Structure-based design of non-peptide, carbohydrazide-based cathepsin K inhibitors

Article abstract of DOI:10.1016/S0968-0896(99)00010-3

Using binding models which were based on the X-ray crystal structure of an amino acid-based active site-spanning inhibitor complexed with cathepsin K, Cbz-leucine mimics have been developed, leading ultimately to the design of a potent cathepsin K inhibitor free of amino acid components. These mimics, which consist of α-substituted biphenylacetyl groups in place of Cbz-leucine moieties, effectively mimic all aspects of the Cbz-leucine moieties which are important for inhibitor binding. The predicted directions of binding for the inhibitors were confirmed by mass spectral analysis of their complexes with cathepsin K, which gave results consistent with acylation of the enzyme and loss of the acylhydrazine portion of the inhibitor which binds on the S' side of the active site. The binding models were found to be very predictive of relative inhibitor potency as well as direction of inhibitor binding. These results strengthen the validity of a strategy involving iterative cycles of structure-based design and inhibitor synthesis and evaluation for the discovery of non-peptide inhibitors. Copyright (C) 1999 Elsevier Science Ltd.

Full text of DOI:10.1016/S0968-0896(99)00010-3

Bioorganic & Medicinal Chemistry 7 (1999) 599±605
Structure-based Design of Non-peptide, Carbohydrazide-based
Cathepsin K Inhibitors
Scott K. Thompson,a,* Stacie M. Halbert,a Renee L. DesJarlais,b,y
Thaddeus A. Tomaszek,c Mark A. Levy,c,{ David G. Tew,c Carl F. Ijamesb
and Daniel F. Vebera
aDepartment of Medicinal Chemistry, SmithKline Beecham Pharmaceuticals, 709 Swedeland Rd., King of Prussia, PA 19406, USA
bDepartment of Physical and Structural Chemistry, SmithKline Beecham Pharmaceuticals, 709 Swedeland Rd., King of Prussia,
PA 19406, USA
cDepartment of Molecular Recognition, SmithKline Beecham Pharmaceuticals, 709 Swedeland Rd., King of Prussia, PA 19406, USA
Received 6 October 1998
AbstractÐUsing binding models which were based on the X-ray crystal structure of an amino acid-based active site-spanning
inhibitor complexed with cathepsin K, Cbz-leucine mimics have been developed, leading ultimately to the design of a potent
cathepsin K inhibitor free of amino acid components. These mimics, which consist of a-substituted biphenylacetyl groups in place
of Cbz-leucine moieties, e€ectively mimic all aspects of the Cbz-leucine moieties which are important for inhibitor binding. The
predicted directions of binding for the inhibitors were con®rmed by mass spectral analysis of their complexes with cathepsin K,
which gave results consistent with acylation of the enzyme and loss of the acylhydrazine portion of the inhibitor which binds on the
S0 side of the active site. The binding models were found to be very predictive of relative inhibitor potency as well as direction of
inhibitor binding. These results strengthen the validity of a strategy involving iterative cycles of structure-based design and inhibitor
synthesis and evaluation for the discovery of non-peptide inhibitors. # 1999 Elsevier Science Ltd. All rights reserved.
Introduction
but contains many groups, such as amino acid residues,
with hydrogen-bonding potential thought to have a
negative impact on oral bioavailability. Based on these
considerations, a key aspect of the design of thera-
peutically useful enzyme inhibitors has been the replace-
ment of amino acid groups by isosteric elements. The
design of such elements has met with much success in
the recent past, owing greatly to the availability of
crystallographic data from enzyme:inhibitor complexes.
We have recently reported the structure-based design
and synthesis of compound 2 (Fig. 1), which contains a
meta-benzyloxy substituted benzoyl group as a mimic
for a benzyloxycarbonyl (Cbz) leucine moiety.5 The
design hypothesis was subsequently con®rmed by X-ray
crystallographic analysis of the complex of compound 2
and cathepsin K. However, the meta-benzyloxy sub-
stituted benzoyl group is an imperfect mimic of the Cbz-
leucine moiety in that it does not mimic the g-methyl
groups of the leucine side chain, as is evidenced by an
apparent second-order rate constant for inactivation of
cathepsin K for compound 2 which is roughly 10-fold
lower than that of compound 1. The X-ray crystal
structure of compound 2 in complex with cathepsin K
also supports the lack of mimicry of the g-methyl
groups of the leucine side chain. The present report
describes the structure-based design and synthesis of a
1,5-diacylcarbohydrazide-based inhibitor which contains
Cysteine proteases have been broadly implicated as tar-
gets for therapeutic intervention (e.g. cancer, arthritis,
viral and parasitic diseases).1 Cathepsin K, a cysteine
protease of the papain superfamily, has been implicated
in the process of bone resorption.2,3 Selective inhibitors
of cathepsin K could therefore be promising therapeutic
agents for the treatment of diseases characterized by
excessive bone loss, such as osteoporosis.
