Dicarboxylic acid azacycle L-prolyl-pyrrolidine amides as prolyl oligopeptidase inhibitors and three-dimensional quantitative structure-activity relationship of the enzyme-inhibitor interactions

Article abstract of DOI:10.1021/jm0500020

A series of dicarboxylic acid azacycle L-prolyl-pyrrolidine amides was synthesized, and their inhibitory activity against prolyl oligopeptidase (POP) from porcine brain was tested. Three different azacycles were tested at the position beyond P3 and six different dicarboxylic acids at the P3 position. L-Prolyl-pyrrolidine and L-prolyl-2(S)-cyanopyrrolidine were used at the P2-Pl positions. The IC₅₀ values ranged from 0.39 to 19000 nM. The most potent inhibitor was the 3,3-dimethylglutaric acid azepane L-prolyl-2(S)- cyanopyrrolidine amide. Molecular docking (GOLD) was used to analyze binding interactions between different POP inhibitors of this type and the POP enzyme. The data set consisted of the novel inhibitors, inhibitors published previously by our group, and well-known reference compounds. The alignments were further analyzed using comparative molecular similarity indices analysis. The binding of the inhibitors was consistent at the P1-P3 positions. Beyond the P3 position, two different binding modes were found, one that favors lipophilic structures and one that favors nonhydrophobic structures.

Full text of DOI:10.1021/jm0500020

4772
J. Med. Chem. 2005, 48, 4772-4782
Dicarboxylic Acid Azacycle L-Prolyl-pyrrolidine Amides as Prolyl
Oligopeptidase Inhibitors and Three-Dimensional Quantitative
Structure-Activity Relationship of the Enzyme-Inhibitor Interactions
Elina M. Jarho,*,† Erik A. A. Walle´n,† Johannes A. M. Christiaans,†,‡ Markus M. Forsberg,§
Jarkko I. Vena¨la¨inen,§ Pekka T. Ma¨nnisto¨,§,| Jukka Gynther,† and Antti Poso†
Departments of Pharmaceutical Chemistry and Pharmacology and Toxicology, P.O. Box 1627, FI-70211 Kuopio, Finland
Received January 3, 2005
A series of dicarboxylic acid azacycle L-prolyl-pyrrolidine amides was synthesized, and their
inhibitory activity against prolyl oligopeptidase (POP) from porcine brain was tested. Three
different azacycles were tested at the position beyond P3 and six different dicarboxylic acids
at the P3 position. L-Prolyl-pyrrolidine and L-prolyl-2(S)-cyanopyrrolidine were used at the
P2-P1 positions. The IC50 values ranged from 0.39 to 19000 nM. The most potent inhibitor
was the 3,3-dimethylglutaric acid azepane L-prolyl-2(S)-cyanopyrrolidine amide. Molecular
docking (GOLD) was used to analyze binding interactions between different POP inhibitors of
this type and the POP enzyme. The data set consisted of the novel inhibitors, inhibitors
published previously by our group, and well-known reference compounds. The alignments were
further analyzed using comparative molecular similarity indices analysis. The binding of the
inhibitors was consistent at the P1-P3 positions. Beyond the P3 position, two different binding
modes were found, one that favors lipophilic structures and one that favors nonhydrophobic
structures.
Introduction
Prolyl oligopeptidase (POP, EC 3.4.21.26) is a member
of the POP family of serine proteases.1 It is a large, 80
kDa enzyme with both a cytosolic and a particulate
form.2,3 POP hydrolyzes oligopeptides at the carboxyl
side of a proline residue, with the exception of an amide
bond between two proline residues and the first two
amide bonds at the amino terminus.4
POP has been related to cognitive disorders because
several substrates of POP, like substance P,5 vaso-
pressin,6-8 and thyroliberin,9 have memory-promoting
Figure 1. Rationale behind the novel inhibitors. (I) A series
effects. In addition, substance P has been reported to
of large symmetrical POP inhibitors.21 (II) A series of small
prevent the neurotoxic effects of the amyloid â peptide,
symmetrical compounds, which were very poor POP inhibitors.
often deposited in the brains of Alzheimer patients.10
(III) A series of the novel POP inhibitors.
In the human brain, the highest POP activity has been
found in the cortex.11,12 POP coexists with several
Recently, the POP of Trypanosoma cruzi has also been
implicated as a therapeutic target. T. cruzi causes
neuropeptide receptors, and it is possibly involved in
inactivation of regulatory neuropeptides.13 Exceptionally
Chagas’ disease, and the parasitic POP is presumably
high POP gene transcriptions have been found in the
involved in the invasion of host cells. Selective inhibitors
hypothalamus and cortex of aged mice,14 and on the
other hand, a stimulating environment has been shown
to decrease POP gene expression in the cortex of mice.15
The inhibition of POP has been shown to prevent scopol-
amine-, ischemia-, and lesion-induced amnesia in
rats.16-18 Furthermore, the inhibition of POP has been
shown to enhance cognition in a model of early Parkin-
sonism in monkeys.19
of the parasitic POP prevented the invasion.20
Our group published recently a series of large, sym-
metrical POP inhibitors (I in Figure 1).21 These inhibi-
tors were very potent, and another series of large
inhibitors were developed from the first set of com-
pounds.22 To diversify the compound set for the three-
dimensional (3D) quantitative structure-activity rela-
tionship (QSAR) study, two series of smaller compounds
were synthesized. The removal of the pyrrolidine-1-
carbonyl group from both ends (II in Figure 1, not
published) decreased strongly the potency. The succinic
acid dipyrrolidine amide was the only compound with
a weak inhibitory activity. (It had an IC50 value of 13
µM against POP from rat brain. Inhibitory activities
against POP from rat brain and porcine brain are
* To whom correspondence should be addressed. Tel: +358-17-
162460. Fax: +358-17-162456. E-mail: Elina.Jarho@uku.fi.
† Department of Pharmaceutical Chemistry.
‡ Present address: Altana Pharma bv, P.O. Box 31, 2130 AA
Hoofddorp, The Netherlands.
§ Department of Pharmacology and Toxicology.
| Present address: Division of Pharmacology and Toxicology, Faculty
of Pharmacy, P.O. Box 56, FI-00014 University of Helsinki, Finland.
10.1021/jm0500020 CCC: $30.25 © 2005 American Chemical Society
Published on Web 06/25/2005

Prolyl Oligopeptidase Inhibitors
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 15 4773
Table 1. Structures and Inhibitory Activities (95% Confidence
Intervals Are Presented in Parentheses) of the Synthesized
Compounds
different dicarboxylic acids coupled to different azacycles
were used at the positions P3 and beyond P3. The
inhibitors were synthesized via the synthetic routes
described in Schemes 1 and 2 using a similar set of
chemical reactions. The building blocks were coupled
to each other by amide bonds, which were synthesized
in two different ways: The carboxylic acid was activated
with pivaloyl chloride and reacted with the amine, or
the dicarboxylic anhydride was reacted with the amine.
The only problematic step was the deprotection of BOC-
2(S)-cyanopyrrolidine. The cyano group reacted to some
extent during the deprotection, and consequently, the
yields of these coupling reactions were low, ranging from
10 to 45% for compounds 11, 12, 17, 19, and 20.
Therefore, the cyano group was introduced in the last
synthetic step of the compounds that were made in the
later part of the study (14, 15, and 18). The com-
pounds in Table 2 were synthesized as described
previously.21,22,24-26
Molecular Modeling. The structure of POP, 1QFS
(native),27 was derived from PDB and the cocrystallized
inhibitor (Z-L-Pro-L-prolinal) was removed from the
crystal structure. The POP inhibitors in Table 1 and
the previously published compounds in Table 2 were
docked into this crystal structure of POP using the
GOLD program.28 The docking area was defined using
the NE atom of ARG643 as a center, with a radius of
20 Å. Unfortunately, current docking programs cannot
automatically take into account a possible covalent
interaction between a ligand and a host. Because several
of the compounds possibly have a covalent interaction,
two violations for steric clashes were allowed. With
these settings, all of the compounds were docked suc-
cessfully into the active site of POP.
The dockings were further analyzed using the
CoMSIA method,23 and a statistically valid 3D QSAR
model was created. Originally, CoMSIA was carried out
by using all five different field types, namely, steric,
electrostatic, hydrophobic, acceptor, and donor fields.
Even this model was statistically valid (q2 close to 0.7
with six components). However, a closer inspection
revealed that most of the model was explained by the
acceptor, electrostatic, and hydrophobic fields, which
were then used in the final models. The selection of the
partial least squares (PLS) components was based on
Scrambling stability test, as proposed by Clark and
Fox.29 This test yields adapted q2, CSDEP and dq**2/
dr2yy values. CSDEP had the optimum with three PLS
components and the dq**2/dr2yy well below 1 (0.91),
indicating low enough PLS component number. The
final CoMSIA model had the following leave-one-out
PLS statistics: q2 ) 0.69, SPRESS ) 0.65, three PLS
components. The final nonvalidated model resulted in
r2 ) 0.91, F value ) 175, and s value ) 0.34.
a The IC50 values were determined against POP from porcine
brain as described in the Experimental Section.
similar in our test system.) These compounds were
not included in the 3D QSAR study. Removal of the
pyrrolidine-1-carbonyl group from only one end, how-
ever, resulted in a new series of potent inhibitors, most
of them having the IC50 value in the nanomolar range
(III in Figure 1).
The 3D QSAR models were created using the com-
parative molecular similarity indices analysis (CoMSIA)
method, which produces information of the favored and
disfavored properties of the examined compound set.23
The compound set consisted of the novel inhibitors
published in this article (Table 1) and the previously
published inhibitors and reference compounds (Table 2).
These data are useful to explain and predict activities
of different POP inhibitors and to find out how different
types of POP inhibitors may bind to the active site of
the enzyme. To our knowledge, this is the first time
when the systematic molecular docking and 3D QSAR
studies have been described for POP inhibitors.
