New prolyl oligopeptidase inhibitors developed from dicarboxylic acid bis(L-prolyl-pyrrolidine) amides

Article abstract of DOI:10.1021/jm030811o

Isophthalic acid bis(L-prolyl-pyrrolidine) amide is a very potent prolyl oligopeptidase inhibitor, but it has a log P value of -0.2, which is very low for a compound targeted to the brain. Therefore, these types of compounds were further modified to improve the structure-activity relationships, with the focus on increasing the log P value. The inhibitory activity against prolyl oligopeptidase from pig brain was tested in vitro. The most promising compounds resulted from replacing the pyrrolidinyl group at the P5 site by cycloalkyl groups, such as cyclopentyl and cyclohexyl groups, and by a phenyl group. These compounds are slightly more potent, and they have a significantly higher log P value. The potency of these compounds was further increased by replacing the pyrrolidinyl group at the P1 site by 2(S)-cyanopyrrolidinyl and 2(S)-(hydroxyacetyl)pyrrolidinyl groups.

Full text of DOI:10.1021/jm030811o

J . Med. Chem. 2003, 46, 4543-4551
4543
New P r olyl Oligop ep tid a se In h ibitor s Develop ed fr om Dica r boxylic Acid
Bis(L-p r olyl-p yr r olid in e) Am id es
Erik A. A. Walle´n,*,† J ohannes A. M. Christiaans,†,# Elina M. J arho,† Markus M. Forsberg,‡
J arkko I. Vena¨la¨inen,‡ Pekka T. Ma¨nnisto¨,‡,§ and J ukka Gynther†,§
Department of Pharmaceutical Chemistry and Department of Pharmacology and Toxicology, University of Kuopio,
P.O. Box 1627, FIN-70211 Kuopio, Finland, and Finncovery, Ltd., Kuopio, Finland
Received February 17, 2003
Isophthalic acid bis(L-prolyl-pyrrolidine) amide is a very potent prolyl oligopeptidase inhibitor,
but it has a log P value of -0.2, which is very low for a compound targeted to the brain.
Therefore, these types of compounds were further modified to improve the structure-activity
relationships, with the focus on increasing the log P value. The inhibitory activity against prolyl
oligopeptidase from pig brain was tested in vitro. The most promising compounds resulted
from replacing the pyrrolidinyl group at the P5 site by cycloalkyl groups, such as cyclopentyl
and cyclohexyl groups, and by a phenyl group. These compounds are slightly more potent, and
they have a significantly higher log P value. The potency of these compounds was further
increased by replacing the pyrrolidinyl group at the P1 site by 2(S)-cyanopyrrolidinyl and 2(S)-
(hydroxyacetyl)pyrrolidinyl groups.
In tr od u ction
important for high inhibitory activity.15 Two well-known
reference compounds are 4-phenylbutanoyl-L-prolyl-
pyrrolidine (SUAM-1221) and benzylcarbamoyl-L-prolyl-
2(S)-(hydroxyacetyl)pyrrolidine (J TP-4819).
The serine protease prolyl oligopeptidase (POP, previ-
ously called prolyl endopeptidase or post-proline cleav-
ing enzyme, EC 3.4.21.26) is a large enzyme of 80 kDa
that preferentially hydrolyzes proline-containing oli-
gopeptides at the carboxyl side of a prolyl residue.1
Many peptide hormones and neuropeptides have one or
more proline residues, and the processing and degrada-
tion of such peptides often require the use of proline-
specific enzymes.2-4 Several substrates of POP, such as
substance P, vasopressin, neurotensin, and thyroliberin,
are implicated in learning and memory.5 In addition, a
low level of substance P is characteristic in the brain of
Alzheimer’s patients, and administration of substance
P is able to block â amyloid-induced neurotoxicity.6
There is no firm evidence of increased POP activity in
Alzheimer’s patients, because rather low enzyme activi-
ties are correlated with the severity of Alzheimer’s
disease.7 On the other hand, this result is thought to
reflect the degree of neuronal damage. It was recently
reported that in rat the expression of POP gene was
decreased when the animals were exposed to an en-
riched environment and the expression was increased
manyfold in the hypothalamus and the cortex in aged
rats.8,9 Therefore, it has been postulated that centrally
acting POP inhibitors that increase neuropeptide con-
centrations10 might be beneficial in patients with cogni-
tive disturbances. Indeed, POP inhibitors have been
shown to reverse scopolamine-induced amnesia in rats
and to improve cognition in MPTP-treated monkeys.11-14
Most described low molecular weight POP inhibitors
have an acyl-L-prolyl-pyrrolidine backbone, wherein a
lipophilic acyl end group has been reported to be
A series of dicarboxylic acid bis(L-prolyl-pyrrolidine)
amides [the compounds consist of two L-prolyl-pyrrol-
idines connected by a dicarboxylic acid unit; the naming
of the compounds is based on the peptide type of
naming; alternatively they could be called dicarboxylic
acid bis(L-proline pyrrolidide) amide], which do not have
the lipophilic acyl end group, were recently described
as new potent POP inhibitors.16 The most active com-
pounds in this series were the isophthalic acid bis(L-
prolyl-pyrrolidine) amide type of compounds 1a -d and
3,3-dimethylglutaric acid bis(L-prolyl-pyrrolidine) amide
1e. The postulated P1-P5 site assignation presented
in Chart 1 is based on the structure-activity relation-
ships at the P1 site, which show that the formyl, cyano,
and hydroxyacetyl groups of compounds 1b-d interact
with the serine residue (Ser554) at the active site of the
enzyme.
The IC50 values for compounds 1a -e were 26, 1.3, 1.5,
0.39, and 13 nM, respectively, against POP from pig
brain. Although compounds 1a -e were potent inhibi-
tors, the log P value of compound 1a was only -0.2. This
is considerably lower than the generally acceptable log
P value between 2 and 3 for compounds that have to
* Corresponding author (e-mail wallen@uku.fi).
† Department of Pharmaceutical Chemistry.
Present address: Altana Pharma, P.O. Box 31, 2130 AA Hoof-
#
ddorp, The Netherlands.
‡ Department of Pharmacology and Toxicology.
§ Finncovery, Ltd.
10.1021/jm030811o CCC: $25.00 © 2003 American Chemical Society
Published on Web 09/04/2003