The structure-based design and X-ray crystallographic
analysis of a new class of active-site spanning cathepsin
K inhibitors which contain a 1,5-diacylcarbohydrazide
sca€old has recently been reported from these labora-
tories.4 Compound 1 (Fig. 1), for example, is a potent,
selective, kinetically irreversible inhibitor of cathepsin K
Key words: Cathepsin K; inhibitor; non-peptide; carbohydrazide;
structure-based design.
*Corresponding author. Tel: +1 610-270-5348; fax: +1 610-270-4451;
ye-mail: Scott_K_Thompson@sbphrd.com
Present address: 3-Dimentional Pharmaceuticals, Inc., 665 Stockton
Drive, Suite 104, Exton, PA 19341, USA.
{
Presently in the Department of Preclinical Pharmacokintics,
SmithKline Beecham Pharmaceuticals, 709 Swedeland Rd., King of
Prussia, PA 19406, USA.
0968-0896/99/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(99)00010-3

600
S. K. Thompson et al. / Bioorg. Med. Chem. 7 (1999) 599±605
was subsequently hydrogenated over Pd/C to provide
carboxylic acid 5. The (R)-isobutyl-substituted 3-
biphenylacetic acid ((R)-6) was prepared in six steps
from 3-bromophenylacetic acid.6
The mono-acyl carbohydrazides 13 and 14 used in the
preparation of the inhibitors were prepared as outlined
in Scheme 2. Preparation of methyl ester 8 was accom-
plished by benzylation of methyl 3-hydroxybenzoate (9).
Hydrazinolysis of methyl esters 77 and 88 provided
acylhydrazines 99 and 10, respectively, which were sub-
sequently treated with phosgene in re¯uxing toluene to
provide 1,3,4-oxadiazol-2-ones 11 and 12. Hydrazino-
lysis of 11 and 12 gave mono-acyl carbohydrazides 13
and 14, respectively.
Figure 1. Structures of compounds 1 and 2.
Compounds 15±18 were prepared as shown in Scheme
3. EDCl-promoted coupling of 13 with (R,S)-5 provided
compounds 15 and 16 as a 1:1 mixture which was sepa-
rated by HPLC. Likewise, EDCl-promoted coupling of
14 with (R,S)-5 gave (R,S)-17, and coupling of 14
with (R)-6 provided (R)-18. The generally low yields
obtained for compounds 15±18 are likely due to incom-
plete elution of the compound from the silica gel col-
umn as a result of their highly polar nature.
a substituted para-biphenylacetyl group in place of the
benzyloxycarbonyl leucine moiety, and which retains
the high potency of compound 1, presumably by com-
plete mimicry of the leucine side chain. We also report
the structure-based design of a substituted meta-biphenyl-
acetyl group as a replacement for the remaining Cbz-
leucine moiety, and the use of this mimic in combina-
tion with the meta-benzyloxy substituted benzoyl group
to construct a potent inhibitor which lacks any amino
acid residues.
Inhibitor design and evaluation
The failure of the meta-benzyloxybenzoyl group in
compound 2 to mimic the g-methyl groups of the S0 side
leucine side chain (nomenclature of Schechter and
Berger)10 appears to be largely a consequence of the ¯at
nature of the benzoyl group. To achieve the desired
mimicry of the entire Cbz-leucine moiety, the use of a
mimic that retains a stereocenter was explored. The posi-
tion of the phenyl portion of the Cbz group relative to the
leucine suggested a para-biphenylacetyl group wherein
substitution of an isobutyl group at the a-position
Results and Discussion
Inhibitor synthesis
The 2-isobutyl-substituted biphenylacetic acid 5 used
in the preparation of the inhibitors was prepared as
outlined in Scheme 1. Sequential treatment of 4-biphenyl-
acetic acid (3) with 2.2 equivalents of LDA and 3-
bromo-2-methylpropene gave carboxylic acid 4, which
Scheme 1. (a) i. 2.2 equiv. LDA (lithium diisopropylamide), THF; ii. 3-bromo-2-methylpropene, 72%; (b) H2, Pd-C, EtOAc, 95%.