Results and Discussion
Molecular Docking. All compounds were success-
fully docked into the binding cavity of POP. When
selected dockings are analyzed together, it is obvious
that most of the compounds occupy in general the same
binding area. Because GOLD produces usually several
alternative binding modes, other binding modes were
also detected, but in all cases, the selection was based
on the scoring function of GOLD. The only exception
Methods
Synthetic Chemistry. A series of dicarboxylic acid
azacycle L-prolyl-pyrrolidine amides was synthe-
sized (Table 1). L-Pro-pyrrolidine and L-Pro-2(S)-cyano-
pyrrolidine were used at the P2-P1 positions, and

4774 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 15
Jarho et al.
Table 2. Structures and Inhibitory Activities of the Previously Published Compounds and the Reference Compounds
a The IC50 values were determined against POP from porcine brain as described in the Experimental Section. b Not published before.

Prolyl Oligopeptidase Inhibitors
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 15 4775
Scheme 1a
a Reagents and conditions: (i) 1. Et3N, pivaloyl chloride/dichloromethane, 0 °C; 2. Et3N, pyrrolidine/dichloromethane. (ii) HCl/EtOAc.
(iii) Succinic acid monomethyl ester pivalic acid anhydride (prepared from succinic acid monomethyl ester and pivaloyl chloride in
dichloromethane at 0 °C), Et3N/dichloromethane. (iv) LiOH/MeOH-H2O. (v) 1. Et3N, pivaloyl chloride/dichloromethane, 0 °C; 2. Et3N,
pyrrolidine or morpholine, or azepane/dichloromethane.
Scheme 2a
a Reagents and conditions: (i) L-Proline methyl ester HCl salt, Et3N/dichloromethane. (ii) 1. Et3N, pivaloyl chloride/dichloromethane,
0 °C; 2. Et3N, pyrrolidine or azepane, or 2(S)-cyanopyrrolidine/dichloromethane. (iii) LiOH/MeOH-H2O. (iv) Azepane/THF. (v) 1. Et3N,
pivaloyl chloride/dichloromethane, 0 °C; 2. Et3N, L-proline methyl ester HCl salt/dichloromethane. (vi) 1. Et3N, ethyl chloroformate/THF,
-10 °C; 2. NH3/H2O. (vii) Et3N, trifluoroacetic anhydride/THF, 0 °C.
was compound 32, where the GOLD-proposed docking
was replaced by the CSCORE-proposed docking. The
binding mode that GOLD proposed to be the best for
compound 32 differed in the P1-P2 area from the rest
of the compounds. To obtain a better mode, a CSCORE
calculation within Sybyl was carried out.
have an aryl-alkanoyl structure, typical to POP inhibi-
tors, at the P3 and beyond P3 positions. Instead, they
have rather hydrophilic P3 and beyond P3 positions, due
to the amide and ketone carbonyl oxygens and the
amide nitrogens. However, their modes of binding do
not seem to differ from that of the typical POP inhibi-
tors. For example, SUAM-1221 (compound 55) occupies
the binding mode A.
The data set consisted of POP inhibitors with varying
inhibitory activities. There was not much structural
variation at the P1 and P2 positions and without
exception, these positions were bound as in the original
crystal structure.27 Most of the variation was at the P3
position and beyond it. However, there is a S3 binding
site for the P3 position, which is occupied by all ligands.
This position is surrounded by the lipophilic residues
ILE591, ALA594, CYS259, and PHE173. Beyond the P3
position, conformational freedom increases and one
consistent binding site was not found. Instead, two
binding modes, which are clearly favored, were found.
Each of them is represented by one inhibitor in Figure
2a. The more populated binding mode is called binding
mode A (compound 15 in Figure 2a). It is occupied by
60% of the ligands, and it is surrounded by GLY254,
CYS255, ARG252, and MET245. The less populated
binding mode is called binding mode B (compound 14
in Figure 2a), and it is occupied by 40% of the ligands.
Both binding modes resulted in equal inhibitory activi-
ties.
CoMSIA. The properties of the CoMSIA model are
in agreement with the known SARs of POP inhibitors
and the 3D structure of the POP enzyme.27 Hydrogen
bond acceptor properties (Figure 2b) are important near
the electrophilic 2(S)-substituent of the P1 pyrrolidine
ring. In the crystal structure, this substituent forms a
hemiacetal bond with SER554 of POP. The thus formed
oxyanion is stabilized by two hydrogen bonds to the
main chain NH group of Asn555 and to the hydroxyl
group of Tyr473.27 The P2 carbonyl oxygen is hydrogen
bonded to ARG643 and the P3 carbonyl oxygen to
Trp595. The P2 hydrogen bond has been proven to be
crucial for the inhibitory activity.32 However, in the
model, these hydrogen bonds seem to have only a small
contribution to the activity. This misinform arises from
the lack of structural variation in these positions within
the data set, and consequently, the real contribution to
the activity cannot be revealed by the model. The
binding mode B includes an area, near ARG128, where
hydrogen bond acceptor properties are also favored. The
formation of the hydrogen bond is not probable due to
The novel inhibitors (Table 1) and most of the previ-
ously synthesized compounds (21-51 in Table 2) do not

4776 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 15
Jarho et al.
Figure 2. Most favored binding modes and the CoMSIA fields. The model compounds 14 and 15 are docked into the active site
of POP. (a) Binding modes A and B are represented with compounds 15 and 14, respectively. ILE591 and ALA594 are in blue,
MET235 is in yellow, and ARG643 is in red. (b) Hydrogen bond acceptor properties. Favored areas are in magenta, and disfavored
areas are in red. (c) Electrostatic properties. The partial negative charge is favored in red areas, and the partial positive charge
is favored in blue areas. (d) Hydrophobic properties. Hydrophobic properties are favored in yellow areas, and hydrophilic properties
are favored in white areas.
internal hydrogen bonds of POP, but the protein surface
is positively charged in this area. The same area also
favors partial negative charge.
The P3 position favors lipophilic aliphatic structures.
Lipophilic aromatic structures with negative π-electrons
are not optimal structures but are tolerated. Beyond the
P3 position, the binding mode A shows a clear hydro-
phobic area. The binding mode B, on the other hand,
favors nonhydrophobic structures near the residue
ARG128.
SARs. Three different azacycles, pyrrolidine, mor-
pholine, and azepane, were studied at the position
beyond P3, resulting in compounds 3, 4, and 5 respec-
tively. A succinic acid residue was located at the P3
position and L-Pro-pyrrolidine at the P2-P1 positions.
The morpholine moiety caused considerably lower po-
tency than the azepane and pyrrolidine moieties. The
replacement by an azepane moiety resulted in a more
potent inhibitor than the replacement by a pyrrolidine
moiety. This difference was also observed in compounds
having a 2(S)-cyanopyrrolidine moiety at the P1 position
(11 and 12). When the flexible succinic acid residue was
changed to the rigid isophthalic acid residue (19 and
20), the azepane and pyrrolidine moieties gave equi-
potent compounds.
The polar oxygen of the morpholine ring is located in
an area that favors lipophilic structures. This explains
the significant decrease in inhibitory activity caused by
morpholine. However, the increased lipophilicity of
azepane does not explain the difference between the IC50
values of compounds 3 and 5. In compound 3, the
carbonyl group next to the pyrrolidine moiety is located
in an area that disfavors hydrogen bond acceptors. The
carbonyl oxygen points toward the enzyme surface,
which lacks hydrogen bond donors to form a bond with,
Hydrogen bond acceptor properties are disfavored in
the area located near the residues that do not have any
hydrogen bond donors, especially the residues ILE591
(blue in Figure 2a), ALA594 (blue in Figure 2a), and
MET235 (yellow in Figure 2a). Any hydrogen bond
acceptor atom that points toward this area is entropi-
cally nonfavorable, since typically all polar atoms of the
bound ligand have a high tendency to interact either
with the receptor residues or with the surrounding
water molecules.
Partial negative charge (Figure 2c) is favorable near
ARG643 (red in Figure 2a), ARG128 and surprisingly
near MET235 (yellow in Figure 2a). While the area near
ARG643 is explained by the P2 carbonyl oxygen of an
inhibitor that forms a hydrogen bond with ARG643, the
area near MET235 does not seem to have a clear
explanation. It may be an artifact arising from the fact
that many potent inhibitors have carbonyl oxygen
located in this area even though the oxygen cannot form
a hydrogen bond with the enzyme. The tendency to favor
partial positive charge over the carbons of the P1 prolyl
ring and partial negative charge over the 2(S)-substitu-
ent (Figure 2c) is easily understood via the polarization
between the substituent and the ring. When this
polarization is strong, the interaction between the
inhibitor and the SER554 becomes stronger.
Hydrophobic properties (Figure 2d) are clearly di-
vided; the hydrophobic nature of the inhibitors is
disfavored at the P1-P2 area, near the residue ARG643.

Prolyl Oligopeptidase Inhibitors
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 15 4777
and this results in an unfavorable entropic effect. In
compound 5, the larger ring size of azepane distorts the
carbonyl oxygen in a different direction, toward an open
cavity. The cavity is large enough to include a water
molecule that may form a hydrogen bond with the
oxygen. In compounds 19 and 20, the rigid isophthalic
acid residue fixes the conformation and the inhibitors
bind in a similar way to the enzyme. The phenyl group
is located in an area that disfavors both lipophilic
structures and negative charge, and this may explain
the lower activity of compounds 19 and 20 as compared
to compound 12.
