Identification of potent and selective cathepsin S inhibitors containing different central cyclic scaffolds

Article abstract of DOI:10.1021/jm401528k

Starting from the weakly active dual CatS/K inhibitor 5, structure-based design supported by X-ray analysis led to the discovery of the potent and selective (>50 000-fold vs CatK) cyclopentane derivative 22 by exploiting specific ligand-receptor interactions in the S2 pocket of CatS. Changing the central cyclopentane scaffold to the analogous pyrrolidine derivative 57 decreased the enzyme as well as the cell-based activity significantly by 24- and 69-fold, respectively. The most promising scaffold identified was the readily accessible proline derivative (e.g., 79). This compound, with an appealing ligand efficiency (LE) of 0.47, included additional structural modifications binding in the S1 and S3 pockets of CatS, leading to favorable in vitro and in vivo properties. Compound 79 reduced IL-2 production in a transgenic DO10.11 mouse model of antigen presentation in a dose-dependent manner with an ED ₅₀ of 5 mg/kg.

Full text of DOI:10.1021/jm401528k

Article
pubs.acs.org/jmc
Identification of Potent and Selective Cathepsin S Inhibitors
Containing Different Central Cyclic Scaffolds
Hans Hilpert,*,† Harald Mauser,† Roland Humm,† Lilli Anselm,† Holger Kuehne,† Guido Hartmann,‡
Sabine Gruener,‡ David W. Banner,§ Joerg Benz,§ Bernard Gsell,§ Andreas Kuglstatter,§ Martine Stihle,§
Ralf Thoma,§ Ruben Alvarez Sanchez,∥ Hans Iding,⊥ Beat Wirz,⊥ and Wolfgang Haap*,†
́
§
†Discovery Chemistry, ‡Cardiovascular and Metabolic Diseases, Discovery Technologies, ∥Drug Metabolism and Pharmacokinetics,
⊥Process Research and Synthesis, Pharma Research and Early Development (pRED), F. Hoffmann-La Roche Ltd., Grenzacherstrasse
124, Basel CH-4070, Switzerland
S
* Supporting Information
ABSTRACT: Starting from the weakly active dual CatS/K inhibitor 5, structure-based design supported by X-ray analysis led to
the discovery of the potent and selective (>50 000-fold vs CatK) cyclopentane derivative 22 by exploiting specific ligand−
receptor interactions in the S2 pocket of CatS. Changing the central cyclopentane scaffold to the analogous pyrrolidine derivative
57 decreased the enzyme as well as the cell-based activity significantly by 24- and 69-fold, respectively. The most promising
scaffold identified was the readily accessible proline derivative (e.g., 79). This compound, with an appealing ligand efficiency (LE)
of 0.47, included additional structural modifications binding in the S1 and S3 pockets of CatS, leading to favorable in vitro and in
vivo properties. Compound 79 reduced IL-2 production in a transgenic DO10.11 mouse model of antigen presentation in a dose-
dependent manner with an ED50 of 5 mg/kg.
Many of the recent CatS inhibitors contain an electrophilic
motif (e.g., a nitrile), which binds covalently and reversibly to
cysteine C25 (vide infra) in the active site, forming a
thioimidate.6 This binding motif has been employed, for
example, in the aminocyclopropane nitrile A by Merck Frosst7
or in the succinamide derivative B8 and more recent triazoles9
both described by Boehringer Ingelheim. Cyclohexyl-substituted
derivatives (e.g., C) were developed as CatK inhibitors at Roche
more than a decade ago10 and served as a basis for the
development of selective CatS inhibitors. Compound C is a
selective CatK inhibitor with modest activity at CatS (CatS/K,
IC50 = 0.640/0.016 μM). To change the selectivity in favor of
CatS, 1,2,4-substituted five-membered scaffolds D (Figure 1)
with three residues attached to the scaffold orientated toward the
S1−S3 pockets of the CatS active site were investigated
(supported by molecular modeling). Selectivity for CatS could
be accomplished by an appropriate extended R2 residue pointing
INTRODUCTION
■
Cathepsin S (CatS), a lysosomal cysteine protease of the papain
family primarily found in B cells, macrophages, and dendritic
cells, plays an important role in the antigen-presentation process.
Its main function is the degradation of the invariant chain Ii
associated with the major histocompatibility complex class II.1
Gene knockout studies in mice showed defects in immune
function, which were also seen with CatS inhibitors,2,3 but the
mice were otherwise healthy and normal in most respects.4,5
From a variety of preclinical results, there is evidence that CatS
inhibition could be beneficial in diseases such as neuropathic pain
as well as rheumatoid arthritis, multiple sclerosis, asthma and
allergy, chronic obstructive pulmonary diseases, emphysema,
Alzheimer’s disease, cancer, atherosclerosis, and obesity.6 CatS as
a target has attracted much attention in the pharmaceutical
industry because of its pleiotropic nature; therefore, strong
efforts have been made to identify potent and selective CatS
inhibitors. Selectivity over other cathepsins is of high importance
to evaluate the pharmacology of CatS inhibitors and to assess
their therapeutic potential in the clinic.
Received: October 1, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/jm401528k | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 1. Known CatS inhibitors A−C and design of new inhibitors D.
a
Scheme 1. Synthesis of R1- and R2-Modified Cyclopentane Derivatives 5 and 20−26
a
Reagents and conditions: (a) CALB L (lipase from Candida antarctica B, Novozymes), aqueous buffer (pH 6.2), 22 °C. (b) 3; morpholine, 5 and
20−23; aminoacetonitrile hydrochloride, EDCI, HOBt, THF, Et3N, 22 °C. (c) DAST, CH2Cl2, 5 to 22 °C. (d) LiOH, MeOH, THF, H2O, 22 °C.
(e) NADP, glucose dehydrogenase GDH 102 (Codexis), ketoreductase KRED-NADP-131(Codexis), aqueous buffer (pH 6.5), 22 °C. (f) MsCl,
Et3N, CH2Cl2, 5 °C. (g) 8; cyclopropylmethylthiol, 9; thiophenole, 10; 2-chlorothiophenol, 11; 2-trifluoromethylthiophenol, NaH, THF, 50 °C. (h)
m-Chloroperbenzoic acid, CH2Cl2, 22 °C. (i) 24; 2-Amino-2-methyl-propionitrile hydrochloride, 25; 1-amino-cyclopropanecarbonitrile
hydrochloride, 26; 1-aminocyclobutane-1-carbonitrile, i-butyl chloroformate, 4-methyl-morpholine, THF, −16 to 22 °C.
into the S2 pocket of CatS, which is significantly smaller in CatK
(vide infra).
CHEMISTRY
■
The syntheses of cyclopentane derivatives 5 and 20−2611 having
modifications at R1 and R2 start with the racemic trans ester 1
(Scheme 1), which is readily prepared from commercial 3,4-
dicarboxy-hexanedioic acid on a multigram scale.15 An efficient
enzymatic hydrolysis of rac-1 gave access to chiral acid 2 as the
formed R,R-enantiomer (ee 95%), requiring the highly selective
lipase from Candida antarctica B. The screening of a large enzyme
library (>100 tested), substrate engineering (Me-, Et-, n-Bu-
diester), and a quite in-depth process optimization afforded this
highly enantioselective and robust enzymatic resolution of rac-1.
Chiral acid 2 was coupled with morpholine using EDCI and
HOBt to give amide 3, which was subjected to diethylamino-
The rigidity of a cyclic scaffold can be of advantage to reduce
the number of rotatable bonds, thereby making the inhibitors less
prone for metabolic break down and more potent if the
appropriate binding conformation is populated. We explored
five-membered rings D such as cyclopentanes a, pyrrolidines b,
and prolines c as homo- and heterocyclic scaffolds. Scaffolds a11
and c12 were already disclosed in patents three years ago, and
more recently Kim et al.13 and F. Hoffmann-La Roche14
described proline scaffold c as well with a keto-heterocycle
attached to it, which binds in the S1 pocket.
B
dx.doi.org/10.1021/jm401528k | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Scheme 2. Synthesis of R3-Modified Cyclopentane Derivatives 33−46
a
Reagents and conditions: (a) 1-amino-cyclopropanecarbonitrile hydrochloride, i-butyl chloroformate, 4-methyl-morpholine, THF, −16 to 22 °C.
(b) NaBH4, THF, −15 to 0 °C, inseparable mixture of 28 and its epimer (structure not shown) was obtained. (c) MsCl, Et3N, CH2Cl2, 5 °C,
inseparable mixture of 29 and its epimer (structure not shown) was obtained. (d) Thiophenole, NaH, THF, 50 °C. (e) m-Chloroperbenzoic acid,
CH2Cl2, 22 °C. (f) LiOH, MeOH, THF, H2O, 22 °C. (g) For amines, see Table 2, EDCI, HOBt, THF, Et3N, 22 °C.
a
Scheme 3. Synthesis of R2-Modified Pyrrolidines 56−58
a
Reagents and conditions: (a) morpholine, EDCI, HOBt, THF, 22 °C. (b) Trifluoroacetic acid, CH2Cl2, 22 °C. (c) 50; phenylsulfonyl chloride, 51;
2-chlorophenylsulfonyl chloride, 52; 2-trifluoromethylphenylsulfonyl chloride, (i-Pr)2EtN, THF, 22 °C. (d) LiOH, MeOH, THF, H2O, 22 °C. (e)
Aminoacetonitrile hydrochloride, EDCI, HOBt, Et3N, CH3CN or DMF, 22 °C.
sulfur trifluoride (DAST), furnishing difluorocylopentane
derivative 4. Hydrolysis of the ester group in 4 to the acid
(structure not shown) using LiOH followed by coupling with
aminoacetonitrile using EDCI/HOBt delivered desired target
compound 5. The reduction of ketone 3 to the required R,R,R-
configured alcohol 6 with NaBH4 was difficult to achieve,
yielding a 1:2 mixture of 6 and its undesired epimer (structure
not shown), which were inseparable by chromatography on
silica. It appears that the attack of the hydride occurred
preferentially from the sterically less hindered site (i.e., opposite
of the larger morpholino substituent in 3), furnishing the
undesired epimer preferentially.
A powerful method to change the course of selectivity was
found by an enzymatic reduction of 3 using ketoreductase
KRED-NADP-131 and the most widely applied cofactor-
recycling system based on D-glucose as final reductant and
glucose dehydrogenase.16,17 This ketoreductase was the single
highly selective screening hit out of >100 enzymes tested,
providing R,R,R-alcohol 6 with a high de of 95%. The subsequent
activation to mesylate 7 enabled the substitution by thioalcohols,
furnishing thioethers 8−11 with inversion of the configuration.
Oxidation of the thioethers to sulfones 12−15 was accomplished
with m-chloroperbenzoic acid, and the ester group was then
hydrolyzed to acids 16−19. Targeted amides 20−23 were
prepared using EDCI/HOBt as the coupling reagents, whereas
access to the sterically more demanding substituted amino-
acetonitriles 24−26 required the activation via a mixed anhydride
using i-butyl chloroformate. The relative and absolute config-
urations of 20−26 were assigned on the basis of an X-ray
structure of 21 bound to mouse CatS.
