Structure activity relationships of 5-, 6-, and 7-methyl-substituted azepan-3-one cathepsin K inhibitors

Article abstract of DOI:10.1021/jm050915u

The syntheses, in vitro characterizations, and rat and monkey in vivo pharmacokinetic profiles of a series of 5-, 6-, and 7-methyl-substituted azepanone-based cathepsin K inhibitors are described. Depending on the particular regiochemical substitution and

Full text of DOI:10.1021/jm050915u

J. Med. Chem. 2006, 49, 1597-1612
1597
Structure Activity Relationships of 5-, 6-, and 7-Methyl-Substituted Azepan-3-one Cathepsin K
Inhibitors
,
†
†
†
†
†
†
Dennis S. Yamashita,* Robert W. Marquis, Ren Xie, Sirishkumar D. Nidamarthy, Hye-Ja Oh, Jae U. Jeong,
†
‡
‡
‡
§
§
§
Karl F. Erhard, Keith W. Ward, Theresa J. Roethke, Brian R. Smith, H-Y. Cheng, Xiaoliu Geng, Fan Lin,
Priscilla H. Offen, Bing Wang, Neysa Nevins, Martha S. Head, R. Curtis Haltiwanger, Amy A. Narducci Sarjeant,
Louise M. Liable-Sands, Baoguang Zhao, Ward W. Smith, Cheryl A. Janson, Enoch Gao, Thaddeus Tomaszek,
Michael McQueney, Ian E. James, Catherine J. Gress, Denise L. Zembryki, Michael W. Lark, and Daniel F. Veber
§
§
§
§
§
§
§
§
§
§
|
||
|
⊥
⊥
⊥
⊥
†
Departments of Medicinal Chemistry; Drug Metabolism and Pharmacokinetics; Musculoskeletal Diseases Biology; Computational, Analytical,
and Structural Sciences; and Screening and Compound Profiling, GlaxoSmithKline, 1250 S. CollegeVille Rd, CollegeVille, PennsylVania 19426
ReceiVed September 14, 2005
The syntheses, in vitro characterizations, and rat and monkey in vivo pharmacokinetic profiles of a series
of 5-, 6-, and 7-methyl-substituted azepanone-based cathepsin K inhibitors are described. Depending on the
particular regiochemical substitution and stereochemical configuration, methyl-substituted azepanones were
identified that had widely varied cathepsin K inhibitory potency as well as pharmacokinetic properties
compared to the 4S-parent azepanone analogue, 1 (human cathepsin K, Ki,app ) 0.16 nM, rat oral bioavailability
)
42%, rat in vivo clearance ) 49.2 mL/min/kg). Of particular note, the 4S-7-cis-methylazepanone analogue,
1
0, had a Ki,app ) 0.041 nM vs human cathepsin K and 89% oral bioavailability and an in vivo clearance
rate of 19.5 mL/min/kg in the rat. Hypotheses that rationalize some of the observed characteristics of these
closely related analogues have been made using X-ray crystallography and conformational analysis. These
examples demonstrate the potential for modulation of pharmacological properties of cathepsin inhibitors by
substituting the azepanone core. The high potency for inhibition of cathepsin K coupled with the favorable
rat and monkey pharmacokinetic characteristics of compound 10, also known as SB-462795 or relacatib,
has made it the subject of considerable in vivo evaluation for safety and efficacy as an inhibitor of excessive
bone resorption in rat, monkey, and human studies, which will be reported elsewhere.
Introduction
Table 1. Structures of Azepanone and Methyl-Substituted Azepanone
Cathepsin Inhibitors
Cathepsin K, a member of the papain-family of cysteine
proteases, is highly expressed by osteoclasts and has been a
therapeutic target of many drug discovery efforts aimed at
identifying agents to treat disease states caused by excessive
1
bone resorption such as osteoporosis. In previously published
papers, we have disclosed several classes of peptidomimetic
2
ketone-based cathepsin K inhibitors and have noted that by
introducing an appropriate conformational constraint in the form
of an azepan-3-one, cathepsin K inhibitory potency, rat oral
bioavailability, and chemical stability could be improved. In
this paper, we describe the effects of methyl-substitution of
azepan-3-one analogues on cathepsin inhibitory potency and
selectivity, chemical stability, cellular efficacy, and rat and
monkey pharmacokinetic profiles.
compd
C4
A
B
C
D
E
F
1
2
3
4
5
6
7
8
9
S
S
S
R
R
S
S
R
R
S
S
R
R
H
Me
H
Me
H
H
H
H
H
H
H
H
Me
H
Me
H
H
H
H
H
H
H
H
H
H
Me
H
Me
H
H
H
H
H
H
H
H
Me
H
Me
H
H
H
H
H
H
H
H
H
H
Me
H
Me
H
H
H
H
H
H
H
H
H
H
H
Me
H
Me
3
As described in ref 3, an X-ray cocrystal structure of
azepanone 1 (human cathepsin K Ki,app ) 0.16 nM, Table 3;
1
1
1
0
1
2
4
2% rat oral bioavailability, Table 5) and human recombinant
H
H
H
H
H
H
H
H
H
H
H
H
cathepsin K was determined. In this cocrystal structure, the
amide group on the azepanone ring adopted an axial orientation
on a chair conformer. This axial amide orientation was different
from the one observed in the small molecule single-crystal X-ray
diffraction study, where the amide group adopted an equatorial
orientation on a different chair conformer (Figure 1 and Table
13
the inhibitory potency of azepanone 1 by lowering the activation
barrier to access this bound chair conformer with an axial amide
orientation. On the basis of the X-ray cocrystal structure of 1
bound to cathepsin K, we hypothesized that methyl substitution
at the 5-, 6-, and 7- positions of the azepanone ring should be
sterically tolerated by the enzyme. We also conjectured that
methyl substitution of these positions should have considerable
effects on ring conformations, rigidity, and chemical stability
with respect to C4 epimerization and might also have an impact
on pharmacokinetic properties such as oral bioavailability.
Owing to the difficulties inherent in prediction of the impact
2
). We were interested in identifying ways to further optimize
*
To whom correspondence should be addressed. Tel.: (610) 917-7830.
Fax: (610) 917-4206. Email: Dennis.S.Yamashita@gsk.com.
†
Department of Medicinal Chemistry.
Department of Drug Metabolism and Pharmacokinetics.
Department of Computational, Analytical, and Structural Sciences.
Department of Screening and Compound Profiling.
Department of Musculoskeletal Diseases Biology.
‡
§
|
⊥
1
0.1021/jm050915u CCC: $33.50 © 2006 American Chemical Society
Published on Web 02/15/2006

1
598 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 5
Yamashita et al.
Grubbs’ RCM catalyst, bis(tricyclohexylphosphine)benzylidine-
ruthenium(IV) dichloride, gave azepine 29. Epoxidation with
mCPBA afforded a mixture of syn- and anti-epoxides 30 and
3
1 (∼1:1 ratio), respectively, which were separated by column
chromatography. Individual epoxides 30 and 31 were treated
with sodium azide to regioselectively and stereospecifically
3
afford azides 32 and 33. Reduction of the azides with triph-
enylphosphine gave amines 34 and 35. Acylation with N-Boc-
L-leucine and HBTU furnished amides 36(a+b) and 37(a+b).
Deprotection of the Boc groups of 36(a+b) and 37(a+b) with
HCl in dioxane followed by acylation with 2-benzofuran
carboxylic acid using HBTU gave the intermediate alcohols 38-
Figure 1. Stereoview comparison of single-molecule X-ray crystal
structure of 1 (top) and human cathepsin K-1 cocrystal (bottom)
structure conformations
(a+b) and 39(a+b). Oxidation with Dess-Martin periodinane
yielded a pair of mixtures of diastereomeric azepanones. Each
pair of diastereomers was further purified by HPLC on an R,R-
Whelk-O chiral phase column to afford the individual azepanones
of the methyl substitutions on preferred conformational prefer-
ence, we synthesized and characterized all of the possible
monomethyl-substituted azepanone analogues, 2-13, shown in
Table 1.
6-9 in an isomerically pure form (>95% diastereomeric purity).
The sequence depicted in Scheme 3 was used for the synthesis
of the 7-methyl-substituted azepanone analogues. Reductive
amination of 5-hexen-2-one 40 with allylamine gave secondary
amine 41. Protection with carbobenzyloxy chloride gave diene
Synthesis of Methyl-Substituted Azepanones
Each position of methyl substitution on the azepanone ring
as well as the diastereomeric forms of each of them presented
unique challenges for chemical synthesis. Three basic chemical
schemes were utilized to generate diastereomeric mixtures that
were separated to give all of the possible forms for biological
and biochemical characterization. Physical characterization of
the individually isolated isomers as described below allowed
exact definition of the geometric and stereochemical nature of
each product.
4
2. Treatment with Grubbs’ RCM catalyst gave azepine 43.
Epoxidation with mCPBA afforded a mixture of anti- and syn-
epoxides 44 and 45 (3:1 ratio), respectively, which were
separated by flash column chromatography. The epoxides were
converted into azido alcohols 46 and 47 by reaction with sodium
azide, along with the regioisomeric azido alcohol 48 which
formed as an 18% impurity from 45. The regio- and stereo-
chemical assignments of 46 and 47 were determined by X-ray
crystallography (ORTEPs and coordinates in Supporting Infor-
mation). Reduction with triphenylphosphine produced primary
amines 49 and 50. Acylation with N-Boc-L-leucine and EDC/
HOBT gave mixtures of amides 51(a+b) and 52(a+b). Hy-
drogenolysis removed the Cbz groups to provide secondary
amines 53(a+b) and 54(a+b), which were sulfonylated with
4
An intramolecular nitroaldol (Henry) cyclization strategy was
employed to synthesize the 5-methylazepanones (2-5), as
shown in Scheme 1. Michael addition of nitromethane to ethyl
crotonate (14) gave 3-methyl-4-nitrobutyrate (15). Reduction
with DIBAL-H provided the aldehyde 16, which was converted
with N-benzylethanolamine in the presence of sodium triac-
etoxyborohydride to nitro alcohol 17. Oxidation of 17 using
DMSO and oxalyl chloride furnished aldehyde 18. Treatment
of the crude intermediate aldehyde 18 with triethylamine
effected the nitroaldol cyclization to give a mixture of diaster-
eomeric racemic azepanols 19. Reduction of the nitro group
with zinc in the presence of hydrochloric acid gave amines 20,
which were coupled with N-Boc-L-leucine in the presence of
EDC and HOBT to deliver amides 21. Reductive removal of
the N-benzyl moiety with hydrogen gas in the presence of 10
mol % Pd on carbon yielded secondary amines 22. Sulfonylation
with 2-pyridylsulfonyl chloride supplied sulfonamides 23.
Removal of the Boc group of 23 under acidic conditions,
followed by EDC/HOBT coupling of the resulting amine salt
with benzofuran-2-carboxylic acid, gave amides 24. Oxidation
of the alcohols with Dess-Martin periodinane provided an
isomeric mixture of the azepanones 2, 3, 4, and 5. The cis-
diastereomers, 2 and 5, were separated from the trans-diaster-
eomers, 4 and 3, by HPLC using an Impaq silica gel column.
Each pair of diastereomers was further subjected to an additional
HPLC purification on a Chiracel OD column to obtain each
individual azepanone in an isomerically pure form (>95%
diastereomeric purity).
2
-pyridinesulfonyl chloride to give sulfonamides 55(a+b) and
56(a+b). Treatment of 55(a+b) and 56(a+b) with HCl removed
the Boc groups, and the resulting primary amines were acylated
with 2-benzofuran carboxylic acid and EDC/HOBT to yield the
alcohols 57(a+b) and 58(a+b). Oxidation with Dess-Martin
periodinane yielded a pair of mixtures of diastereomeric
azepanones. Each pair of diastereomers was purified by HPLC
using an R,R-Whelk-O chiral phase column to afford individual
azepanones 10-13 in an isomerically pure form (>95%
diastereomeric purity).
Physical and Structural Characterization of Azepanones
Small Molecule X-ray Crystallography. As an aid in the
stereochemical assignments for the various diastereomeric
azepanones, high-quality crystals of compounds 3, 5, 6, and 10
were obtained and single-crystal X-ray structures were deter-
mined (ORTEPs and coordinates in Supporting Information).
The ring dihedral angles of the unsubstituted azepanone 1 (ref
3, Supporting Information) along with the 5-methyl-substituted
azepanones 3 and 5, the 6-methyl-substituted azepanone 6, and
the 7-substituted azepanone 10 are listed in Table 2. By
comparison to canonical dihedral angles of chair and twist-boat
6
The 6- and 7-methyl-substituted azepanones 6-13 were
prepared using ring-closing metathesis (RCM) strategies as
shown in Schemes 2 and 3. Toward the preparation of
conformations of cycloheptane, 1, 3, 5, and 6 crystallize in
5
the same chair conformation, while 10 adopts a twist-boat
conformation. In all four compounds examined, the amide group
is oriented equatorially. In 3, 6, and 10, the methyl group
assumes an equatorial orientation, while in 5, the methyl group
is in an axial orientation (Figure 2).
6
-methylazepanone analogues (6-9), racemic 2-methylpent-4-
enoic acid ethyl ester 25 was converted to aldehyde 26 by
DIBAL-H reduction. Reductive amination with allylamine
afforded secondary amine 27. Sulfonylation with 2-pyridine-
sulfonyl chloride provided sulfonamide 28. Treatment of 28 with
Epimerization Rates and Equilibria. The pH-dependent
rates of epimerization of the C4 chiral center for 1, its

SAR of Azepan-3-one Cathepsin K Inhibitors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 5 1599
Scheme 1. Synthesis of 5-Methylazepanonesa
a
Reagents: (a) MeNO2, TMG; (b) DIBAL-H, toluene -78 °C; (c) Na(OAc)3BH, N-benzylethanolamine; (d) oxalyl chloride, DMSO, -78 °C; (e) Et3N,
rt; (f) Zn, MeOH, HCl; (g) N-Boc-L-leucine, EDC, HOBT; (h) H2, Pd/C; (i) 2-pyridinesulfonyl chloride, Et3N; (j) HCl, dioxane; (k) 2-benzofurancarboxylic
acid, EDC, HOBT, Et3N; (l) Dess-Martin periodinane, CH2Cl2
Scheme 2. Synthesis of 6-Methylazepanonesa
a
Reagents: (a) DIBAL-H, toluene -78° C; (b) allylamine; CH2Cl2, NaBH4; (c) 2-pyridinesulfonyl chloride, NMM; (d) bis(tricyclohexylphosphine)ben-
zylidineruthenium(IV) dichloride, CH2Cl2; (e) m-CPBA, CH2Cl2; (f) sodium azide, MeOH, H2O; (g) triphenylphosphine, THF, H2O; (h) N-Boc-L-leucine,
HBTU, NMM, DMF; (i) HCl, dioxane; (j) 2-benzofurancarboxylic acid, HBTU, NMM, DMF; (k) Dess-Martin periodinane, CH2Cl2.
corresponding C4 epimer, and their 2,2,4-trideuterated analogues
were reported in ref 3. Similar HPLC methods, optimized for
each pair of epimers, were developed to evaluate the effect of
methyl substitution at C5 (2, 4), C6 (6, 8), and C7 (11, 13) on
the kinetics and steady-state ratios of the C4 epimers in aqueous
phosphate buffer at pH 11. The C5-substituted compounds 2
and 4 epimerized at a slower rate than unsubstituted azepanone
This increased thermodynamic stability is a fortuitous physical
property because it facilitates synthesis and scale-up of 10 by
allowing for conversion of mixtures of 10 and 12 to predomi-
nantly 10 by a simple epimerization process.
Structure Assignments. A combination of small molecule
X-ray crystallography, epimerization studies, and NMR spec-
troscopic studies were utilized to obtain complete assignment
of the structures of the isomeric compounds 2-12. Representa-
tive azepanones substituted by a methyl group at each of the
5-, 6-, and 7- positions were elucidated by small-molecule
crystallography as described above; the unambiguous assignment
of the azepanone relative and absolute stereochemical configu-
rations were as follows: 3 ) 4S, anti; 5 ) 4R, syn; 6 ) 4S,
syn; 10 ) 4S, syn. The compounds were practically stable to
epimerization at neutral pH, but when subjected to pH 11
buffered conditions, the diastereomers interconverted by C4
epimerization and were separable by HPLC (Scheme 4).
Interconversion of 2 and 4, 3 and 5, 6 and 8, 7 and 9, 10 and
1
and reached a 1:1 ratio from either direction. This slower rate
was anticipated on the basis of the increased steric hindrance
of the methyl group on the formation of the intermediate
7
enolates. The epimerization rate constants increased as the
methyl group was moved from C5 to C6 and to C7, approaching
that of 1. The C6 and C7 substituted compounds reached
equilibrium strongly favoring one epimer over the other, as
shown in Table 3. This reversible unsymmetrical epimerization
is clearly illustrated in Figure 3, where the thermodynamic ratios
of 2S4 and 11S13 are compared. Note that in the 7-methyl
series, the cis diastereomer has the greater cathepsin K inhibitory
potency (10 vs 12) as well as increased thermodynamic stability.