.
Scheme 2. (a) Benzyl chloroformate, Na2CO3, 1,4-dioxane/H2O, 100%; (b) benzyl bromide, K2CO3, acetone, 98%; (c) H2NNH2 H2O, MeOH, 9,
ꢀ
.
100%; 10, 94%; (d) Cl2CO, toluene, 110 C; 11, 96%; 12, 96%; (e) H2NNH2 H2O, MeOH, 13, 60%; 14, 93%.

S. K. Thompson et al. / Bioorg. Med. Chem. 7 (1999) 599±605
601
.
Scheme 3. (a) EDC HCl, 1-HOBT, DMF, 15, 16, 42% combined; (R,S)-17, 53%; (R)-18, 36%.
would function to mimic the leucine side chain as in
compounds 15 and 16 (Scheme 3). Thus, a binding
model of this putative mimic was constructed using
SYBYL version 6.3 and minimized in MAXMIN using
the standard settings. The model was then ®tted inter-
actively into the cathepsin K active site by overlapping
the terminal phenyl group, the carbon bearing the iso-
butyl group and the carbonyl of the model with the
phenyl group, the a-carbon and the carbonyl, respec-
tively, of the S0 side Cbz-leucine moiety of compound 1
from the X-ray crystal structure of compound 1 bound
to cathepsin K.4 Torsional angles were then manually
adjusted to optimize the ®t of the model. Indeed, the
binding model (white) shows very good overlap with
both the phenyl group and side chain of the S0 side
Cbz-leucine moiety when superimposed with inhibitor 1
(Fig. 2). The internal phenyl group of the biphenyl sys-
tem appears to function as a spacer to properly orient
the terminal phenyl group for an aromatic±aromatic
interaction with Trp-184, a key interaction between
cathepsin K and this class of inhibitors, but also
appears to be oriented to participate in a face±face
aromatic±aromatic interaction with Trp-184. While the
aromatic±aromatic interaction between the terminal
phenyl group of the biphenyl system and Trp-184 sug-
gested by the binding model is an edge±face interaction
rather than the face±face interaction observed in the
X-ray crystal structure of the cathepsin K/1 complex,
this type of interaction has been observed on the S0
side of the active site in X-ray crystal structures of other
cathepsin K:inhibitor complexes.4,5,11 Thus the dia-
stereomeric mixture of compounds 15 and 16 was syn-
thesized and subsequently separated into individual
diastereomers. Although compounds 15 and 16 could be
eciently separated, they did, upon standing, undergo
some degree of epimerization at the a-stereocenter of
the substituted biphenylacetyl group (ꢁ10% after 12 h).
The two ``puri®ed'' diastereomers were therefore eval-
uated for cathepsin K inhibition as 90:10 mixtures.
Compounds 15 and 16 were found to inhibit cathepsin
Figure 2. Stereoview of the binding model for the 2-isobutyl-4-biphenylacetyl peptidomimetic (white) superimposed with inhibitor 1 (cyan). Selected
active site residues are colored by atom (C=grey, O=red, N=blue, S=yellow).

602
S. K. Thompson et al. / Bioorg. Med. Chem. 7 (1999) 599±605
K in a time-dependent manner, exhibiting apparent
second-order rate constants for inhibition of 2,900,000
and 650,000 M 1 s 1, respectively (Table 1). The equiva-
lent inhibition constants of compounds 1 and 15 indicate
that the a-isobutyl-para-biphenylacetyl group functions
very e€ectively as a mimic of all aspects of the S0 side
Cbz-leucine moiety which are important for inhibitor
binding. The absolute stereochemistry at the a-stereo-
center is presumably R, corresponding to the natural
amino acid stereochemistry, although this has not been
determined experimentally. Compound 16 exhibits 5-
fold lower potency than compound 15. Although, since
the sample of compound 16 contains about 10% of
compound 15, it is likely that about half of the inhibi-
tory potency exhibited by the 90:10 mixture is due to the
presence of compound 15. A similar decrease in potency
is observed for compound 2 relative to compound 1.