To evaluate variations in the S3 binding pocket, the sequence
of reactions was changed (Scheme 2). Acid R,R-2 was first
converted to amido-cyclopropanecarbonitrile 27 followed by
reduction of the keto group to alcohol 28 using NaBH4. The
selectivity was again driven by the sterically more demanding
amido-cyclopropanecarbonitrile residue, affording this time a 2:1
mixture in favor of desired alcohol 28 and its epimer (structure
not shown), which were again inseparable by chromatography.
Thus, the epimer-mixture was converted via mesylate 29 to
thioether 30, at which stage the separation of the epimers was
C
dx.doi.org/10.1021/jm401528k | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Scheme 4. Synthesis of R2- and R3-Modified Prolines 76−79
a
Reagents and conditions: (a) MsCl, Et3N, CH2Cl2, 5 °C. (b) 61; thiophenol, 62; 2-chlorothiophenol, 63; 2-trifluoromethylthiophenol, NaH, THF,
50 °C. (c) m-Chloroperbenzoic acid, CH2Cl2, 22 °C. (d) LiOH, MeOH, THF, H2O, 22 °C. (e) 1-Amino-cyclopropanecarbonitril hydrochloride, i-
butyl chloroformate, 4-methyl-morpholine, THF, −16 to 22 °C. (f) Trifluoroacetic acid/CH2Cl2 or HCl/1,4-dioxane, 22 °C. (g) 76 and 77; benzoic
acid, 78; cyclohexanecarboxylic acid, EDCI, HOBt, THF or DMF, Et3N, 22 °C. (h) 79; acetic anhydride, CH3CN, (i-Pr)2EtN, 22 °C.
possible. Oxidation of 30 to 31 and hydrolysis yielded targeted
acid 32, which served as the starting material for a series of
compounds, 33−46,11 having R3 modified (Table 2). To confirm
the relative and absolute configurations of 33−46, acid 32 was
reacted with morpholine to give compound 25, which was
identical, according to 1H NMR, with 25 prepared according to
Scheme 1.
enzymes, Table 1). Extending into the S2 pocket with, for
example, the cyclopropylmethyl sulfone residue in 20 resulted in
an approximately 4-fold higher enzymatic inhibition of CatS but,
more importantly, had a profound effect on CatK selectivity,
which improved >100-fold. A number of other alkylsulfones
explored were less active (results not shown). Gratifyingly,
phenylsulfone 21 boosted the IC50 value into the subnanomolar
The synthesis of pyrrolidines 56−58 (Scheme 3) starts from
known chiral acid S,S-47, which is readily accessible on a
multigram scale in three steps.18 Amide formation and
deprotection of the BOC group yielded pyrrolidine 49, which
was converted to sulfonamides 50−52. Hydrolysis of the ethyl
ester and amide formation completed the synthesis of 56−58 in
the same manner as outlined in Scheme 1.
Table 1. SAR of R2 Modifications on Cyclopentane
Derivatives 5 and 20−23 and Pyrrolidines 56−58
Finally, prolines 76−7912 (Scheme 4) were made available
starting with commercially available homochiral hydroxy-proline
derivative S,S-59. Formation of mesylate 60 followed by
substitution with thiophenols and oxidation yielded sulfones
64−66 in an analogous manner to that previously discussed.
Hydrolysis of the methyl ester in 64−66 followed by amide
formation and deprotection of the BOC group furnished proline
derivatives 73−75, which served as the key intermediates for the
introduction of R3 residues, affording final prolines 76−79. The
relative and absolute configurations of 76−79 were assigned on
the basis of an X-ray structure of 75 bound to mCatS (structure
not shown, PDB accession number 4BS5).
a
a
IC50 (μM)
IC50 (μM)
hCatS
enzyme
hCatK
enzyme
selectivity
A20 cell-
b
compd
R2
hCatK/hCatS
based
5
0.830
0.230
0.0007
0.0005
0.0009
0.271
0.012
0.018
0.855
>25
15.5
>25
>25
>25
>25
>25
1
0.925
4.506
0.023
0.003
0.002
4.239
0.208
0.118
20
21
22
23
56
57
58
CH2c-Pr
Ph
>100
>20 000
>20 000
>20 000
>90
2-Cl-Ph
2-CF3-Ph
Ph
RESULTS AND DISCUSSION
■
2-Cl-Ph
2-CF3-Ph
>1000
>1000
We started the SAR exploration with the cyclopentane scaffold
that was devoid of an R2 substituent binding in the S2 pocket of
CatS. Because the S1 and S3 pockets are largely similar in CatS
and CatK, compound 5 turned out to be an unselective dual
CatS/K inhibitor with modest activities (IC50 = 0.85 μM at both
a
IC50 values are the mean of at least two independent experiments.
The errors of the IC50 values are within a factor of 1.6. A20/DO10.11
b
antigen-presentation assay.
D
dx.doi.org/10.1021/jm401528k | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
range (IC50 = 0.0007 μM) with the desired high selectivity of
the R1 substituents adjacent to the nitrile moiety should be
explored (vide infra).
>20 000-fold versus CatK and showed a high activity of IC50
=
0.023 μM in the cellular mouse A20/DO10.11 assay, an assay of
ovalbumin-specific antigen presentation. A variety of mono- and
disubstituted phenyl residues were explored as well, with all
showing uniformly low IC50 values in the enzymatic assay, the
most notable of which were the 2-Cl- and 2-CF3-substituted
phenylsulfones 22 and 23, providing an additional 8- and 12-fold
increase in cellular activity, respectively, compared to unsub-
stituted 21.
The rather hydrophobic S2 pocket is accessible via a small
channel ‘gated’ by the adjacent glycines 288 and 260. In CatK,22
these glycines are replaced by more bulky alanines, thus blocking
access to the S2 pocket. The sulfone linker in 21 is favorably
located in the channel within close van der Waals (vdW)
distances to the glycines 288 and 260 (4.4 and 3.5 Å,
respectively), positioning the terminal phenyl residue of 21 in
a double orthogonal fashion to tyrosines 334 and 193. As a result,
a network of four favorable edge-to-face interactions of the three
aromatic units (3.1−3.8 Å) contributes to the high activity of
IC50 = 0.0007 μM. The rather shallow and hydrophilic S3 pocket
should tolerate many residues and was therefore explored
extensively.
Pyrrolidines 56−58 with optimized R2 phenylsulfone residues
were investigated as well but turned out to be significantly less
active in the enzyme (20−390-fold) as well as cell-based assay
(60−180-fold) compared to cyclopentane counterparts 21−23.
Interestingly, cellular activity increased again from 56−58, as
seen in the cyclopentane series (Table 1).
A possible explanation for the reduced enzymatic binding
affinity of pyrrolidine 56 versus cyclopentane scaffold 21 (IC50 =
0.271 and 0.0007 μM, respectively) can be derived from their X-
ray structures bound to mCatS. Both compounds bind in the
same conformation (Figure 3a) in the active site of mCatS.
X-ray Structures of Cathepsin S Inhibitors. Human19 and
mouse CatS have almost identical amino acid sequences and
topologies (Figure 2) within the active sites in which our
Figure 3. (a) X-ray structures of 21 (green, 1.96 Å, PDB code 4BSQ)
and 56 (orange, 1.47 Å, PDB code 4MZO) in the bound conformation
with mCatS (not shown). (b) Cyclopentane-sulfone scaffold 21 adopts a
low-energy conformation without significant deviation of the dihedral
angles from 60°. (c) Pyrrolidine-sulfonamide 56, however, was found in
a high-energy conformation with dihedral angles deviating significantly
from 60°, resulting in a 390-fold decreased binding affinity compared to
unstrained 21.
Figure 2. X-ray structure of 21 (green) bound to mCatS (light blue) at
1.96 Å resolution (PDB code 4BSQ). The corresponding amino acids of
hCatS are given in parentheses. The nitrile reacts with the catalytic
cysteine 147 to form a thioimidate. Glycines 288 and 260 orient the
phenyl residue of the inhibitor into the S2 pocket in which favorable
edge-to-face interactions with tyrosines 334 and 193 are possible.
Figures 2−4 were prepared with the PyMol program.21
Cyclopentane-sulfone 21 is found in one of the two low-energy
conformations without significant deviation of the dihedral
angles from 60° (Figure 3b). In contrast to 21, pyrrolidine-
sulfonamide 56 adopts a strained conformation in which the
nitrogen lone pair is located in proximity to one of the SO
groups (Figure 3c). Sulfonamides are known to have conforma-
tional preferences in which the nitrogen lone pair bisects the O
SO angle in the Newman projection.23 It has been suggested
that this rotamer is stabilized by the lone pair interaction with the
sulfur d orbital.24 Pyrrolidine-sulfonamide 56, however, cannot
be accommodated in this low-energy conformation in the active
site of CatS but was found in a strained conformation with
dihedral angles deviating significantly from 60°. This energy
consideration is supported by the torsion angles of alkyl-sulfones
versus alkyl-sulfonamides observed in the Cambridge Structural
Database (CSD).25,26 The torsion of cyclopentane-sulfone 21 is
close to the median of 55°; pyrrolidine-sulfonamide 56 torsion is
only found in rare cases, the majority of alkyl-sulfonamides show
a torsion of 180° (see Figure 6, Supporting Information). As a
inhibitors bind. Consequently, IC50’s are similar for both
enzymes (vide infra). Because mouse was the in vivo species,
both mouse and human CatS were cloned, expressed, purified,
and crystallized. In initial parallel trials, it was found that mCatS
often crystallized more readily than hCatS, so hCatS was only
used for crystallization when mCatS failed.
A first glance, Figure 2 shows the nice fit of a five-membered
cyclic scaffold with the three attached residues pointing into the
trigonal cavity (S1−S3) of mCatS, as predicted by molecular
modeling. Catalytic cysteine 147 (hCatS = C25) forms the
central anchoring motif in the S1 binding pocket, which can
interact with a variety of known electrophiles.20 A prominent
electrophile is the nitrile group, which undergoes a reversible
covalent binding in 21 with a short distance of 1.8 Å, forming a
thioimidate anion that is stabilized by a hydrogen bond to Q141.
There is space within the cavity of the S1 pocket, suggesting that
E
dx.doi.org/10.1021/jm401528k | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
consequence, this strain energy is only partially compensated by
the favorable edge-to-face interaction of the aromatic units,
which resulted in reduced enzymatic and cell-based activities by
more than 2 orders of magnitude.
the assay limitations: compound concentrations lower than 0.001
μM are below the enzyme concentration used in the assay,
leading to nonlinear Michaelis−Menten kinetics at low
concentrations.
Another remarkable effect is the approximately 20-fold
increased activity in the enzyme and cell-based assay of
pyrrolidine 57 substituted with a 2-Cl-phenyl residue compared
to unsubstituted phenyl derivative 56 (cf. Table 1). The X-ray
structure of 57 complexed with mCatS (Figure 4) shows the
SAR Exploration of Cylopentane and Proline Scaffolds.
Cyclopentane derivative 21 served as a basis for a limited number
of symmetrical R1,1′ variations in the S1 pocket. From three of the
compounds evaluated (24−26, Table 2), cyclopropyl derivative
25 showed a 2-fold increased cell-based activity compared to 21
and was therefore selected for R3 modifications. Ring-size
variations of various heterocycles is well-tolerated (Table 2); six-
membered derivatives such as dimethyl-morpholine 33 and
piperidines 36 and 37 even showed single-digit nanomolar
activity in the cell-based assay. Interestingly, even small dimethyl-
amide 46 retained substantial activity in the enzyme as well as
cell-based assay (IC50 = 0.006 and 0.091 μM, respectively).