Three aliphatic chains, succinic acid, glutaric acid,
and 3,3-dimethylglutaric acid, were studied at the P3
position, resulting in compounds 12, 14, and 15, respec-
tively. Azepane was used at the position beyond P3 and
L-Pro-2(S)-cyanopyrrolidine at the P2-P1 positions. The
elongation of the chain from succinic acid to glutaric acid
decreased the potency. 3,3-Dimethyl glutaric acid, on
the other hand, increased the potency slightly.
thesized compounds. The removal of the pyrrolidine-1-
carbonyl group from compounds 21 (flexible P3 moiety)
and 37 (rigid P3 moiety) resulted in compounds 3 and
19, respectively. In both cases, the potency decreased
slightly. The replacement of the L-Pro-pyrrolidine group
beyond the P3 position of compounds 21, 22, and 37 with
azepane resulted in compounds 5, 13, and 20, respec-
tively. Compounds 5 and 13 have flexible P3 moieties,
and in both cases, the replacement increased the inhibi-
tory activity. On the other hand, in the case of compound
20 with a rigid P3 moiety, the inhibitory activity
decreased. Compound 20 is also less potent than the
other large compounds of the same series as compound
37 (38-42). Interestingly, also, the small compound 54,
having only a benzoyl group at the P3 position, was
slightly more active than compound 20. The differences
in the inhibitory activities between the novel compounds
and the previously synthesized larger compounds were
rather small and could not be explained with the
CoMSIA model.
Compound 14 (orange in Figure 2a) seems to have a
different binding mode from compounds 12 and 15
(violet in Figure 2a); the molecule turns toward the open
cavity. However, the slight decrease in potency cannot
be explained with the CoMSIA model. On the other
hand, the increased lipophilicity of compound 15 in
comparison to compound 14 may explain the increased
potency in this case.
Three different substitution patterns of the phenylene
moiety were studied at the P3 position, resulting in
compounds 17, 18, and 19. Pyrrolidine was used at the
position beyond P3 and a L-Pro-2(S)-cyanopyrrolidine
group at the P2-P1 positions. The ortho-substituted
inhibitor 17 had strikingly 3 orders of magnitude lower
activity than the para- and meta-substituted analogues
18 and 19, respectively. The para-substitution resulted
in the best inhibitory activity.
A rigid phenylene ring fixes the conformation, and the
inhibitor cannot adjust itself to a more favorable con-
formation. In compound 17, the phenylene ring is
located in an area that favors hydrophilic structures.
In addition, both carbonyl oxygens of the ortho-phthalic
residue are located in the area that disfavors hydrogen
bond acceptors.
Conclusions
The P1 and P2 positions of the docked ligands were
bound to POP in the same way as Z-L-prolyl-L-prolinal
in the crystal structure. A consistent P3 position was
also found. This position is surrounded by lipophilic
residues, and lipophilic aliphatic structures are favored.
Lipophilic aromatic structures with negative π-electrons
can also be used, but then, the substitution pattern
becomes crucial. Beyond the P3 position, the docked
ligands divided into two different binding modes A and
B, which were occupied by 60 and 40% of the inhibitors,
respectively. According to the previously postulated idea
of the lipophilic cavity, the binding mode A favors
lipophilic structures. The binding mode B instead favors
nonhydrophobic structures, and partial negative charge
is favored near ARG128, where the protein surface has
a partial positive charge. The P3 and beyond P3 area is
sufficiently large, and the interactions between POP and
the ligands are weak. This area can accommodate
several different groups without affecting the inhibitory
activity and could be a valuable modification site for the
optimization of ADME properties. The docking studies
with the 3D QSAR models are helpful in the design of
new compounds.
Pyrrolidine and 2(S)-cyanopyrrolidine moieties were
studied at the P1 position of the novel inhibitors.
Inhibitors containing a 2(S)-cyanopyrrolidine group at
the P1 position are very potent POP inhibitors.35 Also
in this study, the 2(S)-cyanopyrrolidine group increased
the potency significantly as compared to a pyrrolidine
group (3 vs 11, 5 vs 12, 13 vs 14, and 16 vs 17). The
interaction of the cyano group with the cystein protease
papain has been studied. Cystein proteases have a
similar catalytic mechanism to serine proteases. The
cyano group forms a covalent, reversible thioimidate
adduct with the thiol group of the active site cys-
teine.36,37 A similar interaction between the hydroxyl
group of the catalytic SER554 and a cyano group is
possible. Recently, the binding kinetics of compound 37
to POP was studied.38 It was found to be slow binding,
and the inhibition was suggested to be a direct binding
process.
Experimental Section
General Methods. 1H and 13C NMR spectra were recorded
on a Bruker AM 400 WB or a Bruker Avance 500 spectrometer.
Chemical shifts (δ) are given in ppm downfield from the
internal standard tetramethylsilane (δ 0.00). Electrospray
ionization mass spectra (ESI-MS) were obtained on a LCQ ion
trap mass spectrometer equipped with an ESI source (Finni-
gan MAT, San Jose, CA). Elemental analyses (C, H, N) were
carried out with a Thermo Quest CE Instruments EA 1110
CHNS-O elemental analyzer. Flash chromatography was
performed on J. T. Bakers silica gel for chromatography (pore
size 60 Å, particle size 50 µm). The silica plate for the
chromatotron was made of Merck silica gel 60 PF254-containing
gypsum (particle size 5-40 µm).
The N-amide bond of a proline residue or a 2-substituted
pyrrolidine moiety has energetically similar cis and trans
isomers (rotamers). These rotamers have slightly different
shifts. This is best seen in the 13C NMR spectra of compound
16 where two rotamers raise two signals for each carbon of
the compound. Minor rotamers that are less than 20% of the
size of the major rotamers are not reported.
The role of the molecular size was studied, and the
novel inhibitors were compared to the previously syn-

4778 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 15
Jarho et al.
General Synthesis Procedure A. A solution of 1.0 equiv
of pivaloyl chloride in dichloromethane (DCM) was added to
a solution of 1.0 equiv of carboxylic acid and 1.1 equiv of
triethylamine in DCM at 0 °C. The reaction mixture was
stirred at 0 °C for 1 h. A solution of 1.1 equiv of triethylamine
and 1.0-1.1 equiv of amine in DCM was slowly added at 0 °C
(if the amine was in the form of a trifluoroacetic acid salt or
HCl salt, then 3.3 equiv of triethylamine was used and the
triethylamine was added separately before the amine was
added). The reaction mixture was stirred 2 h or overnight at
room temperature. The DCM solution was washed with 30%
citric acid (aq), saturated NaCl (aq), and saturated NaHCO3
(aq), dried over anhydrous Na2SO4, and evaporated, yielding
the crude product.
General Synthesis Procedure B. A 1.0-1.5 equiv amount
of LiOH‚H2O was added to a solution of 1.0 equiv of carboxylic
acid methyl ester in 20-30% water in methanol. The reaction
was stirred 2 h or overnight at room temperature. The solvent
(methanol) was evaporated, and the residue was dissolved in
water. The aqueous phase was washed with DCM. The
aqueous phase was made acidic with 2-3 M HCl (aq), and the
product was extracted with DCM. The organic phase was dried
over anhydrous Na2SO4 and evaporated, yielding the product.
General Synthesis Procedure C. A 2-4 mL amount of
ethyl acetate saturated with dry HCl was added to 1.0 mmol
of BOC-protected amine at room temperature. The reaction
mixture was stirred at room temperature for 30 min. Another
way to proceed was to dissolve 1.0 mmol of BOC-protected
amine in 5-10 mL of DCM and add 2-4 mL of trifluoroacetic
acid at 0 °C. The reaction was stirred at 0 °C for 2-2.5 h. In
both cases, the solvent was removed and the product was
finally evaporated in vacuo, yielding the corresponding amine
HCl salt or TFA salt.
Succinic Acid Azepane L-Prolyl-pyrrolidine Amide (5).
The methyl ester group of 2 (560 mg, 2.0 mmol) was hydrolyzed
using LiOH‚H2O (84 mg, 2.0 mmol) in 10 mL of 20% water in
MeOH at room temperature. After 3 h, the solvent was
evaporated and the residue was dissolved in DCM. The organic
phase was dried and evaporated. The obtained product was
coupled with azepane (0.23 mL, 2.0 mmol) according to
procedure A. The product was purified by column chromatog-
1
raphy (20% MeOH in EtOAc). Yield: 350 mg, 50%. H NMR
(500.1 MHz, CDCl3): δ 1.51-1.61 (4 H, m), 1.64-1.77 (4 H,
m), 1.81-2.45 (8 H, m), 2.51-2.64 (2 H, m), 2.80-2.88 (2 H,
m), 3.36-3.90 (10 H, m), 4.65 (0.9 H, dd, J ) 3.8 Hz, 8.4 Hz),
4.71 (0.1 H, dd, J ) 3.1 Hz, 8.6 Hz). 13C NMR (125.1 MHz,
CDCl3): δ 24.1, 24.8, 26.2, 26.9, 27.1, 27.6, 28.0, 28.9, 29.0,
29.6, 45.9, 46.0, 46.3, 47.2, 47.7, 57.8, 170.7, 170.9, 171.6. MS
(ESI, +) m/z 350 [M + H]+. Anal. (C19H31N3O3‚0.2H2O) C, H,
N.
Phthalic Acid Mono-(L-proline methyl ester) Amide
(6). Phthalic acid anhydride (7.4 g, 45 mmol) was added to a
solution of L-proline methyl ester HCl salt (7.4 g, 45 mmol)
and triethylamine (13.8 mL, 99 mmol) in DCM at 0 °C. The
reaction mixture was stirred for 4 h at room temperature. The
organic phase was extracted with saturated NaHCO3 (aq). The
aqueous phase was acidified with 3 M HCl (aq) and extracted
with chloroform. The chloroform phase was dried and evapo-
rated. Yield: 6.53 g, 51%.
Succinic Acid Monopyrrolidine Amide (7a). Succinic
acid monomethyl ester (1.98 g, 15 mmol) and pyrrolidine (1.3
mL, 15 mmol) were coupled according to procedure A. The
product was purified by column chromatography (EtOAc).
