Potent dipeptidylketone inhibitors of the cysteine protease cathepsin K

Article abstract of DOI:10.1016/S0968-0896(99)00011-5

Cathepsin K (EC 3.4.22.38) is a cysteine protease of the papain superfamily which is selectively expressed within the osteoclast. Several lines of evidence have pointed to the fact that this protease may play an important role in the degradation of the bone matrix. Potent and selective inhibitors of cathepsin K could be important therapeutic agents for the control of excessive bone resorption. Recently a series of peptide aldehydes have been shown to be potent inhibitors of cathepsin K. In an effort to design more selective and metabolically stable inhibitors of cathepsin K, a series of electronically attenuated alkoxymethylketones and thiomethylketones inhibitors have been synthesized. The X-ray co-crystal structure of one of these analogues in complex with cathepsin K shows the inhibitor binding in the primed side of the enzyme active site with a covalent interaction between the active site cysteine 25 and the carbonyl carbon of the inhibitor. Copyright (C) 1999 Elsevier Science Ltd.

Full text of DOI:10.1016/S0968-0896(99)00011-5

Bioorganic & Medicinal Chemistry 7 (1999) 581±588
Potent Dipeptidylketone Inhibitors of the Cysteine Protease
Cathepsin K
Robert W. Marquis,a,* Yu Ru,a Dennis S. Yamashita,a Hye-Ja Oh,a Jack Yen,a
Scott K. Thompson,a Thomas J. Carr,a Mark A. Levy,b,y Thaddeus A. Tomaszek,b
Carl F. Ijames,c Ward W. Smith,d Baoguang Zhao,d Cheryl A. Janson,d
Sherin S. Abdel-Meguid,d Karla J. D'Alessio,e M. S. McQueneye and Daniel F. Vebera
aDepartment of Medicinal Chemistry, SmithKline Beecham Pharmaceuticals, 709 Swedeland Rd., King of Prussia, PA 19406, USA
bDepartment of Molecular Recognition, SmithKline Beecham Pharmaceuticals, 709 Swedeland Rd., King of Prussia, PA 19406, USA
cDepartment of Physical and Structural Chemistry, SmithKline Beecham Pharmaceuticals, 709 Swedeland Rd., King of Prussia,
PA 19406, USA
dDepartment of Structural Biology, SmithKline Beecham Pharmaceuticals, 709 Swedeland Rd., King of Prussia, PA 19406, USA
eDepartment of Protein Biochemistry, SmithKline Beecham Pharmaceuticals, 709 Swedeland Rd., King of Prussia, PA 19406, USA
Received 4 December 1998
AbstractÐCathepsin K (EC 3.4.22.38) is a cysteine protease of the papain superfamily which is selectively expressed within the
osteoclast. Several lines of evidence have pointed to the fact that this protease may play an important role in the degradation of the
bone matrix. Potent and selective inhibitors of cathepsin K could be important therapeutic agents for the control of excessive bone
resorption. Recently a series of peptide aldehydes have been shown to be potent inhibitors of cathepsin K. In an e€ort to design
more selective and metabolically stable inhibitors of cathepsin K, a series of electronically attenuated alkoxymethylketones and
thiomethylketones inhibitors have been synthesized. The X-ray co-crystal structure of one of these analogues in complex with
cathepsin K shows the inhibitor binding in the primed side of the enzyme active site with a covalent interaction between the active
site cysteine 25 and the carbonyl carbon of the inhibitor. # 1999 Elsevier Science Ltd. All rights reserved.
Introduction
synthesis of this protein are associated with pycnody-
sostosis, a rare autosomal disorder of bone remodeling.2
Short stature along with dense and brittle bones are
among some of the clinical manifestations of this dis-
ease. Secondly, through the use of a cathepsin K anti-
sense oligonucleotide, Yamamura and co-workers have
determined that the inhibition of m-RNA to protein
translation of this enzyme results in the inhibition of
osteoclastic pit formation in a dose dependent manner.3
The maintenance of healthy bone tissue is the result of a
balance between the formation of the bone matrix by
osteoblasts and the degradation of bone matrix by
osteoclasts. Cathepsin K (EC 3.4.22.38), a cysteine pro-
tease of the papain superfamily, has been found to be
selectively and abundantly expressed within the osteo-
clast.1 This abundant and selective expression suggests
that this protease may play a crucial role in the degra-
dative phase of the remodeling of the bone matrix and
may be an important target for therapeutic intervention
for diseases associated with aberrant resorption of the
bone matrix. Two lines of evidence have emerged which
support the specialized role of cathepsin K in bone
homeostasis. First, Gelb and co-workers have shown
that mutations in humans of the gene encoding for the
In a report from these laboratories, Votta and co-
workers have shown peptidyl aldehydes such as 1 and 2
to be potent inhibitors of cathepsin K (Fig. 1).4 Alde-
hyde 2 was shown to be an inhibitor of bone loss in the
adjuvant-arthritic rat model and to inhibit the increase
of calcium levels in the PTH infused thyroparathyro-
idectomized rat. These data suggest that aldehyde
inhibitors of cathepsin K may be e€ective agents for
slowing the rate of osteoclast mediated bone resorption
associated with diseases such as osteoporosis. Despite
the promising performance of inhibitors such as 2 in the
aforementioned in vitro and in vivo models, their ther-
apeutic utility is limited by metabolic liabilities which
are commonly associated with aldehyde based inhibi-
tors. As such, it is unlikely that these potent inhibitors
Key words: Cysteine protease; cathepsin K; enzyme inhibitor; alkoxy-
methylketone; thiomethylketone.