1
600 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 5
Yamashita et al.
Scheme 3. Synthesis of 7-Methylazepanonesa
a
Reagents: (a) allylamine, pTsOH; (b) NaBH4, MeOH; (c) carbobenzyloxy chloride, Et3N, CH2Cl2; (d) bis(tricyclohexylphosphine)benzylidineruthenium(IV)
dichloride, CH2Cl2; (e) mCPBA, CH2Cl2; (f) sodium azide, MeOH, H2O; (g) triphenylphosphine, THF, H2O; (h) N-Boc-L-leucine, EDC, HOBT, DMF; (i)
H2, Pd/C, MeOH; (j) 2-pyridinesulfonyl chloride, NMM; (k) HCl, dioxane; (l) 2-benzofurancarboxylic acid, HBTU, NMM, DMF; (m) Dess-Martin periodinane,
CH2Cl2
Table 2. Dihedral Angles of 1, 3, 5, 6, 10, Canonical Cycloheptane
Chair (C), Canonical Cycloheptane Twist-boat (TB), 1/Cathepsin K
Cocrystal (1C), and 10/Cathepsin K Co-crystal (10C)
dihedral
N1C2C3C4 9.4
C2C3C4C5 -71.7 -66.7 70.9 -72.0 -55.8 -66.1 -74.6 57 49
C3C4C5C6 82.2 82.5 -81.7 83.3 3.8 83.5 17.9 -67 0.4
1
3
5
6
10
C
TB
1C 10C
-1.3 -3.2 8.5
6.1
0.0
17.9 -73 -70
C4C5C6C7 -60.4 -61.5 59.4 -60.3 72.5 -63.8 64.4 67 -53
C5C6C7N1 60.0 58.2 -60.7 59.5 -45.3 63.8 -45.4 -17 68
C6C7N1C2 -91.4 -86.6 92.1 -91.4 -50.8 -83.5 -45.4 -58 -68
C7N1C2C3 70.7 75.2 -73.9 69.8 76.6 66.1 64.4 92 75
12, and 11 and 13 allowed the unambiguous assignment of
ketones 8 ) 4R, anti and 12 ) 4R, anti.
The remaining structures were assigned by proton NMR
experiments. The structures of the two remaining 5-methyl-
azepanones 2 and 4 were deduced from the methine C4 and C5
proton-proton coupling constants. Matching the corresponding
syn and anti diastereomers 5 (J ) 2.1 Hz) and 3 (J ) 8.5 Hz),
respectively, strongly suggested the assignments as 2 ) 4S, syn
Figure 2. Stereoview comparison of the azepanone cores and sub-
stituents of 5 (enantiomer depicted, top), 6 (middle), and 10 (bottom)
determined by single-crystal X-ray diffraction. Note the axial methyl
group of 5 (enantiomer depicted).
(
J ) 1.8 Hz) and 4 ) 4R, anti (J ) 8.9 Hz). The remaining
6
-methylazepanone assignments were made by proton NOE
experiments. Compounds 6 and 9 (but not 7 and 8) had NOE’s
between the C4 and C6 methine protons, consistent with 6 )
4
S, syn and 9 ) 4R, syn. Compounds 7 and 8 (but not 6 and 9)
of 1 and 10, the NAMFIS (NMR analysis of molecular
8
had NOE’s between the C4 methine and C6 methyl protons,
consistent with 7 ) 4S, anti and 8 ) 4R, anti. Finally, the
remaining 7-methylazepanone assignments were made by
comparing the AB coupling of the C2 methylene protons.
Compounds 10 and 13 had the same coupling constant (JAB )
flexibility in solution) procedure was employed. Briefly, the
NAMFIS method computationally identifies the family of
conformers that best fit NMR-determined NOE and 3JHH
coupling constants. The rationale is that a small flexible
molecule most likely does not exist as a single conformation in
solution but rather as a population of energetically accessible
conformers. Rather than calculating conformer populations or
mole fractions from force field energies, which are especially
unreliable for druglike molecules with multiple long-range polar
functionalities, the computation selects only conformers for
1
9.4 Hz), consistent with 13 ) 4R, syn, whereas 11 and 12 had
the same coupling constant (JAB ) 17.7 Hz), consistent with
1
1 ) 4S, anti.
Conformational Analysis. To assess the solution structures
of 1, 2, 7, and 10 for comparison with the bound conformations

SAR of Azepan-3-one Cathepsin K Inhibitors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 5 1601
Table 3. Inhibition Constants, Physical Properties, and in Vitro Antiresorptive Potency
cathepsin Ki,app (nM)
human
epimerization
MW
(Da)
no. H-bond
clog Pa donor/acceptor
PSA
(A )
thermo
ratio
IC50
(nM)
2 b
k (1/s)c
d
no.
K
L
V
S
B
rat: K
nRot
-
5
1
2
3
4
5
6
7
8
9
0
1
2
3
0.16
2.2
1.8
4.0
ND
ND
ND
ND
350 ND
5.8
300 ND
500
32
>1000 ND
>1000 ND
>1000 ND
60
ND
526.61
540.64
540.64
540.64
540.64
3.08
3.60
3.60
3.60
3.60
3.60
3.60
3.60
3.60
3.60
3.60
3.60
3.60
2/9
2/9
2/9
2/9
2/9
2/9
2/9
2/9
2/9
2/9
2/9
2/9
2/9
138.68
138.68
138.68
138.68
138.68
138.68
138.68
138.68
138.68
138.68
138.68
138.68
138.68
8
8
8
8
8
8
8
8
8
8
8
8
8
6.2 × 10
2.6 × 10
ND
1:1
1:1
ND
1:1
ND
ND
1:5
ND
5:1
9:1
9:1
1:9
1:9
63
20
e
-5
0.0099 0.040 ND
1.5
19
42
29
4.2
52
45
160
0.63
150
10
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
35
ND
ND
22
-
5
3.4 × 10
ND
ND
>1000 540.64
73 540.64
>1000 540.64
660
8.0
680
430
790
-
-
-
-
5
4
5
4
0.14
30
ND
3.9 × 10
ND
4.5
0.041
2.5
1.4
4.0
8.3
ND
13
ND
ND
540.64
540.64
540.64
540.64
540.64
1.8 × 10
ND
6.5 × 10
1
1
1
1
0.068 0.063 1.6
7.9
4.5
3.9
ND
ND
ND
42
92
ND
ND
ND
ND
110 ND
6.0 × 10
a
Daylight Chemical Information Systems, version 4.71. b PSA ) polar surface area. c At pH 11 Osteoclast resorption assay. e ND ) not determined.
d
Table 4. Populations of Axial Conformations from 10 000-Step Monte
Carlo Conformational Searches with the MMFFs/CHCl3 Protocol (8
kcal/mol Cutoff)
conformer populations, conformational searches were carried
out for azepanones 1, 2, 3, 7, and 10 and their methyl
thiohemiketal counterparts (simple models of active site cys-
teine-inhibitor complexes) (Table 4). Of the five compounds
examined, compound 2 with a cis-5-methyl group, is the most
potent cathepsin K inhibitor (Ki ) 0.0099 nM) and has
calculated axial conformer populations of 25% and 47% for
the ketone form and the methyl thiohemiketal form, respectively.
The trans-5-methyl isomer 3 is the least potent inhibitor
examined in this set (Ki ) 1.5 nM) and has only 10% and 7%
of its calculated axial conformer population for the ketone and
methyl thiohemiketal forms, respectively. Compound 1, with
no methyl substitution, has an intermediate level of inhibitory
potency (Ki ) 0.16 nM) and has an intermediate level of
calculated axial conformer populations of 19% for both ketone
and methyl thiohemiketal forms. The trans-6-methyl analogue
7
has approximately the same inhibitory potency (Ki ) 0.14
no. axial
conformers
< 8 kcal/mol < 8 kcal/mol
no. total
conformers
nM) as compound 1 and has calculated axial conformer
populations for both ketone and methyl thiohemiketal forms
comparable to compound 1. However, the cis-7-methyl analogue
molecular
form
fraction
axial
Ki
(nM)
no.
1
azepanone
thiol-reacted
azepanone
thiol-reacted
azepanone
thiol-reacted
azepanone
thiol-reacted
azepanone
thiol-reacted
161
245
201
607
61
102
220
466
96
848
1298
817
1296
627
1432
836
1424
1149
1480
0.190
0.189
0.246
0.468
0.097
0.071
0.263
0.327
0.084
0.189
0.16
1
0, the second most potent inhibitor of this set of compounds
(Ki ) 0.041 nM), has a calculated axial conformer population
2
3
7
0.0099
1.5
of only 8% for the ketone form and 19% for the methyl
thiohemiketal form, lower than expected on the basis of the
other calculated results. So, while the conformational analysis
employed here demonstrates that compounds with larger
calculated axial conformer populations (in either the ketone form
or as the methyl thiohemiketal form) have greater enzyme
inhibitory potency, this trend does not hold for compound 10,
and, therefore, cannot be used to fully rationalize the enzymol-
ogy results.
0.14
1
0
0.041
279
which mole fractions are determined from the NMR data. The
resulting conformer populations were analyzed to assess ring
conformations as well as substituent orientations. This analysis
revealed that the mole fractions of twist-boat conformations of
Cathepsin K X-ray Cocrystallography. High-quality crys-
tals of inhibitor 10 bound to cathepsin K were obtained by the
vapor diffusion method. The structure was determined by
molecular replacement with a model consisting of all protein
atoms from the previously determined cathepsin K/E-64 com-
1
, 2, 7, and 10 were 0.10, 0.02, 0.26, and 0.31, respectively.
Therefore, the predominant conformations of all four compounds
examined were chair, and 2 had very little twist-boat present.
In addition, substituent orientations were very similar for all
four azepanone solution conformer populations. In all of the
cases examined, the amide adopted an equatorial orientation
with no evidence of axial amide orientations within the detection
limits of proton NMR, even though axial amide orientations
were available in the conformer set used as input to the NAMFIS
procedure.
9
plex (PDB ID 1ATK), and the structure was refined to 2.5-Å
resolution. In Figure 4 is shown the X-ray cocrystal structure
of 10 in the cathepsin K active site. The hydrophobic surface
(green), hydrophilic surface (blue), and active lone pairs
(magenta) are depicted. Compound 10 adopts a chair conforma-
tion (Table 2), and both the amide and the methyl groups are
oriented axially. The orientation of these groups is in dramatic
contrast to the preferred diequatorial conformations as observed
in a single molecule X-ray structure. Clearly, the inhibitor
For the purpose of exploring the potential relationships
between cathepsin K inhibitory potencies and calculated axial

1
602 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 5
Yamashita et al.
Figure 3. Percent total of the epimer left or formed vs the elapsed time after the addition of (from left to right, top to bottom) 2, 4, 11, and 13,
respectively. The epimer concentrations at different times were determined by reverse-phase HPLC methods.
Scheme 4. Interconversion of 5-, 6-, and 7-Methylazepanone Diastereomers at pH 11 by Epimerization
undergoes a conformational change on binding to the enzyme.
In addition, the axial methyl group of 10 makes contacts with
the top of the S1′ hydrophobic pocket, and the pyridine ring
binds in the S2′ hydrophobic pocket. A comparison with the
cathepsin K/1 cocrystal structure will be described in the
Discussion and Conclusions section.
Biological and Biochemical Characterization of Azepanones
Cathepsin Inhibition SAR. The rank order of cathepsin K
inhibitory potency of the 4S diastereomers is 5-cis-methyl-
azepanone (2, Table 3, Ki,app ) 0.0099 nM) > 7-cis-methyl-
azepanone (10, Ki,app ) 0.041 nM) > 6-trans-methylazepanone
(7, Ki,app ) 0.14 nM) ) parent azepanone (1, Ki,app ) 0.16 nM).

SAR of Azepan-3-one Cathepsin K Inhibitors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 5 1603
Figure 4. Stereoview of cathepsin K active site cocrystallized with 10 (hydrophobic surface green, hydrophilic surface blue, and hydrogen bond
donor/acceptor surface magenta).
)
3). The substitution of a methyl group on the azepanone ring
had remarkable and unpredictable effects on oral bioavailability,
clearance, and half-life in the rat (Table 5). Unsubstituted
azepanone 1 was a moderate clearance molecule and had a half-
life of 30 min with a volume of distribution three times total
body water. Oral bioavailability of 1 was 42%. In contrast, 5-cis-
methyl analogue 2, the most potent inhibitor of cathepsin K,
was rapidly cleared yet had a half-life of 59 min, probably
related to a volume of distribution approximately 7 times total
body water. Following oral administration, measurable concen-
trations were observed at only one time point in one animal.
Therefore, the oral bioavailability of 2 was below the lower
limit of sensitivity of the test system (<5.4%). The 6-trans-
methyl analogue 7 was cleared at a moderate rate with a half-
life of about 52 min and a volume of distribution approximately
double that of total body water. The oral bioavailability of 7
was 26%. The 7-cis-methyl analogue 10 was a low to moderate
clearance molecule in the rat with a half-life of about 109 min
and a volume of distribution approximately twice total body
water. The oral bioavailability of 10 was outstanding at 89% in
the rat, the highest we had seen of the compounds described
here.
Figure 5. Stereoview comparison of 1 (top) and 10 (bottom) in
cathepsin K-bound conformations.
In the 5-methyl series, the cis-4S-diastereomer 2 was >150-
fold more potent than the trans-4S-diastereomer 3, and 2 was
the most potent cathepsin K inhibitor identified in this study.
In the 6-methyl series, the trans-4S-diastereomer 7 was >200-
fold more potent than the cis-4S-diastereomer 6. In the 7-methyl
series, the cis-4S-diastereomer 10 was >60-fold more potent
than the trans-4S-diastereomer 11. The 4R-diastereomers had
much poorer cathepsin K inhibitory potency (>30-fold) than
the 4S-diastereomers, as was also noted in ref 3 comparing 1
with its 4R-epimer. Quantification of the Ki,apps of the 4R-
diastereomers may have been limited by the inability to
completely purify the samples free of their corresponding, much
more potent 4S-counterparts (although >95% purity was
achieved as measured by analytical HPLC). Compound 10 was
generally a less selective inhibitor than 1 and had increased
inhibitory potency vs human and rat cathepsin K, human
cathepsins L and V, and especially vs human cathepsin B (Table
Compounds 1 and 10 were also evaluated in pharmacokinetic
iv/po crossover studies in monkeys. Compound 1 was a low to
moderate clearance molecule in the monkey with a half-life of
3
6.5 min and a volume of distribution approximately equal to
total body water. The oral bioavailability of 1 was 7% in the
monkey. Compound 10 had a low rate of clearance in the
monkey and a half-life of 168 min with a volume of distribution
approximately equal to total body water. The oral bioavailability
of 10 was 28% in the monkey.
3
). Despite this reduced selectivity, the improved pharmacoki-
netic and development characteristics of 10 discussed below
led to its choice as candidate for human clinical trials.
The percent rat plasma protein binding of 1, 2, 7, and 10
was determined by either equilibrium dialysis or ultrafiltration
and ranged from 95% to 97% (Table 5). Also, permeation rates
in Madin-Darby canine kidney (MDCK) cells and in an
In Vitro Biological Activity. The most potent diastereomers
from each azepanone regioisomeric series (1, 2, 7, and 10) were
tested in the in vitro osteoclast resorption assay.10 Briefly, in
this assay compounds are incubated for 48 h at a range of
concentrations with human osteoclastoma-derived osteoclasts
seeded onto bovine cortical bone slices. The production of type-I
collagen C-telopeptides is measured using a commercial ELISA
assay (Osteometer Biotech, Herlev, Denmark). The mean IC50
of 1 was 63 nM (Table 3, n ) 8, standard deviation ) 29 nM).
Compounds 2, 7, and 10 had IC50s ranging from 20 to 35 nM
11
artificial black-lipid membrane assay were quantified (Table
5
). No significant differences were observed between the
compounds. It is therefore unlikely that protein binding, renal
clearance, or membrane permeation rate play a significant role
in the pharmacokinetic differences observed for these com-
pounds.
Discussion and Conclusions
(
Table 3, n ) 1). Given that the IC50s in this assay can vary as
10
much as ∼3-fold, these activities are considered to be
Attempts to rationalize the differences in cathepsin K
inhibitory potencies of the various methyl-substituted azepanone
inhibitors relative to the conformational accessibility of the
equatorial vs axial amide orientations were made, as this had
been an initial goal of preparing these isomeric series. The 4S-
5-trans-methyl analogue (3), the 4S-6-cis-methyl analogue (6),
approximately equivalent.
Pharmacokinetics, Plasma Protein Binding, and Cell and
Membrane Permeability. The most potent diastereomers from
each azepanone regioisomeric series (1, 2, 7, and 10) were
evaluated in pharmacokinetic iv/po crossover studies in rats (n