This result may mean that the isobutyl group in 16 is
Having designed an e€ective mimic of all aspects of the
S0 side Cbz-leucine moiety, which are important for
inhibitor binding, attention was turned towards the
design of a mimic of the remaining Cbz-leucine moiety.
The slightly less e€ective meta-benzyloxybenzoyl group
present in compound 2 rather than the a-isobutyl-para-
biphenylacetyl group present in 15 was chosen as the
S0 side Cbz-leucine mimic for evaluation of mimics of
the S side Cbz-leucine since it has been shown to bind
exclusively on the S0 side of the active site,4,5 thereby
maximizing the likelihood that the designed S side
Cbz-leucine mimics will bind in the active site as pro-
posed. Furthermore, the choice of this mimic alleviates
concerns about the stereochemical lability of the a-stereo-
center in the a-isobutyl-para-biphenylacetyl group.
Although a binding model of the a-isobutyl-para-
biphenylacetyl group (white), constructed as described
above, suggests that it would ®t into the active site in
place of the S side Cbz-leucine of compound 1 and
e€ectively mimic binding of the leucine side chain in the
P2 binding pocket, it does not place the terminal phenyl
group properly to mimic the interaction of the phenyl
portion of the Cbz-group with Tyr-67 (Fig. 3). Indeed,
racemic compound 17 is about 35-fold less potent than
its Cbz-leucine analogue 2. Further examination of the
binding model shown in Figure 3 suggested that a meta-
biphenylacetyl system (magenta) might allow for the
desired aromatic±aromatic interaction. This group has
previously proven to be e€ective as a mimic of the S side
Cbz-leucine in a structurally related class of 1,3-diamino-
ketone cathepsin K inhibitors.6 Since the model also
suggested that the R stereochemistry at the a-stereo-
center was desirable, the target compound (18) was
synthesized in enantiomerically pure R form. Unlike
compounds 15 and 16, compound 18 was not prone to
racemization at the benzylic stereocenter under the
conditions that promoted epimerization of the corres-
ponding stereocenter in 15 and 16. The greater stability
of the benzylic stereocenter of compound 18 relative to
that of compounds 15 and 16 is not fully understood at
present, but may be a result of the greater ability of the
para phenyl ring in 15 and 16 to stabilize a negative
charge at the benzylic position through resonance.
Nonetheless, compound 18 was found to be equivalent
in potency to compound 2 (Table 1), thus demonstrat-
ing e€ective mimicry of both the side chain and aro-
matic components of the Cbz-leucine moiety.
0
not e€ectively interacting with the P2 binding site.
Table 1. Inhibition of cathepsin K by compounds 1, 2 and 17±20
Compound
R1
R2
kobs/[I]
1 a
(M 1 s
)
1
3,100,000
2
224,000
15
2,900,000
16
17
650,000
Con®rmation of the direction of binding of inhibitors
6,300
Compound 1 exhibits time-dependent kinetics con-
sistent with an essentially irreversible mechanism of
inhibition. Mass spectral and NMR analysis of the
complex of 1 with cathepsin K give results consistent
with formation of an acyl enzyme intermediate with
cleavage of one of the acylhydrazine bonds and loss of
Cbz-leucinylhydrazine as reported previously.4 Mass
spectral analysis of the complex of compound 2 with
cathepsin K gave similar results with exclusive loss of
meta-benzyloxybenzoylhydrazine, indicating a single
direction of binding of the inhibitor.4 Consistent with
this hypothesis, X-ray crystallographic studies of the
complex of 2 with cathepsin K show both halves of the
18
240,000
aAssays were conducted as previously described.12

S. K. Thompson et al. / Bioorg. Med. Chem. 7 (1999) 599±605
603
Figure 3. Stereoview of the binding models for the 2-isobutyl-4-biphenylacetyl peptidomimetic (white) and 2-isobutyl-3-biphenylacetyl peptidomi-
metic (magenta) superimposed with inhibitor 1 (cyan). Selected active site residues are colored by atom (C=grey, O=red, N=blue, S=yellow).
