catalano, J. J.; Deaton, D. N.; Furfine, E. S.; Hassell, A.
M.; McFadyen, R. B.; Miller, A. B.; Miller, L. R.;
Shewchuk, L. M.; Willard, D. H., Jr.; Wright, L. L.
Bioorg. Med. Chem. Lett. 2004, 14, 275.
8. Altmann, E.; Aichholz, R.; Betschart, C.; Buhl, T.; Green,
J.; Lattmann, R.; Missbach, M. Bioorg. Med. Chem. Lett.
2006, 16, 2549.
9. The bound complex coordinates shown are obtained from
molecular dynamic simulations. Starting with the PDB
entry 1MEM for the Cat-k structure, a 700 ps trajectory
was performed on the covalently bound complex in TIP3P
water with periodic boundary conditions. The image is
generated by averaging coordinate snapshots of the last
100 ps.
10. Falgueyret, J. P.; Black, W. C.; Cromlish, W.;
Desmarais, S.; Lamontagne, S.; Mellon, C.; Riendeau,
D.; Rodan, S.; Tawa, P.; Wesolowski, G.; Bass, K.
E.; Venkatraman, S.; Percival, M. D. Anal. Biochem.
2004, 335, 218.
+
H
R
13
14
MeO2S
a
CF3
H
CONH2
R
N
N
H
O
MeO2S
2a-h
H
H2N CN
BocN CONH2
b, c
a
+
1a-d
12
R
R
15a-d
Scheme 1. Reagents and conditions: (a) HATU, Et3N, DMF, 18 h, rt,
75–95% yield; (b) TFAA, pyridine, 30 min, rt, 70–75% yield; (c)
methane sulfonic acid, THF, 18 h, rt, 60–75% yield.
11. Robichaud, J.; Oballa, R.; Prasit, P.; Falgueyret, J. P.;
Percival, M. D.; Wesolowski, G.; Rodan, S. B.; Kimmel,

´
S. Leger et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4328–4332
4332
D.; Johnson, C.; Bryant, C.; Venkatraman, S.; Setti, E.;
Mendonca, R.; Palmer, J. T. J. Med. Chem. 2003, 46,
M.; Truong, V.-L.; Green, M. J.; Hirschbein, B. L.;
Janc, J. W.; Palmer, J. T.; Baskaran, C. WO
075836A2, 2003.
3709.
12. Bayly, C. I.; Black, C.; Leger, S.; Li, C. S.; McKay,
´
D.; Mellon, C.; Gauthier, J. Y.; Lau, C.; Therien,
13. Roy, A.; Gosselin, F.; O’Shea, P. D.; Chen, C. Y. J. Org.
Chem 2006, 71, 4320.