1
604 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 5
Yamashita et al.
Table 5. Rat and Monkey Pharmacokinetic Parameters, Rat Plasma Protein Binding, Cell and Membrane Permeability
rat (monkey)
iv
permeability (×10- cm/s)
5
%
plasma
artificial
no.
T1/2, min
CL, mL/min/kg
Vdss, L/kg
po: %F
protein binding
MDCK
5.7
membrane
1
29.8 ( 2.22
36.5 ( 0.06)
58.5 ( 21.8
49.2 ( 4.5
(13.9 ( 3.4)
80.5 ( 16.5
1.86 ( 0.36
(0.72 ( 0.20)
4.70 ( 2.16
42.1 ( 19.8
(7.3 ( 4.7)
<5.4
97.0 ( 0.1
5.3
6.8
5.6
4.7
(
2
7
94.7 ( 1.5
95.6 ( 0.1
96.8 ( 0.3
5.5
4.9
4.7
a
ND
51.6 ( 12.2
109 ( 8.0
34.4 ( 2.9
1.46 ( 0.20
26.2 ( 10.2
10
19.5 ( 4.0
(11.7 ( 3.1)
1.79 ( 0.50
(0.95 ( 0.38)
89.4 ( 32.3
(27.6 ( 3.5)
(
168 ( 25)
a
ND ) not determined.
and the 4S-7-trans-methyl analogue (11) were 60 to >200-fold
less potent than their corresponding 4S-5-cis-methyl analogue
10 (Ki,app ) 13 nM) and 1 (Ki,app ) 500 nM). Cathepsin K-1
and -10 cocrystal structures were aligned with a publicly
available cathepsin B crystal structure (PDB ID 1CSB).12 In
both cathepsin K and cathepsin B, the S1′ binding pockets are
delineated by a â strand (residues 162-166 in cathepsin K and
residues 197-203 in cathepsin B). In cathepsin B, this â strand
is shifted relative to cathepsin K, reducing the size of the S1′
pocket. We speculate that this reduced size of S1′ in cathepsin
B results in a poor fit of the pyridine of 1 due to steric clashes.
Furthermore, compound 1 lacks a 7-cis-methyl group and
therefore does not form this hydrophobic interaction with S1′
of cathepsin B. In addition, it is plausible that 10 would bind
to cathepsin B in an analogous fashion to its binding mode to
cathepsin K and would therefore make favorable interactions
with the cathepsin B S1′ and S2′ pockets.
The 4S-5-cis-methylazepanone analogue, 2, had <5.4% rat
oral bioavailability and had a rat in vivo clearance rate of 80.5
mL/min/kg, while the 4S-7-cis-methylazepanone analogue, 10,
had 89% rat oral bioavailability and a rat in vivo clearance rate
of 19.5 mL/min/kg. On the basis of the lack of differentiation
of 2 and 10 in plasma protein binding and in cellular and
membrane permeability assays, it is likely that the differences
in oral bioavailability are due to altered rates or efficiency of
first-pass hepatic extraction and/or metabolism. We considered
several hypotheses that might rationalize the differences in the
pharmacokinetic profiles. One simple explanation is a steric
hypothesis, in which the 7-methyl group sterically blocks sites
of metabolism. Another explanation is a ring conformation
hypothesis, in which compound 2 can readily access azepanone
ring conformations that are readily recognized by in vivo first
pass clearance systems (P-glycoproteins, CYP450s, proteases,
etc.), and compounds such as 10 cannot readily access the same
conformations. The small molecule crystal structure of 10
revealed that it crystallized in a twist-boat conformation (Table
2). On the basis of this small molecule X-ray structure, it was
worth considering whether chair azepanone conformations might
be well recognized by in vivo clearance systems and twist-boat
conformations might be poorly recognized. However, on the
basis of the NAMFIS study results, azepanones 1, 2, 7, and 10
all adopted predominantly chair conformations in solution, and
compounds 7 and 10 were not largely differentiated in their
mole fraction of twist-boat populations (26% vs 31%, respec-
tively). Obviously, the chair vs twist-boat hypothesis is too
simplistic, since in an azepanone system there are a large number
of possible chair and twist-boat conformations that are poten-
tially accessible. For instance, in the various X-ray crystal-
lography studies (Table 2), three different chair conformations
and a twist boat conformation were observed: the same chair
in four of the small molecule structures of 1, 2, 5, and 6; a
different chair in the cathepsin K-1 cocrystal structure; yet
(2), 4S-6-trans-methyl analogue (7), and 4S-7-cis-methyl ana-
logue (10). In addition, 2 and 10 were more potent cathepsin K
inhibitors than the parent azepanone 1. We had hypothesized
that the solution population of axial vs equatorial amides might
vary with methyl group substitution of the azepanone ring.
However, the results of the NAMFIS conformational study
revealed that there were no axial amides present with 1, 2, 7,
or 10 in solution within the detection limits of proton NMR.
Therefore, this hypothesis is inconsistent with the experimental
data.
An alternative hypothesis was considered in which methyl
substitution might alter the energy required to interconvert the
equatorial and axial amide orientations. In cases where methyl
substitution stabilizes the equatorial amide, the cathepsin K
inhibitory potency should be decreased; in cases where the
methyl group destabilizes the equatorial amide, the cathepsin
K inhibitory potency should be increased. Indeed, in both sets
of 4S-5- and 6-methylazepanone analogues, the differences in
potency of 3 vs 2 and 6 vs 7 can be rationalized by this
hypothesis. Both 3 and 6 should have greater stabilization of
diequatorial chair conformers relative to the diaxial chair
conformers, and both 2 and 7 should require a lower energy of
interconversion of the two axial-equatorial chair conformers.
However, in the case of the 7-methyl analogues, this hypothesis
is not consistent with the observed potencies. Compound 10
should have greater stabilization of the diequatorial chair
conformer relative to the diaxial chair conformer, but was a
more active cathepsin K inhibitor than 1; compound 11 should
require a lower energy of interconversion of the two axial-
equatorial chair conformers, but was a less potent cathepsin K
inhibitor than 1.
The X-ray cocrystal structure of 10 bound to cathepsin K
provided a plausible explanation for this disparity with the
7
-methylazepanone analogues. When bound to cathepsin K, 10
adopts a chair conformation (10C, Table 2) with both the amide
and the methyl group adopting axial orientations. The methyl
group of 10 occupies the S1′ hydrophobic pocket of the enzyme,
and the pyridine ring interacts with the S2′ hydrophobic pocket
(Figure 4). With compound 1, the pyridine ring occupies the
S1′ hydrophobic pocket and forms an aromatic-aromatic inter-
action with Trp184. Therefore, it appears that favorable interac-
tions with the enzyme counteracted the unfavorable diaxial chair
conformation of 10. Furthermore, the methyl group of 10 affords
an additional interaction with the enzyme that cannot be accessed
by the methyl group of the epimeric analogue 11.
The difference in orientation of the pyridine ring of 10 vs 1
in the cathepsin K cocrystal structures (Figure 5) also provided
insight into the differences in cathepsin B selectivity between

SAR of Azepan-3-one Cathepsin K Inhibitors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 5 1605
another chair in the cathepsin K-10 cocrystal structure; and a
twist-boat in the small molecule X-ray structure of 10. It is
possible that a more sophisticated analysis of particular azepanone
chair and twist-boat conformations accessible by compounds
such as 2 and 10 might begin to rationalize the in vivo
pharmacokinetic results.
In addition to preferred conformations, we also considered a
compound rigidity hypothesis as another way to rationalize the
differences in observed oral bioavailabilities of compounds 1,
5-Methylazepanones. 3-Methyl-4-nitrobutyric Acid Ethyl
Ester (15). To a solution of ethyl 2-crotonate 14 (10 g, 87 mmol)
in nitromethane (23 mL, 438 mmol) was added 1,1,3,3-tetrameth-
ylguanidine (2.0 g, 17 mmol). The solution was stirred at room
temperature for 24 h. Ether (500 mL) was added and the organic
solution was washed with 1 N HCl (100 mL) and dried over sodium
sulfate. The solution was filtered, concentrated, and purified on a
silica gel column (20-30% EtOAc/hexane) to yield the title
1
compound (14 g, 91%): H NMR (CDCl ) δ 4.50 (dd, J ) 12.1,
3
6.2 Hz, 1H), 4.38 (dd, J ) 12.1, 7.2 Hz, 1H), 4.18 (q, J ) 7.1 Hz,
2H), 2.86-2.77 (m, 1H), 2.47 (dd, J ) 16.1, 6.7 Hz, 1H), 2.38
7
)
, and 10. In a previous study of a diverse set of compounds (n
1117), a high probability of good oral bioavailability (>20%)
(
6
dd, J ) 16.1, 6.9 Hz, 1H), 1.29 (t, J ) 7.1 Hz, 3H), 1.13 (d, J )
+
.8 Hz, 3H); MS (ESI) 174.0 (M+H) .
in the rat was observed with compounds with a rotatable bond
count (nRot) of less than 8.13 All of the azepanones discussed
3-Methyl-4-nitrobutyraldehyde (16). To a solution of 3-methyl-
-nitrobutyric acid ethyl ester 15 (1.0 g, 5.71 mmol) in dry toluene
4
here have a nRot ) 8 and therefore lie at or near an inflection
point between high and low probability of good oral bioavail-
ability. In general, methyl substitution of the azepanone should
further decrease the overall ring flexibility, even though rings
are wrongly assumed to be inflexible (nRot ) 0) in this simple
model. 7-Methyl substitution of the azepanone should decrease
the rotational freedom of the 4-sulfonamide substituent. We
speculate that these decreases in rotational freedom might
rationalize the difference of oral bioavailability of 10 compared
to 1. 6-Methyl substitution should have little influence on the
rotational freedom of the ring substituents, and indeed, the trans-
at -78 °C was slowly added DIBAL-H (4 mL, 1.5 M solution in
toluene) so as to maintain the internal temperature below -65 °C.
The reaction was stirred for 2 h at -78 °C. The reaction was then
quenched by slowly adding cold (-78 °C) MeOH while the internal
temperature was kept below -65 °C. The resulting white emulsion
was slowly poured into ice-cold 1N HCl with swirling over 15 min
and the aqueous mixture was then extracted with EtOAc (3×). The
combined organic solution was washed with brine, dried over
sodium sulfate, and concentrated to give the crude product which
was then purified on a silica gel column (25% EtOAc/hexane) to
1
give the pure product as a pale yellow oil (0.73 g, 98%): H NMR
(
3
CDCl ) δ 9.79 (s, 1H), 4.45 (dd, J ) 12.0, 6.4 Hz, 1H), 4.38 (dd,
6-methyl isomer 7 is most like 1 in pharmacokinetic properties.
J ) 12.0, 6.0 Hz, 1H), 2.94-2.85 (m, 1H), 2.69 (dd, J ) 18.2, 6.1
5
-Methyl substitution should decrease the rotational freedom
Hz, 1H), 2.55 (dd, J ) 18.2, 7.0 Hz, 1H), 1.13 (d, J ) 6.9 Hz,
of the 4-amide substituent. Despite having decreased compound
flexibility, 2 has decreased oral bioavailability.
3
H).
2
-[Benzyl(3-methyl-4-nitrobutyl)amino]ethanol (17). To a
solution of 3-methyl-4-nitrobutyraldehyde 16 (0.73 g, 5.57 mmol)
in CH Cl (6.0 mL) were added sodium triacetoxyborohydride (1.57
One particularly interesting aspect of the compounds de-
scribed in this paper is that they are not largely differentiated
by simple calculated diversity metrics (Table 3) such as
Lipinski’s rule-of-five (molecular weight, clog P, number of
2
2
g, 7.40 mmol) and N-benzylethanolamine (0.55 g, 3.67 mmol) at
room temperature. The reaction was stirred for 16 h, whereupon it
was quenched with water, diluted with EtOAc, and washed with
14
hydrogen bond donors or acceptors), number of rotatable
3
aqueous NaHCO and brine. The organic layer was dried over
1
3
15
bonds, or polar surface area. However, the effects on the
compound properties that matter most to medicinal chemists,
namely, potency and in vivo pharmacokinetic parameters such
as oral bioavailabilities and clearance rates, were profound.
Given the close structural similarity of 2 and 10, we speculate
that subtle conformational differences of the azepanone rings
are giving rise to the differences in pharmacokinetic parameters.
In addition, these examples demonstrate the potential for
modulation of pharmacological properties of cathepsin inhibitors
by substituting an azepanone core. Furthermore, such precedents
emphasize the importance of the continued use of in vivo
pharmacokinetic screening in the drug discovery process. The
high potency for inhibition of cathepsin K coupled with the
favorable rat and monkey pharmacokinetic characteristics of
compound 10, also known as SB-462795 or relacatib, has made
it the subject of considerable in vivo evaluation for safety and
efficacy as an inhibitor of excessive bone resorption in rat,
monkey, and human subjects. Results of these numerous in vivo
studies will be reported elsewhere.
sodium sulfate, concentrated, and used directly in the next reac-
1
tion: H NMR (CDCl ) δ 7.41-7.29 (m, 6H), 4.29-4.19 (m, 2H),
3
3
.69-3.59 (m, 4H), 2.79-2.31 (m, 4H), 1.61-1.39 (m, 4H), 0.97
+
(d, J ) 6.9 Hz, 3H); MS (ESI) 265.3 (M + H) .
1-Benzyl-5-methyl-4-nitroazepan-3-ol (19). To a stirring solu-
2 2 2 2
tion of oxalyl chloride (2 M in CH Cl ) (3.38 mL) in CH Cl at
-78 °C was added DMSO (1.25 mL, 17.6 mmol) slowly. After
stirring for 15 min, a solution of alcohol 17 (0.60 g, 2.25 mmol) in
2 2
CH Cl (1 mL) was added slowly. The reaction was continued for
a further 1 h at -78 °C to form the aldehyde 18 in situ.
Triethylamine (4.7 mL, 33.8 mmol) was added, the reaction mixture
was brought to room temperature and quenched with water, and
2 2
the product was extracted into CH Cl . The organic layer was dried
over sodium sulfate, filtered, and concentrated. To the crude product
in THF was added triethylamine and the mixture stirred for 16 h at
room temperature. The crude product was purified on a silica gel
column (1-3% MeOH/DCM) to give title compound as a mixture
1
of diastereomers (0.4 g, 41%, two steps): H NMR (CDCl
7
3
) δ
.39-7.21 (m, 5H), 4.29-4.19 (m, 2H), 3.8 (s, 2H), 2.98-2.51
m, 5H), 1.87-1.79 (m, 2H), 1.10 (d, J ) 6.7 Hz, 3H); MS (ESI)
(
2
+
65.24 (M + H) .
4
-Amino-1-benzyl-5-methylazepan-3-ol (20). To a solution of
Experimental Section
MeOH (56 mL) and 12 N HCl (5.6 mL) was slowly added Zn dust
(0.43 g, 6.47 mmol). Nitroazepanol 19 (171 mg, 0.65 mmol) was
added and the reaction was heated to reflux for 18 h, whereupon it
was concentrated in vacuo to remove the MeOH. The residue was
diluted with EtOAc and water and made basic with solid KOH.
The mixture was washed with brine, dried over sodium sulfate,
filtered, and concentrated to give the title compound as a mixture
General. Except where indicated, materials and reagents were
used as supplied. NMR spectra for compound characterization were
recorded at 400 MHz using a Bruker AC 400 spectrometer. Mass
spectra were taken on a PE Sciex API III instrument using
electrospray (ES) ionization techniques. Elemental analyses were
obtained using a Perkin-Elmer 240C elemental analyzer. Reactions
were monitored by TLC analysis using Analtech silica gel GF or
E. Merck silica gel 60 F-254 thin layer plates. Flash chromatography
was carried out on E. Merck Kieselgel 50 (230-400 mesh) silica
gel. Preparative and analytical HPLC were carried out on either
the Rainin HPXL or Gilson 306 HPLC systems.
+
of diastereomers (120 mg, 80%): MS (ESI) 235.2 (M + H) .
[(S)-1-(1-Benzyl-3-hydroxy-5-methylazepan-4-ylcarbamoyl)-
3-methylbutyl]carbamic Acid tert-Butyl Ester (21). To a solution
of the crude aminoazepanols 20 (1.12 g, 4.76 mmol) in CH
2 2
Cl
(20 mL) were added Boc-L-leucine (1.3 g, 4.76 mmol), EDC (1 g,