inhibitor present in the active site with the inhibitor
bound across both S and S0 sides in a single orientation
and the peptidomimetic portion bound on the S0 side of
the active site. Thus, mass spectral analysis of com-
plexes of these inhibitors with cathepsin K appears to be
a useful analytical tool for prediction of the direction of
inhibitor binding based on the exclusive loss of the
acylhydrazine group that is bound on the S0 side of the
active site. Mass spectral analysis of the complex(es) of
cathepsin K with the mixture of inhibitors 15 and 16
showed exclusive formation of an acyl enzyme inter-
mediate containing Cbz-leucine (Mr=M+306.0) and
exclusive loss of a-isobutyl-para-biphenylacetylhydra-
zine (Mr=283.2 [M+H]+) (Fig. 4). These results are
consistent with the direction of binding predicted by
the design hypothesis. Likewise, the proposed binding
mode of compound 18 was con®rmed by mass spectral
analysis of its complex with cathepsin K which showed
exclusive formation of an acyl enzyme intermediate
containing the a-isobutyl-meta-phenylacetyl group
(Mr=M+309.6) and exclusive loss of meta-benzyloxy-
benzoylhydrazine (Mr=243.2[M+H]+) (Fig. 4). Thus,
the binding models have proven to be very predictive of
relative inhibitor potency as well as direction of inhi-
bitor binding.
Conclusions
Through the use of binding models, we have utilized the
crystal structure of a cathepsin K:inhibitor complex to
design Cbz-leucine replacements which e€ectively mimic
Figure 4. Products of the inhibition of cathepsin K by compounds 15/16 and compound 18.

604
S. K. Thompson et al. / Bioorg. Med. Chem. 7 (1999) 599±605
all aspects of both Cbz-leucine moieties in an amino
acid-based inhibitor (1), ultimately leading to the design
of a potent inhibitor free of amino acid components.
The predicted directions of binding for the inhibitors
were con®rmed by mass spectral analysis of their com-
plexes with cathepsin K, which gave results consistent
with acylation of the enzyme and loss of the acylhy-
drazine portion of the inhibitor which binds on the
S0 side of the active site. The binding models were found
to be very predictive of relative inhibitor potency as well
as direction of inhibitor binding. The strategy involving
iterative cycles of structure-based design and inhibitor
synthesis and evaluation has thus proven to be a viable
approach to obtain potent cathepsin K inhibitors
wherein amino acid elements have been replaced with
non-amino acid surrogates.
d 7.58 (m, 4H), 7.45 (m, 4H), 7.37 (m, 1H), 3.72 (d, 1H),
2.02 (m, 1H), 1.75 (m, 1H), 1.55 (m, 1H), 0.92 (d, 6H).
General procedure for the hydrazinolysis of methyl esters.
To a stirring 0.1 M solution of methyl ester in methanol
was added 10 equivalents of hydrazine hydrate. The
solution was allowed to stir at room temperature for
16 h then concentrated to yield the desired hydrazides.
Spectral data for compounds 9 and 10. N-Benzyloxy-
carbonyl-L-leucinylhydrazide (9). 1H NMR (CDCl3) d
8.32 (s b, 1H), 7.32 (m, 5H), 5.77 (d, 1H), 5.05 (q, 2H),
4.19 (m, 1H), 3.76 (s b, 2H), 1.56 (m, 3H), 0.89 (m, 6H).
1
3-Benzyloxybenzoylhydrazide (10). H NMR (CDCl3) d
7.45±7.27 (m, 9H), 7.24 (m, 1H), 5.12 (s, 2H), 4.10 (s b,
2H).
General procedure for the conversion of hydrazides to
1,3,4-oxadiazol-2-ones. To a stirring 0.2 M solution of
hydrazide in toluene was added 3 equivalents of phos-
gene as a 1.93 M solution in toluene. The solution was
heated at re¯ux for 4 h then concentrated to yield the
desired 1,3,4-oxadiazol-2-ones.
Experimental
Unless otherwise noted, materials were obtained from
commercial suppliers and were used without further
puri®cation. THF, 1,4-dioxane and DMF were anhy-
drous grade (Aldrich). All reactions involving organo-
metallic reagents were conducted under an atmosphere
of dry argon in oven-dried glassware. Electrospray
ionization mass spectra (MS (ESI)) were obtained with
a Sciex API-III triple quadrupole mass spectrometer. 1H
NMR spectra were obtained at 400 MHz. All extracts
were dried over anhydrous magnesium sulfate, and sol-
vents were removed with a Buchi rotary evaporator at
aspirator pressure. Flash chromatography using Kie-
selgel 60 silica gel was performed as previously reported.
