Azepanone-based inhibitors of human cathepsin L

Article abstract of DOI:10.1021/jm0502079

The extension of a previously reported cathepsin K azepanone-based inhibitor template to the design and synthesis of potent and selective inhibitors of the homologous cysteine protease cathepsin L is detailed. Structure-activity studies examining the effect of inhibitor selectivity as a function of the P3 and P2 binding elements of the potent cathepsin K inhibitor 1 revealed that incorporation of either a P3 quinoline-8-carboxamide or a naphthylene-1-carboxamide led to increased selectivity for cathepsin L over cathepsin K. Substitution of the P2 leucine of 1 with either a phenylalanine or a β-naphthylalanine also resulted in an increased selectivity for cathepsin L over cathepsin K. Molecular modeling studies with the inhibitors docked within the active sites of both cathepsins L and K have rationalized the observed selectivities. Optimization of cathepsin L binding by the combination of the P3 naphthylene-1-carboxamide with the P2 β-naphthylalanine provided 15, which is a potent, selective, and competitive inhibitor of human cathepsin L with a K_i = 0.43 nM.

Full text of DOI:10.1021/jm0502079

6870
J. Med. Chem. 2005, 48, 6870-6878
Azepanone-Based Inhibitors of Human Cathepsin L
Robert W. Marquis,*,† Ian James,‡,§ Jin Zeng,† Robert E. Lee Trout,† Scott Thompson,† Attiq Rahman,†
Dennis S. Yamashita,† Ren Xie,† Yu Ru,† Catherine J. Gress,‡,§ Simon Blake,‡,§ Michael A. Lark,‡,§
,#
Shing-Mei Hwang,‡ Thaddeus Tomaszek,| Priscilla Offen, Martha S. Head, Maxwell D. Cummings, and
Daniel F. Veber†,∇
Departments of Medicinal Chemistry, Bone and Cartilage Biology, Molecular Recognition, and Physical and Structural
Chemistry, GlaxoSmithKline, 1250 South Collegeville Road, Collegeville, Pennsylvania 19426
Received March 7, 2005
The extension of a previously reported cathepsin K azepanone-based inhibitor template to the
design and synthesis of potent and selective inhibitors of the homologous cysteine protease
cathepsin L is detailed. Structure-activity studies examining the effect of inhibitor selectivity
as a function of the P3 and P2 binding elements of the potent cathepsin K inhibitor 1 revealed
that incorporation of either a P3 quinoline-8-carboxamide or a naphthylene-1-carboxamide led
to increased selectivity for cathepsin L over cathepsin K. Substitution of the P2 leucine of 1
with either a phenylalanine or a â-naphthylalanine also resulted in an increased selectivity
for cathepsin L over cathepsin K. Molecular modeling studies with the inhibitors docked within
the active sites of both cathepsins L and K have rationalized the observed selectivities.
Optimization of cathepsin L binding by the combination of the P3 naphthylene-1-carboxamide
with the P2 â-naphthylalanine provided 15, which is a potent, selective, and competitive
inhibitor of human cathepsin L with a Ki ) 0.43 nM.
Introduction
The recent completion of the sequencing of the human
genome has revealed the encoding of 11 cysteine pro-
teases of the papain superfamily.1 One member of this
family of cysteine proteases, cathepsin K, has been the
subject of extensive studies reported from these and
other laboratories.2 Orally bioavailable azepanone-based
inhibitors, which were designed to target this enzyme,3
now appear to provide a general template for the
inhibition of the papain family of cysteine proteases.
Human cathepsin L (EC 3.4.22.15), an ubiquitously
expressed lysosomal cysteine endopeptidase of this
family,4 has been implicated in a wide range of intra-
and extracellular pathological and physiological pro-
cesses including antigen presentation,5 ovulation,6 tu-
mor growth and progression,7 osteo and rheumatoid
arthritis,8 as well as bone resorption.9 The cathepsin L
knockout mouse (Ctsl -/-) displays a reduced antigen
presentation, shedding of fur, and a reduction of trabe-
cular bone volume.10,11 Our interest in pursuing selec-
tive inhibitors of this enzyme is derived from the close
homologies of cathepsins K and L as well as reports that
have suggested that cathepsin L may play a role in the
resorption phase of bone remodeling. Indeed, potent
peptide aldehyde inhibitors of cathepsin L have been
reported to attenuate bone resorption both in vitro and
Figure 1.
in the murine thyroparathroidectomized (TPTX) model
of bone turnover.12,13 Additionally, the bone-related
phenotype in the cathepsin L knockout mouse suggests
that potent and selective inhibitors of this protease may
find utility in bringing further clarity to the exact role
of this cysteine protease in bone resorption and other
disparate pathological states.
A series of azepanone-based inhibitors of the osteo-
clast specific cysteine protease cathepsin K have been
reported recently from these laboratories.3 Azepanone
1 (Figure 1) was shown to be a potent inhibitor of
cathepsin K in vitro as well as an inhibitor of bone
resorption in cell-based assays. Analogue 1 has also
been shown to be a potent inhibitor of Ca2+ release in
the TPTX rat and to inhibit bone turnover in the
ovariectomized monkey.14 During the course of struc-
ture-activity studies to explore the effects of substitu-
tion of the P3 benzofuran-2-carboxamide and the P2 iso-
butyl groups of 1, several potent and selective inhibitors
of human cathepsin L were identified. In this paper, we
disclose the optimization of both potency and selectivity
of this azepanone class for the selective inhibition of
cathepsin L. The studies reported in this account
represent the first steps in a systematic, scaffold-based
approach to the selective inhibition of individual mem-
bers of the papain superfamily encoded by the human
* To whom correspondence should be addressed. Tel: 1-610-917-
7368. Fax: 1-610-917-6020. E-mail: Robert.W.Marquis@gsk.com.
† Department of Medicinal Chemistry.
‡ Department of Bone and Cartilage Biology.
§ Current address: Centocor, 145 King of Prussia Rd., Radnor,
PA 19087.
| Department of Molecular Recognition.
Department of Physical and Structural Chemistry.
# Current address: Johnson and Johnson Pharmaceutical Research
and Development, L.L.C., 665 Stockton Drive, Exton, PA 19341.
∇ Current address: 290 Batleson Rd., Ambler, PA 19002.
10.1021/jm0502079 CCC: $30.25 © 2005 American Chemical Society
Published on Web 09/30/2005

Azepanone-Based Inhibitors of Human Cathepsin L
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 22 6871
Scheme 2. Synthesis of Inhibitors 13-15a
Scheme 1. Synthesis of Inhibitors 6-8a
a Reagents and conditions: (a) Boc2O, aqueous NaOH, 1,4-
dioxane. (b) 10% Pd/C, H2, CH3OH. (c) 2-Pyridinesulfonyl chloride,
NaHCO3, CH2Cl2. (d) 4 M HCl/dioxane, CH3OH. (e) N-Boc-
phenylalanine or N-Boc-â-naphthylalanine, EDC, HOBt, CH2Cl2.
(f) Benzofuran-2-carboxylic acid or 1-naphthoic acid, EDC, CH2Cl2.
(g) Dess-Martin periodinane, CH2Cl2. (h) HPLC separation.
a Reagents and conditions: (a) 4 M HCl/dioxane, CH3OH.
(b) RCO2H, EDC, HOBt, CH2Cl2. (c) Pyridine sulfur trioxide, TEA,
DMSO or Dess-Martin periodinane. (d) HPLC separation.
genome as well as those cysteine proteases implicated
in diseases associated with parasitic infection.15
not germane to the discussion presented herein. By
analogy with our characterization of inhibitor 1, we
assume that the potent inhibitors described in this
account possess the C-4 S-stereochemistry. We have also
further confirmed this by the stepwise asymmetric
synthesis of the C-4 S-diastereomer 15 (vida infra).