Yield: 1.75 g, 63%. The methyl ester group of the product was
hydrolyzed using LiOH‚H2O (0.44 g, 10.4 mmol) according to
procedure B. Yield: 1.37 g, 85%.
Isophthalic Acid Monopyrrolidine Amide (7b). Iso-
phthalic acid monomethyl ester (2.25 g, 12.5 mmol) and
pyrrolidine (1.04 mL, 12.5 mmol) were coupled according to
procedure A. The product was purified by column chromatog-
raphy (30% petroleum ether in EtOAc). Yield: 1.75 g, 60%.
The methyl ester group of the product was hydrolyzed using
LiOH‚H2O (470 mg, 11.3 mmol) according to procedure B.
Yield: 1.49 g, 91%.
Terephthalic Acid Monopyrrolidine Amide (7c). Tere-
phthalic acid monomethyl ester (1.08 g, 6.0 mmol) and pyr-
rolidine (0.50 mL, 6.0 mmol) were coupled according to
procedure A. The product was purified by column chromatog-
raphy (EtOAc). Yield: 0.50 g, 36%. The methyl ester group of
the product was hydrolyzed using LiOH‚H2O (135 mg, 3.2
mmol) according to procedure B. Yield: 0.41 g, 87%.
Succinic Acid Monoazepane Amide (8a). Succinic acid
monomethyl ester (1.98 g, 15 mmol) and azepane (1.7 mL, 15
mmol) were coupled according to procedure A. The product was
purified by column chromatography (30% petroleum ether in
EtOAc). Yield: 2.48 g, 78%. The methyl ester group of the
product was hydrolyzed using LiOH‚H2O (540 mg, 12.9 mmol)
according to procedure B. Yield: 2.34 g, 100%.
Isophthalic Acid Monoazepane Amide (8b). Isophthalic
acid monomethyl ester (2.25 g, 12.5 mmol) and azepane (1.41
mL, 12.5 mmol) were coupled according to procedure A. The
product was purified by column chromatography (40% petro-
leum ether in EtOAc). Yield: 2.02 g, 62%. The methyl ester
group of the product was hydrolyzed using LiOH‚H2O (490 mg,
11.7 mmol) according to procedure B. Yield: 1.89 g, 99%.
Glutaric Acid Monoazepane Amide (8c). To a solution
of azepane (4.5 mL, 40 mmol) in 40 mL of THF was added
glutaric anhydride (2.3 g, 20 mmol) in 15 mL of THF dropwise,
and the solution was stirred for 2 days. The reaction mixture
was evaporated, dissolved in 0.025 M NaOH (aq), and washed
with DCM. The aqueous phase was acidified with 3 M HCl,
and the product was extracted with DCM. The combined
organic phases were dried and evaporated. Yield: 1.61 g, 38%.
3,3-Dimethylglutaric Acid Monoazepane Amide (8d).
To a solution of azepane (1.74 mL, 15.4 mmol) in 10 mL of
THF was added 3,3-dimethylglutaric anhydride (1.0 g, 7 mmol)
in 10 mL of THF dropwise at 0 °C. The reaction mixture was
BOC-L-prolyl-pyrrolidine (1). BOC-L-proline (7.97 g, 37.0
mmol) and pyrrolidine (3.09 mL, 37.0 mmol) were coupled
according to procedure A. The product was purified by column
chromatography (2.5-5% MeOH in DCM). Yield: 8.31 g, 84%.
Succinic Acid Monomethyl Ester L-Prolyl-pyrrolidine
Amide (2). BOC-L-prolyl-pyrrolidine (1.98 g, 7.4 mmol) was
deprotected using 30 mL of HCl-saturated ethyl acetate
according to procedure C. Succinic acid monomethyl ester (0.98
g, 7.4 mmol) and the L-prolyl-pyrrolidine HCl salt were coupled
according to procedure A. The product was purified by column
chromatography (5-10% MeOH in EtOAc). Yield: 1.81 g, 86%.
Succinic Acid Pyrrolidine L-Prolyl-pyrrolidine Amide
(3). The methyl ester group of 2 (750 mg, 2.7 mmol) was
hydrolyzed using LiOH‚H2O (170 mg, 4.1 mmol) according to
procedure B. Yield: 360 mg, 1.3 mmol, 45%. The obtained
product was coupled with pyrrolidine (0.11 mL, 1.3 mmol)
according to procedure A. The product was purified by column
chromatography (25% MeOH in EtOAc). Yield: 170 mg, 41%.
1H NMR (500.1 MHz, CDCl3): δ 1.79-2.18 (10 H, m), 2.23-
2.45 (2 H, m), 2.47-2.68 (2 H, m), 2.78-2.92 (2 H, m), 3.39-
3.62 (7 H, m), 3.64-3.84 (3 H, m), 4.68 (dd, 0.9 H, J ) 3.8 Hz,
8.4), 4.77 (dd, 0.1 H, J ) 3.0 Hz, 8.8 Hz). 13C NMR (125.1 MHz,
CDCl3): δ 24.1, 24.4, 24.8, 26.0, 26.2, 28.9, 29.2, 29.3, 45.7,
45.9, 46.2, 46.4, 47.3, 57.8, 170.5, 170.7, 170.8. MS (ESI, +)
m/z 322 [M + H]+. Anal. (C17H27N3O3) C, H, N.
Succinic Acid Morpholine L-Prolyl-pyrrolidine Amide
(4). The methyl ester group of 2 (560 mg, 2.0 mmol) was
hydrolyzed using LiOH‚H2O (84 mg, 2.0 mmol) in 10 mL of
20% water in MeOH at room temperature. After 4 h, the
solvent was evaporated and the residue was dissolved in DCM.
The organic phase was dried and evaporated. The obtained
product was coupled with morpholine (0.18 mL, 2.0 mmol)
according to procedure A. The product was purified by column
chromatography (20-25% MeOH in EtOAc). Yield: 330 mg,
49%. 1H NMR (400.1 MHz, CDCl3): δ 1.80-2.05 (6 H, m),
2.08-2.35 (2 H, m), 2.39-2.64 (2 H, m), 2.78-2.95 (2 H, m),
3.35-3.81 (14 H, m), 4.64 (0.9 H, dd, J ) 3.8 Hz, 8.3 Hz), 4.69
(0.1 H, dd, J ) 3.8 Hz, 8.5 Hz). 13C NMR (125.1 MHz, CDCl3):
δ 24.1, 24.8, 26.2, 27.7, 28.9, 29.3, 42.2, 45.8, 46.1, 46.4, 47.4,
58.0, 66.6, 66.9, 170.9, 171.0, 171.1. MS (ESI, +) m/z 338 [M
+ H]+. Anal. (C17H27N3O4‚0.3H2O) C, H, N.

Prolyl Oligopeptidase Inhibitors
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 15 4779
stirred overnight at room temperature, evaporated, and dis-
solved in 0.025 M NaOH (aq). The water phase was washed
with DCM, acidified with 3 M HCl (aq), and extracted with
DCM. The organic phase was dried and evaporated, yielding
solid product: 1.53 g, 90%.
4.78-4.82 (1 H, m). 13C NMR (125.1 MHz, CDCl3): δ 24.4, 24.8,
25.4, 26.0, 27.7, 29.1, 29.3, 29.7, 45.7, 46.3, 46.4, 46.5, 47.3,
57.5, 118.6, 170.3, 171.2, 171.4. MS (ESI, +) m/z 369 [M +
Na]+. Anal. (C18H26N4O3‚0.1H2O) C, H, N: calcd, 16.09; found,
15.57.
Succinic Acid Pyrrolidine L-Proline Amide (9a). Com-
pound 7a (1.37 g, 8.0 mmol) was coupled with proline methyl
ester HCl salt (1.3 g, 8.0 mmol) according to procedure A. The
product was purified by column chromatography (5% MeOH
in DCM). Yield: 1.45 g, 64%. The methyl ester group of the
product was hydrolyzed using LiOH‚H2O (0.32 g, 7.65 mmol)
according to procedure B. Yield: 1.32 g, 96%.
Isophthalic Acid Pyrrolidine L-Proline Amide (9b).
Compound 7b (1.49 g, 6.8 mmol) and proline methyl ester HCl
salt (1.13 g, 6.8 mmol) were coupled according to procedure
A. The product was purified by column chromatography (10%
MeOH in EtOAc). Yield: 1.83 g, 81%. The methyl ester group
of the product was hydrolyzed using LiOH‚H2O (350 mg, 8.3
mmol) according to procedure B. Yield: 1.58 g, 91%.
Terephthalic Acid Pyrrolidine L-Proline Amide (9c).
Compound 7c (0.41 g, 1.9 mmol) and l-proline methyl ester
HCl salt (0.31 g, 1.9 mmol) were coupled according to proce-
dure A. The product was purified by column chromatography
(10% MeOH in EtOAc). Yield: 0.54 g, 87%. The methyl ester
group of the product was hydrolyzed using LiOH‚H2O (103 mg,
2.5 mmol) according to procedure B. Yield: 0.50 g, 97%.
Phthalic Acid Pyrrolidine L-Proline Amide (9d). Com-
pound 6 (4.25 g, 17.0 mmol) and pyrrolidine (1.6 mL, 18.7
mmol) were coupled according to procedure A. The product was
purified by column chromatography (5-20% MeOH in EtOAc).
Yield: 4.77 g, 85%. The methyl ester group of the product was
hydrolyzed using LiOH‚H2O (906 mg, 21.6 mmol) according
to procedure B. Yield: 3.42 g, 75%.