*Corresponding author. Tel.: +1-610-270-6570;
ye-mail: robert_w_marquis@sbphrd.com
Currently at Department of Preclinical Pharmokinetics, 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)00011-5

582
R. W. Marquis et al. / Bioorg. Med. Chem. 7 (1999) 581±588
Figure 1. The evolution of the alkoxymethylketone and thiomethylketone cathepsin K inhibitors 3 from the aldehyde inhibitors 1 and 2.
could serve as viable therapeutic agents in a disease
target which would require chronic administration. In
an e€ort to circumvent these liabilities a series of a-
heteroatom substituted ketone dipeptides which would
serve to attenuate the reactive nature of the aldehyde
carbonyl moiety were synthesized (3, Fig. 1). Similar
approaches have been employed for the design of potent
inhibitors of the serine protease thrombin5 and for the
cysteine protease ICE.6 In this paper the synthesis and
SAR of a series of potent and selective a-heteroatom
ketone dipeptide inhibitors of the cysteine protease
cathepsin K which are based on the dipeptidyl aldehyde
derivative 1 are detailed.
bromomethyl ketone 910 which was converted to the
hydroxymethyl ketone 10 by reaction with benzoylfor-
mic acid followed by saponi®cation of the intermediate
formate ester (not shown) with aqueous potassium
hydrogen carbonate. Conversion of 10 to the methoxy-
methyl ketone 11 was accomplished by etheri®cation with
silver(I) oxide and methyl iodide in re¯uxing dichloro-
methane. Similarly, etheri®cation of 10 with benzyl iodide
or 4-(bromomethyl)biphenyl provided inhibitors 12 and
13. The phenoxymethyl ketone analogue 14 was synthe-
sized by reaction of 9 with sodium phenoxide in DMF.
The thiomethyl ketones 15, 16, 17 and 18 were synthe-
sized by treatment of 9 with the appropriate sulphide and
KF in DMF at room temperature.
Cathepsin K inhibition data for the ketone inhibitors
are shown in Table 1.11 The Z-leucinyl-leucine methyl
ketone derivative 6 inhibited cathepsin K with a
Ki,app=500 nM as a mixture of diastereomers. This loss
in activity relative to the more reactive aldehyde deri-
vative 3 was probably due to the reduced electrophilicity
of ketones compared to that of aldehydes. Increasing
the reactivity of the ketone inhibitors via the incor-
poration of an a-methoxy group gave analogue 11
which was signi®cantly more active than the methyl
ketone derivative 6. This increased inhibition may be a
re¯ection of the electron withdrawing nature of oxygen
which would serve to increase the reactivity of the car-
bonyl carbon toward addition by the reactive site
cysteine 25 of cathepsin K over that of the ketone ana-
logue 6. In a publication from these laboratories,12 the
X-ray co-crystal structure of inhibitor 11 with papain
revealed that this inhibitor binds in a non-covalent
manner with the phenyl moiety of the carbonylbenzyl-
oxy group binding in the S30 pocket of the enzyme.13
This preference for primed side inhibitor binding was
also seen in the X-ray co-crystal structure of the weakly
time dependent n-propoxymethyl ketone inhibitor 8
Results and Discussion
Synthesis and cathepsin K inhibition data of dipeptidyl
alkoxymethyl and thiomethyl ketone inhibitors
Initial SAR studies focused on determining the e€ects of
the incorporation of both a ketone and various hetero-
atom ketones into the Z-Leu-Leu template of the
aldeyde inhibitor 1. The methyl ketone 6 was prepared
via Dakin±West7 reaction as shown in Scheme 1. Conver-
sion of Cbz-Leu-Leu-OH (4) to the known oxazolinone
58 followed by acylation with acetyl chloride, rearran-
gement and decarboxylation gave the methyl ketone
analogue 6 as a mixture of diastereomers. The synthesis
of a series of alkoxy and thiomethyl dipeptidyl ketones
is also detailed in Scheme 1. Following the literature
procedure,5 Cbz-Leu-Leu-OH (4) was converted to the
a-diazoketone 7 by formation of the mixed anhydride
with iso-butylchloroformate followed by trapping with
diazomethane. Conversion of diazoketone 7 to the n-
propyl ether derivative 8 was e€ected with rhodium
acetate and n-propyl alcohol.9 Alternatively, treatment
of 7 with 30% HBr in acetic acid at 0 ꢀC provided the

R. W. Marquis et al. / Bioorg. Med. Chem. 7 (1999) 581±588
583
Scheme 1. Synthesis of alkoxymethyl and thiomethylketone inhibitors of cathepsin K.