The syntheses of inhibitors 13-15 are outlined in
Scheme 2. Protection of amino alcohol 9 with di-tert-
butyl dicarbonate followed by hydrogenolysis of the
carbonylbenzyloxy protecting group, sulfonylation of the
resulting amine with 2-pyridylsulfonyl chloride, and
removal of the tert-butoxycarbonyl protecting group
under acidic conditions provided the amine salt 10.
Coupling of 10 with either N-Boc-phenylalanine or
N-Boc-â-naphthylalanine and subsequent removal of the
N-Boc protecting group gave intermediates 11 and 12.
Coupling of these amine salts with benzofuran-2-car-
boxylic acid in the presence of EDC followed by oxida-
tion with the Dess-Martin reagent and separation of
the C-4 diastereomers gave 13 and 14. Alternatively,
coupling of intermediate 12 with naphthlene-1-carbox-
ylic acid followed by oxidation and HPLC separation of
the diastereomeric ketones provided inhibitor 15. Here
again, the prevailing assumption is that the faster
eluting, more potent diastereomer 15 possesses the C-4
S-stereochemistry. However, the somewhat equivocal
nature of this critical stereochemical assignment was
viewed as troublesome. In an effort to eliminate all
ambiguities surrounding this issue, inhibitor 15 was
synthesized utilizing the key chiral amino alcohol
intermediate 17 (Scheme 3), which was synthesized in
seven steps from oxazolidinone 16.16 Acylation of 17
with N-Boc-â-naphthylalanine in the presence of EDC
followed by deprotection of the amine under acidic
conditions provided the amine hydrochloride salt 18.
Synthesis Chemistry
The syntheses of inhibitors 6-8 began with sulfona-
mide 2, the synthesis of which has been described
previously (Scheme 1).3 Treatment of 2 with 4 M HCl
in dioxane followed by coupling of the resulting amine
salt with either quinoline-2-carboxylic acid, quinoline-
8-carboxylic acid, or naphthylene-1-carboxylic acid in
the presence of EDC provided the intermediate alcohols
3-5 as mixtures of diastereomers. Oxidation of the
alcohols 3-5 with either the Dess-Martin reagent or
the pyridine sulfur trioxide complex followed by HPLC
separation of the C-4 diastereomeric ketones provided
inhibitors 6-8. In the previously reported account of the
synthesis and characterization of azepanone 1, we had
shown that the diastereomer that eluted first by HPLC
was consistently the more potent inhibitor of cathepsin
K. This observation was true for all members of the
azepanone class of inhibitors. Azepanone 1, the faster
eluting and more potent of the C4 diastereomers, was
shown by small molecule X-ray crystallography to
possess the S-configuration at C-4. Additionally, an
X-ray cocrystal structure of 1 bound within the active
site of cathepsin K also confirmed the critical nature of
the C-4 S-stereochemistry for effective binding. In these
studies, the corresponding C-4 R-diastereomer was
shown to possess at most 1/5000th the potency for
cathepsin K than the C-4 S-diastereomer 1. During the
course of the work detailed in this manuscript, the same
trend emerged whereby the slower eluting diastereo-
mers were also significantly less active than their faster
eluting counterparts. Because of this difference, the
enzyme potencies of the slower eluting distereomers are

6872 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 22
Scheme 3. Enantiospecfic Synthesis of Inhibitor 15a
Marquis et al.
Figure 2. 1H NMR spectroscopic data of inhibitors 7 and 8.
have led to the identification of the potent and selective
inhibitors of human cathepsin L reported in this ac-
count.
Substitution of the P3 benzofuran-2-carboxamide of
a Reagents and conditions: (a) N-Boc-â-naphthylalanine, EDC,
HOBt, CH2Cl2. (b) 4 M HCl/dioxane, CH3OH. (c) 1-Naphthoic acid,
EDC, HOBt, CH2Cl2. (d) Dess-Martin periodinane, CH2Cl2.
azepanone 1 with quinoline-2-carboxamide provided 6,
which is a potent inhibitor of cathepsin K with Ki,app
)
0.89 nM (Table 1). Relative to the benzofuran analogue
1, incorporation of the P3 quinoline-2-carboxamide
leading to 6 has led to a slight increase in selectivity
over cathepsins L (Ki,app ) 5.0 nM; K/L ) 0.18) and S
(Ki,app ) 27 nM; S/L ) 5.4). Substitution with the
isomeric quinoline-8-carboxamide provided analogue 7.
Here, this P3 substitution has resulted in a significant
reduction in potency versus cathepsin K (Ki,app ) 52 nM)
relative to both 1 and 6 while maintaining potency for
cathepsin L (Ki,app ) 8.2 nM; K/L ) 6.3). Again, relative
to azepanone 1, the quinoline-8-carboxamide 7 shows
improved selectivity vs cathepsins S and B with Ki,app
values of 41 and >1000 nM, respectively. Incorporation
of the P3 naphthylene-1-carboxamide provided 8, which
is a potent inhibitor of cathepsin L (Ki,app ) 1.4 nM),
Table 1. Cathepsins L, K, S, and B Inhibition Dataa
Ki,app (nM)
compd
Cat L
Cat K
0.16
0.89
Cat S
Cat B
ratio K/L
1
2.2
4.3
27
41
5.7
8.9
3.1
15.6
500
740
>1000
0.07
0.18
6
5.0
7
8.2
52
6.3
8
1.4
12
10.5
295
>10000
8.6
13
14
15
1.7
193
24
150
6.2
0.57
0.43
517
>20000
a Ki,app values for all newly reported analogues represent the
average of at least three independent titrations.
Acylation of 18 with 1-naphthoic acid followed by
oxidation of the intermediate alcohol provided 15, which
was identical to the active C-4 S-diastereomer described
above in Scheme 2. This chemical correlation provides
further support for the critical role of the C-4 S-
stereochemical assignment assumed in all of the inhibi-
tors described in this account. From this work, we
conclude that the stereochemical preference of the
azepanone inhibitors described above for cathepsin L
is identical to that seen for the previously reported
inhibitors of cathepsin K.
approximately 8-fold selective over cathepsin
K
(Ki,app ) 12 nM). Interestingly, inhibitor 8, which
contains the P3 naphthylene-1-carboxamide, is more
potent than corresponding P3 naphthylene-8-quinoline
carboxamide derivative 7 in all of the enzymes examined
(cat L Ki,app ) 1.4 nM; cat K Ki,app ) 12 nM; and cat S
Ki,app ) 5.7 nM). An explanation for this difference in
1
potencies was found upon inspection of the H NMRs
1
of both 7 and 8. The H NMR of the P3 quinoline-8-
carboxamide analogue 7 (500 MHz, d6-DMSO) shows a
one proton doublet (J ) 7.3 Hz) centered at 11.73 ppm
corresponding to the N-H amide proton as indicated
in Figure 2. Upon dilution, the chemical shift of this
proton was unaltered. These data implicate the exist-
ence of an intramolecular hydrogen bond between the
nitrogen of the quinoline-8-carboxamide moiety and the
amide N-H as shown in Figure 2. This was further
confirmed by a series of NOE difference experiments,
which showed a 2% enhancement of the C-7 proton of
the quinoline moiety upon irradiation of the N-H amide
proton. A smaller enhancement (0.4%) was observed
between the H-2 hydrogen of the quinoline and the N-H
amide proton. The relative magnitude of these enhance-
ments supports the presence of a significant degree of
preorganization of the P3 quinoline-8-carboxamide of
7.19 Alternatively, the corresponding N-H proton of 8
is a one proton doublet (J ) 7.3 Hz) centered at 8.36
ppm. The chemical shift of this proton relative to that
contained within 7 suggests that it is not involved in a
hydrogen-bonded interaction. Difference NOE experi-
ments with 8 indicated that upon irradiation of the
N-H proton there was no enhancement of the C-7
proton of the naphthyl moiety. Again, in combination,
Results and Discussion
Inhibitor Potencies and Selectivities. As has been
shown previously, azepanone 1 is modestly selective for
cathepsin K (Ki ) 0.16 nM; Table 1) vs the highly
homologous cathepsins L (Ki,app ) 2.2 nM; K/L ) 0.07)
and S (Ki,app ) 4.3 nM).3 Not unexpectedly, 1 is very
selective for cathepsin K over the cysteine exopeptidase
cathepsin B (Ki,app ) 500 nM). In an in vitro cell-based
assay of bone resorption, 1 is a potent inhibitor of native
cathepsin K with an IC50 ) 70 nM.3,17 In an in situ
cytochemical assay, a cell-based measure of cathepsin
activity within human tissue sections containing osteo-
clasts, analogue 1 is also an effective inhibitor with an
IC50 ) 80 nM.18 Upon the basis of the promising potency
profile of 1, an array-based approach was employed in
order to probe the effect of substitution of the P3
benzofuran-2-carboxamide of 1 while retaining the
invariant P2 isobutyl moiety. In a like manner, substi-
tution of the P2 iso-butyl group of inhibitor 1 was probed
while the P3 benzofuran-2-carboxamide remained con-
stant. The results from these two independent studies

Azepanone-Based Inhibitors of Human Cathepsin L
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 22 6873
Figure 3. Docking of the P3 quinoline-2-carboxamide 6 and the quinoline-8-carboxamide 7 docked within the active site of human
cathepsin K.