Terminal polar groups such as a NH or OH (in 39, 40, and 45) or
a sulfone (in 41) impaired the cell-based activity remarkably
(IC50 = 0.37 − 1.34 μM). This is presumably due to their reduced
29
permeability (e.g., Peff = 0.2/0.2 × 10−6 cm/s for 40/41
compared to Peff = 4.6 × 10−6 cm/s for 37, as determined in the
parallel artificial-membrane-permeability assay (PAMPA)).
There seems to be ample scope for R3 modifications in the S3
binding pocket, as was expected from its shallow nature (vide
supra); none of them compromised CatK selectivity (33−46,
IC50 > 25 μM).
Figure 4. X-ray structure of 2-Cl-phenyl pyrrolidine 57 (orange, 1.85 Å
resolution, PDB code 4MZS) bound to mCatS (light blue) in the S2
pocket. The corresponding amino acids of hCatS are given in
parentheses. The chlorine atom undergoes a favorable halogen bonding
with the methionine sulfur atom (M194) with a short distance of 3.7 Å
and a close to ideal S···Cl−C angle of 170°, thereby enhancing binding
affinity. In addition, the chlorine atom is in a short distance to the
terminal methyl group of M194 (distance epsilon C···Cl, 3.4 Å),
indicating a favorable vdW interaction.
Keeping the conformation of the sulfone moiety and reducing
the chemical complexity of the cyclopentane series, we next
explored the proline scaffold. Five examples were evaluated (75−
79, Table 3) that incorporated the optimal cyclopropyl residue
R1,1′, 2-Cl- and 2-CF3-phenyl sulfones as R2, and a few simple R3
residues (Ph, cyclohexyl, Me, and H). Gratifyingly, most of the
compounds showed high activity in the enzyme and cell-based
assays, with cyclohexyl derivative 78 being the most potent.
Again, o-Cl-phenylsulfone 77 was approximately 15-fold more
potent in the enzyme and cell-based assays compared to
unsubstituted phenylsulfone 76. In preparation for in vivo
studies in mice, selected examples were also tested against mouse
CatS. As expected, cyclopentane 21 and prolines 75−79 showed
similar activities in h- and mCatS (Table 3) on the basis of their
almost identical amino acid sequences and topologies in the
active site (cf. Figure 2). Furthermore, the compounds were, in
general, highly selective over CatK, L, and B (cf. Tables 2 and 3).
In Vivo Pharmacokinetics (PK) and Pharmacodynamics
(PD) of CatS Inhibitors. With the aim of studying being the
pharmacokinetics and pharmacodynamics of the CatS inhibitors,
the compounds shown in Table 3 were tested in vitro for
enzymatic proteolytic degradation in human and mouse plasma.
Compounds 77 and 78 were unstable in mouse (>90% degraded
after 4 h), whereas 21, 75, and 79 turned out to be sufficiently
stable in mouse (99, 51, and 72% remaining after 4 h,
respectively) and were fully stable in human plasma 5 h after
incubation, rendering them suitable for in vivo PK character-
ization. When tested in mouse PK studies, cyclopentane
derivative 21 exhibited a medium systemic clearance and a
moderate oral bioavailability, resulting in modest exposure levels
upon oral administration (Table 4). In contrast, prolines 75 and
79 showed more suitable PK properties on the basis of their
better bioavailability and higher AUC after po administration and
were thus further profiled for in vivo pharmacology studies.
Transgenic DO10.11 mice (expressing an ovalbumin-specific
T-cell receptor) were challenged to ovalbumin once. The
compound was given 30 min before ovalbumin challenge by oral
same edge-to-face interaction of Y193 with the 2-Cl-phenyl
residue (3.6 and 3.9 Å) as seen in 21 (cf. Figure 2). Contrary to
21, flexible Y334 is no longer undergoing an edge-to-face
interaction, instead it undergoes a face-to-face interaction with
the more bulky 2-Cl-phenyl residue (3.5 and 3.9 Å) as a
consequence of the ortho substituent now occupying the original
position of Y334 seen in 21. Boeckler et al. have predicted the
existence of a favorable halogen bonding between the sulfur atom
of a methionine residue and a chlorine atom on the basis of
quantum mechanical calculations.27 Indeed, there is a remarkably
short distance between the chlorine and the sulfur atom of
methionine 194 of 3.7 Å (i.e., close to the sum of the vdW radii of
the two interacting atoms, 3.6 Å). In addition, the S···Cl−C angle
was found to be 170°, almost representing the ideal angle of 180°
required for optimal interactions of a polarizable atom with the
sigma hole of the Cl atom as found in halogen bonding.27,28 The
chlorine atom is also making a favorable vdW interaction with the
hydrogen atoms of the C-epsilon atom of M194, with the Cl···C-
epsilon distance of 3.4 Å being even shorter than the Cl···S
distance. 2-CF3-phenyl-substituted pyrrolidine 58 shows a
similar effect (i.e., a 15- and 36-fold increase in enzyme and
cell-based activity, respectively, compared to 56). In summary,
shape complementarity, favorable electrostatics, and the move-
ment of Y334 yielding a face-to-face interaction account for this
improved potency over the nonsubstituted phenyl derivatives. In
the cyclopentane series, 21−23, this effect was not evident in the
enzyme (the IC50 = 0.0005−0.0009 μM are virtually identical),
but it was clearly visible in the cell-based assay (8- and 12-fold
increased activity of 22 and 23, respectively, compared to 21).
The absence of the effect in the enzyme assay can be explained by
F
dx.doi.org/10.1021/jm401528k | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 2. SAR of R1 and R3 Modifications on Cylopentane Derivatives 21, 24−26, and 33−46
a
IC50 (μM)
b
compd
R1
R1′
hCatS enzyme
hCatK enzyme
hCatL enzyme
hCatB enzyme
A20 cell-based
21
24
25
26
33
34
35
36
37
38
39
40
41
42
43
44
45
46
H
H
0.0007
0.003
0.0007
0.004
0.0004
0.001
0.003
0.0008
0.0005
0.002
0.011
0.001
0.004
0.026
0.002
0.010
0.007
0.006
15.5
>25
29.7
>25
>25
>25
>25
>25
>25
>25
>25
>25
>25
>25
>25
>25
>25
>25
11.7
>25
12.2
>25
2.7
6.2
0.023
0.043
0.012
0.236
0.001
0.016
0.248
0.005
0.002
0.017
1.340
0.371
0.816
0.391
0.029
0.044
0.572
0.091
Me
Me
>25
8.5
-CH2CH2-
-CH2CH2CH2-
>25
3.2
4.1
13.1
29.7
3.6
22.2
4.2
1.9
0.7
4.7
6.4
2.6
13.3
5.7
4.6
15.8
>25
17.0
33.6
16.2
24.9
38.0
>25
7.1
39.4
63.3
22.9
a
b
IC50 values are the mean of at least two independent experiments. The errors of the IC50 values are within a factor of 1.6. A20/DO10.11 antigen-
presentation assay.
Table 3. hCatS and mCatS Activity of Cyclopentane 21 and Proline Derivatives 75−79 Are Very Similar
a
IC50 (μM)
b
compd
R2
R3
hCatS enzyme
mCatS enzyme
hCatK enzyme
hCatL enzyme
hCatB enzyme
A20 cell-based
21
75
76
77
78
79
0.0007
0.002
0.001
0.006
0.029
0.001
0.0002
0.002
15.5
>25
>25
>25
>25
>25
11.7
>25
>25
6.8
6.2
0.023
0.058
0.217
0.015
0.002
0.019
CF3
H
H
>25
>25
35.2
14.9
Ph−C(O)
Ph−C(O)
Cyclohexyl-C(O)
0.013
Cl
0.0008
0.0003
0.001
Cl
8.3
Cl
Me-C(O)
29.7
47.6
a
b
IC50 values are the mean of at least two independent experiments. The errors of the IC50 values are within a factor of 1.6. A20/DO10.11 antigen-
presentation assay.
administration, and circulating IL-2 levels were determined 4 h
postdosing as a marker of T-cell activation. Proline derivative 75
did not reduce ovalbumin-mediated IL-2 production at a dose as
high as 50 mg/kg despite the high plasma levels of 2053 ng/mL
achieved. The lack of activity was hypothesized to be related to
the very low volume of distribution of the compound (Vdss =
0.13 L/kg), in the range of plasma to extracellular volume. In
contrast, acetamide 79 (Vdss of 0.50 L/kg) showed a dose-
G
dx.doi.org/10.1021/jm401528k | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Table 4. PK Properties of 21, 75, and 79 in Mice
AUC/dose (1 mg) (ng h/mL)
t1/2 (h)
b
c
b
c
compd
iv
po
77
iv
po
Vdss (L/kg)
Cl (mL/min/kg)
F (%)
21
75
79
427
0.3
1.1
0.3
0.9
0.9
0.5
0.50
0.13
0.50
39
1.6
29
18
69
53
10 400
581
7200
309
a
b
c
All values given are the mean of two experiments. For iv, compound (2 mg) was dissolved in aqueous NaCl/N-methyl-2-pyridone. For po,
compound (9 mg) was used as a microsuspension in gelatin/aqueous NaCl.
dependent reduction in IL-2 (Figure 5), with an ED50 of 5 mg/
kg. Free exposure levels at ED50 were estimated to be in the range
of the IC50 determined in the cell-based assay.
interactions of chlorine with the hydrogens of the C-epsilon of
M194 and the movement of Y334 that yielded a face-to-face
interaction boosted the potency in 22 and 57 compared to their
des-chloro counterparts 21 and 56. Proline derivative 79 turned
out to be active in vivo, reducing IL-2 in a transgenic DO10.11
mouse model sensitized to ovalbumin with an ED50 of 5 mg/kg.
Liabilities of the proline series could be overcome in lead
optimization, delivering a clinical candidate that will be discussed
in a future paper.
EXPERIMENTAL SECTION
■
General. All solvents and reagents were obtained from commercial
sources and were used as received. All reactions were followed by TLC
(TLC plates F254, Merck) or LCMS (liquid chromatography−mass
spectrometry) analysis. Silica gel chromatography was either performed
using cartridges packed with silica gel (ISOLUTE Columns, TELOS
Flash Columns) on ISCO Combi Flash Companion or on glass columns
on silica gel 60 (32-60 mesh, 60 Å). 1H NMR spectra were obtained on
Bruker Avance 300 or 400 MHz spectrometer with chemical shifts (δ in
ppm) reported relative to tetramethylsilane or the residual solvent peak
as the internal reference (i.e., CDCl3 = 7.26 ppm, DMSO-d6 = 2.50
Figure 5. Dose-dependent reduction of IL-2 in DO10.11 mice by
proline-derivative 79 4 h after oral postdose (n = 8; mean reduction as
compared to ovalbumin challenge). All compound treatments except
the 0.7 mg/kg group are significantly different than the ovalbumin
challenge group (p < 0.05, ANOVA followed by multiple Dunnett test).