Table 1. Cathepsin K inhibiton data for alkoxymethyl and thio-
methylketone inhibitors
site barely penetrating the P2 pocket. The phenyl group
of inhibitor 8 is seen bound in the P30 pocket where it
forms a ꢀ-stacking interaction with Trp184.
The phenoxymethyl ketone analogue 14 was a potent
inhibitor with a Ki,app=8.0 nM. In addition to rever-
sible addition of the active site cysteine to the ketone
carbonyl of inhibitor 14, one can envision an irrever-
sible alkylation process analogous to that seen by
Krantz and co-workers for the peptidyl (acyloxy)methyl
ketones.14 The phenoxide moiety of inhibitor 14 would
be expected to be a poorer leaving group than the
benzoate of the Krantz inhibitors therefore the reaction
would be expected to be slower. That this does indeed
occur has been con®rmed by mass spectroscopic analy-
sis. Upon incubation of the enzyme with a ®vefold
excess of inhibitor 14 for 2 h at pH 6 the mass spectrum
showed equal amounts of unmodi®ed enzyme and a
mass corresponding to enzyme+inhibitor with the loss
of phenol (374.4 Da). The ability of the phenoxymethyl
ketone 14 to alkylate the enzyme was removed by the
incorporation of a methylene unit which gave weakly
active benzyl ether 12, which under the time course of
the kinetic determination did not show any time depen-
dence. Interestingly, the 4-phenyl benzyl ether derivative
13 was a potent inhibitor with a Ki,app=22 nM. If one
assumes that inhibitor 13 binds to cathepsin K in a
manner analogous to compound 8 then it is dicult to
1
Compound
XR
Ki,app (nM)
kinact/I (M 1 s
)
6*
8
H
500
Ð
60
600
Ð
4100
Ð
Ð
Ð
O(CH2)2CH3
OCH3
OCH2Ph
OCH2Ph-4-Ph
OPh
S(CH2)2CH3
SPh
SCH2Ph
S(CH2)2Ph
11
12
13
14
15
16
17
18
22
IC50=3.7 nM
Ð
Ð
Ð
Ð
22,000
4500
6500
1650
*Analogue 6 was prepared as a mixture of diastereomers at the a-
ketone center (see ref 8).
with cathepsin K (see Figure 2). The X-ray co-crystal
structure suggests that inhibitor 8 forms a covalent
bond with cysteine 25 of cathepsin K. This observation
contrasts the noncovalent binding of the methoxymethyl
ketone 11 seen in the papain co-crystal structure. The n-
propyl group binds on the unprimed side of the active

584
R. W. Marquis et al. / Bioorg. Med. Chem. 7 (1999) 581±588
Other direct comparisons (12 versus 17 and 14 versus
16) are at best tenuous because time-independent inhi-
bition has changed to time-dependent inhibition and no
structural data to indicate the direction of binding for
these particular inhibitors is available.
Synthesis and cathepsin K inhibition data of P1 modi®ed
dipeptidyl alkoxymethyl ketone inhibitors
Prior to the availability of crystallography data, we
sought to further improve the activity of the dipeptidyl
a-alkoxyketones using inhibitor 11 as a basis for fur-
ther SAR studies. Initially the e€ects of substitution at
the then perceived P1 amino acid a to the keto group of
the Z-Leu-Leu-alkoxymethylketone template was stu-
died. The synthesis of these a modi®ed keto compounds
parallels that of derivative 11 and is outlined below in
Scheme 2. The appropriate dipeptide acid 19 was con-
verted to the diazomethyl ketone 20 by formation of the
mixed anhydride followed by reaction with diazomethane.
Treatment of the diazomethyl ketones with 30% HBr
followed by displacement with benzoylformic acid and
subsequent saponi®cation of the formate ester provided
the hydroxymethyl ketones 21. Etheri®cation of 21 was
e€ected with either methyl or ethyl iodide and silver(I)-
oxide. This ®ve step sequence provided the P1 modi®ed
dipeptidyl alkoxymethyl ketones 23±30 (Table 2).
