Arylaminoethyl amides as inhibitors of the cysteine protease cathepsin K - Investigating P1′ substituents

Article abstract of DOI:10.1016/S0960-894X(03)00344-5

Modeling, synthesis and in vitro activities of a series of arylaminoethyl amide based inhibitors of the cysteine protease cathepsin K are described.

Full text of DOI:10.1016/S0960-894X(03)00344-5

Bioorganic & Medicinal Chemistry Letters 13 (2003) 1997–2001
Arylaminoethyl Amides as Inhibitors of0the Cysteine Protease
Cathepsin K—Investigating P1 Substituents
Eva Altmann,a,* Jonathan Greena and Marina Tintelnot-Blomleyb
aArthritis & Bone Metabolism Therapeutic Area, Novartis Pharma AG, CH-4002 Basel, Switzerland
bNervous System Therapeutic Area, Novartis Pharma AG, CH-4002 Basel, Switzerland
Received 23 December 2002; revised 18 March 2003; accepted 25 March 2003
Abstract—Modeling, synthesis and in vitro activities of a series of arylaminoethyl amide based inhibitors of the cysteine protease
cathepsin K are described.
# 2003 Elsevier Science Ltd. All rights reserved.
The identification of cysteine proteases which are selec-
tively expressed in specific tissues has raised consider-
able interest in this class of enzymes as potential
therapeutic targets.1À4 As a result, the search for novel,
potent and selective cysteine protease inhibitors has
evolved into a highly competitive area of research.5 In
thiscontext we have recently reported the discovery of
arylaminoethyl amidesasnovel, non-covalent inhibi-
tors6 of cathepsin K. Cathepsin K,7 a lysosomal cysteine
protease abundant in osteoclasts, exhibits potent col-
lagenolytic activity against type I collagen and is also
involved in the degradation of other important bone
proteins.8 The crucial role cathepsin K plays in bone
resorption is demonstrated by data obtained from
cathepsin K null mice9 which display an osteopetrotic
phenotype in the absence of any other overt pathologi-
cal signs. Mutation of the gene expressing cathepsin K
in humans results in pycnodysostosis,10 a rare bone dis-
ease characterized by increased bone density, osteo-
sclerosis and bone fragility. These observations indicate
that cathepsin K inhibition might be useful for the
treatment of diseases characterized by excessive bone
loss, such as osteoporosis.
substituents and the replacement of the P3 benzyloxy-
carbonyl group by different acyl moieties.
Molecular modeling of lead 1a into the active site of
h cathepsin K (Fig. 2) indicatesaromatic–aromatic
Figure 1. Schematic representation of the assumed binding mode of
inhibitor 1a within the active site of h cathepsin K. The Cbz moiety is
assumed to bind in the S3 pocket, the isobutyl group of leucine within
the hydrophobic S2 binding pocket and the 4-methoxy-phenyl ring
0
extendsinto the S 1 site.
Aspart of our ongoing effortsin the area of arylamino-
ethyl amide-based cathepsin K inhibitors we have now
further probed the SAR around our lead structure ben-
zyloxycarbonyl-l-leucine 4-methoxy-phenylamino-ethyl-
0
amide 1a (Fig. 1) through incorporation of extended P1
*Corresponding author. Tel.: +41-61-696-1227; fax: +41-61-696-
6071; e-mail: eva.altmann@pharma.novartis.com
Figure 2. Compound 1a docked into the h cathepsin K active site.11
0960-894X/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00344-5

1998
E. Altmann et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1997–2001
treatment with the appropriate alcohol, Ph3P and die-
thyl azodicarboxylate. Simultaneouscleavage of both
Boc groupswith 4N HCl in Et O furnished diamines
2
5a–l, whose reaction with Cbz-l-leucine-N-hydroxy-
succinimide ester resulted in exclusive acylation of the
primary amino group to yield the desired secondary
amides 1a–l. Compound 1m wasobtained through
direct acylation of 3 with Cbz-l-leucine-N-hydroxy-
succinimide ester, while phenoxy derivative 1n wassyn-
thesized via Borch reduction of 4-phenoxyaniline with t-
butyl-N-(2-oxoethyl)carbamate followed by cleavage of
the Boc-group and again reaction of the resulting amine
with Cbz-l-leucine-N-hydroxysuccinimide ester. In
order to vary the acyl moiety attached to the leucine
amino group, the Cbz group wasremoved from analo-
gues 1a, b and f by catalytic hydrogenation and the
resulting intermediates 6a, b, f were converted into
inhibitors 7a–h, j–l by BOP-mediated coupling with
benzoic acidsor 1H-indole-2-carboxylic acid ( Scheme 2).