1
ppm). H resonances are reported to the nearest 0.01 ppm. NMR
abbreviations are as follows: s, singlet; d, doublet; t, triplet; q,
quadruplet; quint, quintuplet; sext, sextuplet; m, multiplet; br,
broadened. Coupling constants (J) are reported to the nearest 0.1 Hz.
Purity was analyzed by reverse-phase HPLC performed on Finnigan
LTQ (Thermo Fisher Scientific) and Agilent RRLC 1200 equipment.
Column: Agilent XDB C15, 30 mm × 4.6 mm, 3.5 μm. Analytical
conditions: gradient used, 5 to 95% acetonitrile in water containing 0.1%
trifluoroacetic acid in 3 min. Flow: 4.5 mL/min. UV detector: DAD
190−400 nm. Sample solvent: in water/acetonitrile (8:2). The UV
detection was an averaged signal from wavelengths of 190−400 nm.
Mass spectra were recorded on an SSQ 7000 (Finnigan-MAT)
spectrometer for electron impact ionization. LC-HRMS spectra were
recorded with an Agilent LC system consisting of an Agilent 1290 high-
pressure system, a CTC PAL auto sampler, and an Agilent 6520 QTOF.
The separation was achieved on a Zorbax Eclipse Plus C18 1,7 μm
2.1*30 mm column at 55 °C; eluent A = 0.01% formic acid in water, B =
0.01% formic acid in acetonitrile/2-propanol (8:2); flow, 1 mL/min.
Gradient: 0 min 5% of B, 0.3 min 5% of B, 4.5 min 99% of B, 5 min 99%
of B. The injection volume was 2 μL. All assay compounds had a purity
of >95% according to HPLC. LC−MS (ESI, positive or negative ion)
data were recorded on Waters UPLC-MS systems equipped with Waters
Acquity, a CTC PAL auto sampler, and a Waters SQD single quadrupole
mass spectrometer using ES ionization modes (positive and/or
negative). The separation was achieved on a Zorbax Eclipse Plus C18
1,7 μm 2.1*30 mm column at 50 °C; A = 0.01% formic acid in water, B =
acetonitrile at flow 1; gradient: 0 min 3% B, 0.2 min 3% B, 2 min 97% B,
1.7 min 97% B, 2.0 min 97% B. The injection volume was 2 μL. MS (ESI,
positive or negative ion): FIA (flow injection analysis)-MS were
recorded on a AppliedBiosystem API150 mass spectrometer. Sample
introduction was made with a CTC PAL auto sampler and a Shimadzu
LC-10ADVP pump. The samples were directly flushed to the ESI source
of the mass spectrometer with a flow of 50 μL/min of a mixture of
acetonitrile and 10 mM ammonium acetate (1:1) without a column. The
injection volume was 2 μL.
In the light of the observed in vivo PD effect demonstrated, the
prototype proline derivative 79 certainly looks promising as a
starting point but needs further optimization. Properties to be
improved include plasma stability, metabolic stability, the short
half-life, and the low volume of distribution. Optimization of the
proline series leading eventually to a clinical candidate will be the
subject of a future communication.
SUMMARY AND CONCLUSIONS
■
Structure-based design of weakly active dual CatS/K inhibitor 5
(IC50 = 0.830/0.855 μM) lead to potent and selective
cyclopentane derivative 22 (IC50 = 0.0005/>25 μM, CatS/K)
by utilizing specific interactions in the S2 pocket of CatS. The S3
pocket turned out to be tolerable for R3 modifications, allowing
the tuning of physicochemical properties. Scaffold hopping from
cyclopentane 22 to analogous pyrrolidine derivative 57 reduced
the cellular activity significantly (IC50 = 0.003 and 0.208 uM,
respectively) but was regained with proline derivative 78 (IC50 =
0.002 uM). The X-ray structure of three compounds (21, 56, and
57) bound to mCatS revealed specific favorable and unfavorable
interactions: (1) the high potency of cyclopentane derivative 21
can be explained by a favorable network of edge-to-face
interactions of three aromatic units in the S2 pocket, (2)
conformational strain was significantly reduced in the more
potent cyclopentane-sulfone scaffold 21 compared to pyrroli-
dine-sulfonamide 56, and (3) sigma-hole interactions between a
chlorine and the sulfur atom of M194 in combination with vdW
H
dx.doi.org/10.1021/jm401528k | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
General Procedure A for the Amide Formation from an Acid
and an Amine using EDCI/HOBt. To a solution of the acid (10.0
mmol) in tetrahydrofuran (25 mL) were subsequently added the amine
(15 mmol), triethylamine (70 mmol), 1-hydroxybenzotriazole hydrate
(20 mmol), and 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hy-
drochloride (EDCI, 20 mmol), and stirring was continued at 22 °C
overnight. The mixture was evaporated, the residue partitioned between
ethyl acetate and 1N aqueous HCl, and the organic layer was washed
with saturated aqueous Na2CO3, dried, and evaporated to give the crude
amide.
General Procedure B for the Amide Formation from an Acid
and an Amine using i-Butyl Chloroformate. A solution of the acid
(0.12 mmol) and 4-methyl-morpholine (0.3 mmol) in tetrahydrofuran
(2 mL) was cooled to −16 °C and treated with i-butyl chloroformate (20
mg, 0.14 mmol), and stirring was continued for 30 min. The amine (17
mg, 0.14 mmol) was added, the mixture was allowed to warm to 22 °C,
and stirring was continued overnight. The mixture was partitioned
between aqueous saturated NaHCO3, the organic layer was washed with
hydrochloric acid (1 N), dried, and evaporated, and the residue was
purified by preparative HPLC on a RP-18 column using a gradient of a
mixture of acetonitrile and water to give the amide.
tetrahydrate (12.44 g, 50 mM). After pH adjustment to 6.2 using a
NaOH solution, CALB L (1.94 mL, a liquid formulation of lipase from
Candida antarctica B, Novozymes, Denmark) was added at 22 °C, and
the was pH kept constant at 6.2 under vigorous stirring by automated
addition (pH-stat) of a 1.0 M NaOH solution. After consumption of
118.3 mL (ca. 45% conversion after 42 h reaction time; enantiomeric
excess of formed acid still >95%), dichloromethane (1.0 L) was added,
and the aqueous phase was washed with dichloromethane (3 × 1.5 L),
the pH was set to 3.0 using phosphoric acid (85%), and the aqueous
phase was extracted with ethyl acetate (4 × 1.0 L). The combined ethyl
acetate phases were dried over sodium sulfate and filtered, and the
filtrate was evaporated and dried under high vacuum overnight to give
title compound 2 (24.40 g, 46%) as white crystals. Chemical purity:
99.8% (GC, analyzed as the methylester; same conditions as the ee
determination). Optical purity: 95.2% ee (GC, analyzed as the
methylester on a BGB-176 column, 30 m × 0.25 mm; H2; 100 to 200
°C with 2 °C/min). [α]589 (20 °C) −101.2° (c 1.0; EtOH). 1H NMR
(400 MHz, CDCl3) δ 4.14−4.30 (m, 2H), 3.33−3.50 (m, 2H), 2.65−
2.78 (m, 2H), 2.48−2.61 (m, 2H), 1.29 (t, J = 7.1 Hz, 3H). MS (ESI,
negative ion): 199.1 [M − H]−.
(1R,2R)-2-(Morpholine-4-carbonyl)-4-oxo-cyclopentanecarbox-
ylic Acid Ethyl Ester (3).11 The amide formation of (1R,2R)-4-oxo-
cyclopentane-1,2-dicarboxylic acid monoethyl ester (2) (5.00 g, 25
mmol) and morpholine according to general procedure A gave crude
title compound 3 (5.95 g, 88%) as a pale-yellow oil, which was used in
the next step without further purification. MS: 270.2 (M + H)+.
(1R,2R,4R)-4-Hydroxy-2-(morpholine-4-carbonyl)-cyclopentane-
carboxylic Acid Ethyl Ester (6).11 A mixture of (1R,2R)-2-(morpholine-
4-carbonyl)-4-oxo-cyclopentanecarboxylic acid ethyl ester (3) (7.77 g,
28.85 mmol) in 148 mL of aqueous buffer (289 mg (10 mM) 2-(N-
morpholino)ethanesulfonic acid, 15.5 g (0.5 M) D-glucose monohydrate
[2.7 equ.], and 65 mg (2 mM) magnesium chloride hexahydrate) was
adjusted to pH 6.5. Under stirring, the reduction was started at 22 °C by
the addition of the cofactor NADP (779 mg [0.03 equ.]), the cofactor
regeneration enzyme glucose dehydrogenase (77 mg GDH 102
[Codexis]), and the ketoreductase (387 mg KRED-NADP-131
[Codexis]). During the 21h reaction, the pH was maintained at 6.5 by
the addition of 31.9 mL of 1 M NaOH (pH-stat). Under stirring, the
mixture was adjusted to pH 2.8, saturated with 43.6 g of sodium chloride,
and stirred for at least 10 min after the addition of 17 g of filter aid
(Dicalite) and 300 mL of ethyl acetate. Subsequently, the filter aid was
removed, and the filtrate was adjusted to pH 7.0 prior to extraction. After
phase separation, the aqueous phase was extracted three times with 300
mL of ethyl acetate. The combined organic phases were dried with
sodium sulfate and filtered, the filtrate was evaporated, and the residue
was dried overnight under high vacuum to give title compound 6 (7.1 g,
91%) as a pale-yellow oil. Chiral GC: ee 95.2% [BGB-176, 30m; H2; 2
°C/min, 150 to 220 °C]. 1H NMR (400 MHz, DMSO-d6) δ 4.72 (d, J =
3.9 Hz, 1H), 4.10−4.18 (m, 1H), 4.04 (q, J = 7.1 Hz, 2H), 3.41−3.61 (m,
10H), 3.17 (td, J = 8.1, 9.9 Hz, 1H), 2.19 (ddd, J = 5.4, 9.9, 13.1 Hz, 1H),
1.78−1.88 (m, 1H), 1.57−1.74 (m, 2H), 1.16 (t, J = 7.1 Hz, 3H). MS:
272.1 [M + H]+.
General Procedure C for the Hydrolysis of an Ester to the
Carboxylic Acid. To a solution of the ester (1.35 mmol) in THF (10
mL) were added a solution of LiOH (3.0 mmol) in water (3 mL) and
methanol (3 mL), and stirring was continued at 22 °C until the
conversion was completed. The mixture was evaporated, and the residue
was partitioned between ethyl acetate and hydrochloric acid (0.1 N).
The organic layer was dried and evaporated to give the crude acid.
General Procedure D for the Synthesis of a Mesylates from an
Alcohol. To a solution of the alcohol (10.0 mmol) in dichloromethane
(110 mL) were added at 5 °C triethylamine (40.0 mmol) and
methanesulfonyl chloride (40.0 mmol), and stirring was continued at 5
°C until the conversion was completed. The mixture was partitioned
between aqueous HCl (1 N) and dichloromethane, and the organic layer
was washed with aqueous Na2CO3 solution (half saturated), dried, and
evaporated to give the crude mesylate.
General Procedure E for the Synthesis of a Thioether from a
Mesylate and a Thiol. To a solution of the thiol (0.75 mmol) in THF
(4 mL) was added NaH (55% in oil, 0.75 mmol) at 22 °C, and the
mixture was stirred until gas evolution ceased. A solution of the mesylate
(175 mg, 0.50 mmol) in THF (1 mL) was added, and the mixture was
heated at 50 °C until the conversion was completed. The mixture was
partitioned between aqueous Na2CO3 and EtOAc, and the organic layer
was dried and evaporated to give the crude thioether.