Figure 2. View of the X-ray co-crystal structure of inhibitor 8 (white)
bound in the active site of cathepsin K (blue). Cysteine 25 (yellow) is
attached covalently to the carbonyl group of the inhibitor forming a
hemi-thioketal. The n-propyl ether slightly penetrates the P2 pocket of
the enzyme. The phenyl group of the Cbz is bound in the P30 pocket
forming a ꢀ-stacking interaction with Trp184.
rationalize this increase in potency since no obvious
binding interaction is available for the 4-phenyl group
compared to analogue 12. Alternatively, if inhibitor 13
has bound with the Cbz-Leu moiety on the unprimed
side of the active site, then it would be possible for the
Inhibition data for the a-modi®ed dipeptidyl alkoxy-
methylketones are shown in Table 2. Compound 23, in
which the chiral center formed by the a-isobutyl sub-
stitution of 11 has been inverted, resulted in a ninefold
decrease in activity. Incorporation of glutamic acid gave
a weakly time dependent inhibitor 24. Alternatively,
substitution with the basic lysine residue gave the potent
inhibitor 25. Incorporation of phenylalanine, norleu-
cine, serine and glycine gave analogues 26, 27, 28 and 29
respectively. These modi®cations were not well toler-
ated. Substitution of the leucine of a-methoxy ketone 11
with l-alanine gave the active analogue 30. Diculties
arise in the interpretation of this data because two dis-
tinct binding modes for various alkoxy ketone inhibi-
tors bound to both papain and cathepsin K have been
observed (covalent and non-covalent primed side bind-
ing). Based on our experience with peptide aldehydes we
feel that it is reasonable to expect that a binding mode
0
0
biphenyl group to access both the S2 and S3 subsites
and engage in a ꢀ±ꢀ interaction with Trp184. This type
of interaction could account for the increased anity of
analogue 13 over that of benzyl ether 12. The thio-
methylketone inhibitors 15±18 displayed weak, time
dependent inactivation of the enzyme. This weak time
dependent activity may be a re¯ection of the decreased
electrophilicity of the ketone carbonyl of these analogues
upon substitution of the less electronegative sulfur for
oxygen. The weak time dependent inhibition of the
thiomethylketone inhibitors may also re¯ect the better
leaving group capability of sulfur over oxygen. It is
interesting to note that the sulfur analogue 15 is
approximately ®ve times more potent than its oxygen
counterpart 8. This increased activity may be a result of
the longer C S bond which would allow enhanced
penetration of the n-propyl moiety into the P2 pocket.
Scheme 2. Sythesis of P2 modi®ed alkoxymethylketone inhibitors.

R. W. Marquis et al. / Bioorg. Med. Chem. 7 (1999) 581±588
Table 2. Cathepsin K inhibition data for P1 modi®ed alkoxy ketone inhibitors
585
R0
Ki,app (nM)
kinact/I (M 1 s
)
1
Compound
R
11
CH3
60
Ð
23
24
CH3
4800
Ð
Ð
CH2CH3
850
25
26
CH3
CH3
110
Ð
Ð
1200
27
28
CH3
CH3
490
790
Ð
Ð
29
30
CH3
CH3
1300
150
Ð
Ð
with the Cbz-Leu-Leu on the unprimed side of the active
site cysteine may occur for some analogues even though
we have not yet detected this mode of binding by X-ray
analysis.
cysteine protease cathepsin K. Relative to the previously
described aldehyde inhibitors, the incorporation of sub-
stituted alkoxymethyl and thiomethyl groups provided a
series of analogues with reduced potency versus cathepsin
K. Substitution in the position a to the ketone of the
inhibitors has revealed a lack of speci®c recognition at
this site as seen in the activities of the alanine and lysine
derivatives 30 and 25 respectively. This lack of recogni-
tion may be due to the ability of these, and possibly all
of the inhibitors described in this paper, to bind in dif-
ferent modes within the active site of cathepsin K. In
this series of ketone inhibitors, these multiple binding
modes make the interpretation of SAR data dicult
for all analogues for which there is no X-ray co-crystal
structure data available.
Cathepsin K, B and L selectivities
Selectivity data for four of the more potent analogues
versus cathepsins K, B and L is shown in Table 3.15 The
overall selectivity pro®le of these analogues for cathepsin
K is quite good with the exception of analogue 13 which
is also a potent inhibitor of cathepsin L. The evaluation
of speci®city pro®les is obviously problematic owing to
the potential for di€erent binding modes. For example,
inhibitor 13 may bind in one mode to cathepsin K and
in an alternate mode to cathepsin L.