Spectral data for compounds 11 and 12. (1S)-1-Benzyl-
oxycarbonylamino-3-methyl-1-(1,3,4-oxadiazol-2-on-5-yl)-
1
butane (11). H NMR (CDCl3) d 9.32 (s, 1H), 7.35 (m,
6H), 5.13 (m, 2H), 4.79 (m, 1H), 1.68 (m, 3H), 0.96 (m,
6H).
5-(3-Benzyloxyphenyl)-1,3,4-oxadiazol-2-one (12). 1H
NMR (CDCl3) d 7.41±7.28 (m, 9H), 7.06 (m, 1H), 5.06
(s, 2H).
(RS)-2-(4-Biphenyl)-4-methyl-4-pentenoic acid ((RS)-4).
To
a
stirring solution of N,N-diisopropylamine
General procedure for the conversion of 1,3,4-oxadiazol-
2-ones to acyl carbohydrazides. To a stirring 0.25 M
solution of 1,3,4-oxadiazol-2-one in methanol was added
10 equivalents of hydrazine hydrate. The solution was
allowed to stir at room temperature for 24h then con-
centrated and puri®ed by column chromatography (silica
gel, 8% methanol:dichloromethane) to yield the desired
acyl carbohydrazides.
(537 mg, 5.31 mmol, 0.74 mL) in 5 mL of THF, cooled
to 0 ꢀC, was added dropwise 2.1 mL (5.22 mmol) of a
2.5 M solution of n-butyllithium in hexanes. After stir-
ring for 15 min at 0 ꢀC, the mixture was cooled to
78 ꢀC and a solution of 4-biphenylacetic acid (3)
(500 mg, 2.36 mmol) in 2 mL of THF was added drop-
wise. After warming to 0 ꢀC and again cooling to
78 ꢀC, 3-bromo-2-methylpropene (485 mg, 3.54 mmol,
0.36 mL) was added to the mixture in one portion. After
stirring at 78 ꢀC for 1 h, the reaction was quenched
with 2 mL of water then concentrated. The residue
was re-dissolved in water and extracted with ether
(1Â100 mL). The aqueous layer was acidi®ed (3N HCl)
and extracted with ether (3Â100 mL). The organic lay-
ers were combined, dried (MgSO4), ®ltered and con-
centrated to yield 449 mg (72%) of (RS)-4 as a white
Spectral data for compounds 13 and 14. 2-[N-(N-Benzyl-
oxycarbonyl-L-leucinyl)]carbohydrazide (13). 1H NMR
(CDCl3) d 7.30 (m, 5H), 6.29 (d, 1H), 5.10 (d, 1H), 4.95
(d, 1H), 4.36 (m, 1H), 1.58 (m, 3H), 0.87 (m, 6H).
2-[N-(3-Benzyloxybenzoyl)]carbohydrazide (14). 1H NMR
(CD3OD) d 7.54 (m, 1H), 7.47 (m, 3H), 7.38 (m, 3H),
7.31 (m, 1H), 7.20 (m, 1H), 5.14 (s, 2H).
1
solid: H NMR (CDCl3) d 11.3 (bs, 1H), 7.69 (m, 4H),
7.55 (m, 4H), 7.46 (m, 1H), 4.89 (d, 2H), 4.01 (t, 1H),
3.01 (dd, 1H), 2.59 (dd, 1H), 1.84 (s, 3H).
General procedure for the EDC-promoted coupling of
acyl carbohydrazides with carboxylic acids. To a stirring
0.15 M solution of acyl carbohydrazide in DMF was
added 1.1 equivalents of carboxylic acid, 0.2 equivalents
of 1-hydroxybenzotriazole and 1.1 equivalents of 1-(3-
dimethylaminopropyl)-3-ethylcarbodiimide hydrochlo-
ride. After stirring at room temperature for 16 h, the
solution was diluted with ethyl acetate and washed
successively with saturated aqueous sodium bicarbo-
nate, water (2Â), and brine. The organic layer was
(RS)-2-(4-Biphenyl)-4-methylpentanoic acid ((RS)-5). To
a stirring solution of (RS)-4 (449 mg, 1.69 mmol) in
25 mL of ethyl acetate was added 225 mg of 10% palla-
dium on carbon. After stirring under a balloon of
hydrogen for 16 h, the mixture was ®ltered through
Celite. The ®ltrate was concentrated to yield 430 mg
(95%) of (RS)-5 as an o€ white solid: 1H NMR (CDCl3)

S. K. Thompson et al. / Bioorg. Med. Chem. 7 (1999) 599±605
605
dried, ®ltered and concentrated. The residue was puri®ed
Acknowledgements
by column chromatography (silica gel, 5% methanol:
dichloromethane) to yield the desired diacyl carbohy-
drazides.