General Procedure F for the Oxidation of a Thioether to the
Sulfone. To a solution of the thioether (0.25 mmol) in dichloro-
methane (2.5 mL) was added a solution of m-chloroperbenzoic acid
(70%, 1 mmol) in dichloromethane (2.5 mL), and stirring was
continued at 22 °C until the conversion was completed. The mixture
was vigorously shaken with aqueous NaHSO3 (20%), the organic layer
was washed with saturated aqueous Na2CO3 and water, and the organic
layer was dried and evaporated to give the crude sulfone.
General Procedure G for the Formation of a Sulfonamide
from an Amine. To a solutiuon of the amine (12 mmol) and
diisopropylethylamine (36 mmol) in acetonitrile (30 mL) was added the
sulfonyl chloride (14 mmol) at 22 °C, and stirring was continued at 22
°C overnight. The mixture was partitioned between an aqueous Na2CO3
solution and EtOAc, and the organic layer was washed with hydrochloric
acid (0.1 N), dried, and evaporated to give the crude sulfonamide.
General Procedure H for the Deprotection of the Boc Group
to the Amine. A solution of the Boc-protected amine (2.5 mmol) in
dichloromethane (15 mL) was treated with trifluoracetic acid (25
mmol), and stirring was continued at 22 °C until the conversion was
completed. The solution was evaporated to give the crude amine as the
salt with trifluoro-acetic acid, which was processed in the next step
without further purification.
(1R,2R,4R)-4-Methanesulfonyloxy-2-(morpholine-4-carbonyl)-cy-
clopentanecarboxylic Acid Ethyl Ester (7).11 (1R,2R,4R)-4-Hydroxy-2-
(morpholine-4-carbonyl)-cyclopentanecarboxylic acid ethyl ester (6)
(2.20 g, 10.0 mmol) was converted according to procedure D to the
crude mesylate, which was purified by flash chromatography using
EtOAc to give title compound 7 (2.60 g, 91%) as a pale-yellow oil. 1H
NMR (300 MHz, CDCl3) δ 5.24 (tt, J = 2.5, 5.2 Hz, 1H), 4.16 (q, J = 7.1
Hz, 2H), 3.51−3.75 (m, 9H), 3.34−3.46 (m, 1H), 3.01 (s, 3H), 2.45−
2.59 (m, 1H), 2.25−2.41 (m, 2H), 2.08−2.22 (m, 1H), 1.27 (t, J = 7.1
Hz, 3H). MS: 350.1 [M + H]+.
(1R,2R,4S)-2-(Morpholine-4-carbonyl)-4-phenylsulfanyl-cyclo-
pentanecarboxylic Acid Ethyl Ester (9).11 (1R,2R,4R)-4-Methanesul-
fonyloxy-2-(morpholine-4-carbonyl)-cyclopentanecarboxylic acid ethyl
ester (7) (0.91 g, 2.6 mmol) and thiophenol were converted according
to general procedure E to the crude thioether, which was purified by
flash chromatography using n-hepane/EtOAc (3:2) to give title
(1R,2R)-4-Oxo-cyclopentane-1,2-dicarboxylic Acid Monoethyl
Ester (2).11 Racemic trans-4-oxo-cyclopentane-1,2-dicarboxylic acid
diethyl ester (1)15 (60.44 g, 264.8 mmol) was emulsified under vigorous
stirring in MES buffer (1.16 L) containing 2-(N-morpholino)-
ethanesulfonic acid (1.13 g, 5 mM) and magnesium diacetate
1
compound 9 (0.47 g, 50%) as a pale-yellow oil. H NMR (300 MHz,
CDCl3) δ 7.19−7.41 (m, 5H), 4.12 (q, J = 7.1 Hz, 2H), 3.26−3.74 (m,
I
dx.doi.org/10.1021/jm401528k | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
11H), 2.23−2.47 (m, 2H), 2.09−2.22 (m, 1H), 1.82 (td, J = 9.7, 13.0 Hz,
1H), 1.24 (t, J = 7.1 Hz, 3H). MS: 364.5 [M + H]+.
pentanecarboxylic acid ethyl ester (15) (324 mg, 0.70 mmol) was
converted according to general procedure C to crude acid 19 (286 mg,
94%), which was processed without further purification. MS (ESI,
negative ion): 434.4 [M − H]−.
(1R,2R,4R)-4-Benzenesulfonyl-2-(morpholine-4-carbonyl)-cyclo-
pentanecarboxylic Acid Cyanomethyl-amide (21).11 (1R,2R,4S)-4-
Benzenesulfonyl-2-(morpholine-4-carbonyl)-cyclopentanecarboxylic
acid (17) (369 mg, 0.91 mmol) dissolved in DMF (8 mL) was reacted
according to general procedure A with aminoacetonitrile hydrochloride
to give the crude amide, which was purified by flash chromatography
using EtOAc to give title compound 21 (156 mg, 42%) as a white solid.
1H NMR (300 MHz, CDCl3) δ 7.85−7.96 (m, 2H), 7.64−7.73 (m, 1H),
(1R,2R,4S)-4-(2-Chloro-phenylsulfanyl)-2-(morpholine-4-carbon-
yl)-cyclopentanecarboxylic Acid Ethyl Ester (10).11 (1R,2R,4R)-4-
Methanesulfonyloxy-2-(morpholine-4-carbonyl)-cyclopentanecarbox-
ylic acid ethyl ester (7) (175 mg, 0.5 mmol) and 2-chlorothiophenol
were converted according to general procedure E to the crude thioether,
which was purified by flash chromatography using n-hepane/EtOAc
1
(3:2) to give title compound 10 (113 mg, 57%) as a colorless oil. H
NMR (300 MHz, CDCl3) δ 7.34−7.42 (m, 2H), 7.12−7.25 (m, 2H),
4.13 (q, J = 7.1 Hz, 2H), 3.32−3.79 (m, 11H), 2.29−2.50 (m, 2H),
2.09−2.23 (m, 1H), 1.87 (td, J = 9.6, 13.1 Hz, 1H), 1.25 (t, J = 7.2 Hz,
3H). MS: 398.1 [M + H]+.
7.54−7.63 (m, 2H), 7.07 (t, J = 6.0 Hz, 1H), 4.17 (dd, J = 6.0, 16.0 Hz,
1H), 4.03 (dd, J = 6.0, 16.0 Hz, 1H), 3.34−3.81 (m, 10H), 2.98−3.21
(m, 1H), 2.20−2.49 (m, 4H). LC-HRMS: m/z (M + H)+ calcd for
C19H23N3O5S, 406.1431; found, 406.1436.
(1R,2R,4S)-2-(Morpholine-4-carbonyl)-4-(2-trifluoromethyl-phe-
nylsulfanyl)-cyclopentanecarboxylic Acid Ethyl Ester (11).11
(1R,2R,4R)-4-Methanesulfonyloxy-2-(morpholine-4-carbonyl)-cyclo-
pentanecarboxylic acid ethyl ester (7) (300 mg, 0.86 mmol) and 2-
trifluoromethylthiophenol were converted according to general
procedure E to the crude thioether, which was purified by flash
chromatography using cyclohexane/EtOAc (1:2) to give title
(1R,2R,4R)-4-(2-Chloro-benzenesulfonyl)-2-(morpholine-4-car-
bonyl)-cyclopentanecarboxylic Acid Cyanomethyl-amide (22).11
(1R,2R,4S)-4-(2-Chloro-benzenesulfonyl)-2-(morpholine-4-carbonyl)-
cyclopentanecarboxylic acid (18) (45 mg, 0.09 mmol) dissolved in THF
(0.7 mL) and DMF (0.7 mL) was reacted according to general
procedure A with aminoacetonitrile hydrochloride to give the crude
amide, which was purified by flash chromatography using dichloro-
methane/MeOH (60:1) to give title compound 22 (8 mg, 20%) as a
white solid. 1H NMR (300 MHz, CDCl3) δ 8.09−8.20 (m, 1H), 7.43−
7.64 (m, 3H), 7.20 (t, J = 6.0 Hz, 1H), 4.27−4.42 (m, 1H), 4.17 (dd, J =
6.0, 17.4 Hz, 1H), 4.05 (dd, J = 6.0, 17.4 Hz, 1H), 3.10−3.74 (m, 10H),
2.21−2.45 (m, 4H). LC-HRMS: m/z (M + H)+ calcd for
C19H22ClN3O5S, 440.1041; found, 440.1052.
(1R,2R,4R)-2-(Morpholine-4-carbonyl)-4-(2-trifluoromethyl-ben-
zenesulfonyl)-cyclopentanecarboxylic Acid Cyanomethyl-amide
(23).11 (1R,2R,4S)-2-(Morpholine-4-carbonyl)-4-(2-trifluoromethyl-
benzenesulfonyl)-cyclopentanecarboxylic acid (19) (80 mg, 0.18
mmol) dissolved in DMF (3 mL) was reacted according to general
procedure A with aminoacetonitrile hydrochloride to give the crude
amide, which was purified by flash chromatography using EtOAc to give
title compound 23 (77 mg, 89%) as a white foam. 1H NMR (300 MHz,
CDCl3) δ 8.24−8.33 (m, 1H), 7.91−8.00 (m, 1H), 7.75−7.85 (m, 2H),
6.97 (t, J = 6.0 Hz, 1H), 4.17 (dd, J = 6.0, 17.4 Hz, 1H), 4.05 (dd, J = 6.0,
17.4 Hz, 1H), 3.39−3.75 (m, 10H), 3.02−3.17 (m, 1H), 2.27−2.43 (m,
4H). LC-HRMS: m/z (M + H)+ calcd for C20H22F3N3O5S, 474.1305;
found, 474.1317.
1
compound 10 (304 mg, 82%) as a colorless oil. H NMR (300 MHz,
CDCl3) δ 7.67 (d, J = 7.8 Hz, 1H), 7.55 (d, J = 7.8 Hz, 1H), 7.47 (t, J =
7.8 Hz, 1H), 7.32 (t, J = 7.8 Hz, 1H), 4.12 (q, J = 7.1 Hz, 2H), 3.66 (d, J =
3.2 Hz, 11H), 2.27−2.47 (m, 2H), 2.10−2.24 (m, 1H), 1.88 (td, J = 10.0,
13.0 Hz, 1H), 1.24 (t, J = 7.1 Hz, 3H). MS: 432.3 [M + H]+.
(1R,2R,4S)-4-Benzenesulfonyl-2-(morpholine-4-carbonyl)-cyclo-
pentanecarboxylic Acid Ethyl Ester (13).11 (1R,2R,4S)-2-(Morpho-
line-4-carbonyl)-4-phenylsulfanyl-cyclopentanecarboxylic acid ethyl
ester (9) (783 mg, 2.15 mmol) was converted according to general
procedure F to the crude sulfone, which was purified by flash
chromatography using n-heptane/EtOAc (1:3) to give title compound
1
13 (857 mg, quantitative) as a colorless oil. H NMR (300 MHz,
CDCl3) δ 7.85−7.96 (m, 2H), 7.63−7.72 (m, 1H), 7.53−7.62 (m, 2H),
4.10 (d, J = 7.1 Hz, 2H), 3.25- 3.70 (m, 11H), 2.48−2.63 (m, 1H), 2.06−
2.36 (m, 3H), 1.23 (t, J = 7.1 Hz, 3H). MS: 396.1 [M + H]+.