these data suggest that there is no intramolecular
hydrogen bond within this portion of the molecule.
We conclude that the intramolecular hydrogen bond
contained within 7 results in the reduced potency of this
inhibitor on the cathepsins studied relative to that of
analogue 8. In the X-ray cocrystal structure of several
pyrrolidinone20 and azepanone-based inhibitors bound
within the active site of cathepsin K, we have observed
that this particular N-H of the inhibitors studied is
involved in a critical hydrogen bond with the carbonyl
oxygen of Gly 66 of the protein backbone.21 To promote
effective binding of 7 to cathepsins L, K, and S, the
intramolecular hydrogen bond of the inhibitor must be
disrupted, an energetically unfavorable process. As the
Gly 66 residue is conserved throughout the papain
superfamily of cysteine proteases, it is expected that
there would be a concomitant reduction of inhibitor
potency for all of these enzymes for inhibitor 7 relative
to 8. Indeed, this is reflected consistently in the Ki,app
values of inhibitors 7 and 8 versus all of the cysteine
proteases examined in this study (Table 1).
In an effort to more fully understand the nature of
the binding interactions of the P3 quinoline-8-carboxa-
mide of 7, which have led to the increased selectivity
for cathepsin L relative to the P3 quinoline-2-carboxa-
mide of 6, a series of molecular modeling experiments
were conducted. These studies utilized the previously
published X-ray cocrystal structure of azepanone 1
bound within the active site of human cathepsin K
(1NLJ, 2.0 Å)3 and the X-ray crystal structure of human
pro-cathepsin L (1CS8, 1.80 Å) from which the proregion
had been deleted.22 This modeled structure of human
procathepsin L was utilized in the absence of an
appropriate X-ray crystal structure of mature human
cathepsin L. As seen in Figure 3, docking of 6 (cat K
Ki,app ) 0.89 nM; cat L Ki,app ) 5.0 nM) within the active
site of cathepsin K shows that the terminal aromatic
group of the P3 quinoline-2-carboxamide is capable of
maintaining a critical aromatic-aromatic interaction
with Tyr 67 within the S3 pocket of the active site.23
This interaction mimics the aromatic-aromatic interac-
tion between the phenyl moiety of Tyr 67 of the protein
backbone and the P3 benzofuran-2-carboxamide of the
potent cathepsin K inhibitor 1 bound within the active
site. Docking studies with inhibitor 7 (cat K Ki,app ) 52
nM; cat L Ki,app ) 8.2 nM) reveal that the aromatic-
aromatic interaction between the Tyr 67 of cathepsin
K and the quinoline-8-carboxamide moiety is partially
eliminated. Rotation of quinoline-8-carboxamide group
of 7 toward Tyr 67 creates a steric clash between these
two moieties which would not promote an effective
binding interaction. Alternatively, rotation of the quino-
line-8-carboxamide away from Tyr 67 (not pictured) in
order to alleviate this poor interaction leads to unfavor-
able steric interactions with Asp 61 and Gly 64 of the
protein backbone. This poor interaction between the P3
quinoline-8- carboxamide of 7 and the Tyr 67 coupled
with the steric crowding of the inhibitor with Asp 61
and Gly 64 represents what is believed to be a plausible
explanation for the reduced potency of this compound
for cathepsin K.
A comparison of the X-ray crystal structures as well
as the primary amino acid sequences of cathepsins K
and L reveal two critical differences in amino acid
residues within the S3 binding pockets of these enzymes
that may account for the similar cathepsin L potencies
of 1, 6, and 7. The first of these differences is that Tyr
67 in the S3 binding pocket of cathepsin K is replaced
by Leu 69 in cathepsin L (see Figures 4 and 5). The
presence of the hydrophobic Leu 69 residue in cathepsin
L would eliminate the possibility of an aromatic-
aromatic interaction within the S3 pocket of this
enzyme. This was seen to be a key interaction in the
docking studies and in the X-ray cocrystal structure of
1 bound in the active site of cathepsin K. Second, a
comparison of the aromatic residues within the S3
binding pockets of cathepsins K and L shows that the
aromatic group within the S3 pocket of cathepsin L is
Tyr 72 which is located at the end of this pocket. As
discussed above, the aromatic residue in the S3 binding
pocket of cathepsin K is Tyr 67 which is located at the
bottom of this hydrophobic pocket (compare Figures

6874 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 22
Marquis et al.
Figure 4. Docking of the P3 quinoline-2-carboxamide 6 and the quinoline-8-carboxamide 7 docked within the active site of human
procathepsin L (for clarity, the proregion has been deleted).
engage Tyr 72 in a similar interaction. However, in this
instance, the P3 quinoline-8-carboxamide moiety may
be able to remain buried within the S3 pocket since the
iso-butyl side chain of Leu 69 is smaller and should
possess the appropriate conformational flexibility to
permit the binding of this group.
Investigation of the effect of the substitution of the
P2 iso-butyl moiety of 1 on inhibitor selectivities re-
vealed that azepanone 13, which incorporates a P2
benzyl group, was 6-fold more selective for cathepsin L
over cathepsin K (cat K Ki,app ) 10.2 nM; cat L Ki,app
1.7 nM), 5-fold selective over cathepsin S (Ki,app ) 8.9
nM), and 113-fold selective over cathepsin B (Ki,app
)
)
193 nM). Extension of the P2 aromatic group via the
incorporation of the P2 â-naphthylalanine to provide 14
further increased selectivity for cathepsin L via a
diminution in potency for cathepsin K (cat K Ki,app
290 nM; cat L Ki,app ) 0.57 nM; cat K/L ) 517). The
potency of 14 for the inhibition of cathepsin S (Ki,app
)
)
3.1 nM) was similar to that of inhibitor 13. The potency
of 14 for the inhibition of cathepsin B (Ki,app ) 42 nM)
increased 7-fold relative to the P2 phenylalanine ana-
logue 13.
Bro¨mme and co-workers have investigated the effect
that substitution of the P2 amino acid residue plays in
determining selectivities of cathepsins K, L, and S on
rate ratios for a series of irreversible vinyl sulfone
inhibitors.24 These studies and others have shown that
the hydrophobic S2 binding pocket of cathepsin K is
specific for leucine.25 On the basis of this work as well
as modeling studies, the loss in potency for cathepsin
K upon incorporation of the P2-benzyl or naphthyl
moieties is likely the result of a steric clash of these
groups with Leu 160 and/or Leu 209 within the S2
pocket of the active site. These residues serve to
effectively limit access of larger amino acid residues
(Figure 5) into the S2 binding pocket. The corresponding
residues of the S2 pocket of cathepsin L are Ala 214
and Met 161. Here, the smaller alanine side chain in
combination with the conformationally more flexible
Met 161 make the S2 pocket of cathepsin L larger and
Figure 5. Overlay of the P3 and P2 binding pockets within
the active sites of cathepsin L (orange) and cathepsin K (aqua)
highlighting the residues within these pockets that are critical
in determining the specificities for these cysteine proteases.