The authors thank Karl Erhard for the HPLC separa-
tion of compounds 15 and 16.
Spectral data for diacyl carbohydrazides. (20R)-2-[N-(N-
Benzyloxycarbonyl-L-leucinyl)]-20-[N0-[2-(4-biphenyl)-4-
methylpentanoyl)]]carbohydrazide and (20S)-2-[N-(N-
benzyloxycarbonyl-L-leucinyl)]-20 -[N0 -[2-(4-biphenyl)-4-
References and Notes
1. McKerrow, J. H.; James, M. N. G., Eds.; Perspectives in
Drug Discovery and Design 1997, 6, 1±125.
2. Drake, F. H.; Dodds, R. A.; James, I. E.; Connor, J. R.;
Debouck, C.; Richardson, S.; Lee-Rykaczewski, E.; Coleman,
L.; Rieman, D.; Barthlow, R.; Hastings, G.; Gowen, M. J.
Biol. Chem. 1996, 271, 12511±12516.
3. Bossard, M. J.; Tomaszek, T. A.; Thompson, S. K.; Ame-
gadzie, B. Y.; Hanning, C. R.; Jones, C.; Kurdyla, J. T.;
McNulty, D. E.; Drake, F. H.; Gowen, M.; Levy, M. A. J.
Biol. Chem. 1996, 271, 12517±12524.
4. Thompson, S. K.; Halbert, S. M.; Bossard, M. J.; Tomaszek,
T. A.; Levy, M. A.; Zhao, B.; Smith, W. W.; Abdel-Meguid, S.
S.; Janson, C. A.; D'Alessio, K. J.; McQueney, M. S.; Ame-
gadzie, B. Y.; Hanning, C. R.; DesJarlais, R. L.; Briand, J.;
Sarkar, S. K.; Huddleston,M. J.; Ijames, C. F.; Carr, S. A.;
Garnes, K. T.; Shu, A.; Heys, J. R.; Bradbeer, J.; Zembryki,
D.; Lee-Rykaczewski, L.; James, I. E.; Lark, M. W.; Drake, F.
H.; Gowen, M.; Gleason, J. G.; Veber, D. F. Proc. Natl. Acad.
Sci., USA 1997, 94, 14249±14254.
1
methylpentanoyl)]]carbohydrazide (15 and 16). H NMR
(CDCl3/CD3OD) d 7.48 (m, 5H), 7.35 (m, 5H), 7.23 (m,
4H), 4.92 (m, 2H), 4.08 (m, 1H), 3.54 (m, 1H), 1.91 (m,
1H), 1.74±1.40 (m, 5H), 0.84 (m, 12H).
Compounds 15 and 16 were separated by HPLC
(Sumipax OA-3100, 4.6Â150 mm, 80:20 hexane:ethanol,
1.0 mL min 1). Compound 15: retention time=5.9 min;
compound 16: retention time=8.1 min.
(20RS)-2-[N-(3-Benzyloxybenzoyl)]-20-[N0-[2-(4-biphenyl)-
1
4-methylpentanoyl)]]carbohydrazide ((RS)-17). H NMR
(CDCl3/CD3OD) d 7.45 (m, 4H), 7.35 (m, 9H), 7.26 (m,
4H), 7.04 (dd, 1H), 4.98 (s, 2H), 3.59 (t, 1H), 1.92 (m,
1H), 1.71 (m, 1H), 1.44 (m, 1H), 0.85 (m, 6H).
5. Thompson, S. K.; Smith, W. W.; Zhao, B.; Halbert, S. M.;
Tomaszek, T. A.; Tew, D. G.; Levy, M. A.; Janson, C. A.;
D'Alessio, K. J.; McQueney, M. S.; Kurdyla, J.; Jones, C. S.;
DesJarlais, R. L.; Abdel-Meguid, S. S.; Veber, D. F. J. Med.