Analogues 7i and 7m were prepared by reaction of
amine 3 with Boc-l-leucine (EDC/HOBt) followed by
treatment of the coupling productswith neat formic
acid and subsequent BOP-mediated coupling of the
resulting amines with 4-ethyl benzoic acid and 1H-
indole-2-carboxylic acid, respectively.
Figure 3. Compound 1m docked into the h cathepsin K active site.11
interaction between Tyr 67 and the Cbz phenyl ring in
S3 and tight binding of the isobutyl side chain of leucine
into the lipophilic S2 pocket. Furthermore, from this
model we hypothesized that modifying 1a to extend
0
deeper into the S1 site could improve potency and/or
selectivity of this class of compounds (Fig. 3).
In thiscommunication we report on the effectsof these
modificationson the potency and selectivity for cathep-
sin K inhibition.
The synthesis of these analogues proceeded through the
common intermediate 4, which wasobtained through
decarboxylative ring opening of 2-oxazolidinone13 with
4-benzyloxyaniline 2 (Scheme 1). Bis-Boc protection of
the resulting 2-aminoethyl-4-benzyloxy aniline 3 fol-
lowed by hydrogenolytic removal of the benzyl moiety
provided the free phenol, which was O-alkylated with
different bromidesin the preesnce of Cs 2CO3 or by
The effectsof 4-alkoxy-substituentson the aminophenyl
ring on inhibitory potency and the selectivity profile are
summarized in Table 1. Compounds 1a–f, in which the
ether substituent R1 wasvaried, exhibit ismilar IC
50
valuesin the range of 36–81 nM. The preesnce of a
phenoxy substituent in compound 1n leadsto a 6-fold
reduced potency ascompared to the correpsonding
cyclohexyloxy derivative 1d. In contrast, the activity of
Scheme 1. (a) 2-oxazolidinone, 2-(2-methoxyethoxy)ethanol, 160–175 ꢀC, 16 h, 20%; (b) Boc2O, N-ethyl-diisopropylamine, DMF, room temp, 18 h,
84%; (c) H2, Pd/C (10%), MeOH, room temp, 3 h, 98%; (d) R1-Br, Cs2CO3, DMF, 16–20 h, 70–98% (5a–c, e, j–l) or R1-OH, Ph3P, DEAD, THF,
5–6 h, 52–68% (5d,f–i); (e) 4N HCl in Et2O, room temp, 2 h, 92–98%; (f) Cbz-Leu-OSu, N-ethyl-diisopropylamine, DMF, room temp, 16 h, 59–
91%. For identities of substitutents R1 cf. Table 1.
Scheme 2. (a) H2, Pd/C (10%), MeOH, room temp, 1 h, 93–98%; (b) R2-COOH, benzotriazol-1-yloxy-tris(dimethylamino)phosphonium hexa-
fluorophosphate (BOP), N-ethyl-diisopropylamine, DMF, room temp, 63–95%.