Experimental
Conclusions
Crystallization of the complex of cathepsin K with
inhibitor 8. Crystals of mature activated cathepsin K
complexed with the inhibitor grew from a protein solution
In this paper we have detailed a series of alkoxy-
methylketones which are potent inhibitors of the

586
R. W. Marquis et al. / Bioorg. Med. Chem. 7 (1999) 581±588
Table 3. Selectivity of alkoxymethyl ketone inhibitors versus
cathepsins K, B and L
(3S)-3-[(N-Benzyloxycarbonyl)-L-leucinyl]amino-5-methyl-
2-hexanone (6). Acetyl chloride (43 mL, 0.61 mmol) was
added dropwise to a solution of 1-(4-isobutyl-5-oxo-4,5-
dihydro-oxazol-2-yl)-3-methyl-butyl-carbamic acid benzyl
ester8 (5) (200 mg, 0.56 mmol) and triethylamine (155 mL,
1.11mmol) in THF (1.0mL) at rt and was stirred over-
night. The reaction mixture was diluted with EtOAc
(20 mL) and extracted with H2O (5 mL); the combined
organics were dried with magnesium sulfate, ®ltered, then
concentrated in vacuo. The crude product was then dis-
solved in 1:1 AcOH:pyridine (1.0 mL) and was heated
to 110 ꢀC for 1 h. The reaction mixture was cooled to rt,
diluted with EtOAc (50 mL) and extracted with 1N HCl
(3Â20 mL); the combined organics were dried with mag-
nesium sulfate, ®ltered, concentrated in vacuo and chro-
matographed (silica gel, 1±5% MeOH:CH2Cl2) to yield
the title compounds as a mixture of diastereomers of
unknown relative con®gurations (44 mg, 22%): 1H
NMR (CDCl3, 400 MHz): d 7.35±7.27 (m, 5H), 6.55±6.45
(m, 1H), 5.12 (s, 2H), 4.27±4.17 (m, 1H), 4.20±4.06 (m,
1H), 2.18 (s, 3H), 1.70±1.35 (m, 6H), 0.99±0.94 (m, 12H).
Compound
Cat K
Ki,app (nM)
Cat B
Ki,app (nM)
Cat L
Ki,app (nM)
11
13
25
31
60
22
110
55
>1000
1310
2700
513
63
1700
Kobs/I=135
Kobs/I=57
of 10 mg/mL mixed with equal volume of a precipitant
solution of 30% PEG 8000 in 0.1 M Na citrate, pH 4.5
containing 0.2 M Li2SO4. Crystals of the complex are
tetragonal, space group P43212, with cell constants of
a=57.6 A, and c=131.2 A. The crystals contain one
molecule in the asymmetric unit. X-ray di€raction data
were measured from a single crystal using a Siemens
2-D position-sensitive detector on a Siemens rotating
anode generator operating at 5 KW. The structure was
determined at 2.5 A resolution by molecular replace-
ment as implemented in X-PLOR. The structure
determination was carried out using X-PLOR with a
model consisting of all protein atoms from the pre-
viously determined cathepsin K-E64 structure.16 The
model was re®ned by rigid-body re®nement using X-
PLOR, and the resulting phases were used to calculate
Fourier maps with coecients jFo-Fcj and j2Fo-Fcj,
into which the atomic model of the inhibitor was built
using the molecular graphics program FRODO. Con-
ventional positional re®nement was used to re®ne the
structure during model building. Several cycles of map
®tting and re®nement were carried out to a ®nal Rc of
0.245.
(3S)-3-[(N-Benzyloxycarbonyl)-L-leucinyl]amino-1-bromo-
5-methyl-2-hexanone (9). 1-Methyl-3-nitro-1-nitrosogua-
nidine (5.9 g) in ether (200 mL) was cooled to 0 ꢀC. 40%
potassium hydroxide is added slowly and the diazo-
methane was allowed to collect in the ether solution
for 30 min at 0 ꢀC. Cbz-l-leucinyl-l-leucine (4.0 gm,
10.6 mmol) was stirred in tetrahydrofuran at 40 ꢀC.
N-Methyl morpholine (1.07 g, 10.6 mmol, 1.16 mL) and
isobutyl chloroformate (1.45 g, 10.6 mmol, 1.38 mL) are
added. The mixture was stirred at 40 ꢀC for 15 min
and then ®ltered into a cold ¯ask to remove precipitated
salts. To the ®ltered solution was added an excess of the
previously prepared diazomethane solution and the
mixture was allowed to stand at 0 ꢀC for 16 h. An excess
of 30% HBr in acetic acid was added at 0 ꢀC and the
reaction was allowed to stir until complete consumption
of the a-diazoketone was observed (TLC analysis). The
solution was then washed successively with 1.0 N citric
acid, saturated aqueous sodium bicarbonate (carefully),
and brine. The solution was dried over sodium sulfate,
®ltered, and evaporated to give the 9 as a clear solid
Incubation experiments with inhibitor 14 and cathepsin
K. Cathepsin K, 27 uM in MES bu€er at pH 6.0 was
incubated with inhibitor 14 by adding 6.5 nmol of 14 in
20% DMSO:water to 1.3 nmol protein so that the ®nal
protein concentration was 13.5 uM and the DMSO
concentration was 10%. After vortexing, the solution
was allowed to stand at room temperature for 2 h before
being loaded onto a peptide trap cartridge (Michrom),
washed with 40 uL A, and back-eluted onto a 1Â40 mm
Aquasil C18 reverse phase HPLC column (Keystone)
for LC/MS analysis with a gradient from 5% to 95% B
in 10 min at a ¯ow rate of 60 uL/min. Solvent A was
0.02% TFA:water, and B was 0.018% TFA:90%
MeCN:10% water. The mass spectrometer (Sciex API
III) was scanned from m/z 800 to 1800 once every 3.95 s.