3-5). The relative locations of the aromatic amino acid
residues present in the S3 binding pockets of cathepsins
L and K suggest that cathepsin L may be less sensitive
to the differences between the P3 aromatic moieties of
inhibitors 1, 6, and 7. The similar cathepsin L potencies
of these azepanones suggest this to be true (Table 1).
Docking of 6 and 7 within the active site of cathepsin L
(Figure 4) shows that the quinoline-2-carboxamide of 6
may be capable of a ring-ring interaction with Tyr 72
of cathepsin L. The conformation required for the
binding of the quinoline-8-carboxamide of 7 could not

Azepanone-Based Inhibitors of Human Cathepsin L
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 22 6875
for the increased cathepsin L/K selectivity suggesting
that the quinoline-8-carboxamide and, by extension, the
closely related naphthylene-1-carboxamide, cannot ef-
fectively engage Tyr 67 within the S3 pocket of cathep-
sin K in a critical aromatic-aromatic interaction.
Additionally, the incorporation of the P3 quinoline-8-
carboxamide was shown to result in a steric clash with
amino acid residues Asp 61 and Gly 64 within this same
binding pocket. Replacement of the P2 leucine moiety
of azepanone 1 with either a phenylalanine or a â-naph-
thylalanine also resulted in a loss of potency vs cathep-
sin K while retaining potencies for both cathepsins L
and S. This increase in cathepsin L and S selectivity vs
cathepsin K was related to the inability of the smaller
more restrictive S2 binding pocket of cathepsin K to
accommodate the larger phenyl or naphthyl moieties.
Both cathepsins L and S, which possess larger S2
binding pockets, permitted the binding of these aromatic
groups.
These potent and selective inhibitors should play a
useful role in helping to further define the role of
cathepsin L in both normal and pathological physiology.
Through the incorporation of the appropriate P3, P2,
and P1′ specificity determinants, we anticipate that this
azepanone template may be extended to the selective
inhibition of many members of the papain family of
cysteine proteases.
Figure 6. Lineweaver-Burk plot of inhibitor 15.
permit access of the phenyl and naphthyl moieties of
inhibitors 13 and 14.
From the above-mentioned structure-activity rela-
tionship and modeling studies, azepanones 8 and 14
were identified as possessing individually the optimal
binding elements required for the P2 and P3 pockets of
human cathepsin L. A combination of the naphthylene-
1-carboxamide of 8 with the P2 of the â-naphthylalanine
of 14 has led to azepanone 15, which is a 0.43 nM
inhibitor of cathepsin L with greater than 20000-fold
selectivity over cathepsin K (Ki,app ) >10000 nM). It is
36-fold selective for cathepsin L over cathepsin S
(Ki,app ) 15.6 nM) and 340-fold selective over cathepsin
B (Ki,app ) 150 nM). The Lineweaver-Burk plot (Figure
6) showed a constant 1/v axis intercept with increasing
substrate concentrations, indicative of a competitive
mode of inhibition of cathepsin L by 15.
As reported previously, inhibitor 15 was devoid of
activity in the in situ cyctochemical assay when evalu-
ated at concentrations up to 1 µM and was also inactive
in an in vitro assay of bone resorption, which employs
osteoclasts harvested from osteoclastomas up to a
concentration of 300 nM.3,26,27 In contrast, inhibitor 1
(cathepsin Ki ) 0.16 nM) has been reported to be a
potent inhibitor in both the in situ cytochemical assay
(IC50 ) 70 nM) and the human osteoclast resorption
assay (IC50 ) 80 nM). These results have suggested that
it is the inhibition of the latter enzyme that is required
for the attenuation of bone resorption by small molecule
inhibitors rather than the inhibition of cathepsin L.
It is also now recognized that the previously reported
selective peptide-derived aldehyde inhibitors of cathep-
sin L, which, in part, formed the basis for the hypothesis
of this protease’s importance in bone resorption, are in
actuality at least as potent inhibitors of human cathep-
sin K, a property not explored in these earlier studies.28
Experimental Section
General. Materials and reagents were used as supplied.
Nuclear magnetic resonance spectra were recorded at either
250, 400, or 500 MHz. Mass spectra were taken on a PE Syx
API III instrument using electrospray (ES) ionization tech-
niques. 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 60 (230-400 mesh) silica
gel.
Quinoline-2-carboxylic Acid {(S)-1-[3-Oxo-1-(pyridine-
2-sulfonyl)azepan-4-ylcarbamoyl]-3-methyl-butyl}-
amide (6). To a solution of 2 (1.20 g, 2.48 mmol) in CH3OH
(5.0 mL) was added 4 M HCl in dioxane (5 mL). This mixture
was maintained at room temperature for approximately 2 h
whereupon it was concentrated to provide 1.1 g of (S)-2-amino-
4-methyl-pentanoic acid [3-hydroxy-1-(pyridine-2-sulfonyl)-
azepan-4-yl]amide hydrochloride as a white powder, which was
used in the following step with no further purification. To a
solution of the amine hydrochloride (0.57 g, 1.25 mmol) in CH2-
Cl2 (10 mL) were added TEA (0.44 mL, 3.12 mmol), EDC (0.26
g, 1.38 mmol), HOBt (0.17 g, 1.25 mmol), and quinoline-2-
carboxylic acid (0.22 g, 1.25 mmol). This mixture was main-
tained at room temperature until complete consumption of the
starting material was observed as determined by TLC analysis.
The mixture was concentrated, and the residue was dissolved
in EtOAc and washed with saturated NaHCO3 and brine. The
organic layer was dried (MgSO4), filtered, and concentrated.
Column chromatography (4:1 EtOAc:hexanes) of the residue
Conclusions
In this paper, we have detailed the extension of the
azepanone template of cathepsin K inhibitors to the
design and synthesis of potent and selective inhibitors
of the highly homologous cysteine protease cathepsin
L. Structure-activity studies investigating substitution
of the P3 benzofuran-2-carboxamide of the potent ca-
thepsin K inhibitor 1 led to the discovery that incorpo-
ration of a P3 quinoline-8-carboxamide or naphthylene-
1-carboxamide resulted in a diminution of cathepsin K
inhibition while retaining potent cathepsin L activity.
Molecular modeling experiments were able to account
1
provided 0.67 g (64%) of 3 as a white powder. H NMR (400
MHz, CDCl3): δ 8.74-8.52 (m, 2H), 8.36 (m, 2H), 8.22 (m, 1H),
8.05-7.51 (m, 5H), 6.72 (m, 1H), 4.63 (m, 1H), 4.15 (m, 1H),
4.05-3.24 (m, 6H), 2.03-1.65 (m, 8H), 1.07 (m, 6H). MS (ESI)
540.4 (M + H)+. To a solution of alcohol 3 (0.36 g, 0.67 mmol)
in DMSO (2.0 mL) were added TEA (0.93 mL, 6.68 mmol) and
pyridine sulfur trioxide complex (0.53 g, 3.34 mmol). This
mixture was maintained at room temperature for approxi-
mately 2 h whereupon it was diluted with EtOAc and washed
with saturated NaHCO3 and brine, dried (MgSO4), filtered,
and concentrated. Column chromatography (2:1 EtOAc:hex-
anes) provided 0.29 g (81%) of the diastereomeric mixture as

6876 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 22
Marquis et al.