Succinic Acid Azepane L-Prolyl-2(S)-cyanopyrrolidine
Amide (12). BOC-2(S)-cyanopyrrolidine (1.3 g, 6.8 mmol) was
deprotected using 27 mL of trifluoroacetic acid in 70 mL of
DCM according to procedure C. Compound 10a (2.01 g, 6.8
mmol) and the 2(S)-cyanopyrrolidine trifluoroacetic acid salt
were coupled according to procedure A. The product was
purified by column chromatography (8-10% MeOH in EtOAc).
1
Yield: 320 mg, 13%. H NMR (400.1 MHz, CDCl3): δ 1.47-
1.59 (4 H, m), 1.62-1.76 (4 H, m), 1.92-2.06 (2 H, m), 2.10-
2.33 (6 H, m), 2.45-2.58 (2 H, m), 2.76-2.89 (2 H, m), 3.35-
3.87 (8 H, m), 4.58 (1 H, dd, J ) 3.7 Hz, 8.1 Hz), 4.80-4.83 (1
H, m). 13C NMR (125.1 MHz, CDCl3): δ 24.8, 25.4, 26.9, 27.1,
27.6, 27.9, 28.7, 29.0, 29.5, 29.7, 46.0, 46.3, 46.5, 47.3, 47.7,
57.5, 118.6, 171.2, 171.3, 171.4. MS (ESI, +) m/z 375 [M +
H]+. Anal. (C20H30N4O3‚0.2H2O) C, H, N.
Glutaric Acid Azepane L-Prolyl-pyrrolidine Amide
(13). Compound 10c (0.51 g, 1.6 mmol) was coupled with
pyrrolidine (0.14 mL, 1.7 mmol) according to procedure A. The
product was purified by column chromatography (15% MeOH
in EtOAc) to give the product as colorless oil. Yield: 0.34 g,
57%. 1H NMR (500.1 MHz, CDCl3): δ 1.51-1.58 (4 H, m),
1.65-1.74 (4 H, m), 1.78-2.29 (10 H, m), 2.31-2.48 (4 H, m),
3.35-3.72 (9 H, m), 3.80-3.85 (1 H, m), 4.57 (0.1 H, dd, J )
2.8 Hz, 8.8 Hz), 4.64 (0.9 H, dd, J ) 3.8 Hz, 8.4 Hz). 13C NMR
(125.1 MHz, CDCl3): δ 20.4, 24.1, 24.9, 26.2, 26.9, 27.2, 27.7,
28.9, 29.2 32.3, 33.8, 45.9, 45.9, 46.3, 47.4, 47.9, 57.7, 170.7,
171.4, 172.3. MS (ESI, +) m/z 364 [M + H]+. Anal. (C20H33N3O3‚
0.1H2O) C, H, N.
Glutaric Acid Azepane L-Prolyl-L-proline Amide. Com-
pound 10c (0.38 g, 1.2 mmol) and l-proline methyl ester HCl
salt (0.20 g, 1.2 mmol) were coupled according to procedure
A. The product was purified by column chromatography (10-
20% MeOH in EtOAc) to give the product as oil. Yield: 0.42
g, 82%. The methyl ester group of the obtained product was
hydrolyzed using LiOH‚H2O (63 mg, 1.5 mmol) according to
procedure B to give the white solid product. Yield: 0.39 g, 95%.
Glutaric Acid Azepane L-Prolyl-L-prolinamide Amide.
To a solution of glutaric acid azepane l-prolyl-L-proline amide
(170 mg, 0.42 mmol) and Et3N (59 µL, 0.42 mmol) in 8 mL of
THF was added ethyl chloroformate (40 µL, 0.42 mmol) in 4
mL of THF dropwise at -10 °C. After 20 min, 25% NH3 (aq)
(143 µL, 2.1 mmol) was added and the reaction mixture was
stirred overnight at room temperature. The reaction mixture
was evaporated, dissolved in DCM, and filtered. The filtrate
was washed with saturated NaHCO3 (aq), dried, and evapo-
rated yielding the white solid product (150 mg, 88%).
Glutaric Acid Azepane L-Prolyl-2(S)-cyanopyrrolidine
Amide (14). To a solution of glutaric acid azepane L-prolyl-
L-prolinamide amide (150 mg, 0.37 mmol) and Et3N (155 µL,
1.11 mmol) in 8 mL of THF was added trifluoroacetic anhy-
dride (80 µL, 0.56 mmol) in 4 mL of THF dropwise at 0 °C.
The reaction mixture was stirred 2 h at room temperature,
evaporated, and dissolved in DCM. The organic phase was
washed with 30% citric acid (aq), dried, and evaporated. The
product was purified by column chromatography (10% MeOH
in EtOAc) yielding colorless oil (76 mg, 53%). 1H NMR (500.1
MHz, CDCl3): δ 1.52-1.59 (4 H, m), 1.66-1.73 (4 H, m), 1.90-
2.06 (4 H, m), 2.09-2.31 (6 H, m), 2.35-2.47 (4 H, m), 3.39-
3.44 (2 H, m), 3.45-3.64 (4 H, m), 3.66-3.71 (1 H, m), 3.85-
3.90 (1 H, m), 4.55 (0.9 H, dd, J ) 4.2 Hz, 8.6 Hz), 4.60-4.62
(0.1 H, m), 4.65-4.68 (0.1 H, m), 4.79-4.81 (0.9 H, m). 13C
NMR (125.1 MHz, CDCl3): δ 20.3, 25.0, 25.4, 26.9, 27.2, 27.7,
28.8, 29.1, 29.7, 32.1, 33.7, 45.9, 46.3, 46.5, 47.4, 47.8, 57.4,
118.6, 171.4, 171.7, 172.1. MS (ESI, +) m/z 389 [M + H]+. Anal.
(C21H32N4O3‚0.2H2O) C, H, N.
Succinic Acid Azepane L-Proline Amide (10a). Com-
pound 8a (2.34 g, 11.7 mmol) was coupled with proline methyl
ester HCl salt (1.9 g, 11.7 mmol) according to procedure A.
The product was purified by column chromatography (5%
MeOH in DCM). Yield: 2.22 g, 61%. The methyl ester group
of the product was hydrolyzed using LiOH‚H2O (0.45 g, 10.7
mmol) according to procedure B. Yield: 2.01 g, 95%.
Isophthalic Acid Azepane L-Proline Amide (10b). Com-
pound 8b (1.89 g, 7.6 mmol) and proline methyl ester HCl salt
(1.27 g, 7.6 mmol) were coupled according to procedure A. The
product was purified by column chromatography (10% MeOH
in EtOAc). Yield: 2.08 g, 76%. The methyl ester group of the
product was hydrolyzed using LiOH‚H2O (370 mg, 8.7 mmol)
according to procedure B. Yield: 1.67 g, 83%.
Glutaric Acid Azepane L-Proline Amide (10c). Com-
pound 8c (1.61 g, 7.5 mmol) and L-proline methyl ester HCl
salt (1.24 g, 7.5 mmol) were coupled according to procedure
A. The product was purified by column chromatography (5-
10% MeOH in EtOAc) to give the product as colorless oil.
Yield: 1.8 g, 74%. The methyl ester group of the product (1.8
g, 5.6 mmol) was hydrolyzed using LiOH‚H2O (0.35 g, 8.3
mmol) according to procedure B. Yield: 1.55 g, 90%.
3,3-Dimethylglutaric Acid Azepane L-Proline Amide
(10d). Compound 8d (0.50 g, 2.1 mmol) and L-proline methyl
ester HCl salt (0.34 g, 2.1 mmol) were coupled according to
procedure A. The product was purified by column chromatog-
raphy (25-50% EtOAc in petroleum ether). Yield: 0.55 g, 75%.
The methyl ester group of the product (0.55 g, 1.6 mmol) was
hydrolyzed using LiOH‚H2O (0.10 g, 2.4 mmol) according to
procedure B. Yield: 0.52 g, 96%.
Succinic Acid Pyrrolidine L-Prolyl-2(S)-cyanopyrroli-
dine Amide (11). BOC-2(S)-cyanopyrrolidine (0.96 g, 4.9
mmol) was deprotected using 20 mL of trifluoroacetic acid in
50 mL of DCM according to procedure C. 2(S)-cyanopyrrolidine
TFA salt was coupled with 9a (1.32 g, 4.9 mmol) according to
procedure A. The product was purified by column chromatog-
1
raphy (10% MeOH in EtOAc). Yield: 170 mg, 10%. H NMR
(500.1 MHz, CDCl3): δ 1.81-1.86 (2 H, m), 1.91-2.05 (4 H,
m), 2.12-2.32 (6 H, m), 2.42-2.54 (2 H, m), 2.73-2.88 (2 H,
m), 3.36-3.48 (4 H, m), 3.60-3.63 (1 H, m), 3.66-3.75 (2 H,
m), 3.82-3.86 (1 H, m), 4.59 (1 H, dd, J ) 4.0 Hz, 8.4 Hz),
3,3-Dimethylglutaric Acid Azepane L-Prolyl-L-proline
Amide. Compound 10d (0.52 g, 1.5 mmol) and L-proline
methyl ester HCl salt (0.26 g, 1.5 mmol) were coupled
according to procedure A. The product was purified by column

4780 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 15
Jarho et al.
chromatography (10% MeOH in EtOAc). Yield: 0.54 g, 78%.
The methyl ester group of the obtained product was hydrolyzed
using LiOH‚H2O (76 mg, 1.8 mmol) according to procedure B.
Yield: 0.54 g, 100%.
3,3-Dimethylglutaric Acid Azepane L-Prolyl-L-prolin-
amide Amide. To a solution of 3,3-dimethylglutaric acid
azepane L-prolyl-L-proline amide (0.54 g, 1.2 mmol) and Et3N
(0.17 mL, 1.2 mmol) in THF was added ethyl chloroformate
(0.12 mL, 1.3 mmol) in THF dropwise at -10 °C. After 20 min,
25% NH3 (aq) (84 µL, 1.2 mmol) was added and the reaction
mixture was allowed to warm to room temperature and stirred
overnight. The reaction mixture was evaporated, dissolved in
DCM, and washed with saturated NaHCO3 (aq). The organic
phase was dried and evaporated. Yield: 450 mg, 84%.
group of the product was hydrolyzed using LiOH‚H2O (69 mg,
1.6 mmol) according to procedure B. Yield: 0.43 g, 95%.