(1R,2R,4S)-4-(2-Chloro-benzenesulfonyl)-2-(morpholine-4-car-
bonyl)-cyclopentanecarboxylic Acid Ethyl Ester (14).11 (1R,2R,4S)-4-
(2-Chloro-phenylsulfanyl)-2-(morpholine-4-carbonyl)-cyclopentane-
carboxylic acid ethyl ester (10) (100 mg, 0.25 mmol) was converted
according to general procedure F to the crude sulfone, which was
purified by flash chromatography using n-heptane/EtOAc (1:3) to give
title compound 14 (92 mg, 85%) as a colorless oil. 1H NMR (300 MHz,
CDCl3) δ 8.13−8.18 (m, 1H), 7.43−7.63 (m, 3H), 4.18−4.33 (m, 1H),
4.12 (q, J = 7.1 Hz, 2H), 3.28−3.74 (m, 10H), 2.50−2.65 (m, 1H),
2.02−2.38 (m, 3H), 1.24 (t, J = 7.1 Hz, 3H). MS: 430.5 [M + H]+.
(1R,2R,4S)-2-(Morpholine-4-carbonyl)-4-(2-trifluoromethyl-ben-
zenesulfonyl)-cyclopentanecarboxylic Acid Ethyl Ester (15).11
(1R,2R,4S)-2-(Morpholine-4-carbonyl)-4-(2-trifluoromethyl-phenyl-
sulfanyl)-cyclopentanecarboxylic acid ethyl ester (11) (304 mg, 0.70
mmol) was converted according to general procedure F to the crude
sulfone, which was purified by flash chromatography using EtOAc to
give title compound 15 (324 mg, 99%) as a colorless oil. 1H NMR (300
MHz, CDCl3) δ 8.24−8.32 (m, 1H), 7.88−7.98 (m, 1H), 7.73−7.83 (m,
2H), 4.12 (q, J = 7.2 Hz, 2H), 3.84−4.01 (m, 1H), 3.25−3.76 (m, 10H),
2.62 (ddd, J = 7.9, 9.8, 13.8 Hz, 1H), 2.27−2.43 (m, 1H), 2.05−2.22 (m,
2H), 1.23 (t, J = 7.2 Hz, 3H). MS: 464.1 [M + H]+.
(1R,2R,4S)-4-Benzenesulfonyl-2-(morpholine-4-carbonyl)-cyclo-
pentanecarboxylic Acid (17).11 (1R,2R,4S)-4-Benzenesulfonyl-2-
(morpholine-4-carbonyl)-cyclopentanecarboxylic acid ethyl ester (13)
(853 mg, 2.16 mmol) was converted according to general procedure C
to crude acid 17 (780 mg, 98%), which was processed without further
purification. MS (ESI, negative ion): 366.3 [M − H]−.
(1R,2R,4S)-4-(2-Chloro-benzenesulfonyl)-2-(morpholine-4-car-
bonyl)-cyclopentanecarboxylic Acid (18).11 (1R,2R,4S)-4-(2-Chloro-
benzenesulfonyl)-2-(morpholine-4-carbonyl)-cyclopentanecarboxylic
acid ethyl ester (14) (80 mg, 0.19 mmol) was converted according to
general procedure C to crude acid 18 (93 mg, quantitative), which was
processed without further purification. MS: 402.1 [M + H]+.
(2S,4S)-Methyl-4-methanesulfonyloxy-N-Boc-pyrrolidine-2-car-
boxylate (60).30 (2S,4S)-Methyl-4-hydroxy-N-Boc-pyrrolidine-2-car-
boxylate (59) (1.00 g, 4.1 mmol) was converted according to general
procedure D to the crude mesylate, which was purified by flash
chromatography using cyclohexane/EtOAc (1:2) to give title
1
compound 60 (1.33 g, quantitative) as a pale-yellow oil. H NMR
(300 MHz, CDCl3) δ 5.18−5.29 (m, 1H), 4.35−4.59 (m, 1H), 3.72−
3.85 (m, 2H), 3.76 (s, 3H), 3.01 (s, 3H), 2.43−2.59 (m, 2H), 1.48 and
1.43 (s each, 9H). MS: 324.4 [M + H]+.
(2S,4R)-Methyl-4-(2-chloro-phenylsulfanyl)-N-Boc-pyrrolidine-2-
carboxylate (62).12 (2S,4S)-Methyl-4-methanesulfonyloxy-N-Boc-pyr-
rolidine-2-carboxylate (60) (1.00 g, 3.1 mmol) and 2-chlorothiophenol
were converted according to general procedure E to the crude thioether,
which was purified by flash chromatography using EtOAc to give title
1
compound 62 (0.644 g, 70%) as a white solid. H NMR (300 MHz,
CDCl3) δ 7.36−7.46 (m, 2H), 7.26 (s, 2H), 4.35−4.55 (m, 1H), 3.84−
4.02 (m, 2H), 3.74 (s, 3H), 3.34−3.55 (m, 1H), 2.18−2.43 (m, 2H),
1.45 and 1.42 (s each, 9H). MS: 372.1 [M + H]+.
(2S,4R)-Methyl-4-(2-chloro-benzenesulfonyl)-N-Boc-pyrrolidine-
2-carboxylate (65).12 (2S,4R)-Methyl-4-(2-chloro-phenylsulfanyl)-N-
Boc-pyrrolidine-2-carboxylate (62) (0.64 g, 1.7 mmol) was converted
according to general procedure F to the crude sulfone, which was
purified by flash chromatography using cyclohexane/EtOAc (1:1) to
give title compound 65 (0.60 g, 86%) as a colorless oil. 1H NMR (300
MHz, CDCl3) δ 8.05−8.18 (m, 1H), 7.55−7.67 (m, 2H), 7.45−7.54 (m,
1H), 4.32−4.62 (m, 2H), 3.57−4.02 (m, 5H), 2.58−2.98 (m, 1H),
2.09−2.41 (m, 1H), 1.45 and 1.40 (s each, 9H). MS: 404.5 [M + H]+.
(2S,4R)-4-(2-Chloro-benzenesulfonyl)-N-Boc-pyrrolidine-2-car-
boxylic Acid (68).12 (2S,4R)-Methyl-4-(2-chloro-benzenesulfonyl)-N-
(1R,2R,4S)-2-(Morpholine-4-carbonyl)-4-(2-trifluoromethyl-ben-
zenesulfonyl)-cyclopentanecarboxylic Acid (19).11 (1R,2R,4S)-2-
(Morpholine-4-carbonyl)-4-(2-trifluoromethyl-benzenesulfonyl)-cyclo-
J
dx.doi.org/10.1021/jm401528k | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Boc-pyrrolidine-2-carboxylate (65) (3.60 g, 8.9 mmol) was hydrolyzed
according to general procedure C to crude acid 68 (3.47 g, quantitative),
which was used in the next step without purification. MS (ESI, negative
ion): 388.3 [M − H]−.
sites at the 5′ and 3′ ends, respectively, were used to clone the gene into a
pET24 vector. No tag was added to the N- or C-terminus. The protein
was produced in Escherichia coli strain BL21 (DE3)RIL at 37 °C in LB
medium by induction with IPTG (isopropyl thio-β-D-galactoside).
Isolation and Solubilization of Inclusion Bodies (IBs). Cell pellets
(20g) were resuspended in 50 mM Tris-HCl (120 mL), pH 8.0,
supplemented with 2 mM ethylendiaminetetraacetic acid (EDTA), 5%
saccharose, 30 mg/mL DNase I, and 30 mg/mL RNase I (fresh). After
cell disruption and centrifugation, IBs were washed with 50 mM Tris-
HCl, pH 8.0, containing 2 mM EDTA and 0.1% Triton X-100 and
centrifuged again. The washing procedure was repeated twice with a
second buffer (50 mM Tris-HCl, pH 8.0, and 1 M urea). The final pellet
was solubilized in 50 mM Tris-HCl (15 mL), pH 8.0, 5 mM EDTA, 6 M
guanidinium hydrochloride, 150 mM NaCl, and 10 mM dithiothreitol
(DTT) and centrifuged at 4 °C, and the protein concentration of mCatS
was determined via HPLC.
(2S,4R)-4-(2-Chloro-benzenesulfonyl)-2-(1-cyano-cyclopropylcar-
bamoyl)-N-Boc-pyrrolidine (71).12 (2S,4R)-4-(2-Chloro-benzenesul-
fonyl)-N-Boc-pyrrolidine-2-carboxylic acid (68) (5.0 g, 12.8 mmol) and
1-amino-cyclopropanecarbonitril hydrochloride were converted accord-
ing to general procedure B to the crude amide, which was purified by
flash chromatography using n-heptane/EtOAc (gradient from 1:0 to
1
1:1) to give title compound 71 (4.70 g, 87%) as a colorless foam. H
NMR (300 MHz, CDCl3) δ 8.08−8.14 (m, 1H), 7.89 (br. s., 1H), 7.57−
7.67 (m, 2H), 7.46−7.54 (m, 1H), 4.36−4.54 (m, 2H), 4.00 (dd, J = 6.3,
12.1 Hz, 1H), 3.61 (dd, J = 7.9, 11.5 Hz, 1H), 2.57−2.74 (m, 1H), 2.26−
2.44 (m, 1H), 1.44−1.54 (m, 11H), 1.15−1.23 (m, 2H). MS: 454.4 [M
+ H]+.
(2S,4R)-4-(2-Chloro-benzenesulfonyl)-2-(1-cyano-cyclopropylcar-
bamoyl)-pyrrolidine; Salt with Hydrochloric Acid (74).12 A solution of
(2S,4R)-4-(2-chloro-benzenesulfonyl)-2-(1-cyano-cyclopropylcarba-
moyl)-N-Boc-pyrrolidine (71) (408 mg, 0.9 mmol) in HCl/1,4-dioxane
(2 M, 4 mL) was stirred at 22 °C overnight, the suspension was filtered,
and the residue washed with diethyl ether and dried to give amine
hydrochloride 74 (280 mg, 80%) as a white solid. MS: 354.1 [M + H]+.
(2S,4R)-1-Acetyl-4-(2-chloro-benzenesulfonyl)-pyrrolidine-2-car-
boxylic acid (1-Cyano-cyclopropyl)-amide (79).12 To a mixture of
(2S,4R)-4-(2-chloro-benzenesulfonyl)-2-(1-cyano-cyclopropylcarba-
moyl)-pyrrolidine (salt with hydrochloric acid) (74) (293 mg, 0.75
mmol) in acetonitrile (15 mL) were added diisopropylethylamine (582
mg, 4.5 mmol) and acetic anhydride (368 mg, 3.6 mmol), and stirring
was continued at 22 °C for 3 h. The mixture was partitioned between
saturated aqueous Na2CO3 and EtOAc, and the organic layer was
washed with 1 N aqueous HCl, dried, and evaporated to give the crude
amide, which was purified by flash chromatography using n-heptane/
EtOAc (gradient, 2:3 to 1:5) to give title compound 79 (167 mg, 56%)
as a colorless foam. 1H NMR (300 MHz, CDCl3) δ 8.10−8.15 (m, 1H),
7.97 (s, 1H), 7.60−7.68 (m, 2H), 7.47−7.55 (m, 1H), 4.55−4.75 (m,
2H), 4.05−4.17 (m, 1H), 3.77−3.86 (m, 1H), 2.62−2.75 (m, 1H),
2.25−2.35 (m, 1H), 2.16 (s, 3H), 1.45−1.55 (m, 2H), 1.15−1.25 (m,
2H). LC-HRMS: m/z (M + H)+ calcd for C17H18ClN3O4S, 396.0779;
found, 396.0779.