1
606 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 5
.76 mmol), and HOBT (0.13 g, 0.96 mmol). This mixture was
Yamashita et al.
4
(m, 5H), 0.93-0.85 (m, 3H), 0.80 (d, J ) 7.58 Hz, 2H); MS (ESI)
+
stirred at room temperature for 3 h, whereupon it was diluted with
EtOAc and washed with an aqueous solution of sodium bicarbonate.
The organic layer was dried over sodium sulfate, filtered, and
concentrated. The crude product was purified on a silica gel column
541.2 (M + H) ; analytical HPLC, Impaq silica gel 10 µm column,
4.6 × 250 mm 60 A, 93% CH
2 2
Cl /THF, 1.0 mL/min, UV 235 nm,
retention time ) 9.1 min, >95% purity. Anal. (C27
32 4 6
H N O S) C,
59.98; H, 5.97; N, 10.36. Found: C, 59.29; H, 5.98; N, 10.14. 5:
1
(50% EtOAc/hexane) to remove polar impurities, and the mixture
3
H NMR (CDCl ) δ 8.69 (d, J ) 4.55 Hz, 1H), 7.99-7.91(m, 2H),
of diastereomers was carried on to the next reaction (0.12 g,
7.69 (d, J ) 7.58 Hz, 1H), 7.58-7.50 (m, 2H), 7.45 (td, J ) 7.83,
1.26 Hz, 1H), 7.34-7.26 (m, 1H), 7.19 (d, J ) 6.06 Hz, 1H), 7.09
(d, J ) 8.08 Hz, 1H), 5.30 (dd, J ) 6.06, 2.78 Hz, 1H), 4.80-4.72
(m, 1H), 4.65 (dd, J ) 19.33, 1.64 Hz, 1H), 3.93 (s, 1H), 3.78 (d,
J ) 19.20 Hz, 1H), 3.00 (s, 1H), 2.50 (s, 1H), 2.38 (s, 1H), 1.86
(s, 1H), 1.78-1.70 (m, 3H), 1.67 (s, 1H), 1.02 (dd, J ) 6.32, 2.27
+
2
8%): MS (ESI) 448.4 (M + H) .
(S)-1-(3-Hydroxy-5-methylazepan-4-ylcarbamoyl)-3-methyl-
butyl]carbamic Acid tert-Butyl Ester (22). To a solution of amides
[
2
1
1 (10 g, 22.4 mmol) in MeOH and EtOAc (10 mL, 1:3) was added
0% Pd/C (2 g, 1.88 mmol) at room temperature. This mixture
+
was shaken for 16 h on a Parr hydrogenation apparatus at 45 psi
of hydrogen gas. The reaction mixture was filtered through a pad
of Celite and concentrated to give the title compound which was
Hz, 6H), 0.84 (d, J ) 7.33 Hz, 3H); MS (ESI) 541.2 (M + H) ;
analytical HPLC, Impaq silica gel 10 µm column, 4.6 × 250 mm
2 2
60 A, 93% CH Cl /THF, 1.0 mL/min, UV 235 nm, retention time
used in the next reaction without further purification: MS (ESI)
) 13.4 min, >95% purity. Anal. (C27
32 4 6
H N O S) C, 59.98; H, 5.97;
+
1
3
58.4 (M + H) .
N, 10.36. Found: C, 59.16; H, 5.93; N, 10.10. 4: H NMR (CDCl )
3
{
(S)-1-[3-Hydroxy-5-methyl-1-(pyridine-2-sulfonyl)azepan-4-
ylcarbamoyl]-3-methylbutyl}carbamic Acid tert-Butyl Ester
23). To a solution of amines 22 (6 g, 16.8 mmol) in CH Cl (30
δ 8.70 (d, J ) 4.29 Hz, 1H), 7.98-7.90 (m, 2H), 7.66 (d, J ) 7.83
Hz, 1H), 7.55-7.48 (m, 3H), 7.45-7.40 (m, 1H), 7.32-7.26 (m,
1H), 7.15 (d, J ) 8.34 Hz, 1H), 7.02 (d, J ) 8.08 Hz, 1H), 5.17-
5.11 (m, 1H), 4.82 (td, J ) 8.65, 5.18 Hz, 1H), 4.59 (dd, J ) 18.44,
1.52 Hz, 1H), 4.03-3.94 (m, 1H), 3.81 (d, J ) 18.19 Hz, 1H),
2.93-2.83 (m, 1H), 1.90-1.82 (m, 3H), 1.79-1.70 (m, 3H), 1.14
(d, J ) 6.82 Hz, 3H), 1.05-0.96 (m, 6H); MS (ESI) 541.2 (M +
(
2
2
mL) were added 2-pyridinesulfonyl chloride (3 g, 16.9 mmol) and
triethylamine (3 mL, 22.5 mmol). The reaction was allowed to stir
at room temperature for 16 h, whereupon it was washed with
NaHCO . The organic layer was dried over sodium sulfate, filtered,
3
concentrated, and purified on a silica gel column (80% EtOAc/
+
H) ; analytical HPLC, Impaq silica gel 10 µm column, 4.6 × 250
hexane) to yield the title compound as a mixture of diastereomers
2 2
mm 60 A, 93% CH Cl /THF, 1.0 mL/min, UV 235 nm, retention
+
(
5.36 g, 48%, two steps): MS (ESI) 499.1 (M + H) .
time ) 16.1 min, >95% purity. Anal. (C27
5.97; N, 10.36. Found: C, 59.87; H, 6.00; N, 9.99. 3: H NMR
(CDCl ) δ 8.76-8.71 (m, 1H), 8.02-7.93(m, 2H), 7.68 (d, J )
32 4 6
H N O S) C, 59.98; H,
1
Benzofuran-2-carboxylic Acid {(S)-3-Methyl-1-[5-methyl-3-
hydroxy-1-(pyridine-2-sulfonyl)azepan-4-ylcarbamoyl]butyl}-
amide (24). To a solution of sulfonamides 23 (5.36 g, 11.6 mmol)
in MeOH (2 mL) was added 4 M HCl/dioxane (25 mL) and the
mixture stirred for 2 h at room temperature. The solvent and an
excess amount of HCl were removed in vacuo, and the residue
was azeotropically dried twice with toluene to yield the primary
amine as a hydrochloride salt (5.37 g), which was used in the next
reaction without further purification. To a solution of the intermedi-
3
7.33 Hz, 1H), 7.56-7.52 (m, 2H), 7.50 (d, J ) 1.01 Hz, 1H), 7.44
(td, J ) 7.83, 1.26 Hz, 1H), 7.34-7.27 (m, 1H), 7.14 (d, J ) 8.59
Hz, 1H), 6.91 (d, J ) 8.59 Hz, 1H), 5.20-5.13 (m, 1H), 4.81-
4.71(m, 2H), 4.06-3.96 (m, 1H), 3.88 (d, J ) 18.44 Hz, 1H), 2.92-
2.82 (m, 1H), 1.88-1.83 (m, 2H), 1.81-1.69 (m, 4H), 1.08 (d, J
) 6.82 Hz, 3H), 1.00 (dd, J ) 6.06, 3.54 Hz, 6H); MS (ESI) 541.2
+
(M + H) ; analytical HPLC, Impaq silica gel 10 µm column 4.6
ate amine (0.66 g, 1.26 mmol) in CH
carboxylic acid (0.24 g, 1.51 mmol), EDC (0.29 g, 1.51 mmol),
HOBT (0.04 g, 0.29 mmol), and Et N (1 mL). The reaction mixture
Cl
2 2
were added 2-benzofuran
× 250 mm 60 A, 93% CH
2 2
Cl /THF, 1.0 mL/min, UV 235 nm,
retention time ) 19.5 min, >95% purity. Anal. (C27
32 4 6
H N O S) C,
59.98; H, 5.97; N, 10.36. Found: C, 59.02; H, 6.06; N, 10.01.
6-Methylazepanones. Allyl-(2-methylpent-4-enyl)amine (27).
To a solution of 2-methylpent-4-enoic acid ethyl ester 25 (7.1 g,
50 mmol) was added dropwise a solution of DIBAL-H (1.0 M in
hexanes, 75 mL) at -78 °C over 1 h. After addition, the reaction
mixture was stirred at -78 °C for another 1 h. The reaction was
quenched with saturated aqueous ammonium chloride (10 mL) and
4% aqueous HCl and then was extracted with EtOAc (3 × 100
3
was stirred at room temperature for 3 h, whereupon it was washed
with an aqueous solution of sodium bicarbonate. The organic layer
was dried over sodium sulfate, filtered, and concentrated. The crude
product was purified on a silica gel column (80% EtOAc/hexane)
to yield the title compound (0.55 g, 62%) as a mixture of
diastereomers that was carried on to the next step: MS (ESI) 565.08
+
(
M + Na) .
Benzofuran-2-carboxylic Acid {(S)-3-Methyl-1-[5-methyl-3-
4
mL). The combined organic extracts were dried over MgSO ,
oxo-1-(pyridine-2-sulfonyl)azepan-4-ylcarbamoyl]butyl}amide
filtered, and concentrated by rotary evaporation. The crude product
was used in the next reaction without further purification. 2-Methyl-
4-pentenal 26 (3.3 g, 33.7 mmol) was dissolved in CH Cl (100
2 2
mL). To this solution was added allylamine (2.9 g, 50.5 mmol).
Molecular sieves (4 Å, 5 g) were used to absorb the water generated
during the reaction. The mixture was stirred at room temperature
overnight. The reaction mixture was concentrated by rotary
evaporation and the crude product was used in the next reaction
without further purification. Allyl-(2-methylpent-4-enylidene)amine
(3.2 g, 23.4 mmol) was diluted in 50 mL of MeOH. To the above
(
2-5). To a solution of alcohols 24 (0.15 g, 0.27 mmol) in CH
2 2
Cl
(2 mL) was added Dess-Martin periodinane (0.17 g, 0.41 mmol).
The reaction was stirred at room temperature for 1 h, diluted with
CH Cl , and then washed with aqueous sodium thiosulfate solution,
2
2
aqueous sodium bicarbonate solution, and brine. The organic layer
was washed, dried over sodium sulfate, filtered, concentrated, and
purified on a silica gel column (80% EtOAc/hexane) to provide
the title compound as a mixture of four diastereomers (0.1 g, 67%).
The cis-diastereomers 2 and 5 were separated from the trans-
diastereomers 4 and 3 by HPLC on an Impaq silica gel 10 µm
4
solution was added NaBH (1.0 g, 26.3 mmol) at 0 °C. After
column (4.6 × 250 mm 60 A, 93% CH
2
Cl
2
/THF, 1.0 mL/min, UV
addition, the mixture was stirred at room temperature for 5 h. The
reaction mixture was concentrated and the residue was partitioned
between EtOAc and 20% aqueous NaOH. The organic layer was
2
35 nm). Individual diastereomers from each pair were then
separated by HPLC using Chiralcel OD (4.6 × 250 mm, 80%
hexane/EtOH, 1 mL/min, UV 235) to provide individual diaster-
eomers as white powders: 2 (32 mg, 21.5%, 8.1 min), 5 (17 mg,
dried over Na
2 4
SO , filtered, and concentrated by rotary evaporation
1
to give the title compound (1.5 g, 48%): H NMR (CDCl
3
) δ 5.93-
1
1
7
7
1.4%, 10.9 min) and 4 (7 mg, 4.7%, 8.9 min), 3 (10 mg, 6.7%,
5.68 (m, 2H), 5.18-4.92 (m, 4H), 3.21 (d, J ) 5.2 Hz, 2H), 2.60-
2.40 (m, 2H), 1.97-1.65 (m, 2H), 0.92 (d, J ) 7.6 Hz, 3H).
Pyridine-2-sulfonic Acid Allyl(2-methylpent-4-enyl)amide (28).
Allyl(2-methyl-pent-4-enyl)amine 27 (1.0 g, 7.2 mmol) and 4-me-
1
2.9 min). 2: H NMR (CDCl ) δ 8.73 (d, J ) 4.04 Hz, 1H), 8.02-
3
.93 (m, 2H), 7.68 (d, J ) 7.58 Hz, 1H), 7.57-7.49 (m, 1H), 7.48-
.42 (m, 1H), 7.33-7.28 (m, 1H), 7.18 (d, J ) 6.57 Hz, 1H), 7.15
(
d, J ) 8.59 Hz, 1H), 5.32 (dd, J ) 6.82, 2.78 Hz, 1H), 4.81-4.73
thylmorpholine (1.7 g, 17.2 mmol) were mixed in 30 mL of CH
2
-
(m, 1H), 3.93 (dd, J ) 14.91, 1.26 Hz, 1H), 3.81 (d, J ) 19.20 Hz,
Cl . 2-Pyridinesulfonyl chloride (1.53 g, 8.6 mmol) was added
2
1
H), 3.02-2.93 (m, 1H), 2.46 (dd, J ) 7.33, 3.03 Hz, 1H), 2.39-
.28 (m, 1H), 1.79-1.69 (m, 3H), 1.34-1.24 (m, 3H), 1.05-0.97
slowly to the solution while it was cooled in an ice-water bath.
After addition, the reaction mixture was stirred at room temperature
2