1
a white powder. Separation of this mixture by preparative
HPLC (50:50 hexanes:ethanol, 10 µ 100 Å (R,R) Whelk-O, 25
cm × 21.1 mm ID) gave 0.07 g of the desired faster eluting
S-diastereomer. 1H NMR (400 MHz, CDCl3): δ 8.72 (d, J )
8.22 Hz, 1H), 8.68 (d, J ) 4.76 Hz, 1H), 8.34 (d, J ) 8.46 Hz,
1H), 8.30 (d, J ) 8.46 Hz, 1H), 8.22 (d, J ) 8.53 Hz, 1H), 7.96
(d, J ) 7.68 Hz, 1H), 7.93 (td, J ) 7.74, 1.60 Hz, 1H), 7.89 (d,
J ) 8.20 Hz, 1H), 7.79 (t, J ) 7.67 Hz, 1H), 7.64 (t, J ) 7.49
Hz, 1H), 7.51 (m, 1H), 7.13 (d, J ) 6.38 Hz, 1H), 5.12 (m, 1H),
4.75 (d, J ) 19.8 Hz, 1H), 4.72 (m, 1H), 4.09 (d, J ) 14.7 Hz,
1H), 3.80 (d, J ) 19.0 Hz, 1H), 2.65 (t, J ) 12.5 Hz, 1H), 2.22
(m, 1H), 2.12 (m, 1H), 1.80 (m, 4H), 1.43 (m, 1H), 1.01 (d, J )
4.41 Hz, 3H), 0.99 (d, J ) 4.28 Hz, 3H). 13C NMR (125.77 MHz,
CDCl3): δ 205.8, 170.9, 164.2, 157.2, 150.2, 148.8, 146.1, 138.2,
137.9, 130.4, 129.6, 129.4, 128.2, 127.7, 127.0, 122.4, 119.0,
58.7, 57.9, 52.1, 51.3, 41.3, 31.7, 28.3, 24.9, 22.9, 22.1. MS (ES)
538.2 (M + H)+. Elemental analysis (C27H31N5O5S) C, H, N.
Quinoline-8-carboxylic Acid {(S)-1-[3-Oxo-1-(pyridine-
2-sulfonyl)azepan-4-ylcarbamoyl]-3-methyl-butyl}-
amide (7). 1H NMR (500 MHz, CDCl3): δ 11.72 (d, 1H), 8.97
(dd, 1H), 8.86 (dd, 1H), 8.70 (d, 1H), 8.29 (dd, 1H), 7.95-7.92
(m, 3H), 7.69 (t, 1H), 7.53-7.51 (m, 2H), 7.34 (d, 1H), 5.13 (m,
1H), 4.81 (q, 1H), 4.73, d, 1H), 4.14 (d, 1H), 3.79 (d, 1H), 2.67
(dt, 1H), 2.23 (dd, 1H), 2.18-2.08 (m, 1H), 1.91-1.81 (m, 4H),
1.43 (m, 1H), 1.01 (dd, 6H). 13C NMR (62.9 MHz, d6-DMSO):
205.73, 171.7, 166.1, 157.3, 150.2, 149.5, 145.6, 138.1, 137.7,
134.1, 132.2, 128.5, 128.1, 126.9, 126.5, 122.4, 121.0, 58.6, 57.8,
52.7, 51.3, 40.7, 31.7, 28.3, 25.1, 23.1, 22.1. MS (ESI) 538.2
(M + H)+. Elemental analysis (C27H31N5O5S) C, H, N.
Naphthoic-1-carboxylic Acid {(S)-1-[3-Oxo-1-(pyridine-
2-sulfonyl)azepan-4-ylcarbamoyl]-3-methyl-butyl}-
amide (8). 1H NMR (500 MHz, d6-DMSO): δ 8.76 (d, 1H), 8.66
(d, 1H), 8.23 (1H, d), 8.22 (d, 1H), 8.12 (dd, 1H), 8.02 (m, 3H),
7.74 (ddd, 1H), 7.56 (m, 4H), 4.86 (m, 1H), 4.63 (m, 1H), 4.43
(d, 1H), 3.86 (d, 2H), 2.87 (m, 1H), 1.86-1.80 (m, 3H), 1.66
(m, 1H), 1.63-1.57 (m, 3H), 0.96 (d, 3H), 0.92 (d, 3H). 13C NMR
(62.9 MHz, d6-DMSO): 205.56, 171.66, 168.62, 156.25, 150.47,
139.08, 134.58, 133.07, 129.76, 128.14, 127.62, 126.69, 126.21,
125.44, 125.30, 124.95, 122.17, 57.84, 57.21, 51.48, 50.13,
40.29, 39.99, 39.83, 39.66, 39.33, 39.16, 38.99, 29.77, 28.34,
24.48, 23.18, 21.36. MS (ESI) 540.2 (M + H)+.
powder. H NMR (400 MHz, CDCl3, as a mixture of diaster-
eomers): δ 8.71 (m, 1H), 7.95 (m, 2H), 7.68 (m, 1H), 7.51-
7.42 (m, 4H), 7.32-7.26 (m, 6H), 6.54 (m, 1H), 4.89 (m, 1H),
3.77-3.59 (m, 3H), 3.41 (m, 2H), 3.30-3.20 (m, 3H), 1.84 (m,
4H), 1.67 (m, 2H). MS (ESI) 563.0 (M + H)+. To a solution of
the diastereomeric alcohols (0.33 g, 0.59 mmol) in CH2Cl2 (6.0
mL) was added Dess-Martin periodinane (0.37 g, 0.88 mmol).
The reaction was stirred under argon at room temperature
for 1.5 h whereupon it was diluted with CH2Cl2 and washed
with 10% aqueous Na2S2O3 and two portions of saturated
NaHCO3, dried (Na2SO4), filtered, and concentrated. Column
chromatography (1:2 hexanes:ethyl acetate) followed by pre-
parative HPLC (50:50 hexanes:ethanol, 10 µ 100 Å (R,R)
Whelk-O, 25 cm × 21.1 mm ID) provided 0.074 g of the desired
ketone 13 (22.5%) as a white powder. 1H NMR (400 MHz,
CDCl3): δ 8.67 (d, J ) 4.74 Hz, 1H), 7.96 (d, J ) 7.79 Hz, 1H),
7.92 (td, J ) 7.63, 1.46 Hz, 1H), 7.66 (d, J ) 7.77 Hz, 1H),
7.52 (m, 2H), 7.46 (s, 1H), 7.43 (t, J ) 7.25 Hz, 1H), 7.35-
7.22 (m, 7H), 6.54 (d, J ) 5.89 Hz, 1H), 5.00 (m, 1H), 4.84 (dd,
J ) 14.2, 8.06 Hz, 1H), 4.68 (dd, J ) 19.1, 1.30 Hz, 1H), 4.09
(d, J ) 15.0 Hz, 1H), 3.73 (d, J ) 19.0 Hz, 1H), 3.27 (dd, J )
13.5, 6.05 Hz, 1H), 3.08 (dd, J ) 13.5, 8.36 Hz, 1H), 2.62 (t,
J ) 12.6 Hz, 1H), 2.18 (m, 2H), 1.81 (m, 1H), 1.31 (m, 1H). 13
C
NMR (125.77 MHz, CDCl3): δ 205.0, 169.5, 158.4, 157.2, 154.8,
150.2, 148.1, 138.2, 136.1, 129.2, 128.9, 127.5, 127.3, 127.1,
127.0, 123.7, 122.7, 122.4, 112.0, 110.9, 58.5, 57.9, 54.6, 51.3,
39.2, 31.7, 28.3. MS (ESI) 561.0 (M + H)+.
Benzofuran-2-carboxylic Acid {(S)-1-[3-Oxo-1-(pyri-
dine-2-sulfonyl)azepan-4-ylcarbamoyl]-2-naphthylen-2-
yl}amide (14). 1H NMR (400 MHz, CDCl3): δ 8.66 (d, J )
4.43 Hz, 1H), 7.97 (d, J ) 7.75 Hz, 1H), 7.92 (td, J ) 7.68,
1.39 Hz, 1H), 7.82 (m, 3H), 7.71 (s, 1H), 7.66 (d, J ) 7.76 Hz,
1H), 7.54-7.36 (m, 8H), 7.28 (t, J ) 7.53 Hz, 1H), 6.53 (d, J )
5.99 Hz, 1H), 5.04 (m, 1H), 4.92 (dd, J ) 14.0, 8.03 Hz, 1H),
4.47 (d, J ) 18.4 Hz, 1H), 4.06 (d, J ) 15.0 Hz, 1H), 3.62 (d,
J ) 19.0 Hz, 1H), 3.44 (dd, J ) 13.5, 5.84 Hz, 1H), 3.23 (dd,
J ) 13.5, 8.32 Hz, 1H), 2.56 (t, J ) 12.5 Hz, 1H), 2.12 (m, 2H),
1.78 (m, 1H), 1.29 (m, 1H). 13C NMR (125.77 MHz, CDCl3): δ
204.8, 169.5, 158.4, 157.2, 154.8, 150.2, 148.1, 138.2, 133.6,
133.5, 132.6, 128.7, 128.0, 127.6, 127.5, 127.2, 127.1, 127.0,
126.3, 125.8, 123.7, 122.7, 122.5, 112.0, 111.0, 58.4, 57.8, 54.6,
51.2, 39.3, 31.7, 28.2. MS (ESI) 611.2 (M + H)+.