Cathepsin S, K, L, and B Enzyme Inhibition Assay. Enzyme
activity was measured by observing the increase in fluorescence intensity
caused by cleavage of a peptide substrate containing a fluorophore
whose emission is quenched in the intact peptide. Assay buffer: 100 mM
potassium phosphate, pH 6.5, EDTA-Na 5 mM, Triton X-100 0.001%,
and DTT 5 mM. Enzymes (all at 1 nM): human and mouse cathepsin S
and mouse CatK, CatL, and CatB. Substrate (20 μM): Z-Val-Val-Arg-
AMC except for CatK, which uses Z-Leu-Arg-AMC (both from
Bachem). Final volume: 100 μL. Excitation 360 nm, emission 465 nm.
Enzyme was added to the substance dilutions in 96-well microtiter
plates, and the reaction was started with substrate. Fluorescence
emission was measured over 20 min, during which time a linear increase
was observed in the absence of inhibitor. IC50 values were calculated by
standard methods.
Folding and Reoxidation of Mouse Procathepsin S. The IB solution
was diluted to a protein concentration of 4.4 mg/mL and subsequently
diluted 1:100 using a sample pump with a flow rate of 0.6 mL/min in 50
mM Tris-HCl buffer, pH 8.5, 500 mM L-arginine, 0.01% BRIJ 35, 100
mM NaCl, 80 mL catalase (1 mg/mL), 10 mM glutathione (GSH), and
1 mM glutathione disulfide (GSSG) at 4 °C. The mixture was stirred for
48 h at 4 °C, filtered, concentrated, and dialyzed at 4 °C against 25 mM
sodium phosphate buffer, pH 7.0, and 500 mM NaCl overnight.
Purification of mCatS. The protein solution was adjusted to a final
concentration of 1500 mM (NH4)2SO4 and centrifuged, the supernatant
was loaded on a HIC Toyopearl-Butyl650 column equilibrated with 10
mM Tris-HCl, pH 8.0, 1 M NaCl, 1.5 M (NH4)2SO4, and 0.01% NaN3,
and the column was eluted with a linear gradient to 10 mM Tris-HCl,
pH 8.0, and 0.01% NaN3. The pooled fractions were concentrated and
loaded on a Superdex 75 (XK26/100) column equilibrated with 20 mM
sodium acetate at pH 4.5, 300 mM NaCl, 5 mM DTT, 2.5 mM EDTA,
0.01% NaN3, and 0.01% BRIJ35. Autoactivation was performed during
size-exclusion chromatography, and fractions containing mCatS were
pooled and concentrated around 20-fold with a 10 kDa cutoff
ultrafiltration membrane.
Active-Site Cysteine Oxidation. The integrity of the active-site
cysteine was analyzed by titration with the covalent, irreversible inhibitor
N-morpholinourea-leucine-homophenylalanine-vinylsulfone-phenyl
(LHVS). Initially, oxidation of this cysteine appeared to be a problem,
but this was overcome by the inclusion of catalase in the refolding step.
By taking care to degas buffers and to work and store protein under
argon, the amount of mCatS with an oxidized active-site cysteine could
be reduced to around 30%.
Removal of Oxidized CatS by (S−S) Covalent Chromatography.
Concentrated and activated mCatS was loaded onto a Fast Desalting
column HR10/10 (GE Heathcare) equilibrated in 100 mM sodium
acetate buffer, pH 4.0, and 1 mM EDTA. The DTT-free mCatS fraction
was subsequently applied to a Thiopropyl Sepharose 6B column (2.6 ×
5 cm, GE Healthcare). The part of the protein sample with an oxidized
active-site cysteine was collected in the column flow-through. Active
mCatS was eluted in 25 mM sodium acetate buffer, pH 4.5, 300 mM
NaCl, 10% glycerin, 0.01% BRIJ35, and 2.5 mM EDTA containing 10
mM DTT. The mCatS pool was concentrated and passed over a Fast
desalting HR10/10 equilibrated in 100 mM sodium acetate, pH 4.5,
containing 5 mM DTT and 5 mM EDTA. Aliquots showing over 97%
active, nonoxidized mCatS were stored under argon in sealed tubes at
−80 °C.
Cocrystallization of mCatS in Complex with 21, 56, 57, and 75.
Protein at a concentration of 2 μM in 100 mM sodium acetate pH 4.5, 5
mM DTT, 5 mM EDTA, and 0.01% NaN3 was incubated with ligand in a
10-fold molar excess overnight at 4 °C under argon. Prior to
crystallization experiments, the protein was concentrated to 21−36
mg/mL and centrifuged at 20 000g. The crystallization droplets for 21
and 75 were set up at 22 °C by mixing 0.25 μL of protein solution with
0.25 μL of reservoir solution (Index Screen, Hampton Research) in
microbatch experiments under Al’s oil (Hampton Research). For the
cocrystallization of 56 and 57, vapor-diffusion experiments in sitting-
drop set ups were used. For compound 21, crystals were obtained out of
2 M (NH4)2SO4 (pH 5.5) as precipitant. Complexes with compounds
Cell-Based Assay. A20 cells (ATCC, no. 30-2001) were plated at 3
× 105 cells/ml in 100 μL/well of a U-bottom 96-well plate. Compounds
were added, and cells were incubated for 1 h before the addition of
ovalbumin to a final concentration of 5 mg/mL (Sigma Aldrich). After
16 h, the cells were washed three times with PBS, and 2 × 105 DO10.10
T-cell hybridoma cells/ml were added. After 24 h, IL-2 production was
measured by Luminex single-bead analysis using two anti-IL-2
antibodies (BD, cat. nos. 554424 and 554426).
LogD Determination. LogD determination was carried out as
described.31
Cell Permeability Determination. Cell permeability was
determined as described.29
Protein Crystallography. Cloning and Protein Expression. The
mouse Cathepsin S (mCatS) gene without a secretion signal (coding for
amino acids A22−I340 with L97S compared to UniProt O70370) was
amplified from lung cDNA by PCR. NdeI and NotI restriction enzyme
K
dx.doi.org/10.1021/jm401528k | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
56 and 57 crystallized out of 25% PEG 3350 (pH 7.5 and 6.5,
respectively), and the complex with compound 75 crystallized out of 3
M sodium citrate (pH 3.5). Before data collection, crystals were
transferred to crystallization buffer supplemented with 20% glycerol or
20% ethylene glycol and flash-frozen in liquid N2. Data collection and
evaluation and structure determination were performed as described
with minor modifications.28
Pharmacokinetic Experiments in Mice and Plasma Stability
Studies. Animals and Housing Conditions. Animal experiments were
conducted during the light phase and under ad libitum conditions.
Animal housing and experimental procedures were in line with ethical
and legal guidelines and were authorized by local veterinary authorities.
In Vivo Experiment. C57Bl/6 male mice (8 week old; 24−29 g;
Charles River Laboratories, Germany GmbH) received doses of the test
compound by po gavage (9 mg/kg as an aqueous microsuspension in
gelatin/NaCl (7.5%/0.62%)) and iv bolus (2 mg/kg in aqueous NaCl/
30% 1-methyl-2-pyrrolidone). The nominal concentration of the
compound in the formulation was adjusted to reach a 4 mL/kg oral
dosage or a 2 mL/kg iv bolus. Terminal blood samples (500 mL, n = 2
per time point) were taken at different time points postdosing by
decapitation and collected in precooled EDTA-coated tubes. The
samples were kept on ice and immediately centrifuged at 4 °C to obtain
plasma. Samples were stored at −20 °C until analysis.
Plasma in Vitro stability. Heparinized pooled human or NMRI
mouse plasma samples were spiked with 1000 ng/mL of the compound
of interest. Incubations were performed in a thermostatted orbital
shaker at 37 °C. Aliquots (n = 3 per time point) were taken after 2 min, 1
h, and 4 h of incubation and immediately processed for analysis.
LC−MS/MS Compound Analysis. Quantification of compound levels
in plasma was accomplished by means of LC−MS/MS analysis. Plasma
(50 mL) was precipitated in three volumes of acetonitrile containing d6-
midazolam (0.2 mg/mL) as an internal standard, mixed, and centrifuged
(10 min, 4 °C, 5850g). The supernatant was diluted 1:1 with water and
injected onto a Phenomenex Gemini C18 3 μm 2.0·30 mm column
operating with mobile phases A (0.2% formic acid in water containing
5% acetonitrile) and B (acetonitrile) at 0.3 mL/min flow. The outlet of
the column was coupled to an API3000 mass spectrometer (AB Sciex,
Brugg, Switzerland) with TurboIonSpray source. Detection was carried
out using multiple reactions monitoring mode with positive ion
detection to follow compound-specific transitions.
AUTHOR INFORMATION
Corresponding Authors
*Phone: +41 61 688 72 64. E-mail: hans.hilpert@roche.com
(H.H.).
*Phone: +41 61 688 68 93. E-mail: wolfgang.haap@roche.com
(W.H.).
Notes
■
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank P. Bourdeaux, M. Marti, and S. Thomi for synthesis, P.
Hartman, A. Araujo Del Rosario, and R. Ecabert for screening, J.
Schneider for NMR, and C. Bartelmus for mass spectroscopy
support.
ABBREVIATIONS USED
■
AMC, 7-amino-4-methyl-coumarin BOC, t-butoxycarbonyl;
DAST, diethylaminosulfur trifluoride; DMF, N,N-dimethylfor-
mamide, DMSO, dimethyl sulfoxide; DTT, dithiothreitol; EDCI,
1-ethyl-3-(3-dimethylaminopropyl)carbodiimid; EtOAc, ethyl
acetate; EtOH, ethanol; HOBt, 1-hydroxybenzotriazol; MeOH,
methanol; NADP, nicotinamide adenine dinucleotide phos-
phate; TEA, triethylamine; TFA, trifluoro-acetic acid; THF,
tetrahydrofuran; Z, benzyloxycarbonyl
REFERENCES
■
(1) Turk, B.; Turk, V.; Turk, D. Structural and functional aspects of
papain-like cysteine proteinases and their protein inhibitors. Biol. Chem.
1997, 378, 141−150.
(2) Podolin, P. L.; Bolognese, B. J.; Carpenter, D. C.; Davis, T. G.;
Johanson, R. A.; Fox, J. H.; Long, E.,, III; Dong, X.; Marquis, R. W.;
LoCastro, S. M.; Terfloth, G. J.; Kurali, E.; Peterson, J. J.; Smith, B. R.;
McQueney, M. S.; Yamashita, D. S.; Capper-Spudich, E. A. Inhibition of
invariant chain processing, antigen-induced proliferative responses, and
the development of collagen-induced arthritis and experimental
autoimmune encephalomyelitis by a small molecule cysteine protease
inhibitor. J. Immunol. 2008, 180, 7989−8003.