1
(4.10 g). H NMR (400 MHz, CDCl3) d 7.34 (m, 5H),
6.51 (d, 1H), 5.15 (d, 1H), 5.10 (s, 2H), 4.78 (m, 1H),
4.20 (m, 1H), 4.04 (dd, 2H), 1.63 (m, 6H), 0.93 (m,
1
12H). IR (thin ®lm): 1716, 1700, 1670 cm
.
(3S)-3-[(N-Benzyloxycarbonyl)-L-leucinyl]amino-1-(1-prop-
oxy)-5-methyl-2-hexanone (8). (3S)-3-[(N-Benzyloxy-
carbonyl)-l-leucinyl]amino-1-diazo-5-methyl-2-hexanone
(7) (150mg, 0.37mmol) was dissolved in 1-propanol
(2.5 mL) and rhodium acetate (2mg) was added and the
reaction was stirred at rt for 2 h. The reaction mixture
was concentrated and chromatographed (silica gel, 20%
EtOAc:hexanes) to yield the n-propoxyketone 8 as a
white solid (59 mg, 37%): 1H NMR (400MHz, CDCl3) d
7.4±7.15 (m, 5H), 6.4 (d, 1H), 5.5±4.0 (m, 13H), 5.01 (s,
2H), 3.4 (m, 2H), 1.6 (m, 6H), 0.9 (m, 12H) MS(ES)
M+H+=435, M+ NH4+=452, 2M+H+=869.6.
Experimental procedures
Nuclear magnetic resonance spectra were recorded at
either 250 or 400 MHz using, respectively, a Bruker AM
250 or Bruker AC 400 spectrometer. Fourier transform
infrared (FTIR) spectra were recorded on a Nicolet
Impact 400 D infrared spectrometer. Mass spectra were
taken on PE Sciex API III or Micromass platform
API-only instruments, using electrospray (ES) ionization
techniques. Analtech Silica Gel GF and E. Merck Silica
Gel 60 F-254 thin layer plates were used for thin layer
chromatography. Flash chromatography was carried out
on E. Merck Kieselgel 60 (230±400 mesh) silica gel.
(3S)-3-[(N-Benzyloxycarbonyl)-L-leucinyl]amino-1-hydroxy-
5-methyl-2-hexanone (10). The bromomethylketone 9

R. W. Marquis et al. / Bioorg. Med. Chem. 7 (1999) 581±588
587
(1.3 g, 2.9 mmol) and benzoylformic acid (0.515 g,
3.5 mmol) were stirred in dimethylformamide (10 mL)
and treated with potassium ¯uoride (0.25 g, 4.4 mmol).
The mixture was stirred for 16 h at room temperature
and partitioned between ethyl acetate and water. The
organic layer was dried over sodium sulfate, ®ltered,
and evaporated to give a residue which was chromato-
graphed (silica gel, 25% ethyl acetate in hexane) to give
the formate ester as a white solid (1.04 g). The formate
ester (1.02 g, 1.9 mmol) was stirred vigorously in a mix-
ture of tetrahydrofuran (100 mL) and 1.0 M potassium
bicarbonate (100 mL) at room temperature for 18 h. The
layers were separated and the aqueous layer was
extracted with ethyl acetate. The organic layers were
combined, dried over sodium sulfate, ®ltered, and eva-
stirred for an additional 2 h. The mixture was neutralized
with 0.3M potassium bisulfate and the solvent removed
in vacuo. The residue obtained was partitioned between
ethyl acetate and water and the organic extracts were
washed with brine. The solution was dried over magne-
sium sulfate, ®ltered, and evaporated to give a residue
which was chromatographed (silica gel, 25% ethyl ace-
1
tate in hexane) to give 13 as a white solid (0.48 g). H
NMR (400MHz, CDCl3) d 7.26 (m, 7H), 6.92 (t, 1H),
6.80 (d, 2H), 6.42 (d, 1H), 5.13 (d, 1H), 4.98 (s, 2H), 4.91
(m, 1H), 4.60 (d, 2H), 4.12 (m, 1H), 1.58 (m, 4H), 1.39
(m, 2H), 0.84 (m, 12H); MS(ES) M+H+=469.2,
2M+H+=937.6.
(3S)-3-[(N-Benzyloxycarbonyl)-L-leucinyl]amino-1-(phenyl-
thio)-5-methyl-2-hexanone (16). Thiophenol (21 mg,
20 mL, 0.190 mmol) was added to a solution of 9 (70 mg,
0.154 mmol) in DMF (0.5 mL) at rt. Then solid potas-
sium ¯uoride (10 mg, 0.172 mmol) was added, and the
reaction mixture was stirred overnight. Concentration
of the reaction mixture in vacuo followed by column
chromatography (silica gel, 25% EtOAc:hexane) of the
residue gave the title compound as a white solid (52 mg,
67%): 1H NMR (CDCl3, 400 MHz): d 7.38±7.22 (m,
10H), 6.55 (br d, J=7.7 Hz, 1H), 5.24 (br d, J=8.2 Hz,
1H), 5.12 (s, 2H), 4.85±4.78 (m, 1H), 4.22±4.15 (m, 1H),
3.78 (s, 2H), 1.70±1.40 (m, 6H), 0.95 (d, J=6.4 Hz, 6H),
0.90 (d, J=6.4 Hz, 6H). MS(ES) M+H+=485.5,
M+Na+=507.5, 2M+Na+=991.7.