Benzofuran-2-carboxylic Acid {(S)-1-[3-Oxo-1-(pyri-
dine-2-sulfonyl)azepan-4-ylcarbamoyl]-2-phenyl-ethyl}-
amide (13). To a solution of 10 (1.5 g, 4.88 mmol) in CH2Cl2
(25 mL) were added N-Boc phenylalanine (1.30 g, 4.88 mmol),
EDC (1.03 g, 5.37 mmol), HOBt (0.66 g, 4.88 g), and TEA (1.7
mL, 12.2 mmol). This mixture was maintained at room
temperature until complete consumption of the starting mate-
rial was observed whereupon it was concentrated. The residue
was diluted with EtOAc and washed with saturated NaHCO3
and brine. The organic layer was dried (MgSO4), filtered, and
concentrated. Column chromatography (4:1 EtOAc:hexanes)
of the residue provided 1.47 g (58%) of {(S)-1-[3-hydroxy-1-
(pyridine-1-sulfonyl)azepan-4-ylcarbamoyl]-2-phenyl-ethyl}-
carbamic acid tert-butyl ester as a white powder. MS (ESI)
519.2 (M + H)+. To a solution of {(S)-1-[3-hydroxy-1-(pyridine-
1-sulfonyl)azepan-4-ylcarbamoyl]-2-phenyl-ethyl}carbamic acid
tert-butyl ester (1.2 g, 2.32 mmol) in methanol (23 mL) was
added a solution of 4 M HCl in dioxane (5.8 mL). This mixture
was maintained at room temperature overnight whereupon it
was concentrated and azeotropically dried with toluene (3×)
to provide 1.12 g of 11 as a white powder, which was used in
the following step with no further purification. MS (ESI) 419.4
(M + H)+. To a mixture of 11 (0.20 g, 0.44 mmol), 2-benzo-
furancarboxylic acid (0.077 g, 0.48 mmol), EDC (0.095 g, 0.50
mmol), and HOBt (0.067 g, 0.50 mmol) were added CH2Cl2
(4.5 mL) and TEA (0.25 mL, 1.8 mmol). The reaction mixture
was stirred under argon at room temperature for 22 h. The
reaction was diluted with ethyl acetate and washed succes-
sively with saturated K2CO3 and brine. The combined aqueous
layers were then back extracted with ethyl acetate. The
combined organic layers were dried over MgSO4, filtered, and
concentrated. Column chromatography (4% CH3OH:CH2Cl2)
yielded 0.17 g (68%) of the intermediate alcohol as a white
Naphthalene-2-carboxylic Acid {(S)-2-Naphthalen-2-
yl-1-[(S)-3-oxo-1-(pyridine-2-sulfonyl)azepan-4-ylcarbam-
oyl]ethyl}amide (15). To a mixture of the amine hydrochlo-
ride 12 (0.51 g, 1.0 mmol), 1-naphthoic acid (0.18 g, 1.0 mmol),
EDC (0.21 g, 1.1 mmol), and HOBt (0.15 g, 1.1 mmol) were
added CH2Cl2 (10 mL) and TEA (0.56 mL, 4.0 mmol). The
mixture was maintained under argon at room temperature for
18 h whereupon it was diluted with ethyl acetate and washed
with saturated K2CO3 and brine. The combined aqueous layers
were then back extracted with ethyl acetate. The combined
organic layers were dried over MgSO4, filtered, and concen-
trated. Column chromatography (4% CH3OH:CH2Cl2) yielded
0.42 g (67%) of the diastereomeric alcohols as a white powder.
1H NMR (400 MHz, CDCl3, as a mixture of diastereomers): δ
8.66 (m, 1H), 7.95-7.79 (m, 9H), 7.47 (m, 6H), 7.35 (m, 1H),
7.23 (m, 1H), 6.82 (m, 1H), 5.18 (m, 1H), 3.79 (m, 1H), 3.62
(m, 1H), 3.51-3.26 (m, 5H), 1.70 (m, 4H), 1.58-1.42 (m, 3H).
MS (ESI) 623.0 (M + H)+. To a solution of the alcohols (0.41
g, 0.66 mmol) in CH2Cl2 (6.5 mL) was added Dess-Martin
periodinane (0.45 g, 1.0 mmol). The reaction was maintained
under argon at room temperature for 1.5 h whereupon it was
diluted with CH2Cl2 and washed with 10% aqueous Na2S2O3
and two portions of saturated NaHCO3, dried (Na2SO4),
filtered, and concentrated. Column chromatography (4% CH3-
OH/CH2Cl2) followed by preparative HPLC (30:70 hexanes:
ethanol, 10 µ 100 Å (R,R) Whelk-O, 25 cm × 21.1 mm ID)
1
provided 0.11 g of the ketone 15 as a white powder. H NMR
(400 MHz, CDCl3): δ 8.67 (d, J ) 4.93 Hz, 1H), 7.98 (m, 2H),
7.94 (m, 1H), 7.89 (d, J ) 8.23 Hz, 1H), 7.84 (m, 3H), 7.77 (m,
1H), 7.72 (s, 1H), 7.53 (m, 2H), 7.46 (m, 4H), 7.40 (m, 1H),
7.29 (m, 1H), 6.84 (d, J ) 6.13 Hz, 1H), 6.65 (d, J ) 7.80 Hz,
1H), 5.10 (m, 2H), 4.61 (dd, J ) 19.1, 1.22 Hz, 1H), 4.11 (d,

Azepanone-Based Inhibitors of Human Cathepsin L
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 22 6877
J ) 14.8 Hz, 1H), 3.73 (d, J ) 19.0 Hz, 1H), 3.42 (m, 2H), 2.64
(m, 1H), 2.24 (m, 1H), 2.16 (m, 1H), 1.84 (m, 1H), 1.38 (m,
1H). 13C NMR (125.77 MHz, CDCl3): δ 205.1, 169.8, 169.2,
157.2, 150.2, 138.2, 134.2, 134.0, 133.8, 132.6, 130.9, 130.0,
128.7, 128.2, 128.1, 127.7, 127.6 (2), 127.2 (2), 127.0, 126.4,
126.3, 125.9, 125.2, 125.2, 124.6, 122.4, 58.5, 57.8, 55.0, 51.3,
39.0, 31.8, 28.3. MS (ES) 621.0 (M + H)+. Elemental analysis
(C35H32N4O5S‚0.5H2O) C, H, N.
References
(1) (a) Venter, C. J.; et al. The sequence of the human genome.
Science 2001, 291, 1304-1351. (b) Lander, E. S.; et al. Initial
sequencing and analysis of the human genome. Nature 2001,
409, 860-921. (c) Bro¨mme, D.; Kaleta. Thiol-depedent cathep-
sins: Pathophysiological implications and recent advances in
inhibitor design. J. Curr. Pharm. Des. 2002, 8, 1639-1658.
(2) (a) Marquis, R. W. Inhibition of the cystiene protease cathepsin
K. Ann. Rep. Med. Chem. 2004, 39, 79-98. (b) Deaton, D. N.;
Kumar, S. Prog. Med. Chem. 2004, 42, 245-375. (c) Veber, D.