SAR of Azepan-3-one Cathepsin K Inhibitors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 5 1607
overnight. The reaction mixture was washed with 10% aqueous
evaporated and the remaining material was azeotroped twice with
toluene (100 mL). The resulting thick oil was dissolved in MeOH
and treated with HCl in ether to adjust pH to acidic. Additional
ether was added and the solution turned cloudy to give the title
NaHCO
3
and brine and then was purified by column chromatog-
raphy to give the title compound as a colorless oil (1.2 g, 60%):
1
H NMR (CDCl
.47 (m, 1H), 5.80-5.60 (m, 2H), 5.15-4.92 (m, 4H), 4.00-3.90
m, 2H), 3.20-3.06 (m, 2H), 2.19-2.12 (m, 1H), 1.91-1.85 (m,
3
) δ 8.72-8.69 (m, 1H), 8.00-7.75 (m, 2H), 7.49-
7
(
compound as a white precipitate that was collected by filtration
1
(0.27 g, 71%): H NMR (CD
3
OD) δ 8.72 (d, J ) 4.8 Hz, 1H),
+
2
H), 0.89 (d, J ) 6.4 Hz, 3H); MS (ESI) 281.2 (M + H) .
-Methyl-1-(pyridine-2-sulfonyl)-2,3,4,7-tetrahydro-1H-
azepine (29). Pyridine-2-sulfonic acid allyl(2-methylpent-4-enyl)-
amide 28 (1.2 g, 4.3 mmol) was dissolved in CH Cl (100 mL).
8.14-7.99 (m, 2H), 7.70-7.66 (m, 1H), 3.85-3.69 (m, 2H), 3.38-
3
3.22 (m, 3H), 3.10-3.04 (m, 2H), 2.08-2.00 (m, 1H), 1.82-1.66
+
(m, 2H), 1.02 (d, J ) 6.8 Hz, 3H); MS (ESI) 286.0 (M + H) .
2
2
4-Amino-6-methyl-1-(pyridine-2-sulfonyl)azepan-3-ol (35).
After carefully degassing with argon, bis(tricyclohexylphosphine)-
benzylidineruthenium(IV) dichloride (0.35 g, 0.43 mmol) was added
under argon. The mixture was then refluxed for 2 h before the
reaction mixture was concentrated by rotary evaporation. The
product was purified by column chromatography (5%-20% EtOAc/
hexane) to give the title compound (0.9 g, 83% yield): 1H NMR
4-Azido-6-methyl-1-(pyridine-2-sulfonyl)azepan-3-ol 33 (0.48 g,
1.54 mmol) was dissolved in THF (50 mL) and H O (0.2 mL).
2
Triphenylphosphine (0.61 g, 2.32 mmol) was added to this solution.
The reaction mixture was stirred at 45 °C overnight. THF was
evaporated and azeotroped by toluene (2 × 100 mL). The resulting
thick oil was dissolved in MeOH and treated with HCl in ether to
adjust pH to acidic. Additional ether was added and the solution
(
7
(
2
CDCl
.49 (m, 1H), 5.79-5.60 (m, 2H), 4.00 (d, J ) 4.8 Hz, 2H), 3.65
dd, J ) 12.8, 4.0 Hz, 1H), 3.22 (dd, J ) 13.3, 9.1 Hz, 1H), 2.30-
.0+5 (m, 3H), 0.96 (d, J ) 6.4 Hz, 3H); MS (ESI) 253.2 (M +
H) .
3
) δ 8.70 (d, J ) 4.4 Hz, 1H), 8.00-7.75 (m, 2H), 7.52-
turned cloudy to give the title compound as a white precipitate that
1
was collected by filtration (0.27 g, 71%): H NMR (CDCl
3
) δ 8.73
(d, J ) 4.8 Hz, 1H), 8.13-7.99 (m, 2H), 7.70-7.66 (m, 1H), 3.84-
3.80 (m, 1H), 3.74-3.66 (m, 1H), 3.38-3.21 (m, 3H), 3.10-3.06
(m, 2H), 2.05 (brs, 1H), 1.85-1.66 (m, 2H), 1.02 (d, J ) 6.7 Hz,
5
-Methyl-3-(pyridine-2-sulfonyl)-8-oxa-3-azabicyclo[5.1.0]-
+
octane (30 and 31). To a solution of 3-methyl-1-(pyridine-2-
sulfonyl)-2,3,4,7-tetrahydro-1H-azepine 29 (1.3 g, 5.16 mmol) in
CH Cl (50 mL) were added NaHCO (1.3 g, 15.5 mmol) and
2 2 3
mCPBA (2.67 g, 15.5 mmol) in portions. The reaction was stirred
at room temperature for 4 h before working up by washing with
3H); MS (ESI) 286.0 (M + H) .
{(S)-1-[3-Hydroxy-6-methyl-1-(pyridine-2-sulfonyl)azepan-4-
ylcarbamoyl]-3-methylbutyl}carbamic Acid tert-Butyl Ester
(36a, 36b). 4-Amino-6-methyl-1-(pyridine-2-sulfonyl)azepan-3-ol
34 HCl salt (0.11 g, 0.59 mmol) was dissolved in 5 mL of DMF.
To this solution were added Boc-L-leucine (0.22 g, 0.88 mmol),
HBTU (0.34 g, 0.90 mmol), and N-methylmorpholine (0.24 g, 2.4
mmol). The reaction mixture was stirred at room temperature
overnight. DMF was removed under high vacuum. The residue was
2 3
15% aqueous NaOH, saturated aqueous K CO and brine, followed
by drying over Na SO . After filtration, the organic solution was
2
4
concentrated by rotary evaporation, and the two isomers were
separated on column chromatography (30%-40% EtOAc/hexane).
30 (cis-isomer, 230 mg, 17%), 31 (trans-isomer, 200 mg, 14%),
2 3
diluted in EtOAc and washed with H O, 10% aqueous NaHCO ,
and a mixture of 30 and 31 (570 mg, 46%) were used in next steps,
and brine. Purification by column chromatography (50% EtOAc/
hexane) gave the title compound as a mixture of diastereomers (0.2
1
carried on separately. 30: H NMR (CDCl ) δ 8.70 (d, J ) 4.8
3
g, 68%): 1H NMR (CDCl
Hz, 1H), 8.00-7.75 (m, 2H), 7.54-7.48 (m, 1H), 4.41-4.37 (m,
3
) δ 8.75 (d, J ) 4.4 Hz, 1H), 8.0-7.75
1
0
H), 3.93-3.90 (m, 1H), 3.34-2.00 (m, 6H), 1.45-1.35 (m, 1H),
(m, 2H), 7.5 6-7.50 (m, 1H), 5.10-5.00 (m, 1H), 4.15-2.90 (m,
10H), 2.10-1.48 (m, 14H), 1.05-1.00 (m, 9H); MS (ESI) 499.1
+
1
.88 (d, J ) 7.0 Hz, 3H); MS (ESI) 269.0 (M + H) . 31: H NMR
+
(CDCl
3
) δ 8.68-8.66 (m, 1H), 7.98-7.87 (m, 2H), 7.49-7.46 (m,
(M + H) .
1
3
H), 4.06-3.98 (m, 1H), 3.60-3.55 (m, 1H), 3.35-3.22 (m, 2H),
{(S)-1-[3-Hydroxy-6-methyl-1-(pyridine-2-sulfonyl)azepan-4-
ylcarbamoyl]-3-methylbutyl}carbamic Acid tert-Butyl Ester
(37a, 37b). 4-Amino-6-methyl-1-(pyridine-2-sulfonyl)azepan-3-ol
35 HCl salt (0.27 g, 0.76 mmol) was dissolved in 5 mL of DMF.
To this solution were added Boc-L-leucine (0.23 g, 0.91 mmol),
HBTU (0.35 g, 0.91 mmol), and N-methylmorpholine (0.39 g, 3.0
mmol). The reaction mixture was stirred at room temperature
overnight. DMF was removed under high vacuum. The residue was
diluted in EtOAc and washed with H O, 10% NaHCO , and brine.
.14-3.06 (m, 2H), 2.10-1.90 (m, 3H), 1.02 (d, J ) 6.6 Hz, 3H);
+
MS (ESI) 269.2 (M + H) .
4
-Azido-5-methyl-1-(pyridine-2-sulfonyl)azepan-3-ol (32).
5-Methyl-3-(pyridine-2-sulfonyl)-8-oxa-3-azabicyclo[5.1.0]octane
30 (230 mg, 0.86 mmol) was dissolved in 8 mL of MeOH and 2
2 3
mL of H O. NaN (170 mg, 2.6 mmol) and ammonium chloride
(140 mg, 2.6 mmol) were added to the above solution. The reaction
mixture was refluxed overnight. After the removal of MeOH, the
residue was diluted in EtOAc and washed with 10% aqueous
2
3
Purification by column chromatography (50% EtOAc/hexane) gave
the title compound as a mixture of diastereomers (0.35 g, 92%):
NaHCO
tography (1% MeOH/CH
4%): 1H NMR (CDCl
) δ 8.69 (d, J ) 4.8 Hz, 1H), 8.04-7.94
m, 2H), 7.56-7.53 (m, 1H), 4.00-2.95 (m, 7H), 2.23-2.14 (m,
H), 1.90-1.74 (m, 2H), 0.98 (d, J ) 6.9 Hz, 3H); MS (ESI) 312.2
3
and brine. The product was purified by column chroma-
1
2
Cl ) to give the title compound (170 mg,
2
H NMR (CDCl ) δ 8.70 (d, J ) 4.0 Hz, 1H), 7.99-7.92 (m, 2H),
3
6
3
7.55-7.48 (m, 1H), 6.46-6.42 (m, 1H), 4.60 (brs, 1H), 4.11-
4.02 (m, 1H), 3.88-3.38 (m, 5H), 3.02-2.95 (m, 3H), 2.05 (brs,
1H), 1.68-1.63 (m, 4H), 1.45 (s, 9H), 0.97-0.93 (m, 9H); MS
(
1
+
+
(M + H) .
(ESI) 499.1 (M + H) .
4
-Azido-5-methyl-1-(pyridine-2-sulfonyl)azepan-3-ol (33).
Benzofuran-2-carboxylic Acid {(S)-1-[3-Hydroxy-6-methyl-
1-(pyridine-2-sulfonyl)azepan-4-ylcarbamoyl]-3-methylbutyl}-
amide (38a, 38b). {(S)-1-[3-Hydroxy-6-methyl-1-(pyridine-2-
sulfonyl)azepan-4-ylcarbamoyl]-3-methylbutyl}carbamic acid tert-
butyl ester 36 (0.18 g, 0.36 mmol) was dissolved in 4 M HCl in
dioxane (2.8 mL, 11.2 mmol). The mixture was stirred at room
temperature for 2 h before solvents and an excess amount of HCl
were removed by rotary evaporation. The resulting white solid was
dissolved in 5 mL of DMF. To the solution were added benzofuran-
2-carboxylic acid (0.074 g, 0.46 mmol), HBTU (0.174 g, 0.46
mmol), and N-methylmorpholine (0.15 g, 1.52 mmol). The reaction
mixture was stirred at room temperature overnight. DMF was then
removed and the residue was dissolved in EtOAc (50 mL) and
5-Methyl-3-(pyridine-2-sulfonyl)-8-oxa-3-azabicyclo[5.1.0]octane
31 (700 mg, 2.6 mmol) was dissolved in 16 mL of MeOH and 4
2 3
mL of H O. NaN (510 mg, 7.8 mmol) and ammonium chloride
(420 mg, 7.8 mmol) were added to the solution. The resulting
mixture was refluxed overnight. After evaporation of MeOH, the
residue was diluted in EtOAc and washed with 10% aqueous
NaHCO
3
and brine. The product was purified by column chroma-
Cl ) to give the title compound (530 mg,
6%): 1H NMR δ 8.69-8.67 (m, 1H), 8.05-7.70 (m, 2H), 7.53-
.50 (m, 1H), 3.90-3.45 (m, 5H), 2.88-2.80 (m, 1H), 2.08-1.50
tography (1% MeOH/CH
6
7
2
2
+
(
m, 4H), 0.95 (d, J ) 6.7 Hz, 3H); MS (ESI) 312.2 (M + H) .
-Amino-6-methyl-1-(pyridine-2-sulfonyl)azepan-3-ol (34).
-Azido-6-methyl-1-(pyridine-2-sulfonyl)azepan-3-ol (32, 0.48 g,
.54 mmol) was dissolved in THF (50 mL) and H O (0.2 mL).
4
4
1
washed with 10% aqueous NaHCO
3
(50 mL × 2) and brine (50
2
mL). The combined organics were concentrated by rotary evapora-
tion. Purification by column chromatography (70% EtOAc/hexane)
gave the title compound as a mixture of diastereomers (0.15 g, 92%,
Triphenylphosphine (0.61 g, 2.32 mmol) was added to this solution.
The reaction mixture was stirred at 45 °C overnight. THF was

1
608 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 5
Yamashita et al.
two steps): 1H NMR (CDCl
) δ 8.74-8.65 (m, 1H), 7.93-7.87
compound (0.25 g, 78%). The diastereomers were separated by
HPLC using a Whelk-O (R,R) chiral prep column, 21.1 × 250 mm,
40% EtOH/hexane, 20 mL/min, UV 235 to provide individual
diastereomers as white powders: 9 (100 mg, 31%, 14.2 min), 6
3
(
5
0
m, 2H), 7.64-7.23 (m, 8H), 4.77-4.73 (m, 1H), 4.18-3.19 (m,
H), 3.01-2.80 (m, 1H), 1.90-1.44 (m, 5H), 0.99-0.94 (m, 8H),
+
.88-0.84 (m, 3H); MS (ESI) 543.2 (M + H) .
(100 mg, 31%, 16.7 min). 9: 1H NMR (CDCl
) δ 8.69 (d, J ) 4.8
Benzofuran-2-carboxylic Acid {(S)-1-[3-Hydroxy-6-methyl-
-(pyridine-2-sulfonyl)azepan-4-ylcarbamoyl]-3-methylbutyl}-
3
1
Hz, 1H), 7.98-7.91 (m, 2H), 7.66 (d, J ) 7.6 Hz, 1H), 7.54-7.01
(m, 7H), 5.23-5.19 (m, 1H), 4.82-4.73 (m, 2H), 3.98-3.96 (m,
1H), 3.83 (d, J ) 4.8 Hz, 1H), 2.36-2.31 (m, 1H), 2.17-2.04 (m,
3H), 1.74-1.73 (m, 2H), 1.25-1.14 (m, 1H), 1.00 (d, J ) 4.4 Hz,
amide (39). {(S)-1-[3-Hydroxy-6-methyl-1-(pyridine-2-sulfonyl)-
azepan-4-ylcarbamoyl]-3-methylbutyl}carbamic acid tert-butyl ester
37a and 37b (0.35 g, 0.70 mmol) were dissolved in 4 M HCl in
+
dioxane (3.5 mL, 14 mmol). The reaction mixture was stirred at
room temperature for 2 h before solvents and an excess amount of
HCl were removed by rotary evaporation. The resulting white solid
was dissolved in 5 mL of DMF. To the solution were added
benzofuran-2-carboxylic acid (0.15 g, 0.91 mmol), HBTU (0.35 g,
6H), 0.87 (d, J ) 6.0 Hz, 3H); MS (ESI) 541.2 (M + H) ; analytical
HPLC, Whelk-O (R,R) 5 µm, 4.6 × 250 mm, 40% EtOH/hexane,
1.0 mL/min, UV 215 nm, retention time ) 12.0 min, >95% purity.
32 4 6
Anal. (C27H N O S) C, 59.98; H, 5.97; N, 10.36. Found: C, 59.72;
H, 5.52; N, 10.21. 6: δ 8.64 (d, J ) 5.2 Hz, 1H), 7.93-7.87 (m,
2H), 7.63 (d, J ) 8.0 Hz, 1H), 7.49-7.37 (m, 4H), 7.28-7.17 (m,
3H), 5.21-5.17 (m, 1H), 4.80-4.70 (m, 2H), 3.97-3.94 (m, 1H),
3.80-3.69 (m, 1H), 2.36-2.30 (m, 1H), 2.15-2.11 (m, 3H), 1.83-
1.70 (m, 2H), 1.25-1.16 (m, 1H), 0.98 (d, J ) 5.2 Hz, 6H), 0.89
0.91 mmol), and N-methylmorpholine (0.3 g, 3.0 mmol). The
mixture was stirred at room temperature overnight. DMF was then
removed and the residue was dissolved in EtOAc (50 mL) and
washed with 10% aqueous NaHCO
mL). The combined organics were concentrated by rotary evapora-
tion. Purification by column chromatography (70% EtOAc/hexane)
gave the title compounds (0.32 g, 84%, two steps): H NMR
3
(50 mL × 2) and brine (50
+
(d, J ) 6.0 Hz, 3H); MS (ESI) 541.2 (M + H) ; analytical HPLC,
Whelk-O (R,R) 5 µm, 4.6 × 250 mm, 40% EtOH/hexane, 1.0 mL/
min, UV 215 nm, retention time ) 13.9 min, >95% purity.
7-Methylazepanones. Allyl-(1-methylpent-4-enyl)amine (41).
Hex-5-en-2-one 40 (9.8 g, 11.6 mL, 100 mmol) was added to a
stirred solution of allylamine (8.55 mmol, 11.25 mL, 150 mmol),
1
(
8
1
CDCl
3
) δ 8.69-8.66 (m, 1H), 8.02-7.86 (m, 2H), 7.68-7.03 (m,
H), 4.75-4.68 (m, 1H), 3.79-3.38 (m, 7H), 3.10-2.87 (m, 1H),
.9+6-1.64 (m, 5H), 1.06-0.86 (m, 9H); MS (ESI) 543.2 (M +
H) .
2 2
4 Å molecular sieves (52 g), and p-TsOH (10 mg) in CH Cl (200
Benzofuran-2-carboxylic Acid {(S)-3-Methyl-1-[(4S,6S)-6-
methyl-3-oxo-1-(pyridine-2-sulfonyl)azepan-4-ylcarbamoyl]butyl}-
amide (7) and Benzofuran-2-carboxylic Acid {(S)-3-Methyl-1-
mL) and the mixture was stirred overnight and then filtered. The
reaction mixture was concentrated by rotary evaporation and was
used in the next reaction without further purification (13 g, 95%).
To a stirred solution of allyl(1-methylpent-4-enylidene)amine (6.5
g, 47 mmol) in MeOH (100 mL) at 0 °C was added sodium
borohydride (2.7 g, 71 mmol) portionwise. The reaction mixture
was stirred for 30 min and then warmed to room temperature.
Approximately 90 mL of MeOH was removed from the reaction
mixture by rotary evaporation. The resultant reaction mixture was
diluted with ether (200 mL) and then washed with water and brine.
[
(4R,6R)-6-methyl-3-oxo-1-(pyridine-2-sulfonyl)azepan-4-
ylcarbamoyl]butyl}amide (8). To a solution of benzofuran-2-
carboxylic acid {(S)-1-[3-hydroxy-6-methyl-1-(pyridine-2-sulfonyl)-
azepan-4-ylcarbamoyl]-3-methylbutyl}amide 38a and 38b (0.15 g,
0
(
.28 mmol) in 5 mL of CH
0.176 g, 0.42 mmol) at room temperature. The solution was stirred
for 2 h. An additional 50 mL of CH Cl was added and then the
mixture was washed with 10% aqueous NaHCO and brine.
2 2
Cl was added Dess-Martin periodinane
2
2
4
The combined organics were dried over MgSO , filtered, and
3
Purification by column chromatography (50% EtOAc/hexane) gave
the title compounds (0.13 g, 87%). The diastereomers were
separated by HPLC using a Whelk-O (R,R) chiral prep column (21.1
concentrated by rotary evaporation to give a pale yellow liquid that
was used in the next reaction without further purification (5.2 g,
80%).
×
250 mm, 50% EtOH/hexane, 20 mL/min, UV 235) to provide
individual diastereomers as white powders: 7 (45 mg, 30%, 13.9
min), 8 (40 mg, 27%, 16.9 min). 7: 1H NMR (CDCl
) δ 8.69 (d,
Allyl(1-methylpent-4-enyl)carbamic Acid Benzyl Ester (42).
Carbobenzyloxy chloride (9.56 g, 8 mL, 55.6 mmol) was added
dropwise to a stirred solution of allyl(1-methyl-pent-4-enyl)amine
41 (7 g, 50 mmol) and triethylamine (5.8 g, 8.0 mL, 57.5 mmol)
3
J ) 4.0 Hz, 1H), 7.94-7.90 (m, 2H), 7.65 (d, J ) 8.0 Hz, 1H),
7
3
.52-7.02 (m, 7H), 5.28-5.02 (m, 1H), 4.78-4.60 (m, 1H), 3.88-
.50 (m, 2H), 3.31-3.19 (m, 1H), 3.02-2.97 (m, 1H), 2.10-1.45
in CH
room temperature and was stirred for 2 h. The reaction mixture
was diluted with CH Cl (100 mL), followed by washing with water
and brine. The combined organics were dried over MgSO , filtered,
2 2
Cl (100 mL) at 0 °C. The reaction mixture was warmed to
(
m, 6H), 1.20 (d, J ) 6.6 Hz, 3H), 0.98 (d, J ) 4.4 Hz, 6H); MS
2
2
+
(
4
ESI) 541.2 (M + H) ; analytical HPLC, Whelk-O (R,R) 5 µm,
4
.6 × 250 mm, 50% EtOH/hexane, 1.0 mL/min, UV 215 nm,
concentrated by rotary evaporation and then were purified by
column chromatograpy (silica gel, 4% EtOAc/hexane) to give the
retention time ) 12.9 min, >95% purity. Anal. (C27
32 4 6
H N O S) C,
+
5
9.98; H, 5.97; N, 10.36. Found: C, 60.01; H, 5.94; N, 10.11. 8:
title compound (8.9 g, 65%): MS (ESI) 274.2 (M + H) .
1
H NMR (CDCl
3
) δ 8.64 (d, J ) 4.0 Hz, 1H), 7.95-7.88 (m, 2H),
.65 (d, J ) 8.0 Hz, 1H), 7.53-7.38 (m, 3H), 7.30-7.25 (m, 2H),
.08-7.04 (m, 2H), 5.20-5.15 (m, 1H), 4.78-4.70 (m, 1H), 4.45
2-Methyl-2,3,4,7-tetrahydroazepine-1-carboxylic Acid Benzyl
7
7
Ester (43). Allyl(1-methylpent-4-enyl)carbamic acid benzyl ester
42 (1.04 g, 3.8 mmol) was dissolved in CH Cl (10 mL) and a
2 2
(d, J ) 12 Hz, 1H), 3.86 (d, J ) 12.0 Hz, 1H), 3.59-3.50 (m,
stream of argon gas was bubbled into the reaction mixture for 10
min. Bis(tricyclohexylphosphine)benzylidineruthenium(IV) dichlo-
ride (Strem Chemicals, 22 mg, 0.027 mmol) was added and the
reaction mixture was refluxed for 2 h. Additional bis(tricyclohexyl-
phosphine)benzylidineruthenium(IV) dichloride (11 mg, 0.014
mmol) was added and the reaction mixture was refluxed for an
additional 1.5 h. The reaction mixture was cooled to room
temperature under argon overnight, concentrated by rotary evapora-
tion, and then purified by column chromatography (silica gel, 5%
1
1
5
2
H), 3.08-3.02 (m, 1H), 2.10-2.00 (m, 2H), 1.87-1.61 (m, 4H),
.20 (d, J ) 6.2 Hz, 3H), 0.99 (d, J ) 4.0 Hz, 6H); MS (ESI)
+
41.2 (M + H) ; analytical HPLC, Whelk-O (R,R) 5 µm, 4.6 ×
50 mm, 50% EtOH/hexane, 1.0 mL/min, UV 215 nm, retention
time ) 15.9 min; > 95% purity.
Benzofuran-2-carboxylic Acid {(S)-3-Methyl-1-[(4S,6S)-6-
methyl-3-oxo-1-(pyridine-2-sulfonyl)azepan-4-ylcarbamoyl]butyl}-
amide (9) and Benzofuran-2-carboxylic Acid {(S)-3-Methyl-1-
1
[
(4R,6R)-6-methyl-3-oxo-1-(pyridine-2-sulfonyl)azepan-4-
EtOAc/hexane) to give the title compound (0.83 g, 89%): H NMR
ylcarbamoyl]butyl}amide (6). To a solution of benzofuran-2-
carboxylic acid {(S)-1-[3-hydroxy-6-methyl-1-(pyridine-2-sulfonyl)-
azepan-4-ylcarbamoyl]-3-methylbutyl}amide 38 (0.32 g, 0.89 mmol)
3
(CDCl ) δ 7.35-7.20 (m, 5H), 5.65-5.60 (1H, m), 5.15-5.12 (m,
2H), 4.45-4.05 (m, 2H), 3.63-5.57 (m, 1H), 2.25-2.10 (m, 2H),
1.90-1.60 (m, 2H), 1.15-1.12 (m, 3H); MS (ESI) 246.2 (M +
+
in 5 mL of CH
.89 mmol) at room temperature. The solution was stirred for 2 h.
An additional 50 mL of CH Cl was added and then the mixture
was washed with 10% aqueous NaHCO and brine. Purification
by column chromatography (50% EtOAc/hexane) gave the title
2
Cl
2
was added Dess-Martin periodinane (0.38 g,
H) .
0
4-Methyl-8-oxa-3-azabicyclo[5.1.0]octane-3-carboxylic Acid
Benzyl Ester (44 and 45). mCPBA (1.05 g, 57-86% pure) was
added to a solution of 2-methyl-2,3,4,7-tetrahydroazepine-1-car-
2
2
3
2 2
boxylic acid benzyl ester 43 (0.83 g, 3.39 mmol) in CH Cl (45