Terephthalic Acid Pyrrolidine L-Prolyl-L-prolinamide
Amide. The compound was prepared from terephthalic acid
pyrrolidine L-prolyl-L-proline amide (0.43 g, 1.0 mmol) and 25%
NH3 (aq) (71 µL, 1.0 mmol) according to the method for
preparation of 3,3-dimethylglutaric acid azepane l-prolyl-L-
prolinamide amide. The product was purified by column
chromatography (10-30% MeOH in EtOAc). Yield: 260 mg,
61%.
Terephthalic Acid Pyrrolidine L-Prolyl-2(S)-cyano-
pyrrolidine Amide (18). Compound 18 was prepared from
terephthalic acid pyrrolidine L-prolyl-L-prolinamide amide
(0.26 g, 0.63 mmol) according to the method for the preparation
of 15. The reaction time was 2 h. The product was purified by
column chromatography (30-50% ACN in EtOAc), yielding
3,3-Dimethylglutaric Acid Azepane L-Prolyl-2(S)-cyano-
pyrrolidine Amide (15). To a solution of 3,3-dimethylglutaric
acid azepane L-prolyl-L-prolinamide amide (0.45 g, 1.0 mmol)
and Et3N (0.43 mL, 3.1 mmol) in THF was added trifluoro-
acetic anhydride (0.22 mL, 1.6 mmol) in THF dropwise at 0
°C. The reaction mixture was stirred for 6 h at room temper-
ature, and the reaction was then quenched with 5 mL of water.
The reaction mixture was evaporated, dissolved in DCM, and
washed with 30% citric acid (aq), saturated NaCl (aq), and
saturated NaHCO3 (aq). The organic phase was dried and
evaporated. The product was purified by column chromatog-
raphy (30-50% ACN in EtOAc), yielding white solid: 250 mg,
58%. 1H NMR (500.1 MHz, CDCl3): δ 1.17 (3 H, s), 1.18 (3 H,
s), 1.51-1.58 (4 H, m), 1.67-1.73 (4 H, m), 1.90-1.98 (2 H,
m), 2.09-2.28 (6 H, m), 2.35-2.64 (4 H, m), 3.40-3.75 (7 H,
m), 3.82-3.87 (1 H, m), 4.56 (0.9 H, dd, J ) 4.3 Hz, 8.4 Hz),
4.73-4.86 (1.1 H, m). 13C NMR (125.1 MHz, CDCl3): δ 25.0,
25.4, 26.8, 27.0, 27.7, 28.5, 28.6, 28.7, 29.1, 29.7, 34.0, 42.1,
43.8, 45.9, 46.2, 46.5, 48.1, 48.4, 57.4, 118.7, 171.1, 171.4, 171.5.
MS (ESI, +) m/z 417 [M + H]+. Anal. (C23H36N4O3‚0.1H2O) C,
H, N.
Phthalic Acid Pyrrolidine L-Prolyl-pyrrolidine Amide
(16). Compound 9d (0.90 g, 2.8 mmol) was coupled with
pyrrolidine (0.26 mL, 3.1 mmol) according to procedure A. The
product was purified by column chromatography (10% MeOH
in EtOAc). Yield: 860 mg, 83%. 1H NMR (500.1 MHz, CDCl3):
δ 1.49-1.58 (1 H, m), 1.65-1.74 (1 H, m), 1.81-2.06 (8 H, m),
2.14-2.29 (1.5 H, m), 2.48-2.52 (0.5 H, m) 3.00-3.05 (0.5 H,
m), 3.14-3.89 (9.5 H, m), 4.69 (0.5 H, dd, J ) 3.3 Hz, 8.2 Hz),
4.79 (0.5 H, dd, J ) 6.2 Hz, 8.3 Hz), 7.29-7.42 (3.5 H, m),
7.51-7.55 (0.5 H, m). 13C NMR (125.1 MHz, CDCl3) δ 23.3,
24.0, 24.2, 24.5, 24.6, 25.2, 25.9, 25.9, 26.0, 26.2, 29.2, 30.5,
45.6, 45.6, 45.7, 45.7, 46.0, 46.3, 46.8, 48.8, 48.9, 49.8, 57.6,
59.4, 125.7, 126.4, 127.4, 127.5, 128.5, 128.7, 128.9, 128.9,
135.2, 135.6, 135.7, 135.8, 168.6, 168.7, 168.8, 169.0, 170.3,
171.1. MS (ESI, +) m/z 370 [M + H]+. Anal. (C21H27N3O3‚
0.2H2O) C, H, N.
1
yellowish solid: 145 mg, 59%. H NMR (500.1 MHz, CDCl3):
δ 1.86-2.33 (12 H, m), 3.22-3.26 (0.1 H, m), 3.33 (0.2 H, t, J
) 6.6 Hz), 3.38 (1.8 H, t, J ) 6.6 Hz), 3.50-3.55 (1 H, m),
3.60-3.91 (4.0 H, m), 3.70-4.02 (0.9 H, m), 4.27-4.29 (0.1 H,
m), 4.39-4.41 (0.1 H, m), 4.75 (0.9 H, dd, J ) 6.1 Hz, 8.0 Hz),
4.85-4.87 (0.9 H, m), 7.37-7.39 (0.2 H, m), 7.43-7.45 (0.2 H,
m), 7.54-7.55 (1.8 H, m), 7.58-7.60 (1.8 H, m). 13C NMR (125.1
MHz, CDCl3): δ 24.4, 25.4, 25.6, 26.4, 29.0, 29.8, 46.2, 46.5,
46.6, 49.5, 50.2, 58.0, 118.6, 127.1, 127.3, 137.3, 139.0, 168.8,
168.9, 171.1. MS (ESI, +) m/z 395 [M + H]+. Anal. (C22H26N4O3‚
0.5H2O) C, H, N.
Isophthalic Acid Pyrrolidine L-Prolyl-2(S)-cyano-
pyrrolidine Amide (19). BOC-2(S)-cyanopyrrolidine (980 mg,
5.0 mmol) was deprotected using 20 mL of trifluoroacetic acid
in 50 mL of DCM according to procedure C. Compound 9b (1.58
g, 5.0 mmol) and the 2(S)-cyanopyrrolidine trifluoroacetic acid
salt were coupled according to procedure A. The product was
purified by column chromatography (1.5-3% MeOH in DCM).
1
Yield: 730 mg, 37%. H NMR (500.1 MHz, CDCl3): δ 1.82-
2.13 (7 H, m), 2.15-2.32 (5 H, m), 3.21-3.43 (2 H, m), 3.53-
3.90 (5 H, m), 3.98-4.02 (1 H, m), 4.25-4.27 (0.2 H, m), 4.55-
4.57 (0.2 H, m), 4.75 (0.8 H, dd, J ) 6.3 Hz, 8.1 Hz), 4.85-
4.87 (0.8 H, m), 7.43-7.47 (1 H, m), 7.58-7.62 (2 H, m), 7.68-
7.69 (1 H, m). 13C NMR (100.6 MHz, CDCl3): δ 24.5, 25.4, 25.7,
26.4, 29.0, 29.8, 46.2, 46.5, 46.6, 49.6, 50.3, 58.0, 118.6, 125.8,
128.5, 128.6, 129.0, 136.1, 137.5, 168.8, 168.8, 171.1. MS (ESI,
+) m/z 395 [M + H]+. Anal. (C22H26N4O3‚0.2H2O) C, H, N.
Isophthalic Acid Azepane L-Prolyl-2(S)-cyanopyrroli-
dine Amide (20). BOC-2(S)-cyanopyrrolidine (940 mg, 4.8
mmol) was deprotected using 19 mL of trifluoroacetic acid in
50 mL of DCM according to procedure C. Compound 10b (1.67
g, 4.8 mmol) and the 2(S)-cyanopyrrolidine trifluoroacetic acid
salt were coupled according to procedure A. The product was
purified by column chromatography (1.5-2.5% MeOH in
1
DCM). Yield: 920 mg, 45%. H NMR (500.1 MHz, CDCl3): δ
1.53-1.66 (6 H, m), 1.78-2.32 (10 H, m), 3.21-3.35 (2 H, m),
3.53-3.90 (5 H, m), 3.97-4.02 (1 H, m), 4.24-4.26 (0.2 H, m),
4.53 (0.2 H, dd, J ) 3.3 Hz, 6.8 Hz), 4.74 (0.8 H, dd, J ) 6.4
Hz, 7.9 Hz), 4.85-4.87 (0.8 H, m), 7.42-7.45 (2 H, m), 7.56-
7.60 (2 H, m). 13C NMR (100.6 MHz, CDCl3): δ 25.4, 25.7, 26.5,
27.2, 27.8, 29.0, 29.4, 29.8, 46.3, 46.5, 46.6, 49.8, 50.3, 58.0,
118.6, 125.3, 127.9, 128.3, 128.6, 136.2, 137.6, 168.8, 170.6,
171.1. MS (ESI, +) m/z 423 [M + H]+. Anal. (C24H30N4O3‚
0.1H2O) C, H, N.
Phthalic Acid Pyrrolidine L-Prolyl-2(S)-cyanopyrroli-
dine Amide (17). BOC-2(S)-cyanopyrrolidine (980 mg, 5.0
mmol) was deprotected using 10 mL of trifluoroacetic acid in
50 mL of DCM according to procedure C. Compound 9d (1.58
g, 5.0 mmol) and the 2(S)-cyanopyrrolidine trifluoroacetic acid
salt were coupled according to procedure A. The product was
purified with a silica plate chromatotron (2% MeOH in DCM).