F.; Drake, F. H.; Gowen, M. The new partnership of genomics
and chemistry for accelerated drug discovery. Curr. Opin. Chem.
Biol. 1997, 1, 151-156.
(3) Marquis, R. W.; Ru, Y.; LoCastro, S. M.; Zeng, J.; Yamashita,
D. S.; Oh, H.-J.; Erhard, K. E.; 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.
(4) (a) Turk, B.; Turk, D.; Turk, V. Lysosomal cysteine proteases:
More than scavengers. Biochim. Biophys. Acta 2000, 1477, 98-
11. (b) Kirschke, H.; Langner, J.; Wiederranders, B.; Ansorge,
S.; Bohley, P.; Cathepsin, L. A new proteinase from rat-liver
lysosomes. Eur. J. Biochem. 1977, 74, 293-301.
(5) Nakagawa, T.; Roth, W.; Wong, P.; Nelson, A.; Farr, A.;
Deussing, J.; Villadangos, J. A.; Ploegh, H.; Peters, C.; Rudensky,
A. Y. Cathesin, L. Critical role in Ii degradation and CD4 T cell
selection in the thymus. Science 1998, 280, 450-453.
(6) Robker, R. L.; Russell, D. L.; Espey, L. L.; Lyndon, J. P.;
O’Malley, B. W.; Richards, J. S. Progesterone-regulated genes
in the ovulation process: ADAMTS-1 and cathepsin L proteases.
Proc. Natl. Acad. Sci. 2000, 97, 4689-4694.
(7) Kirschke, H.; Eerola, R.; Hopsu-Havu, V. K.; Bro¨mme, D.; Vuorio,
E. Antisense RNA inhibition of cathepsin L expression reduces
tumorigenicity of malignant cells. Eur. J. Cancer 2000, 36, 787-
795 and references therein.
(8) Iwata, Y.; Mort, J. S.; Tateishi, H.; Lee, E. R. Macrophage
cathepsin L, a factor in the erosion of subchondral bone in
rheumatoid arthritis. Arthritis Rheum. 1997, 40, 499-509.
(9) (a) Furuyama, N.; Fujisawa, Y. Distinct roles of cathepsin K and
cathepsin L in osteoclastic bone resorption. Endocr. Res. 2000,
26, 189-204. (b) Kakegawa, H.; Nikawa, T.; Kamioka, H.;
Sumitani, K.; Kawata, T.; Drobnic-Kosorok, M.; Lenarcic, B.;
Turk, V.; Katunuma, N. Participation of cathepsin L on bone
resorption. FEBS Lett. 1993, 321, 247-250.
(10) Benavides, F.; Venables, A.; Klug, H. P.; Glasscock, E.; Ruden-
sky, A.; Gomez, M.; Palenzuela, N. M.; Guenet, J.-L.; Richie, E.
R.; Conti, C. J. The CD4-T cell-deficient mouse mutation nackt
(nkt) involves a deletion in the cathepsin L (Ctsl) gene. Immu-
nogenetics 2001, 53, 233-242.
(11) Potts, W.; Bowyer, J.; Jones, H.; Tucker, D.; Freemont, A. J.;
Millest, A.; Martin, C.; Vernon, W.; Neerunjun, D.; Slynn, G.;
Harper, F.; Maciewicz, R. Cathepsin-L deficient mice exhibit
abnormal skin and bone development and show increased
resistance to osteoporosis following ovariectomy. Int. J. Exp.
Pathol. 2004, 85, 85-96.
{(S)-1-[(3R,4S)-3-Hydroxy-1-(pyridine-2-sulfonyl)azepan-
4-ylcarbamoyl]-2-naphthylen-2-yl-ethyl}carbamic Acid
tert-Butyl Ester (18). To a solution of amino alcohol 17 (0.065
g, 0.211 mmol), EDC (0.045 g, 0.235 mmol), HOBt (0.031 g,
0.229 mmol), and Boc-L-naphthylalanine (0.067 g, 0.212 mmol)
in CH2Cl2 (2.4 mL) was added triethylamine (0.12 mL, 0.861
mmol). The reaction was stirred at room temperature under
argon for 4 days. After concentration of the reaction mixture
in vacuo, column chromatography (2-3% CH3OH/CH2Cl2)
1
produced 0.091 g (76% yield) of 18 as a white solid. H NMR
(CDCl3, 400 MHz): δ 8.62 (d, J ) 4.24 Hz, 1H), 8.00 (d, J )
7.77 Hz, 1H), 7.93 (t, J ) 7.46 Hz, 1H), 7.79 (m, 3H), 7.65 (s,
1H), 7.52 (dd, J ) 7.38, 4.88 Hz, 1H), 7.46 (m, 2H), 7.36 (d,
H ) 7.97 Hz, 1H), 6.51 (d, J ) 7.15 Hz, 1H), 5.07 (m, 1H),
4.41 (m, 1H), 4.05 (m, 1H), 3.77 (m, 1H), 3.56 (m, 1H), 3.35-
3.16 (m, 5H), 1.84 (m, 1H), 1.67 (m, 2H), 1.57 (m, 1H), 1.49 (s,
9H). 13C NMR (125.77 MHz, CDCl3): δ 170.3, 157.0, 149.6,
138.5, 134.1, 133.4, 132.4, 128.4, 128.0, 127.7, 127.5, 127.3,
126.9, 126.2, 125.7, 123.3, 69.3, 55.8, 54.6, 48.4, 47.9, 38.9, 28.2,
26.4, 24.0. MS (ESI) 569.4 (M + H)+.
Naphthalene-2-carboxylic Acid {(S)-2-Naphthalen-2-
yl-1-[(S)-3-oxo-1-(pyridine-2-sulfonyl)azepan-4-ylcarbam-
oyl]ethyl}amide (15). To a solution of 18 (0.08 g, 0.14 mmol)
in CH3OH (1.8 mL) was added HCl (0.37 mL, 4.0N in dioxane).
This mixture was maintained under argon for 19 h whereupon
it was concentrated in vacuo and dried azeotropically three
times with toluene. The crude product was carried to the next
step with no further purification. To a solution of the amine
hydrochloride (0.071 g, 0.141 mmol) in CH2Cl2 were added
triethylamine (0.08 mL, 0.574 mmol), EDC (0.031 g, 0.162
mmol), HOBt (0.022 g, 0.163 mmol), and 1-naphthoic acid
(0.025 g, 0.145 mmol) in CH2Cl2. This mixture was maintained
at room temperature for 24 h whereupon it was concentrated
in vacuo. Column chromatography (2%-5% CH3OH/CH2Cl2)
produced 0.050 g (57% yield over two steps) of the intermediate
1
alcohol as a white solid. H NMR (CDCl3, 400 MHz): δ 8.64
(d, J ) 4.23 Hz, 1H), 8.03 (d, J ) 7.82 Hz, 1H), 7.95 (m, 2H),
7.88 (d, J ) 8.17 Hz, 1H), 7.85 (m, 3H), 7.74 (m, 2H), 7.51 (m,
2H), 7.46 (m, 5H), 7.38 (m, 1H), 7.26 (m, 1H), 6.72 (d, J )
8.13 Hz, 1H), 6.67 (d, J ) 7.81 Hz, 1H), 5.08 (dd, J ) 14.8,
7.32 Hz, 1H), 4.14 (dd, J ) 15.5, 5.04 Hz, 1H), 3.83 (m, 1H),
3.64 (m, 1H), 3.40 (m, 4H), 3.25 (d, J ) 15.5 Hz, 1H), 1.87 (m,
1H), 1.78 (m, 1H), 1.62 (m, 2H). 13C NMR (125.77 MHz,
CDCl3): δ 169.8, 169.2, 156.9, 149.6, 138.5, 133.6, 132.5, 130.8,
129.9, 128.6, 128.2, 127.7, 127.5, 127.3, 127.1, 127.0, 126.4,
126.3, 125.8, 125.2, 124.6, 123.4, 69.4, 55.0, 54.8, 48.2, 47.8,
39.0, 26.5, 24.0. MS(ESI) 623.4 (M + H)+. To a solution of the
alcohol (0.040 g, 0.064 mmol) in CH2Cl2 (1.0 mL) was added
Dess-Martin periodinane (0.044 g, 0.104 mmol). This mixture
was maintained at room temperature for 45 min whereupon
it was diluted with CH2Cl2 and washed with 10% Na2S2O3
(aqueous) and two times with saturated NaHCO3. The com-
bined NaHCO3 layers were back extracted with CH2Cl2, and
the combined organic layers were dried over Na2SO4, filtered,
and concentrated. Column chromatography (1:1 hexanes:ethyl
acetate) of the residue yielded 0.036 g (90%) of 15 as a white
solid.