SAR of Azepan-3-one Cathepsin K Inhibitors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 5 1609
mL) at 0 °C. The reaction mixture was stirred for 0.5 h and then
was warmed to room temperature. Additional m-chloroperbenzoic
acid (0.3 g, 57-86% pure) was added and the reaction mixture
was stirred for 2 h at room temperature. The reaction mixture was
concentrated by rotary evaporation, 80 mL of 10% EtOAc/hexane
was added, and the reaction mixture was filtered. The filtrate was
concentrated by rotary evaporation and then was purified by column
chromatography (silica gel, 20% EtOAc/hexane) to give cis-4-
mixture was then azeotroped twice with toluene by rotary evapora-
tion. The resulting oil was dissolved in MeOH and HCl in Et O
2
and the resulting salt was collected following filtration and was
used in the next reaction without further purification (1.7 g, 100%).
5-((S)-2-tert-Butoxycarbonylamino-4-methylpentanoylamino)-
6-hydroxy-2-methylazepane-1-carboxylic Acid Benzyl Ester
(51a, 51b). EDC (0.33 g, 1.73 mmol) was added to a solution of
Boc-L-leucine hydrate (0.43 g, 1.7 mmol), diisopropylethylamine
(0.22 g, 0.3 mL, 1.7 mmol), HOBT (0.25 g, 1.85 mmol), and
5-amino-6-hydroxy-2-methylazepane-1-carboxylic acid benzyl ester
49 (0.5 g, 1.6 mmol) in DMF (10 mL). The reaction mixture was
stirred overnight at room temperature and then was diluted with
methyl-8-oxa-3-azabicyclo[5.1.0]octane-3-carboxylic acid benzyl
1
ester 44 (0.44 g, 50%) [ H NMR (CDCl
5
3
(
3
) δ 7.42-7.22 (m, 5H),
.13 (brs, 2H), 4.50-4.15 (m, 2H), 3.28 (brd, J ) 16.0 Hz, 1H),
.12-2.95 (m, 1H), 2.15-1.70 (m, 2H), 1.41-1.38 (m, 2H), 1.06
+
brd, J ) 6.6 Hz, 3H); MS (ESI) 262.0 (M + H) .] and trans-4-
EtOAc (100 mL), washed with H O (3 × 50 mL) and brine (50
2
methyl-8-oxa-3-azabicyclo[5.1.0]octane-3-carboxylic acid benzyl
mL), dried over magnesium sulfate, filtered, concentrated by rotary
1
ester 45 (0.15 g, 17%) [ H NMR (CDCl
3
) δ 7.37-7.30 (m, 5H),
evaporation, and purified by column chromatography (silica gel,
50% EtOAc/hexane) to yield the title compound as a mixture of
5
2
(
.19-5.11 (m, 2H), 4.31-4.05 (m, 2H), 3.44-3.25 (m, 1H), 3.07-
1
.94 (m, 2H), 2.28-2.16 (m, 1H), 1.90-1.70 (m, 1H), 1.55-1.32
diastereomers (0.78 g, 100%): H NMR (CDCl ) δ 7.40-7.29 (m,
3
+
m, 2H), 1.04-1.01 (m, 3H); MS (ESI) 262.0 (M + H) .].
5H), 6.75 (brd, 1H), 5.18-5.11 (m, 2H), 5.05 (brs, 1H), 4.15-
3.72 (m, 5H), 3.06 (d, J ) 16.0 Hz, 1H), 1.90-1.40 (m, 7H), 1.40
(s, 9H); 1.12-1.05 (m, 3H), 0.97-0.87 (m, 6H); MS (ESI) 492.0
trans-5-Azido-6-hydroxy-2-methylazepane-1-carboxylic Acid
Benzyl Ester (46). Sodium azide (0.56 g, 8.6 mmol) was added to
a solution of trans-4-methyl-8-oxa-3-azabicyclo[5.1.0]octane-3-
carboxylic acid benzyl ester 44 (0.75 g, 2.87 mmol) and ammonium
chloride (0.46 g, 8.6 mmol) in MeOH (5 mL) and H O (0.5 mL)
2
and then was refluxed for 6 h. The reaction mixture was
concentrated by rotary evaporation and then was diluted with water
+
(M + H) .
5-((S)-2-tert-Butoxycarbonylamino-4-methylpentanoylamino)-
6-hydroxy-2-methylazepane-1-carboxylic Acid Benzyl Ester
(52a, 52b). EDC (1.0 g, 5.2 mmol) was added to a solution of
Boc-L-leucine hydrate (1.31 g, 5.26 mmol), diisopropylethylamine
(0.74 g, 1.0 mL, 5.75 mmol), HOBT (0.77 g, 5.69 mmol), and
5-amino-6-hydroxy-2-methylazepane-1-carboxylic acid benzyl ester
50 (1.5 g, 4.77 mmol) in DMF (30 mL). The reaction mixture was
stirred overnight at room temperature and then was diluted with
(5 mL) and extracted with EtOAc (10 mL). The organic layer was
then washed with water and brine, dried over MgSO , filtered,
4
concentrated by rotary evaporation, and purified by column
chromatography (silica gel, 20% EtOAc/hexane) to yield the title
compound (0.7 g, 80%): 1H NMR (CDCl
) δ 7.39-7.30 (m, 5H),
.15 (s, 2H), 4.10-3.67 (m, 5H), 3.10 (d, J ) 16.0 Hz, 1H), 1.90-
.80 (m, 2H), 1.80-1.53 (m, 2H), 1.09 (d, J ) 6.6 Hz, 3H); MS
3
EtOAc (100 mL), washed with H O (3 × 50 mL) and brine (50
2
5
1
mL), dried over magnesium sulfate, filtered, concentrated by rotary
evaporation, and purified by column chromatography (silica gel,
+
1
(ESI) 305.2 (M + H) .
60% EtOAc/hexane) to yield the title compound (2.1 g, 90%):
NMR (CDCl
H
5
-Azido-6-hydroxy-2-methylazepane-1-carboxylic Acid Ben-
3
) δ 7.45-7.31 (m, 5H), 6.72-6.49 (m, 1H), 5.20-
zyl Ester (47). Sodium azide (1.8 g, 27.7 mmol) was added to a
solution of cis-4-methyl-8-oxa-3-azabicyclo[5.1.0]octane-3-car-
boxylic acid benzyl ester 45 (2.4 g, 9.2 mmol) and ammonium
4.93 (m, 2H), 4.12-3.45 (m, 5H), 2.98-2.85 (m, 1H), 2.02-1.30
(m, 7H), 1.40 (s, 9H), 1.12-1.07 (m,3H), 0.98-0.90 (m, 6H).
[(S)-1-(3-Hydroxy-7-methylazepan-4-ylcarbamoyl)-3-methyl-
butyl]carbamic Acid tert-Butyl Ester (53a, 53b). 5-((S)-2-tert-
Butoxycarbonylamino-4-methylpentanoylamino)-6-hydroxy-2-me-
thylazepane-1-carboxylic acid benzyl ester (51a, 51b) (0.77 g, 1.57
mmol) was dissolved in EtOAc (27.5 mL) and MeOH (5.5 mL).
Ten percent Pd/C (0.39 g) was added and the reaction mixture was
stirred overnight under a balloon filled with hydrogen gas. The
reaction mixture was filtered through Celite, concentrated by rotary
evaporation, and used in the next reaction without further purifica-
2
chloride (1.48 g, 27.7 mmol) in MeOH (16 mL) and H O (1.6 mL)
and then was refluxed overnight. The reaction mixture was concen-
trated by rotary evaporation and then was diluted with water (5
mL) and extracted with EtOAc. The organic layer was then washed
4
with water and brine, dried over MgSO , filtered, and concentrated
by rotary evaporation. The crude product was stirred with ether
and filtered to give the title compound 47 (1.4 g, 50%). The mother
liquors were concentrated by rotary evaporation and the residue
was purified by chromatography (silica gel, 20% EtOAc/hexane)
+
tion (0.56 g, 99%): MS (ESI) 358.1 (M + H) .
to yield the title compound, 47 (0.36 g, 13%; combined yield )
[(S)-1-(3-Hydroxy-7-methylazepan-4-ylcarbamoyl)-3-methyl-
butyl]carbamic Acid tert-Butyl Ester (54a, 54b). 5-((S)-2-tert-
Butoxycarbonylamino-4-methylpentanoylamino)-6-hydroxy-2-me-
thylazepane-1-carboxylic acid benzyl ester (52a, 52b) (2.1 g, 4.28
mmol) was dissolved in EtOAc (75 mL) and MeOH (15 mL). 10%
Pd/C (1.05 g) was added and the reaction mixture was stirred
overnight under a balloon filled with hydrogen gas. The reaction
mixture was filtered through Celite, concentrated by rotary evapora-
tion, and used in the next reaction without further purification (1.53
1
1
7
(
.76 g, 63%): H NMR (CDCl
3
) (shown as two rotamers): δ 7.40-
.28 (m, 5H), 5.20-5.12 (m, 2H), 4.20-3.98 (m, 1H), 3.90-3.75
m, 1H), 3.65-3.43 (m, 1H), 3.23-3.10 (m, 1H), 2.87 and 2.83 (d
each, J ) 10.8 Hz, 1H), 2.55 and 2.29 (d each, J ) 2.8 Hz, 1H),
2
3
(
3
(
.10-1.90 (m, 2H), 1.50-1.25 (m, 2H), 1.10 (m, 3H); MS (ESI)
+
05.2 (M + H) ; and the regioisomer, 48 (0.5 g, 18%): NMR
3
CDCl ) δ 7.39-7.27 (m, 5H), 5.30-5.05 (m, 2H), 4.16-3.90 (m,
H), 3.59-3.50 (m, 1H), 3.35 (d, J ) 15.3 Hz, 1H), 1.88-1.60
+
+
m, 4H), 1.13-1.08 (m, 3H); MS (ESI) 305.2 (M + H) .
g, 99%): MS (ESI) 358.2 (M + H) .
5
-Amino-6-hydroxy-2-methylazepane-1-carboxylic Acid Ben-
[(S)-1-Benzenesulfonyl-3-hydroxy-7-methylazepan-4-ylcar-
bamoyl)-3-methylbutyl]carbamic Acid tert-Butyl Ester (55a,
55b). 2-Pyridinesulfonyl chloride (0.6 g, 3.4 mmol) was added to
a solution of [(S)-3-hydroxy-7-methylazepan-4-ylcarbamoyl)-3-
methylbutyl]carbamic acid tert-butyl ester 53a, 53b (1.0 g, 2.8
mmol) and N-methylmorpholine (0.45 mL, 4.1 mmol) in CH Cl
zyl Ester (49). Triphenylphosphine (1.94 g, 7.4 mmol) was added
to a solution of 5-azido-6-hydroxy-2-methylazepane-1-carboxylic
acid benzyl ester 46 (1.5 g, 4.93 mmol) in THF (185 mL) and H
0.7 mL) and then was heated to 45 °C overnight. The solvent was
removed, the residue was azeotroped twice with toluene, and the
resulting oil was dissolved in MeOH and HCl in Et O. Additional
Et O was added and the resulting salt was collected following
2
O
(
2
2
2
(35 mL) and was stirred at room temperature for 1 h. The reaction
2
mixture was diluted with EtOAc (100 mL), washed with H O and
2
filtration and was used in the next reaction without further
purification (1.4 g, 90%).
brine, dried over magnesium sulfate, filtered, concentrated by rotary
evaporation, and purified by column chromatography (silica gel,
5
-Amino-6-hydroxy-2-methylazepane-1-carboxylic Acid Ben-
2.5% MeOH/CH
diastereomers (0.9 g, 64%): H NMR (CDCl ) (as a mixture of
2 2
Cl ) to yield the title compound as a mixture of
1
zyl Ester (50). Triphenylphosphine (2.13 g, 8.14 mmol) was added
3
to a solution of 5-azido-6-hydroxy-2-methylazepane-1-carboxylic
two diastereomers): δ 8.71-8.68 (m, 1H), 8.08-7.92 (m, 2H),
7.57-7.49 (m,1H), 6.76, 6.67 (brs each, 1H), 5.15-4.96 (m, 2H),
4.18-3.87 (m, 4H), 3.74-3.63 (m, 1H), 3.46-3.36 (m, 1H), 1.96-
acid benzyl ester 47 (1.6 g, 5.43 mmol) in THF (185 mL) and H
2
O
(0.7 mL) and then was heated to 45 °C overnight. The reaction