Chem. 1998, 41, 3923±3927.
6. DesJarlais, R. L.; Yamashita, D. S.; Oh, H.-J.; Uzinskas,
I. N.; Erhard, K. F.; Allen, A. C.; Haltiwanger, R. C.; Zhao,
B; Smith, W. W.; Abdel-Meguid, S. S.; D'Alessio, K. J.; Jan-
son, C. A.; McQueney, M. S.; Tomaszek, T. A.; Levy, M. A.;
Veber, D. F. J. Am. Chem. Soc. 1998, 120, 9114±9115.
7. Yamada, T.; Isono, N.; Inui, A.; Miyazawa, T.; Kuwata,
S.; Watanabe, H. Bull. Chem. Soc. Jpn. 1978, 51, 1897±1898.
8. Subramanyam, C.; Bell, M. R.; Ferguson, E; Gordon,
R. G.; Dunlap, R. P.; Franke, C.; Mura, A. J. Bioorg. Med.
Chem. Lett. 1995, 5, 319±324.
(20R)-2-[N-(3-Benzyloxybenzoyl)]-20-[N0-[2-(3-biphenyl)-4-
methylpentanoyl)]]carbohydrazide ((R)-18). 1H NMR
(CDCl3/CD3OD) d 7.48 (m, 2H), 7.41±7.24 (m, 15H),
7.03 (m, 1H), 4.96 (s, 2H), 3.59 (t, 1H), 1.92 (m, 1H),
1.69 (m, 1H), 1.43 (m, 1H), 0.81 (m, 6H).
Mass spectral analysis of compounds 15/16 and
compound 18
Cathepsin K (27 mM in MES bu€er at pH 6.0) was
activated as previously described13 and stored at 70 to
40 ꢀC. LC/MS analysis revealed the observed mole-
cular weights of 23645.4 and 23700.3 for the N-terminal
sequences beginning RAPD and GRAPD, in approxi-
mately a 5:4 ratio (calculated masses are 23645.6 and
23702.7 Da, respectively). Incubations were performed
by adding 5 nmol compound in DMSO to 1 nmol pro-
tein and sucient water so that the ®nal protein con-
centration was 13.5 mM and the DMSO concentration
was 5% (compounds 15 and 16) or 10% (compound
18). After vortexing, the solutions were allowed to stand
at room temperature for 60 and 90 min, respectively.
Then, they were loaded onto a peptide trap cartridge
(Michrom), washed with 50 mL of 0.1% TFA/water
(solvent A), and back-eluted onto a 1Â50 mm Inertsil
C18 reversed phase HPLC column with a gradient from
5% to 95% solvent B (0.1% TFA:90% MeCN:10%
water) in 10 min. The mass spectrometer was scanned
from m/z 200 to 1800 once every 3.9 s.
9. Szarvasi, E.; Fontaine, L.; Betbeder-Matibet, A. J. Med.
Chem. 1973, 16, 281±287.
10. Schechter, I.; Berger, A. Biochem. Biophys. Res. Commun.
1967, 27, 157±162.
11. Yamashita, D. S.; Smith, W. W.; Zhao, B.; Janson, C. A.;
Tomaszek, T. A.; Bossard, M. J.; Levy, M. A.; Oh, H.-J.;
Carr, T. J.; Thompson, S. K.; Ijames, C. F.; Carr, S. A.;
McQueney, M. S.; D'Alessio, K. J.; Amegadzie, B. Y.; Hanning,
C. R.; Abdel-Meguid, S. S.; DesJarlais, R. L.; Gleason, J. G.,
Veber, D. F. J. Am. Chem. Soc. 1997, 119, 11351.
12. Votta, B.; Levy, M. A.; Badger, A.; Bradbeer, J.; Dodds,
R. A.; James, I. E.; Thompson, S.; Bossard, M. J.; Carr, T.;
Connor, J. R.; Tomaszek, T. A.; Szewczuk, L.; Drake, F. H.;
Veber, D. F.; Gowen, M. J. Bone Miner. Res. 1997, 12, 1396.
13. McQueney, M. S.; Amegadzie, B. Y.; D'Alessio, K. D.;
Hanning, C. R.; McLaughlin, M. M.; McNulty, D.; Carr, S.
A.; Ijames, C.; Kurdyla, J.; Jones, C. S. (1997) J. Biol. Chem.
272, 13955±13960.