(12) (a) Woo, J.-T.; Yamaguchi, K.; Hayami, T.; Kobori, T.; Sigeizumi,
S.; Sugimoto, K.; Kondo, K.; Tsuji, T.; Ohba, Y.; Tagami, K.;
Sumitani, K. Suppresive effect of N-(benzyloxycarbonyl)-L-
phenylalanyl-L-tyrosinal on bone resorption in vitro and in vivo.
Eur. J. Pharmacol. 1996, 300, 131-135. (b) Millest, A. J.; Breen,
S. A.; Loveday, B. E.; Clarkson, P. N.; Sipson, C. A.; Waterton,
J. C.; Johnstone, D. Effects of an inhibitor of cathepsin L on bone
resorption in thyroparathroidectomized and ovariectomized rats.
Bone 1997, 20, 465-471.
(13) Yasuma, T.; Oi, S.; Choh, N.; Nomura, T.; Furuyama, N.;
Nishimura, A.; Fujisawa, Y.; Sohda, T. Synthesis of peptide
aldehyde derivatives as selective inhibitors of human cathepsin
L and their inhibitory effect on bone resorption. J. Med. Chem.
1998, 41, 4301-4308.
(14) Stroup, G. B.; Lark, M. L.; Veber, D. F.; Bhattacharyya, A.;
Blake, S.; Dare, L. C.; Erhard, K. F.; Hoffmann, S. J.; James, I.
E.; Marquis, R. W.; Ru, Y.; Vasko-Moser, J. A.; Smith, B. R.;
Tomaszek, T. A.; 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, 1739-
1746.
(15) (a) Mottram, J. C.; Brooks, D. R.; Coombs, G. H. Roles of cysteine
proteinases of trypanosomes and Leishmania in host-parasite
interactions. Curr. Opin. Microbiol. 1998, 1, 455. (b) McKerrow,
J. H.; McGrath, M. E.; Engel, J. C. The cysteine protease of
Trypanasoma cruzi as a model for antiparasitic drug design.
Parasitol. Today 1995, 11, 279-282.
Supporting Information Available: Elemental analysis
and HPLC data. This material is available free of charge via
the Internet at http://pubs.acs.org.

6878 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 22
Marquis et al.
(16) Trout, R. E. L.; Marquis, R. W. Asymmetric synthesis of a potent
azepanone-based inhibitor of the cysteine protease cathepsin K.
Tetrahedron Lett. 2005, 46, 2799-2801.
(17) James, I. E.; Lark, M. W.; Zembryki, D.; Lee-Rykaczewski, E.
V.; Hwang, S.-M.; Tomaszek, T. A.; Belfiore, P.; Gowen, M.
Development and characterization of a human in vitro resorption
assay: Demonstration of utility using novel antiresorptive
agents. J. Bone Miner. Res. 1999, 14, 1562-1569.
(18) Dodds, R. A.; James, I. E.; Rieman, D.; Hwang, S.-M.; Ahern,
R.; Conner, J. R.; Thompson, S.; Veber, D.; Drake, F. H.; Holmes,
S.; Lark, M. W.; Gowen, M. Human osteoclast cathepsin K is
processed intracellularly prior to attachment and bone resorp-
tion. J. Bone Miner. Res., 2001, 16, 478-486.
(19) Definitive data for the presence of a similar intramolecular
hydrogen bond was observed in the X-ray crystal structure of
the closely related P3 quinoline-8-carboxamide 16.
cysteine residue has added to the sterically more conjested re
face of the ketone carbonyl moiety as was observed in the X-ray
cocrystal structures of several azapanone-based inhibitors bound
within the active site of cathepsin K.
(24) Bro¨mme, D.; Klaus, J. L.; Okamoto, K.; Rasnick, D.; Palmer, J.
T. Peptidyl vinyl sulphones: A new class of potent and selective
cysteine protease inhibitors. Biochem. J. 1996, 315, 85-89.
(25) (a) McGrath, M. E.; Eakin, A. E.; Barnes, M. G.; Bro¨mme, D.
Crystal structure of human cathepsin K complexed with a potent
inhibitor. Nat. Struct. Biol. 1997, 5, 104-109. (b) Zhao, B.;
Janson, C. A.; Amegadzie, B. Y.; D’Alessio, K.; Griffin, C.;
Hanning, C. R.; Jones, C.; Kurdyla, J.; McQueney, M. S.; 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, 109-111.
(26) James, I. E.; Marquis, R. W.; Blake, S. M.; Hwang, S. M.; Gress,
C. J.; Ru, Y.; Zembryki, D.; Yamashita, D. S.; McQueney, M. S.;
Tomaszek, T. A.; Oh, H.-J.; Gowen, M.; Veber, D. F.; Lark, M.
W. Potent and selective inhibitors of cathepsin L do not inhibit
human osteoclast resorption in vitro. J. Biol. Chem. 2001, 276,
11507-11511.
(27) A correction to ref 26 has been published. James, I. E.; Marquis,
R. W.; Blake, S. M.; Hwang, S. M.; Gress, C. J.; Ru, Y.; Zembryki,
D.; Yamashita, D. S.; McQueney, M. S.; Tomaszek, T. A.; Oh,
H.-J.; Gowen, M.; Veber, D. F.; Lark, M. W. Potent and selective
inhibitors of cathepsin L do not inhibit human osteoclast
resorption in vitro. J. Biol. Chem. 2003, 278, 32484.
(28) Our analysis of the selectivities of the potent cathepsin L peptide
aldehyde inhibitors 16 and 17, reported in refs 12a and 13,
respectively, revealed that these analogues were equally potent
inhibitors of both cathepsins K and S in addition to the potent
inhibition of cathepsin L. These results, which were not reported
in the referenced accounts, suggests that the conclusions regard-
ing the mode by which these inhibitors may have elicited their
antiresorptive effect were ambiguous.
(20) Marquis, R. W.; Ru, Y.; Zeng, J.; Trout, R. E. L.; LoCastro, S.
M.; Gribble, A. D.; Witherington, J.; Fenwick, A. E.; Garnier,
B.; Tomaszek, T. A.; Tew, D.; Hemling, M. E.; Quinn, C. J.;
Smith, W. W.; Zhao, B.; McQueney, M. S.; Janson, C. A.;
D’Alessio, K.; Veber, D. F. Cyclic ketone inhibitors of the cysteine
protease cathepsin K. J. Med. Chem. 2001, 44, 725-736.
(21) Within these series, N-methylation of this amide N-H moiety
resulted in significant losses of potency vs the cathepsins studied.
(22) (a) Coulombe, R.; Grochulski, P.; Sivaraman, J.; Menard, R.;
Mort, J. S.; Cygler, M. Structure of human procathepsin L
reveals the molecular basis of inhibition of the prosegment.
EMBO J. 1996, 15, 5492-5503. (b) Groves, M. R.; Coulombe,
R.; Jenkins, J.; Cygler, M. Structural basis for specificity of
papain-like cysteine protease proregions toward their cognate
enzymes proteins. Struct., Funct., Genet. 1998, 32, 504.
(23) In the absence of X-ray cocrystal data of either 6 or 7 within
the active site of cathepsin L, we are assuming that these
inhibitors orient themselves on the unprimed side of the active
site in a similar fashion as inhibitor 1 within the active site of
cathepsin K. Additionally, we anticipate that the active site
JM0502079