1
610 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 5
Yamashita et al.
1
6
.40 (m, 7H), 1.45 (s, 9H), 1.14-1.05 (m, 3H), 0.97-0.85 (m,
and was stirred at room temperature for 45 min. The solution was
washed with 10% aqueous NaHCO , 10% aqueous Na S O , and
3 2 2 5
+
H); MS (ESI) 499.0 (M + H) .
[
(S)-1-Benzenesulfonyl-3-hydroxy-7-methylazepan-4-ylcar-
brine. Purification by column chromatography (60% EtOAc/hexane)
gave the title compound as a mixture of diastereomers (0.38 g,
94%). The diastereomers were separated by HPLC using a Whelk-O
(R,R) chiral prep column (21.1 × 250 mm, 50% EtOH/hexane, 20
mL/min, UV 235) to provide individual diastereomers as white
powders: 10 (128 mg, 32%, 10.5 min), 13 (112 mg, 28%, 13.4
bamoyl)-3-methylbutyl]carbamic Acid tert-Butyl Ester (56a,
6b). 2-Pyridinesulfonyl chloride (0.72 g, 4.0 mmol) was added to
5
a solution of [(S)-3-hydroxy-7-methylazepan-4-ylcarbamoyl)-3-
methylbutyl]carbamic acid tert-butyl ester 54a, 54b (1.0 g, 2.8
mmol), sodium bicarbonate (0.84 g, 10 mmol) in CH
and H O (10 mL) at room temperature. After stirring for 45 min,
the reaction mixture was washed with H O and brine, dried over
magnesium sulfate, filtered, concentrated by rotary evaporation, and
purified by column chromatography (silica gel, 2.5% MeOH/CH
Cl ) to yield the title compound as a mixture of diastereomers (1.2
g, 86%): H NMR (CDCl
δ 8.74-8.71 (m, 1H), 8.04-7.86 (m, 2H), 7.48-7.45 (m, 1H),
2 2
Cl (35 mL),
min). 10: 1H NMR (CDCl
) δ 8.72 (d, J ) 4.5 Hz, 1H), 8.0 (d, J
2
3
) 7.8 Hz, 1H), 7.93 (ddd, J ) 7.7, 7.7, 1.5 Hz, 1H), 7.65 (d, J )
7.8 Hz, 1H), 7.56 (m, 2H), 7.45 (s, 1H), 7.42 (dd, J ) 8.1, 7.3 Hz,
1H), 7.28 (dd, J ) 7.6, 7.4 Hz, 1H), 7.10 (d, J ) 8.4 Hz, 1H), 6.85
(d, J ) 6.4 Hz, 1H), 5.15 (m, 1H), 4.77 (d, J ) 19.5 Hz, 1H), 4.68
(m, 1H), 4.40 (m, 1H), 3.86 (d, J ) 19.5 Hz, 1H), 2.20-2.08 (m,
2
2
-
2
1
3
) (as a mixture of two diastereomers):
+
2H), 1.78-1.40 (m, 5H), 0.98 (m, 9H); MS (ESI) 541.2 (M + H) ;
6.43-6.37 (m, 1H), 5.03, 4.88 (brd each, 1H), 4.13-4.00 (m, 2H),
3.90-3.60 (m, 3H), 3.07-2.97 (m, 1H), 2.10-2.01 (m, 1H), 1.82-
1.42 (m, 6H), 1.40 (s, 9H), 0.96-0.88 (m, 9H).
analytical HPLC, ULMO-(S,S) 5 µm, 4.6 × 250 mm, 50% EtOH/
hexane, 1.0 mL/min, UV 215 nm, retention time ) 7.4 min, >95%
purity. Anal. (C27
C, 59.90; H, 5.99; N, 10.26. 13: H NMR (CDCl ) δ 8.68 (d, J )
32 4 6
H N O S) C, 59.98; H, 5.97; N, 10.36. Found:
1
Benzofuran-2-carboxylic Acid {(S)-1-[3-Hydroxy-7-methyl-
-(pyridine-2-sulfonyl)azepan-4-ylcarbamoyl]-3-methylbutyl}-
3
1
4.6 Hz, 1H), 8.04 (d, J ) 7.8 Hz, 1H), 7.92 (ddd, J ) 7.7, 7.7, 1.4
Hz, 1H), 7.68 (d, J ) 7.8 Hz, 1H), 7.58 (d, J ) 8.4 Hz, 1H), 7.52
(m, 2H), 7.42 (t, J ) 7.3 Hz, 1H), 7.29 (t, J ) 7.4 Hz, 1H), 7.05
(m, 2H), 5.12 (m, 1H), 4.75-4.68 (m, 2H), 4.43 (m, 1H), 3.83 (d,
J ) 19.4 Hz, 1H), 2.25-2.12 (m, 2H), 1.88-1.40 (m, 5H), 0.98
amide (57a, 57b). HCl in dioxane (4.0 M, 15 mL) was added to a
stirred solution of [(S)-1-benzenesulfonyl-3-hydroxy-7-methylazepan-
4-ylcarbamoyl)-3-methylbutyl]carbamic acid tert-butyl ester 55a,
55b (0.9 g, 1.8 mmol) in MeOH (15 mL) at room temperature.
+
The reaction mixture was stirred for 2 h, concentrated by rotary
evaporation, and was used in the next reaction without further
purification. The intermediate amines were dissolved in DMF (10
mL). Diisopropylethylamine (0.48 g, 0.65 mL, 3.7 mmol), HOBT
(m, 9H); MS (ESI) 541.2 (M + H) ; analytical HPLC, ULMO-
(S,S) 5 µm, 4.6 × 250 mm, 50% EtOH hexane, 1.0 mL/min, UV
215 nm, retention time ) 8.5 min, >95% purity.
Benzofuran-2-carboxylic Acid {(S)-3-Methyl-1-[7-methyl-3-
oxo-1-(pyridine-2-sulfonyl)azepan-4-ylcarbamoyl]butyl}amide
(11, 12). Dess-Martin periodinane (0.2 g, 0.47 mmol) was added
to a solution of benzofuran-2-carboxylic acid {(S)-1-[3-hydroxy-
7-methyl-1-(pyridine-2-sulfonyl)azepan-4-ylcarbamoyl]-3-
(0.25 g, 1.85 mmol), 2-benzofurancarboxylic acid (0.3 g, 1.85
mmol), and EDC (0.35 g, 1.85 mmol) were added, and the reaction
mixture was stirred at room temperature overnight. The reaction
mixture was diluted with EtOAc (100 mL), washed with H O and
2
brine, dried over magnesium sulfate, filtered, concentrated by rotary
evaporation, and purified by chromatography (silica gel, 2.5%
methylbutyl}amide 58 (0.15 g, 0.28 mmol) in CH
was stirred at room temperature for 45 min. The solution was
washed with 10% aqueous NaHCO , 10% aqueous Na , and
2 2
Cl (3 mL) and
MeOH/CH
diastereomers (0.8 g, 82% for two steps): H NMR (CDCl
mixture of two diastereomers): δ 8.68-8.63 (m, 1H), 8.08-8.03
m, 1H), 7.97-7.86 (m, 1H), 7.67-7.63 (m, 1H), 7.58-7.42 (m,
2
Cl
2
) to yield the title compound as a mixture of
3
2 2 5
S O
1
3
) (as a
brine. Purification by column chromatography (60% EtOAc/hexane)
gave the title compound as a mixture of diastereomers (0.14 g,
94%). The diastereomers were separated by HPLC using a Whelk-O
(R,R) chiral prep column (21.1 × 250 mm, 50% EtOH/hexane, 20
mL/min, UV 235) to provide individual diastereomers as white
powders: 11 (46 mg, 31%, 10.3 min), 12 (53 mg, 35%, 12.5 min).
(
3
1
3
0
H), 7.35-7.27 (m, 1H), 7.6-7.05 (m, 1H), 6.94, 6.83 (brd each,
H), 4.71-4.63 (m, 1H), 4.18-3.89 (m, 2H), 3.74-3.65 (m, 1H),
.58-3.40 (m, 1H), 1.87-1.66 (m, 6H), 1.50-1.27 (m, 1H), 1.26-
+
11: 1H NMR (CDCl
.92 (m, 9H); MS (ESI) 543.0 (M + H) .
3
) δ 8.87 (d, J ) 4.7 Hz, 1H), 8.05 (d, J ) 7.8
Benzofuran-2-carboxylic Acid {(S)-1-[3-Hydroxy-7-methyl-
-(pyridine-2-sulfonyl)azepan-4-ylcarbamoyl]-3-methylbutyl}-
amide (58a, 58b). HCl in dioxane (4.0 M, 4 mL) was added to a
stirred solution of [(S)-1-benzenesulfonyl-3-hydroxy-7-methylazepan-
-ylcarbamoyl)-3-methylbutyl]carbamic acid tert-butyl ester 56a,
6b (0.26 g, 0.52 mmol) in MeOH (4 mL) at room temperature.
The reaction mixture was stirred for 2 h, concentrated by rotary
evaporation, and used in the next reaction without further purifica-
tion. The resultant amines were dissolved in DMF (4 mL).
Diisopropylethylamine (0.19 g, 0.26 mL, 1.5 mmol), HOBT (0.08
g, 0.6 mmol), 2-benzofurancarboxylic acid (0.3 g, 1.85 mmol), and
EDC (0.12 g, 0.62 mmol) were added, and the reaction mixture
was stirred at room temperature overnight. The reaction mixture
2
was diluted with EtOAc (100 mL), washed with H O and brine,
dried over magnesium sulfate, filtered, concentrated by rotary
evaporation, and purified by column chromatography (silica gel,
Hz, 1H), 7.95 (ddd, J ) 7.8, 7.8, 1.4 Hz, 1H), 7.67 (d, J ) 7.8 Hz,
1H), 7.60-7.45 (m, 4H), 7.43 (t, J ) 7.4 Hz, 1H) 7.30 (t, J ) 7.4
Hz, 1H), 7.10 (d, J ) 8.6 Hz, 1H), 4.90-4.80 (m, 2H), 4.45 (d, J
) 17.9 Hz, 1H), 4.15 (d, J ) 17.9 Hz, 1H), 3.90-3.80 (m, 1H),
2.17-2.10 (m, 1H), 2.0-1.90 (m, 1H), 1.90-1.70 (m, 4H), 1.65-
1.55 (m, 1H), 1.20 (d, J ) 6.8 Hz, 3H), 0.97 (d, J ) 6.2 Hz, 3H),
1
4
5
+
0.95 (d, J ) 6.1 Hz, 3H); MS (ESI) 541.2 (M + H) ; analytical
HPLC, ULMO-(S,S) 5 µm, 4.6 × 250 mm, 50% EtOH/hexane,
1.0 mL/min, UV 215 nm, retention time ) 6.8 min, >95% purity.
12: 1H NMR (CDCl
) δ 8.65 (d, J ) 4.5 Hz, 1H), 7.90-7.88 (m,
3
2H), 7.68 (d, J ) 7.8 Hz, 1H), 7.57 (d, J ) 8.4 Hz, 1H), 7.55 (s,
1H) 7.46-7.34 (m, 2H), 7.30 (t, J ) 7.4 Hz, 1H), 7.18 (d, J ) 8.6
Hz, 1H), 7.07 (d, J ) 5.7 Hz, 1H), 4.85-4.80 (m, 1H), 4.70-4.65
(m, 1H), 4.35 (d, J ) 17.7 Hz, 1H), 4.05-3.95 (m, 2H), 2.20-
1.80 (m, 7H), 1.13 (d, J ) 7.0 Hz, 3H), 1.00 (d, J ) 6.3 Hz, 6H);
+
MS (ESI) 541.2 (M + H) ; analytical HPLC, ULMO-(S,S) 5 µm,
8
0% EtOAc/hexane) to yield the title compound as a mixture of
4.6 × 250 mm, 50% EtOH/hexane, 1.0 mL/min, UV 215 nm,
1
diastereomers (0.23 g, 88% for two steps): H NMR (CDCl
3
) (as
retention time ) 13.7 min; >95% purity.
a mixture of two diastereomers): 8.71-8.64 (m, 1H), 8.08-7.95
m, 1H), 7.95-7.86 (m, 1H), 7.73-7.62 (m, 1H), 7.58-7.20 (m,
Determination of Cathepsin K Proteolytic Catalytic Activity.
All assays for human cathepsin K, L, V, S, B, and rat cathepsin K
were carried out with recombinant enzymes. Standard assay
conditions for the determination of kinetic constants used a
fluorogenic peptide substrate, typically Cbz-Phe-Arg-AMC, and
were determined in 100 mM sodium acetate at pH 5.5 containing
20 mM cysteine and 5 mM EDTA. Stock substrate solutions were
prepared at concentrations of 10 or 20 mM in DMSO with 20 µM
final substrate concentration in the assays. All assays contained
10% DMSO. Independent experiments found that this level of
DMSO had no effect on enzyme activity or kinetic constants. All
(
5
3
H), 7.16-7.11 (m, 1H), 4.75-4.65 (m, 1H), 4.17-3.65 (m, 4H),
.12-3.03 (m, 1H), 1.98-1.37 (m, 7H), 1.30-0.91 (m, 9H); MS
+
(
ESI) 542.3 (M + H) .
Benzofuran-2-carboxylic Acid {(S)-3-Methyl-1-[7-methyl-3-
oxo-1-(pyridine-2-sulfonyl)azepan-4-ylcarbamoyl]butyl}amide
10, 13). Dess-Martin periodinane (0.48 g, 1.18 mmol) was added
to a solution of benzofuran-2-carboxylic acid {(S)-1-[3-hydroxy-
-methyl-1-(pyridine-2-sulfonyl)azepan-4-ylcarbamoyl]-3-
methylbutyl}amide 57a, 57b (0.4 g, 0.74 mmol) in CH Cl (10 mL)
(
7
2
2