(3) Biroc, S. L.; Gay, S.; Hummel, K.; Magill, C.; Palmer, J. T.; Spencer,
D. R.; Sa, S.; Klaus, J. L.; Michel, B. A.; Rasnick, D.; Gay, R. E. Cysteine
protease activity is up-regulated in inflamed ankle joints of rats with
adjuvant-induced arthritis and decreases with in vivo administration of a
vinyl sulfone cysteine protease inhibitor. Arthritis Rheum. 2001, 44,
703−711.
(4) Nakagawa, T. Y.; Brissette, W. H.; Lira, P. D.; Griffiths, R. J.;
Petrushova, N.; Stock, J.; McNeish, J. D.; Eastman, S. E.; Howard, E. D.;
Clarke, S. R. M.; Rosloniec, E. F.; Elliott, E. A.; Rudensky, A. Y. Impaired
invariant chain degradation and antigen presentation and diminished
collagen-induced arthritis in cathepsin S null mice. Immunity 1999, 10,
207−217.
IL-2 Reduction in DO10.11 Transgenic Mice. Animals and
Housing Conditions. Animals were maintained in a 12 h:12 h light/
dark cycle, with lights on starting at 6 a.m., and experiments were
conducted during the light phase. Animal housing and experimental
procedures were in line with ethical and legal guidelines and were
authorized by local veterinary authorities.
Experiment. DO10.11 BALBc mice were used to test the role of CatS
in antigen presentation. These mice express a transgenic TCR
recognizing an ovalbumin peptide. Challenge of the mice with 1.0 mg
iv low-endotoxin ovalbumin results in rapid production of interleukin 2
by T cells. Increased levels of IL-2 were detected at the 4 h time point in
plasma using Luminex bead technology and two anti-IL-2 antibodies
(BD nos. 554424 and 554426).
(5) Shi, G.-P.; Villadangos, J. A.; Dranoff, G.; Small, C.; Gu, L.; Haley,
K. J.; Riese, R.; Ploegh, H. L.; Chapman, H. A. Cathepsin S required for
normal MHC class II peptide loading and germinal center development.
Immunity 1999, 10, 197−206.
(6) Lee-Dutra, A.; Wiener, D. K.; Sun, S. Cathepsin S inhibitors: 2004−
2010. Expert Opin. Ther. Pat. 2011, 21, 311−337.
ASSOCIATED CONTENT
■
S
* Supporting Information
(7) Gauthier, J. Y.; Black, W. C.; Courchesne, I.; Cromlish, W.;
Desmarais, S.; Houle, R.; Lamontagne, S.; Li, C. S.; Masse, F.; McKay, D.
J.; Ouellet, M.; Robichaud, J.; Truchon, J.-F.; Truong, V.-L.; Wang, Q.;
Percival, M. D. The identification of potent, selective, and bioavailable
cathepsin S inhibitors. Bioorg. Med. Chem. Lett. 2007, 17, 4929−4933.
(8) Bekkali, Y.; Thomson, D. S.; Betageri, R.; Emmanuel, M. J.; Hao,
M.-H.; Hickey, E.; Liu, W.; Patel, U.; Ward, Y. D.; Young, E. R. R.;
Nelson, R.; Kukulka, A.; Brown, M. L.; Crane, K.; White, D.; Freeman,
D. M.; Labadia, M. E.; Wildeson, J.; Spero, D. M. Identification of a novel
class of succinyl-nitrile-based cathepsin S inhibitors. Bioorg. Med. Chem.
Lett. 2007, 17, 2465−2469.
Preparation details and analytical data for compounds 5, 20, 24−
26, 33−46, 56−58, and 74−78; and distribution of torsion
angles observed for alkyl-sulfones and alkyl-sulfonamides in the
small-molecule crystal structures database. This material is
available free of charge via the Internet at http://pubs.acs.org.
Accession Codes
Crystal structure coordinates and structure factors have been
deposited in the PDB under accession numbers 4BSQ (21),
4MZO (56), 4MZS (57), and 4BS5 (75).
L
dx.doi.org/10.1021/jm401528k | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(9) Moss, N.; Xiong, Z.; Burke, M.; Cogan, D.; Gao, D. A.; Haverty, K.;
Heim-Riether, A.; Hickey, E. R.; Nagaraja, R.; Netherton, M.; O’Shea,
K.; Ramsden, P.; Schwartz, R.; Shih, D.-T.; Ward, Y.; Young, E.; Zhang,
Q. Exploration of cathepsin S inhibitors characterized by a triazole P1-
P2 amide replacement. Bioorg. Med. Chem. Lett. 2012, 22, 7189−7193.
(10) Gabriel, T.; Pech, M.; Rodriguez, S. R. M. β-Amino acid nitrile
derivatives useful for the treatment of diseases which are associated with
cysteine proteases. Patent WO2001096285A1, 2001.
(11) Banner, D.; Ceccarelli, S. M.; Grether, U.; Haap, W.; Hilpert, H.;
Kuehne, H.; Mauser, H.; Plancher, J.-M.; Sanchez, R. A. Novel
cyclopentane derivatives as cysteine protease cathepsin inhibitors and
their preparation and use in the treatment of diseases. Patent
WO2010142650A1, 2010.
Hartmann, G.; Isshiki, Y.; Morikami, K.; Shimma, N.; Haap, W.; Banner,
D. W.; Diederich, F. Halogen bonding at the active sites of human
cathepsin L and MEK1 kinase: Efficient interactions in different
environments. ChemMedChem 2011, 6, 2048−2054.
(29) Kansy, M.; Senner, F.; Gubernator, K. Physicochemical high
throughput screening: Parallel artificial membrane permeation assay in
the description of passive absorption processes. J. Med. Chem. 1998, 41,
1007−1010.
(30) Tsunoda, H.; Fukumura, K. Process for producing ester or
alcohol. Patent WO2007148435A1, 2007.
(31) Hilpert, H.; Guba, W.; Woltering, T. J.; Wostl, W.; Pinard, E.;
Mauser, H.; Mayweg, A. V.; Rogers-Evans, M.; Humm, R.;
Krummenacher, D.; Muser, T.; Schnider, C.; Jacobsen, H.; Ozmen,
L.; Bergadano, A.; Banner, D. W.; Hochstrasser, R.; Kuglstatter, A.;
David-Pierson, P.; Fischer, H.; Polara, A.; Narquizian, R. β-Secretase
(BACE1) inhibitors with high in vivo efficacy suitable for clinical
evaluation in Alzheimer’s disease. J. Med. Chem. 2013, 56, 3980−3995.
(12) Sanchez, R. A.; Banner, D.; Ceccarelli, S. M.; Grether, U.; Haap,
W.; Hartman, P.; Hartmann, G.; Hilpert, H.; Kuehne, H.; Mauser, H.;
Plancher, J.-M. Preparation of proline dipeptidyl nitrile derivatives as
cathepsin, particularly cathepsin S and L, inhibitors. Patent
US20100267722A1, 2010.
(13) Kim, M.; Jeon, J.; Song, J.; Suh, K. H.; Kim, Y. H.; Min, K. H.; Lee,
K.-O. Synthesis of proline analogues as potent and selective cathepsin S
inhibitors. Bioorg. Med. Chem. Lett. 2013, 23, 3140−3144.
(14) Banner, D.; Haap, W.; Kuhn, B.; Luebbers, T.; Peters, J.-U.;
Schulz-Gasch, T. Preparation of substituted 4-(phenylsulfonyl)-
pyrrolidine-2-carboxamide derivatives as cathepsin inhibitors. Patent
WO2013076063A1, 2013.
(15) Mittendorf, J.; Benet-Buchholz, J.; Fey, P.; Mohrs, K.-H. Efficient
asymmetric synthesis of β-amino acid BAY 10-8888/PLD-118, a novel
antifungal for the treatment of yeast infections. Synthesis 2003, 136−
140.
(16) Kataoka, M.; Kita, K.; Wada, M.; Yasohara, Y.; Hasegawa, J.;
Shimizu, S. Novel bioreduction system for the production of chiral
alcohols. Appl. Microbiol. Biotechnol. 2003, 62, 437−445.
(17) Kosjek, B.; Nti-Gyabaah, J.; Telari, K.; Dunne, L.; Moore, J. C.
Preparative asymmetric synthesis of 4,4-dimethoxytetrahydro-2H-
pyran-3-ol with a ketone reductase and in situ cofactor recycling using
glucose dehydrogenase. Org. Process Res. Dev. 2008, 12, 584−588.
(18) Rodriguez, S. R. M.; Wirz, B.; Iding, H. Chemoenzymatic
preparation of non-racemic N-Boc-pyrrolidine-3,4-dicarboxylic acid 3-
ethyl esters and their 4-hydroxymethyl derivatives. Tetrahedron:
Asymmetry 2003, 14, 1547−1551.
(19) McGrath, M. E.; Palmer, J. T.; Bromme, D.; Somoza, J. R. Crystal
structure of human cathepsin S. Protein Sci. 1998, 7, 1294−1302.
(20) Oballa, R. M.; Truchon, J.-F.; Bayly, C. I.; Chauret, N.; Day, S.;
Crane, S.; Berthelette, C. A generally applicable method for assessing the
electrophilicity and reactivity of diverse nitrile-containing compounds.
Bioorg. Med. Chem. Lett. 2007, 17, 998−1002.
(21) The PyMOL Molecular Graphics System, version 1.5; Schrodinger,
̈
LLC; http://www.pymol.org.
(22) McGrath, M. E.; Klaus, J. L.; Barnes, M. G.; Broemme, D. Crystal
structure of human cathepsin K complexed with a potent inhibitor. Nat.
Struct. Biol. 1997, 4, 105−109.
(23) Brameld, K. A.; Kuhn, B.; Reuter, D. C.; Stahl, M. Small molecule
conformational preferences derived from crystal structure data. A
medicinal chemistry focused analysis. J. Chem. Inf. Model. 2008, 48, 1−
24.
(24) Menziani, M. C.; Cocchi, M.; De, B. P. G. Electronic and
electrostatic aspects of carbonic anhydrase inhibition by sulfonamides. J.
Mol. Struct.: THEOCHEM 1992, 88, 217−229.
(25) Allen, F. H. The Cambridge Structural Database: A quarter of a
million crystal structures and rising. Acta Crystallogr., Sect. B 2002, 58,
380−388.
(26) Cambridge Crystallographic Data Centre Home Page. http://
www.ccdc.cam.ac.uk.
(27) Wilcken, R.; Zimmermann, M. O.; Lange, A.; Zahn, S.; Kirchner,
B.; Boeckler, F. M. Addressing methionine in molecular design through
directed sulfur-halogen bonds. J. Chem. Theory Comput. 2011, 7, 2307−
2315.
(28) Hardegger, L. A.; Kuhn, B.; Spinnler, B.; Anselm, L.; Ecabert, R.;
Stihle, M.; Gsell, B.; Thoma, R.; Diez, J.; Benz, J.; Plancher, J.-M.;
M
dx.doi.org/10.1021/jm401528k | J. Med. Chem. XXXX, XXX, XXX−XXX