E. Altmann et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1997–2001
1999
Table 1. Inhibition of rh cathepsins K, L and S catalytic activity—Variation of the alkoxy substituent
Compd
R1
h Cath K
IC50 (mM)a
h Cath L
IC50 (mM)b
h Cath S
IC50 (mM)c
1a
0.070
4.0
4.0
1b
1c
1d
1e
1f
0.036
>10 (41%)
6.0
1.2
4.4
0.081
0.043
10
1.2
0.069
>10 (40%)
>10 (43%)
>10 (31%)
>10 (40%)
10
2.8
0.049
1.1
1m
1n
1g
1h
1i
0.054
9.5
0.280
5.8
0.152
3.4
0.190
10
4.6
>1.0 (43%)
>1.0 (27%)
>1.0 (40%)
0.855
10
4.6
1j
>10 (29%)
>10 (43%)
>10 (45%)
>10 (34%)
9.5
1k
1l
5.7
aInhibition of recombinant human (rh) cathepsin K activity in a fluoresence assay using 48 mM Cbz-Phe-Arg-AMC as substrate in 100 mM
NaH2PO4, 1 mM EDTA, 20 mM Tween 80, 2 mM DTT, pH 7.
bInhibition of rh cathepsin L activity using 3 mM Cbz-Phe-Arg-AMC as substrate in 100 mM NaOAc, 1 mM EDTA, 0.005% Brig 35, 1 mM DTT,
pH 5.5.
cInhibition of rh cathepsin S activity using 11 mM Cbz-Leu-Leu-Arg-AMC as substrate in 100 mM NaOAc, 1 mM EDTA, 0.01% Triton X-100,
1mM DTT, pH 5.5. Data represent mean of 2 experiments performed in duplicate, individual data points in each experiment were within a 3-fold
range with each other.
benzyloxy-containing analogue 1m (IC50 for cathepsin
K of 54 nM) issimilar to that of cyclopentylmethyloxy
and cyclohexyloxy derivatives 1f and 1d. These data
selectivity profile of all Cbz-derivatives studied (>400
fold for cathepsin K over cathepsins L and 176–fold for
cathepsin K over S).
0
show clearly that extension into the S1 site with lipo-
0
philic substituents does not improve the potency of our
inhibitors. However, compounds with larger alkoxy
groups display high selectivity for cathepsin K over
cathepsin L14 while selectivity against cathepsin S14 is
not affected. We identified 1m asexhibiting the bets
Incorporation of a heteroatom into the P1 subunit, such
asthe replacement of the benzyloxy- by a 4-pyr-
idinylmethoxy substituent (1m!1g) and the substitu-
tion of a 4-tetrahydro-pyranyloxy moiety for the
cyclohexyloxy group (1d!1h) led to slightly reduced

2000
E. Altmann et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1997–2001
0
Table 2. Inhibition of rh cathepsins K, L and S catalytic activity—Variations in P3/P1
Compd
R2
R1
h Cath Ka
h Cath Lb
h Cath Sc
IC50 (mM)
IC50 (mM)
IC50 (mM)
7a
7b
7c
7d
7e
7f
7g
7h
7i
Phenyl
CH3
CH3
CH3
CH3
0.490
0.372
0.044
0.419
0.009
0.15
<0.03 (67%)
<0.03 (52%)
0.13
0.32
0.07
0.58
0.47
1.55
0.48
0.53
0.66
3-Methylphenyl
4-Methylphenyl
3-Methoxyphenyl
4-Methoxyphenyl
4-Methoxyphenyl
4-Methoxyphenyl
4-Ethylphenyl
4-Ethylphenyl
1H-Indol-2-yl
CH3
0.80
0.58
0.75
0.03
1.3
0.09
CH2CH(CH3)2
CH2(C5H9)
CH3
CH2(C6H5)
CH3
<0.003 (71%)
<0.003 (56%)
<0.003 (80%)
<0.003 (72%)
<0.003
4.9
0.16
7j
7k
7l
7m
1H-Indol-2-yl
1H-Indol-2-yl
1H-Indol-2-yl
CH2CH(CH3)2
CH2(C5H9)
CH2(C6H5)
<0.003 (71%)
0.003
0.004
>10 (22%)
>10 (27%)
6.9
>10 (46%)
7.5
4.5
aInhibition of recombinant human (rh) cathepsin K activity in a fluoresence assay using 48 mM Cbz-Phe-Arg-AMC as substrate in 100 mM
NaH2PO4, 1 mM EDTA, 20 mM Tween 80, 2 mM DTT, pH 7.