1
porated to give 10 as a white solid (0.768 g). H NMR
(CDCl3, 400 MHz): d 7.35±7.31 (m, 5H), 6.74 (d,
J=7.6 Hz, 1H), 5.35 (d, J=7.6 Hz, 1H), 5.10 (s, 2H),
4.7±4.6 (m, 1H), 4.36 (s, 2H), 4.25±4.18 (m, 1H), 3.2 (br
s, 1H), 1.7±1.43 (m, 6H), 0.93 (d, J=6.0 Hz, 3H), 0.91
(2d, J=6.0 Hz, 6H), 0.90 (d, J=6.0 Hz, 3H); IR (thin
®lm): 3307, 1716, 1699, 1652 cm 1; MS(ES) M+H+=
393.2, 2M+H+=785.4.
(3S)-3-[(N-Benzyloxycarbonyl)-L-leucinyl]amino-1-methoxy-
5-methyl-2-hexanone (11). Methyl iodide (0.32 mL,
5.1 mmol) was added to a solution of the a-hydroxy-
ketone 10 (0.4 g, 1.0 mmol) and silver(I) oxide (0.46 g,
2.0 mmol) in methylene chloride (30 mL), and the reaction
mixture was re¯uxed for 7 h. Additional methyl iodide
(1.0 mL, 15.5 mmol) was added and the reaction mixture
was stirred at rt over 2 days. The reaction mixture was
®ltered, concentrated in vacuo, then chromatographed
(silica gel, 25% EtOAc:hexane) to give the title com-
Acknowledgements
The authors would like to acknowledge Dr. Michael L.
Moore for his critical review of this manuscript.
1
pound as a white solid (0.17 g, 42%): H NMR (CDCl3,
400 MHz): 1H NMR (CDCl3, 360 MHz): d 7.35±7.24
(m, 5H), 6.35±6.33 (d, 1H, J=7.74 Hz), 5.06 (s, 2H),
4.81±4.75 (m, 1H), 4.16 (br s, 1H), 4.12 (AB, JAB=10.3,
ÁdAB=0.097, 2H), 3.40 (s, 3H), 1.60±1.30 (m, 6H), 0.92
(d, 9H, J=6.2 Hz), 0.90 (d, 3H, J=6.47 Hz); IR (KBr):
3301, 1731, 1694, 1646 cm 1; MS(ES) M+H+=407.5,
M+Na=429.5, 2M+H+. Anal. calcd for C22H34N2O5:
C, 65.00; H, 8.43; N, 6.89. Found: C, 65.09; H, 8.61; N,
6.93.
References and Notes
1. (a) Drake, F. H.; Dodds, R. A.; James, I. A.; Connor, J. R.;
Debouck, C.; Richardson, S.; Lee-Rykaczewski, L.; Coleman,
L.; Riemann, D.; Barthlow, R.; Hastings, G.; Gowen, M. J.
Biol. Chem. 1996, 271, 12511±12516. (b) Bromme, D.; Oka-
moto, K. Biol. Chem. Hoppe-Seyler 1996, 376, 379±384. (c)
Shi, G.; Chapman, H. A.; Bhairi, S. M.; DeLeeuw, C.; Reddy,
V. Y.; Weiss, S. W. FEBS Letter 1995, 357, 129±134. (d)
Littlewood-Evans, A.; Kokubo, T.; Ishibashi, O.; Inaoka, T.;
Wlodarski, B.; Gallager, J. A.; Bilbe, G. Bone 1997, 20, 18±86.
2. (a) Gelb, B. D.; Moissoglu, K.; Zhang, J.; Martignetti, J. A.;
Bromme, D.; Desnick, R. J. Biochem Mol. Med. 1996, 59,
200±206. (b) Johnson, M. R.; Polymeropoulos, M. H.; Vos,
H. L.; Ortiz de Luna, R. I.; Francomano, C. A. Genome Res.
1996, 6, 1050±1055. (c) Gelb, B. D.; Shi, G.-P.; Chapman,
H. A.; Desnick, R. J. Science 1996, 273, 1236±1238.
(3S)-3-[(N-Benzyloxycarbonyl)-L-leucinyl]amino-1-benzoyl-
oxy-5-methyl-2-hexanone (12). The benzyl ether analo-
gue 12 was synthesized in the same manner as that
described for the synthesis of 11 except substituting
benzyliodide for methyliodide. Analogue 12 was pre-
pared as a white solid (0.17 g). 1H NMR (400 MHz,
CDCl3) d 7.24 (m, 5H), 6.60 (d, 1H), 5.23 (d, 1H), 5.01
(s, 2H), 4.79 (m, 1H), 4.51 (d, 2H), 4.08 (m, 3H), 1.53
(m, 6H), 0.82 (m, 12H); MS(ES) M+H+=483.2, M+
NH4+=500.2, 2M+H+=965.4.