1
Yield: 250 mg, 13%. H NMR (400.1 MHz, CDCl3): δ 1.65-
2.12 (8 H, m), 2.15-2.33 (4 H, m), 3.11-3.37 (2 H, m), 3.42-
3.73 (5 H, m), 3.79-3.94 (1 H, m), 4.43-4.45 (0.3 H, m), 4.72
(0.7 H, dd, J ) 6.3 Hz, 8.3 Hz), 4.77 (0.3 H, dd, J ) 3.4 Hz, 8.2
Hz), 4.81-4.84 (0.7 H, m), 7.26-7.33 (1 H, m), 7.34-7.44 (2
H, m), 7.46-7.51 (1 H, m). 13C NMR (100.6 MHz, CDCl3): δ
24.6, 25.3, 25.4, 26.0, 29.2, 29.7, 45.7, 46.3, 46.6, 49.0, 49.7,
57.4, 118.6, 126.5, 127.2, 129.1, 129.1, 135.4, 135.7, 168.6,
168.9, 171.0. MS (ESI, +) m/z 395 [M + H]+. Anal. (C22H26N4O3‚
0.2H2O) C, H, N.
TerephthalicAcidPyrrolidineL-Prolyl-L-prolineAmide.
Compound 9c (0.50 g, 1.6 mmol) and l-proline methyl ester
HCl salt (0.27 g, 1.6 mmol) were coupled according to proce-
dure A. The product was purified by column chromatography
(5-10% MeOH in EtOAc). Yield: 0.45 g, 66%. The methyl ester
In Vitro Assay of POP Inhibitory Activity. The whole
porcine brains, excluding cerebellum and most of the brain
stem, were placed in liquid nitrogen within 30 min after the
animals were killed and stored at -80 °C until homogenized.
The brains were homogenized in 3 volumes (w/v) of ice-cold
0.1 M sodium-potassium phosphate buffer (pH 7.0), and the
homogenates were centrifuged for 20 min at 4 °C at 10000g.
The supernatants were pooled and stored in small aliquots at
-80 °C until used. The supernatant was thawed in ice just
before it was used in the activity assay and diluted in a ratio
1:2 with homogenization buffer.
Inhibitors were dissolved in DMSO to 0.01 M, and further
dilutions were made in assay buffer. The highest final DMSO
concentration in all but one case (compound 16) was 0.1% or

Prolyl Oligopeptidase Inhibitors
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 15 4781
below (at inhibitor concentration of 10 µM), which caused
negligible POP inhibition.39 For compound 16, the highest final
DMSO concentration was 1% (at inhibitor concentration of 100
µM), which inhibits POP by 40%.39 Because the measured IC50
value was 19 µM, it can be seen that the inhibition was caused
by the investigated compound and not by DMSO.
In the microplate assay, 10 µL of the enzyme preparation
was preincubated with 460 µL of 0.1 M sodium-potassium
phosphate buffer (pH 7.0) and 5 µL of a solution of the
compound dissolved in DMSO and diluted with 0.1 M sodium-
potassium phosphate buffer at 30 °C for 30 min. The controls
contained 10 µL of enzyme preparation and 465 µL of 0.1 M
sodium-potassium phosphate buffer (pH 7.0). The reaction
was initiated by adding 25 µL of 4 mM Suc-Gly-Pro-7-amido-
4-methylcoumarin dissolved in 0.1 M sodium-potassium
phosphate buffer (pH 7.0), and the mixture was incubated at
30 °C for 60 min. The reaction was terminated by adding 500
µL of 1 M sodium acetate buffer (pH 4.2).
The formation of 7-amido-4-methylcoumarin was deter-
mined fluorometrically with microplate fluorescence reader
(excitation at 360 nm and emission at 460 nm). The final
concentration of the compounds in the assay mixture varied
from 10-12 to 10-4 M.
The inhibitory activities (percent of control) were plotted
against the log concentration of the compound, and the IC50
value was determined by nonlinear regression utilizing Graph-
Pad Prism 3.0 software.
Molecular Modeling. Unless noted otherwise, molecular
modeling was performed using Sybyl 6.9.0 and 6.9.140 with
default options. To prepare the enzyme structure, the POP
structure was derived from PDB (ID code 1QFS, resolution
2.0 Å). Prior to docking, the cocrystallized inhibitor was
removed from the structure, and the hydrogen atoms were
added using Sybyl Biopolymer module. Because the X-ray
structure may also contain energetic tensions, an energetically
favorable conformation of the suggested active site was
searched by carrying out a molecular mechanics minimization
using the MMFF9441 force field. Nitrogen NE2 of histidine 680
was not protonated while ND1 was protonated thus allowing
interaction with ASP641.
The inhibitor structures (total number 52) were created
using automatic translation from in-house ISIS database to
mol2-files, using CONCORD42 software. After that, it was
verified that the inhibitors had right configurations. The
resulting structures were minimized using the MMFF94S force
field as implemented in Sybyl. The following structures were
docked into the active site of POP model using the GOLD28,43
program, version 2.2, with standard default settings unless
otherwise mentioned. The docking area was defined using NE
atom of ARG643 as a center, with a radius of 20 Å. Unfortu-
nately, current docking programs cannot automatically take
into account a possible covalent interaction between a ligand
and a host. Because several of the compounds possibly have a
covalent interaction, two violations for steric clashes were
allowed. The docking procedure was repeated 10 times for each
inhibitor using default parameters. All dockings, except for
compound 32, were ranked using GOLD score, and the best
alternatives were visually inspected. For compound 32, a
CSCORE calculation within Sybyl was carried out.
The selected docking modes were further analyzed using the
CoMSIA method. The CoMSIA descriptor fields (steric, elec-
trostatic, hydrophobic, acceptor, donor) were calculated in a
3D cubic box extending 4 Å beyond the aligned inhibitors.
Atomic point charges for inhibitors were calculated using the
Gasteiger-Hu¨ckel model as implemented in Sybyl. pIC50
values were used as dependent variables in the PLS statistical
analysis of the resulting data. The predictive value of the PLS
model and the optimum number of the PLS components were
evaluated by the progressive scrambling method.29 The final
noncross-validated model was developed using the optimal
number of components that had both the highest scrambled
q2 value and the smallest value of the estimated cross-
validated standard error with the slope of q2 under 1.1.
Acknowledgment. We thank Tiina Koivunen,
Jaana Leskinen, and Pa¨ivi Sutinen for their excellent
technical assistance and Dr. Seppo Auriola and Unni
Tengvall (M.Sc.) for performing the ESI-MS analyses.
We also thank the National Technology Agency in
Finland, Orion Pharma Oy, and the Finnish Cultural
Foundation of Northern Savo for financial support.
Supporting Information Available: Elemental analyses
of the end products and 1H NMR data for the intermediate
products. This material is available free of charge via the
Internet at http://pubs.acs.org.
References
(1) Barrett, A. J.; Rawlings, N. D. Oligopeptidases, and the emer-
gency of the prolyl oligopeptidase family. Biol. Chem. Hoppe
Seyler 1992, 373, 353-360.
(2) O’Leary, R. M.; O’Connor, B. Identification and localisation of a
synaptosomal membrane prolyl endopeptidase from bovine
brain. Eur. J. Biochem. 1995, 227, 277-283.
(3) Vanhoof, G.; Goossens, F.; Hendriks, L.; De Meester, I.; Hen-
driks, D.; Vriend, G.; Van Broeckhoven, C.; Scharpe´, S. Cloning
and sequence analysis of the gene encoding human lymphocyte
prolyl endopeptidase. Gene 1994, 149, 363-366.
(4) Koida, M.; Walter, R. Post-proline cleaving enzyme. Purification
of this endopeptidase by affinity chromatography. J. Biol. Chem.
1976, 251, 7593-7599.
(5) Huston, J. P.; Haseno¨hrl, R. U. The role of neuropeptides in
learning: Focus on the neurokinin substance P. Behav. Brain
Res. 1995, 66, 117-127.
(6) Dietrich, A.; Allen, J. D. Vasopressin and memory. I. The
vasopressin analogue AVP4-9 enhances working memory as
well as reference memory in the radial arm maze. Behav. Brain
Res. 1997, 87, 195-200.
(7) de Wied, D.; Gaffori, O.; van Ree, J. M.; de Jong, W. Central
target for the behavioural effects of vasopressin neuropeptides.
Nature 1984, 308, 276-278.
(8) Dietrich, A.; Allen, J. D. Vasopressin and memory. II. Lesions
to the hippocampus block the memory enhancing effects of
AVP4-9 in the radial maze. Behav. Brain Res. 1997, 87, 201-
208.
(9) Miyamoto, M.; Hirai, K.; Takahashi, H.; Kato, K.; Nishiyama,
M.; Okada, H.; Nagaoka, A. Effects of sustained release formula-
tion of thyrotropin-releasing hormone on learning impairments
caused by scopolamine and AF64A in rodents. Eur. J. Pharma-
col. 1993, 238, 181-189.
(10) Yankner, B. A.; Duffy, L. K.; Kirschner, D. A. Neurotrophic and
neurotoxic effects of amyloid beta protein: Reversal by tachy-
kinin neuropeptides. Science 1990, 250, 279-282.
(11) Irazusta, J.; Larrinaga, G.; Gonzalez-Maeso, J.; Gil, J.; Meana,
J. J.; Casis, L. Distribution of prolyl endopeptidase activities in
rat and human brain. Neurochem. Int. 2002, 40, 337-345.
(12) Kato, T.; Okada, M.; Nagatsu, T. Distribution of post-proline
cleaving enzyme in human brain and the peripheral tissues. Mol.
Cell. Biochem. 1980, 32, 117-121.