bInhibition of rh cathepsin L activity using 3 mM Cbz-Phe-Arg-AMC as substrate in 100 mM NaOAc, 1 mM EDTA, 0.005% Brig 35, 1 mM DTT,
pH 5.5.
cInhibition of rh cathepsin S activity using 11 mM Cbz-Leu-Leu-Arg-AMC as substrate in 100 mM NaOAc, 1 mM EDTA, 0.01% Triton X-100, 1
mM DTT, pH 5.5. Data represent mean of 2 experiments performed in duplicate, individual data points in each experiment were within a 3-fold
range with each other.
inhibitory potency. Asexpected from modeling, the
introduction of a 1-methyl- piperidinyloxy, a dimethyl-
amino-, a pyrrolidinyl- or imidazolyl-ethoxy sub-
highly homologuescat0hepsinsL and S. The appropriate
combination of P3/P1 subunits results in highly potent
cathepsin K inhibitors with an excellent selectivity
profile.
0
stituent in the P1 -fragment resulted in significantly less
potent inhibitors( 1i–1l), most likely due to the polar
0
group which cannot adjust within the hydrophobic S1
pocket.
Acknowledgements
As illustrated by the results summarized in Table 2,
replacement of the Cbz group by appropriate acyl moi-
etiesafforded a esriesof highly potent cathepisn K
inhibitors( 7e–m). Replacement of the Cbz moiety by an
unsubstituted benzamide (1a!7a) led to a loss in activ-
ity and selectivity. Introduction of a meta substituent at
the benzamide moiety (7b, 7d) reversed the selectivity
profile in favour of cathepsins S and L. However, a
4-substituted benzamide or 1H-indolyl moiety as the P3
substituent provides analogues with activities in the low
nanomolar range. Within thisesriesof inhibitorsthe
The authorsthank M. Bpising, D. Huerzeler, G.
Schreiber and Y. Seltenmeyer for excellent technical
assistance.
References and Notes
1. Marquis, R. W. Annu. Rep. Med. Chem. 2000, 35, 309.
2. Leung, D.; Abbenante, G.; Fairlie, D. P. J. J. Med. Chem.
2000, 43, 305.
3. Veber, D. F.; Thompson, S. K. Curr. Opin. Drug Discov.
Dev. 2000, 3, 362.
4. Bromme, D. Drug News Perspect 1999, 12, 73.
0
size of the P1 substituent had an impact on the selec-
tivity profile. Inhibitors 7e, h and j all incorporating a
4-methoxy substituent on the aminophenyl ring are the
least selective for cathepsin K over cathepsins L and S.
5. Recent publications on cysteine protease inhibitors: (a)
Falgueyret, J. P.; Oballa, R. M.; Okamoto, O.; Weso-
lowski, G.; Aubin, Y; Rydzewski, R. M.; Prasit, P.;
Riendeau, D.; Rodan, S. B.; Percival, M. D. J. Med.
Chem. 2001, 44, 94. (b) Marquis, R. W.; Ru, Y.; LoCastro,
S. M.; Zeng, J.; Yamashita, D. S.; Oh, H.-J.; Erhard, K. F.;
Davis, L. D.; Tomaszek, T. A.; Tew, D.; Salyers, K.; Proksch,
J.; Ward, K.; Smith, B.; Levy, M; Cummings, M. D.; Halti-
wanger, R. C.; Trescher, G.; Wang, B; Hemling, M. E.; Quinn,
C. J.; Cheng, H.-Y.; Lin, F.; Smith, W. W.; Janson, C. A.;
Zhao, B.; McQueney, M. S.; D’Alessio, K.; Lee, C.-P.; Mar-
zulli, A.; Dodds, R. A.; Blake, S.; Hwang, S.-M.; James, I. E;
Gress, C. J.; Bradley, B. R.; Lark, M. W.; Gowen,
M.; Veber, D. F. J. Med. Chem. 2001, 44, 725. (c)
Marquis, R. W.; Ru, Y.; LoCastro, S. M.; Zeng, J.; Trout,
R. E. L.; LoCastro, S. M.; Gribble, A. D.; Witherington, J;
Fenwick, A. E.; Garnier, B.; Tomaszek, T; Tew, D; Hemling,
D; Quinn, C. J.; Smith, W. W.; Zhao, B.; McQueney, M. S.;
Janson, C. A.; D’Alessio, K; Veber, D. F. J. Med. Chem. 2001,
44, 1380. (d) Huang, L.; Lee, A.; Ellman, J. A. J. Med. Chem.