3. Inui, T.; Ishibashi, O.; Inaoka, T.; Origane, Y.; Kumegawa,
M.; Kokubo, T.; Yamamura, T. J. Biol. Chem. 1997, 272,
8109±8112.
4. Votta, B. J.; Levy, M. A.; Badger, A.; Bradbeer, J.; Dodds,
R. A.; James. I. A.; Thompson, S. T.; 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±
1406.
5. (a) Jones, D. M.; Atrash, B.; Teger-Nilsson, A. C.; Gyzander,
E.; Deinum, J.; Szelke, M. Lett. Pep. Sci. 1995, 2, 147. (b)
Jones, D. M.; Atrash, B.; Teger-Nilsson, A. C.; Gyzander, E.;
Deinum, J.; Szelke, M. Proc. Eur. Pept. Symp. 23rd (1995),
(3S)-3-[(N-Benzyloxycarbonyl)-L-leucinyl]amino-5-methyl-
1-phenoxy-2-hexanone (14). Phenol (0.138 gm, 1.5 mmol)
and sodium hydride (0.036 gm, 1.5 mmol) were stirred
in dimethylformamide at 20 ꢀC under an argon atmo-
sphere for 15 min and bromomethylketone 9 (0.66 g,
1.5 mmol) was added. The reaction was stirred at 20 ꢀC
for 30min and it was warmed to room temperature and

588
R. W. Marquis et al. / Bioorg. Med. Chem. 7 (1999) 581±588
899±900. Ed.: Hernani M. L. S., ESCOM, Leiden. (c) Jones,
M. D.; Atrash, B.; Ryder, H.; Teger-Nilsson, A.-C.; Gyzan-
der, E.; Szelke, M. J. Enzyme Inhibition, 1995, 9, 43±60.
6. Majalli, A. M. M.; Chapman, K. T.; MacCoss, M.;
Thornberry, N. A.; Peterson, E. P. Bioorg. Med. Chem. Lett.
1994, 4, 1965±1968.
11. Testing of the inhibitors was conducted with puri®ed
recombinant cathepsin K as described in Bossard, M. J.;
Tomaszek, T. A.; Thompson, S. K.; Amegadzie, B. Y.; Han-
ning, 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.
7. For a review on the Dakin±West reaction see: (a) Bucha-
nan, G. L. Chem. Soc. Rev. 1998, 17, 91±109. See also: (b)
Steglich, W.; Holfe, G. Tetrahedron Lett. 1968, 13, 1619. (c)
Smissman, E. E.; Terada, A.; El-Antably, S. J. Med. Chem.
1976, 19, 165.
8. Yagasawa, S.; Urakami, M. Tetrahedron Lett. 1996, 37,
7557.
9. Jones, G. B.; Moody, C. J.; Padwa, A.; Kassir, J. M. J.
Chem. Soc. Perkin Trans. 1991, 1, 1721.
12. LaLonde, J. M.; Zhao, B.; Smith, W. W.; Janson, C. A.;
DesJarlais, R. L.; Carr, T. J.; Thompson, S. K.; Yamashita,
D. S.; Veber, D. F.; Abdel-Meguid, S. S. J. Med. Chem.
accepted for publication.
13. The amino acid nomenclature of Schecter, I.; Berger, A.
Biochem. Biophys. Res. Comm. 1967, 27, 157.
14. (a) Krantz, A.; Copp, L. J.; Coles, P. J.; Smith, R. A.;
Heard, S. B. Biochemistry, 1991, 30, 4678±4687. (b) Krantz, A.
Methods Enzymol. 1994, 244, 656.
10. (a) Carr, T. J.; DesJarlais, R. L.; Gallagher, T. F.; Halbert,
S. M.; Oh, H.-J.; Thompson, S. K.; Veber, D. F.; Yamashita,
D. S.; Yen, J. WO 9716433 A1 970509. (b) Dolle, R. E.;
Graybill, T. L.; Osifo, I. K.; Harris, A. L.; Miller, M. S. WO
9641638 A1 961227.
15. Cathepsins B and L were purchased from Calbiochem.
16. Zhao, B.; Janson, C. A.; Amegadzie, B. Y.; D'Alessio, K.;
Grin, C.; Hanning, C. R.; Jones, C.; Kurdyla, J.; McQueney,
M; Qui, X.; Smith, W. W.; Abdel-Meguid, S. S. Nature Struct.
Biol. 1997, 4, 109±111.