SAR of Azepan-3-one Cathepsin K Inhibitors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 5 1611
z-gradient at 25 °C. Samples were dissolved in CDCl (99.96%
assays were conducted at ambient temperature. Product fluorescence
3
(
excitation at 360 nM, emission at 460 nM) was monitored with a
deuterium) with an approximated concentration of 10 mM for each
sample. H NMR assignments were achieved according to con-
ventional 1D proton and 2D gradient COSY and gradient HSQC
experiments. Gradient COSY was collected with 16 scans, 2K data
1
Perceptive Biosystems Cytofluor II fluorescent plate reader. Product
progress curves were generated over 20-30 min following forma-
tion of AMC product.
Potential inhibitors were evaluated using the progress curve
method. Assays were carried out in the presence of variable
concentrations of test compound. Reactions were initiated by
addition of enzyme to buffered solutions of inhibitor and substrate.
Data analysis was conducted according to one of two procedures,
depending on the appearance of the progress curves in the presence
of inhibitors. For those compounds whose progress curves were
linear, apparent inhibition constants (Ki,app) were calculated accord-
ing to eq 116
points (t
Hz in both dimensions. For gradient HSQC experiments, 128 t
increments of 2048 complex points with 16 scans each with a
spectral width of 4997.5 Hz in F and 20105.6 Hz in F dimension
were acquired. 2D NOESY spectra were collected with 32 scans,
2K data points (t ) and 200 increments (t ) with a spectral width of
2 1
) and 128 increments (t ) with a spectral width of 4997.5
1
2
1
2
1
4997.5 Hz in both dimensions, at mixing times of 50, 150, 200,
350, 500, 600, and 700 ms. NOE buildup curves were found to be
linear within the tested mixing times (up to 700 ms). Interproton
distances were derived on the basis of the NOE measurements from
the NOESY spectra acquired with a mixing time of 500 ms for all
four compounds. J coupling constants were measure from the proton
spectra. In addition to the conformational searches carried out for
azepanones 1, 2, 7, and 10, searches were also carried out for
azepanone 3 as well as the thiol-adducts of azepanones 1, 2, 3, 7,
and 10, with methylmercaptan bonded to the original azepanone
V ) V A/[K (1 + I/Ki,app) + A]
(1)
m
a
where V is the velocity of the reaction with maximal velocity V
m
,
,
A is the concentration of substrate with Michaelis constant of K
and I is the concentration of inhibitor.
a
For those compounds whose progress curves showed the down-
ward curvature characteristic of time-dependent inhibition, the data
from individual sets was analyzed to give kobs according to eq 2
3
carbonyl carbon, resulting in an sp -hybridized carbon center with
a hydroxyl group (thiohemiketal) attached with the stereochemistry
observed in the cathepsin K/azepanone cocrystal structures.
Crystallization and Structure Solution of the Complex of
Cathepsin K with the Inhibitor 10. Crystals of mature activated
cathepsin K at a concentration of 14 mg/mL in complex with 10
were grown at 10 °C by the vapor diffusion method in sitting drops
from a solution of 28% PEG 400, 0.1 M Hepes buffer at pH 7.5
[
AMC] ) V t + (V - V ) [1 - exp(-kobst)]/kobs
(2)
ss
0
ss
where [AMC] is the concentration of product formed over time t,
is the initial reaction velocity and Vss is the final steady-state
V
0
rate. Values for kobs were then analyzed as a linear function of
inhibitor concentration to generate an apparent second-order rate
constant (kobs/inhibitor concentration or kobs/[I]) describing the time-
dependent inhibition. This kinetic treatment has been fully de-
scribed.17
2
containing 0.2 M CaCl . Crystals of the complex are hexagonal,
space group P6122, with cell constants of a ) 74.84 Å and c )
339.56 Å. There are two protein-10 complexes in the asymmetric
unit. The crystal was flash-frozen in a nitrogen cold stream at -180
°C for the data collection. The diffraction data were measured at
beamline 17-ID in the facilities of the Industrial Macromolecular
Crystallography Association Collaborative Access Team (IMCA-
CAT) at the Advanced Photon Source. These facilities were
established by the companies of the Industrial Macromolecular
Crystallography Association through a contract with Illinois Institute
of Technology (IIT), executed through IIT’s Center for Synchrotron
Radiation Research and Instrumentation. The structure was deter-
mined by molecular replacement with a model consisting of all
protein atoms from the previously determined cathepsin K/E-64
HPLC Method for Epimerization Study. The rates of C-4
epimerization were determined by HPLC analysis. The chromato-
graphic conditions (1) for 2, 4, 11, and 13 were as follows: column,
XTerra RP18, 3.5 µm, 2.1 × 100 mm (Waters Corp., Milford, MA);
elution, isocratic with 47% A (acetonitrile), 53% B (20mM pH 7.4
phosphate); flow rate, 0.48 mL/min. The chromatographic condi-
tions (2) for 1 (and epi-1) were as follows: column, Inertsil ODS3,
5
µm, 4.6 × 250 mm (MetaChem Technologies Inc, Torrance, CA);
elution, 50% A and 50% B; flow rate, 1 mL/min. An Agilent Model
100 HPLC system with a diode array detector and ChemStation
1
9
software was used for chromatographic data acquisition and
processing. A 10 µL portion of DMSO stock solution containing
the compound of interest (10 mM) was added into 1 mL of pH 11
phosphate buffer. The solution was quickly transferred to a HPLC
sample vial and placed in the HPLC autosampler. Aliquots of the
solution were injected immediately and subsequently at designated
time intervals, and the chromatographic data were exported into
an EXCEL table. The first-order forward rate constant, k, for the
disappearance of the starting epimer can be determined on the basis
of the following equation:
complex (PDB ID 1ATK). The structure was refined at 2.5-Å
resolution with Rc of 0.24 and Rfree of 0.28.
Pharmacokinetic Study Protocols. In vivo rat and monkey
pharmacokinetic experiments were conducted as iv × po crossover
studies using N ) 3 rats or monkeys per study using standard
techniques. All studies were reviewed and approved by the GSK
IACUC prior to initiation. Briefly, male Sprague-Dawley rats
received the test compounds as a 1-2 mg/kg 0.5-h intravenous
infusion via an indwelling femoral vein catheter; the same rats
received a 2-4 mg/kg oral bolus gavage of the same compound 2
days later. Male cynomolgus monkeys received the test compounds
as a 1-2 mg/kg 1-h intravenous infusion via a temporary saphenous
vein catheter; the same monkeys received a 2-4 mg/kg oral bolus
gavage of the same compound 7 days later. All dosages were
administered as aqueous solutions containing <2% DMSO and
<20% hydroxypropyl-â-cyclodextrin. Blood samples were obtained
from a lateral tail vein (rats) or a femoral vein (monkeys) at timed
intervals after each dosage and were centrifuged to obtain 50-µL
plasma aliquots. LC/MS/MS was used to quantify drug concentra-
tions in each sample, with a lower limit of quantitation that ranged
from 1.0 to 10 ng/mL. Standard noncompartmental techniques were
used to calculate pharmacokinetic parameters from the resulting
concentration versus time data.19 In vitro plasma protein binding
experiments were conducted using either equilibrium dialysis or
ultrafiltration, according to standard techniques.20
C(t)/C(0) ) [R/(1 + R)]e-kt[(R+1)/R] + [1/(1 + R)]
C(0) and C(t) are concentrations at time ) 0 and elapsed time t. R
is the equilibrium ratio of product to reactant, calculated on the
basis of HPLC peak areas for the epimer pairs at a long elapsed
time.
NAMFIS and Conformational Analysis. Conformers for 1, 2,
7
, and 10 were generated by carrying out a 10 000-step Monte Carlo
18
search in MacroModel 7.1 using the MMFFs force field and
chloroform implicit solvent model with an 8 kcal/mol cutoff. NOE
and coupling constant information were obtained for protons at the
chiral centers, amide nitrogens, pyridinyl group, and positions C2,
C4, C5, C6, and C7. The NAMFIS algorithm was applied, and a
best solution of conformers for each molecule was obtained. All
NMR experiments for conformational analysis (1, 2, 7, and 10)
were performed on a Varian Inova 500 MHz spectrometer equipped
with a 5 mm inverse detection probe with actively shielded
Madin-Darby Canine Kidney Cell Permeability. The in vitro
cell permeability of selected compounds was determined using
Madin-Darby canine kidney (MDCK, American Type Culture

1
612 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 5
Yamashita et al.
Collection) cells. MDCK cells were maintained in modified Eagle’s
medium (MEM) containing 10% fetal bovine serum in an atmo-
sphere of 5% CO and 95% relative humidity at 37 °C. Cells were
2
passaged at 80-90% confluence (every 3-4 days) using trypsin-
EDTA solution (0.05% trypsin + 0.53 mM EDTA). MDCK cells
were used between passage numbers 60-80. All cell culture
reagents were purchased from either Sigma or Invitrogen Corp.
For the transport assay, cells were seeded on 12-well Transwells
(3) 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.; Haltiwanger, 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.; Marzulli, 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. Azepanone-Based Inhibitors of Human and Rat Cathepsin K. J.
Med. Chem. 2001, 44, 1380-1395.
(0.4 µm polycarbonate filters, Corning Costar Corp.) at 100 000
cells per well. Cells were fed 24 h after seeding, and the assay was
performed 48 h later. Immediately prior to the experiment, cell
medium was removed and replaced with transport buffer (Hank’s
balanced salt solution (HBSS) containing 25 mM glucose and 25
mM N-(2-hydroxyethyl)piperazine-N-(2-ethanesulfonic acid, HEPES).
Monolayers were allowed to equilibrate for approximately 1 h. The
transport buffer was then removed and a dosing solution containing
compound at a concentration of 10 µM in transport buffer was added
to the apical (A) side of the monolayer. Blank transport buffer was
added to basolateral (B) side of monolayer. DMSO was used as
the initial solvent for compound dissolution, and then was diluted
with transport buffer. The final DMSO concentration was 1% (v/
v) in all dose solutions. Each treatment was done in duplicate.
Propranolol and mannitol were included as controls.
(4) Luzzio, F. A. The Henry Reaction: Recent Examples. Tetrahedron
2001, 57, 915-945.
(
(
5) Grubbs, R. H. Olefin Metathesis. Tetrahedron 2004, 60, 7117-7140.
6) Hendrickson, J. B. Molecular Geometry. V. Evaluation of Functions
and Conformations of Medium Rings J. Am. Chem. Soc. 1967, 89,
7036-7043. Favini, G. Conformational Analysis of Seven-Membered
Cyclic Systems. J. Mol. Struct 1983, 93, 139-152.
7) Gula, M. J.; Vitale, D. E.; Dostal, J. M.; Trometer, J. D.; Spencer,
T. A. Evaluation of 1,3-Diaxial Steric Hindrance to Proton Abstrac-
tion Alpha to Carbonyl Groups. J. Am. Chem. Soc. 1988, 110, 4400-
4405.
(
(8) Nevins, N.; Cicero, D.; Snyder, J. P. A Test of the Single-
Conformation Hypothesis in the Analysis of NMR Data for Small
Polar Molecules: A Force Field Comparison. J. Org. Chem. 1999,
6
4 (11), 3979-3986; Snyder, J. P.; Nevins, N.; Cicero, D. O.; Jansen,
J. The Conformations of Taxol in Chloroform. J. Am. Chem. Soc.
000, 122 (4), 724-725.
After 1 h, aliquots of solutions in the donor and receiver
compartments were diluted with acetonitrile and analyzed by LC/
MS/MS. HPLC was accomplished using a Phenomenex Synergi
Hydro-RP 4 µm, 2.1 × 50 mm column and a mobile phase
consisting of 10 mM ammonium acetate (pH 6.8) and methanol.
Compound product ion was generated using a Sciex API300 in
turboionspray mode. Concentrations of each compound in apical
and basolateral samples were determined from a peak area versus
concentration standard curve. Standards were prepared from serial
dilution of the initial dosing solution.
2
(
9) Zhao, B.; Janson, C. A.; Amegadzie, B. Y.; D’Alessio, K.; Griffin,
C.; Hanning, C. R.; Jones, C.; Kurdyla, J.; McQueney, M.; Qui, X.;
Smith, W. W.; Abdel-Meguid, S. S. Crystal Structure of Human
Osteoclast Cathepsin K Complex with E-64. Nat. Struct. Biol. 1997,
4 (2), 109-111.
(10) James, I. E.; Lark, M. W.; Zembryki, D.; Lee-Rykaczewski, E. V.;
Hwang, S. M.; Tomaszek, T. A.; Belfiore, P.; Gowen, M. Develop-
ment and Characterization of a Human in Vitro Resorption Assay:
Demonstration of Utility Using Novel Antiresorptive Agents. J. Bone
Miner. Res. 1999, 14, 1562-1569.
11) Thompson, M.; Lennox, R. B.; McClelland, R. A. Structure and
Electrochemical Properties of Microfiltration Filter-lipid Membrane
Systems. Anal. Chem. 1982, 54, 76-81; Kansy, M.; Senner, F.;
Gubernator, K. Physicochemical High Throughput Screening: Paral-
lel Artificial Membrane Permeation Assay in the Description of
Passive Absorption Processes. J. Med. Chem. 1998, 41, 1007-1010.
12) Turk, D.; Podobnik, M.; Popovic, T.; Katunuma, N.; Bode, W.;
Huber, R.; Turk, V. Crystal Structure of Cathepsin B Inhibited with
CA030 at 2.0-Å Resolution: A Basis for the Design of Specific
Epoxysuccinyl Inhibitors. Biochemistry 1995, 34, 4791-4797.
An apparent permeability coefficient (Papp) was calculated from
the LC/MS/MS-determined concentration in the basolateral com-
partment (CBL) and the nominal 10 µM concentration in the donor
compartment, according to the equation below (units ) nm/s),
where 3600 s is the total time for the measurement of compound
(
2
flux and 1.13 cm is the area of the Transwell filter.
(
2
Papp ) (CBL × 1.5 mL)/3600 s/1.13 cm /10 µM
The coordinates for the complex of cathepsin K and 10 have
been deposited in the Protein Data Bank, accession number 2FTD.
(13) Veber, D. F.; Johnson, S. R.; Cheng, H.-Y.; Smith, B. R.; Ward, K.
W.; Kopple, K. D. Molecular Properties That Influence the Oral
Bioavailability of Drug Candidates. J. Med. Chem. 2002, 45, 2615-
Acknowledgment. The authors would like to acknowledge
Wanda Bodnar, Jo Salisbury, and Tim Tippin for conducting
the MDCK permeability experiments.
2623.
(14) Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J.
Experimental and Computational Approaches To Estimate Solubility
and Permeability in Drug Discovery and Development Settings. AdV.
Drug. Del. ReV. 2001, 46, 3-26.
15) Ertl, P.; Rohde, B.; Selzer, P. Fast Calculation of Molecular Polar
Surface Area as a Sum of Fragment-Based Contributions and Its
Application to the Prediction of Drug Transport Properties. J. Med.
Chem. 2000, 43, 3714-3717.
Supporting Information Available: ORTEPs, crystal data, and
coordinates for compounds 3, 5, 6, 10, 46, and 47. This material is
available free of charge via the Internet at http://pubs.acs.org.
(
References
(
1) Stroup, G. B.; Lark, M. W.; Veber, D. F.; Bhattacharyya, A.; Blake,
S.; Dare, L. C.; Erhard, K. F.; Hoffman, S. J.; James, I. E.; Marquis,
R. W.; Ru, Y.; Vasko-Moser, J. A.; Smith, B. R.; Tomaszek, T.;
Gowen, M. Potent and Selective Inhibition of Human Cathepsin K
Leads to Inhibition of Bone Resorption In Vivo in a Nonhuman
Primate. J. Bone Miner. Res. 2001, 16 (10), 1739-1746; Yamashita,
D. S.; Dodds, R. A. Cathepsin K and the Design of Inhibitors of
Cathepsin K. Curr. Pharm. Des. 2000, 6 (1), 1-24; Marquis, R. W.
Inhibition of the Cysteine Protease Cathepsin K. Annu. Rep. Med.
Chem. 2004, 39, 79-98.
(16) Brandt, M.; Levy, M. A. 3â-Hydroxy-∆5-steroid Dehydrogenase/3-
Keto- ∆5-steroid Isomerase from Bovine Adrenals: Mechanism of
Inhibition by 3-Oxo-4-aza Steroids and Kinetic Mechanism of the
Dehydrogenase. Biochemistry 1989, 28, 140-148.
(
17) Morrison, J. F.; Walsh, C. T. The Behavior and Significance of Slow-
Binding Enzyme Inhibitors. AdV. Enzymol. Relat. Areas Mol. Biol.
1988, 61, 201-301.
(
18) Mohamadi, F.; Richards, N. G. J.; Guida, W. C.; Liskamp, R.; Lipton,
M.; Caufield, C.; Chang, G.; Hendrickson, T.; Still, W. C.
MacroModelsAn Integrated Software System for Modeling Organic
and Bioorganic Molecules Using Molecular Mechanics. J. Comput.
Chem. 1990, 11 (4), 440-467.
(
2) 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.; D’Alessio, K. J.;
Amegadzie, B. Y.; Hanning, C. R.; Abdel-Meguid, S.; DesJarlais,
R. L.; Gleason, J. G.; Veber, D. F. Structure and Design of Potent
and Selective Cathepsin K Inhibitors. J. Am. Chem. Soc. 1997, 119,
(19) Gabrielsson, J.; Weiner, D. Pharmacokinetic and Pharmacodynamic
Data Analysis: Concepts and Applications; Apotekarsocieteten:
Stockholm, 1997.
(20) Pacifici, G. M.; Viani, A. Methods of Determining Plasma and Tissue
Binding of Drugs. Pharmacokinetic Consequences. Clin. Pharma-
cokinet. 1992, 23, 449-468.
(
46), 11351-11352; Marquis, R. W.; Yamashita, D. S.; Ru, Y.;
LoCastro, S. M.; Oh, H.-J.; Erhard, K. F.; DesJarlais, R. L.; Head,
M. S.; Smith, W. W.; Zhao, B.; Janson, C. A.; Abdel-Meguid, S. S.;
Tomaszek, T. A.; Levy, M. A.; Veber, D. F. Conformationally
Constrained 1,3-Diamino Ketones: A Series of Potent Inhibitors of
the Cysteine Protease Cathepsin K. J. Med. Chem. 1998, 41 (19),
3563-3567.
JM050915U