(13) Bellemere, G.; Vaudry, H.; Mounien, L.; Boutelet, I.; Jegou, S.
Localization of the mRNA encoding prolyl endopeptidase in the
rat brain and pituitary. J. Comput. Neurol. 2004, 471, 128-
143.
(14) Jiang, C. H.; Tsien, J. Z.; Schultz, P. G.; Hu, Y. The effects of
aging on gene expression in the hypothalamus and cortex of
mice. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 1930-1934.
(15) Rampon, C.; Jiang, C. H.; Dong, H.; Tang, Y. P.; Lockhart, D.
J.; Schultz, P. G.; Tsien, J. Z.; Hu, Y. Effects of environmental
enrichment on gene expression in the brain. Proc. Natl. Acad.
Sci. U.S.A. 2000, 97, 12880-12884.
(16) Shishido, Y.; Furushiro, M.; Tanabe, S.; Nishiyama, S.; Hash-
imoto, S.; Ohno, M.; Yamamoto, T.; Watanabe, S. ZTTA, a
postproline cleaving enzyme inhibitor, improves cerebral
ischemia-induced deficits in a three-panel runway task in rats.
Pharmacol. Biochem. Behav. 1996, 55, 333-338.
(17) Shishido, Y.; Furushiro, M.; Tanabe, S.; Taniguchi, A.; Hash-
imoto, S.; Yokokura, T.; Shibata, S.; Yamamoto, T.; Watanabe,
S. Effect of ZTTA, a prolyl endopeptidase inhibitor, on memory
impairment in a passive avoidance test of rats with basal
forebrain lesions. Pharm. Res. 1998, 15, 1907-1910.
(18) Yoshimoto, T.; Kado, K.; Matsubara, F.; Koriyama, N.; Kaneto,
H.; Tsura, D. Specific inhibitors for prolyl endopeptidase and
their anti-amnesic effect. J. Pharmacobiodyn. 1987, 10, 730-
735.
(19) Schneider, J. S.; Giardiniere, M.; Morain, P. Effects of the prolyl
endopeptidase inhibitor S 17092 on cognitive deficits in chronic
low dose MPTP-treated monkeys. Neuropsychopharmacology
2002, 26, 176-182.

4782 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 15
Jarho et al.
(20) Grellier, P.; Vendeville, S.; Joyeau, R.; Bastos, I. M.; Drobecq,
H.; Frappier, F.; Teixeira, A. R.; Schre´vel, J.; Davioud-Charvet,
E.; Sergheraert, C.; Santana, J. M. Trypanosoma cruzi prolyl
oligopeptidase Tc80 is involved in nonphagocytic mammalian
cell invasion by trypomastigotes. J. Biol. Chem. 2001, 276,
47078-47086.
(21) Walle´n, E. A. A.; Christiaans, J. A. M.; Forsberg, M. M.;
Vena¨la¨inen, J. I.; Ma¨nnisto¨, P. T.; Gynther, J. Dicarboxylic acid
bis(L-prolyl-pyrrolidine) amides as prolyl oligopeptidase inhibi-
tors. J. Med. Chem. 2002, 45, 4581-4584.
(22) Walle´n, E. A. A.; Christiaans, J. A. M.; Jarho, E. M.; Forsberg,
M. M.; Vena¨la¨inen, J. I.; Ma¨nnisto¨, P. T.; Gynther, J. New prolyl
oligopeptidase inhibitors developed from dicarboxylic acid bis-
(l-prolyl-pyrrolidine) amides. J. Med. Chem. 2003, 46, 4543-
4551.
(23) Klebe, G.; Abraham, U.; Mietzner, T. Molecular similarity indices
in a comparative analysis (CoMSIA) of drug molecules to
correlate and predict their biological activity. J. Med. Chem.
1994, 37, 4130-4146.
(24) Walle´n, E. A. A.; Christiaans, J. A. M.; Saarinen, T. J.; Jarho,
E. M.; Forsberg, M. M.; Vena¨la¨inen, J. I.; Ma¨nnisto¨, P. T.;
Gynther, J. Conformationally rigid N-acyl-5-alkyl-L-prolyl-
pyrrolidines as prolyl oligopeptidase inhibitors. Bioorg. Med.
Chem. 2003, 11, 3611-3619.
(25) Walle´n, E. A. A.; Christiaans, J. A. M.; Saario, S. M.; Forsberg,
M. M.; Vena¨la¨inen, J. I.; Paso, H. M.; Ma¨nnisto¨, P. T.; Gynther,
J. 4-Phenylbutanoyl-2(S)-acylpyrrolidines and 4-phenylbutanoyl-
L-prolyl-2(S)-acylpyrrolidines as prolyl oligopeptidase inhibitors.
Bioorg. Med. Chem. 2002, 10, 2199-2206.
(26) Jarho, E. M.; Vena¨la¨inen, J. I.; Huuskonen, J.; Christiaans, J.
A. M.; Garcia-Horsman, J. A.; Forsberg, M. M.; Ja¨rvinen, T.;
Gynther, J.; Ma¨nnisto¨, P. T.; Walle´n, E. A. A. A cyclopent-2-
enecarbonyl group mimics proline at the P2 position of prolyl
oligopeptidase inhibitors. J. Med. Chem. 2004, 47, 5605-5607.
(27) Fu¨lo¨p, V.; Bo¨cskei, Z.; Polga´r, L. Prolyl oligopeptidase: An
unusual beta-propeller domain regulates proteolysis. Cell 1998,
94, 161-170.
(31) Portevin, B.; Benoist, A.; Re´mond, G.; Herve´, Y.; Vincent, M.;
Lepagnol, J.; De Nanteuil, G. New prolyl endopeptidase inhibi-
tors: in vitro and in vivo activities of azabicyclo[2.2.2]octane,
azabicyclo[2.2.1]heptane, and perhydroindole derivatives. J.
Med. Chem. 1996, 39, 2379-2391.
(32) Yoshimoto, T.; Tsuru, D.; Yamamoto, N.; Ikezawa, R.; Furukawa,
S. Structure activity relationship of inhibitors specific for prolyl
endopeptidase. Agric. Biol. Chem. 1991, 55, 37-43.
(33) Toide, K.; Iwamoto, Y.; Fujiwara, T.; Abe, H. JTP-4819: A novel
prolyl endopeptidase inhibitor with potential as a cognitive
enhancer. J. Pharmacol. Exp. Ther. 1995, 274, 1370-1378.
(34) Wilk, S.; Orlowski, M. Inhibition of rabbit brain prolyl endopep-
tidase by n-benzyloxycarbonyl-prolyl-prolinal, a transition state
aldehyde inhibitor. J. Neurochem. 1983, 41, 69-75.
(35) Tanaka, Y.; Niwa, S.; Nishioka, H.; Yamanaka, T.; Torizuka,
M.; Yoshinaga, K.; Kobayashi, N.; Ikeda, Y.; Arai, H. New potent
prolyl endopeptidase inhibitors: synthesis and structure-activ-
ity relationships of indan and tetralin derivatives and their
analogues. J. Med. Chem. 1994, 37, 2071-2078.
(36) Liang, T. C.; Abeles, R. H. Inhibition of papain by nitriles:
mechanistic studies using NMR and kinetic measurements.
Arch. Biochem. Biophys. 1987, 252, 626-634.
(37) Brisson, J. R.; Carey, P. R.; Storer, A. C. Benzoylamidoaceto-
nitrile is bound as a thioimidate in the active site of papain. J.
Biol. Chem. 1986, 261, 9087-9089.
(38) Vena¨la¨inen, J. I.; Juvonen, R. O.; Garcia-Horsman, J. A.; Walle´n,
E. A. A.; Christiaans, J. A. M.; Jarho, E. M.; Gynther, J.;
Ma¨nnisto¨, P. T. Slow-binding inhibitors of prolyl oligopeptidase
with different functional groups at the P1 site. Biochem. J. 2004,
382, 1003-1008.
(39) Vena¨la¨inen, J. I.; Juvonen, R. O.; Forsberg, M. M.; Garcia-
Horsman, J. A.; Poso, A.; Walle´n, E. A. A.; Gynther, J.; Ma¨nnisto¨,
P. T. Substrate-dependent, nonhyperbolic kinetics of pig brain
prolyl oligopeptidase and its tight binding inhibition by JTP-
4819. Biochem. Pharmacol. 2002, 64, 463-471.
(40) Tripos Inc. SYBYL; S. H. R.: St. Louis, MO.
(41) Halgren, T. Maximally diagonal force constants in dependent
angle-bending coordinates. 2. Implications for the design of
empirical force fields. J. Am. Chem. Soc. 1990, 112, 4710-4723.
(42) Pearlman, R. S. Concord User’s Manual; Distributed by Tripos
Inc.: St. Louis, MO, 2004.
(28) Jones, G.; Willett, P.; Glen, R. C.; Leach, A. R.; Taylor, R.
Development and validation of a genetic algorithm for flexible
docking. J. Mol. Biol. 1997, 267, 727-748.
(29) Clark, R. D.; Fox, P. C. Statistical variation in progressive
scrambling. J. Comput.-Aided Mol. Des. 2004, 18, 563-576.
(30) Arai, H.; Nishioka, H.; Niwa, S.; Yamanaka, T.; Tanaka, Y.;
Yoshinaga, K.; Kobayashi, N.; Miura, N.; Ikeda, Y. Synthesis of
prolyl endopeptidase inhibitors and evaluation of their structure-
activity relationships: In vitro inhibition of prolyl endopeptidase
from canine brain. Chem. Pharm. Bull. 1993, 41, 1583-1588.
(43) Jones, G.; Willett, P.; Glen, R. C. Molecular recognition of
receptor sites using a genetic algorithm with a description of
desolvation. J. Mol. Biol. 1995, 245, 43-53.
JM0500020