2002, 45, 676.
0
Extended and lipophilic P1 substituents such as an iso-
butyloxy, a methylcyclopentyloxy- or a benzyloxy moi-
ety in compounds 7f, g, i, k–m were found to be
significantly more selective inhibitors for cathepsin K.
Out of thisesriescompound
7k emerged asthe mots
attractive analoge investigated, which apart from high
potency for cathepsin K inhibition exhibits the most
favorable selectivity profile (>3300-fold for cathepsin K
over cathepsins L and S).
In conclusion we have described the synthesis and in
vitro activitiesof a seriesof Nꢀ-benzyloxycarbonyl- and
Nꢀ-acyl-L-leucine(2-phenylaminoethyl)amide
deriva-
substituents.
0
tiveswhich incorporate 0 extended P
1
Expanded lipophilic P1 moitiesdo not improve the
potency of our inhibitors, however they generally lead
to an increased specificity for cathepsin K over the two

E. Altmann et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1997–2001
2001
6. Altmann, E.; Renaud, J.; Green, J.; Farley, D.; Cutting, B.;
Jahnke, W. J. Med. Chem. 2002, 45, 2352.
10. Hou, W.-S.; Bromme, D.; Zhao, Y.; Mehler, E.; Dushey,
C.; Weinstein, H.; Miranda, C. S.; Fraga, C.; Greig, F.; Carey,
J.; Rimoin, D. L.; Desnick, R. J.; Gelb, B. D. J. Clin. Invest.
1999, 103, 731.
11. Compound 1a and 1m were docked in the cathepsin K/
vinylsulfone inhibitor co-crystal structure (PDB Refc. 1MEM)
using MCDOCK in the QPX12 program.
12. McMartin, C.; Bohacek, R. S. J. Comp.-aided Mol. Des.
1997, 11, 333.
13. Pointdexter, G. S.; Owen, D. A.; Dolen, P. L.; Woo, E. J.
Org. Chem. 1992, 57, 6257.
7. (a) Inaoka, T.; Bilbe, G.; Ishibashi, O.; Tezuka, K.;
Kumegawa, M.; Kokubo, T. Biochem. Biophys. Res. Commun.
1995, 206, 89. (b) Drake, F. H.; Dodds, R. A.; James, I. E.;
Connor, J. R.; Debouck, C.; Richardson, S.; Lee-Rykac-
zewski, E.; Coleman, L.; Rieman, D.; Barthlow, R.; Hastings,
G.; Gowen, M. J. Biol. Chem. 1996, 271, 12511.
8. Garnero, P.; Borel, O.; Byrjalsen, I.; Ferreras, M.; Drake,
F. H.; McQueney, M. S.; Foged, N. T.; Delmas, P. D.;
Delaisse, J.-M. J. Biol. Chem. 1998, 273, 32347.
9. Saftig, P.; Hunziker, E.; Wehmeyer, O.; Jones, S.; Boyde,
A.; Rommerskirch, W.; Moritz, J. D.; Schu, P.; Von Figura,
K. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 13453.
14. Cathepsin K shares the highest homology in DNA and
amino acid sequence to cathepsins L and S. Bromme, D.;
Okamoto, K. Biol. Chem. Hoppe-Seyler 1995, 376, 379.