4544 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 21
Walle´n et al.
Ch a r t 1
The starting point for the development of new POP
inhibitors was compound 1a . Although this compound
was slightly less potent and to some extent also less
druglike than SUAM-1221, it represented a new type
of POP inhibitor that had the potential to be developed
further. To study the structure-activity relationships
of compound 1a further, some modifications of the
molecular structure were made. The L-prolyl groups at
the P2 and P4 sites were replaced by acyclic R-amino
acid residues. The pyrrolidinyl group at the P5 site was
replaced by different alkylamino, cycloalkyl, alkyl, and
phenyl groups. In combination with the replacements
at the P5 site the pyrrolidinyl group at the P1 site was
also replaced symmetrically by some of these groups.
All compounds were tested in vitro against POP from
pig brain, and the log P values for a selection of them
were determined.
penetrate the blood-brain barrier in order to reach the
brain. Therefore, further investigations were carried out
with the focus on improving the log P value of this type
of compound.
Syn th etic Ch em istr y. The synthetic procedures for
the compounds are presented in Scheme 1. The earlier
presented synthetic route for compound 1a could be
used only for its symmetrical glycyl analogue,16 because
this route may cause racemization for other R-amino
acids other than glycine and L-proline. The other sym-
metrical compounds were synthesized by reacting 2
equiv of the R-aminoacyl derivative AX with isophthal-
oyl dichloride.
In the same in vitro assay SUAM-1221 and J TP-4819
had IC50 values of 2.2 and 0.2 nM, respectively. The log
P values for SUAM-1221 and J TP-4819 were 1.8 and
0.2, respectively. In vivo, these two compounds have
been reported to improve performance in memory- and
learning-related tests, such as the passive avoidance
test for rats with scopolamine-induced amnesia.17,18 For
a compound that penetrates the blood-brain barrier,
J TP-4819 has a remarkably low log P value. Therefore,
the log P value should be used only as a first prediction
of the blood-brain barrier penetration.
The synthesis of the unsymmetrical compounds with
a central isophthaloyl group was started from iso-
Sch em e 1a
a
(a) AX, NaOH/diethyl ether (dichloromethane), water; (b) glycine, NaOH/diethyl ether, water; (c) (1) pivaloyl chloride, Et3N/
dichloromethane, (2) pyrrolidine, Et3N/dichloromethane; (d) (1) pivaloyl chloride, Et3N/dichloromethane, (2) AX, Et3N/dichloromethane;
(e) LiOH/MeOH, water; (f) (1) pivaloyl chloride, Et3N/dichloromethane, (2) L-proline methyl ester, Et3N/dichloromethane; (g) (1) pivaloyl
chloride, Et3N/dichloromethane, (2) Y (mono-Boc-piperazine and 2(S)-(acetoxyacetyl)pyrrolidine were used in the coupling reaction when
Y was piperazinyl and 2(S)-(hydroxyacetyl)pyrrolidinyl; the protecting groups had to be removed to obtain the end product),
Et3N/dichloromethane; (h) 2(S)-(cyclopentanecarbonyl)pyrrolidine, Et3N/tetrahydrofuran.

Prolyl Oligopeptidase Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 21 4545
Ta ble 1. Inhibitory Activity of the Compounds of Series 2 against POP from Pig Brain (95% Confidence Intervals Given in
Parentheses)
compd
R, R′ (R-amino acid)
R′′, R′′′ (R-amino acid)
IC50, nM
1a
2a
2b
2c
2d
2e
-(CH2)3-
(L-Pro)
(Gly)
(L-Ala)
(L-Phe)
(L-Met)
(L-Ala)
-(CH2)3-
(L-Pro)
(Gly)
(L-Ala)
(L-Phe)
(L-Met)
(L-Pro)
26 (21-32)
640 (490-820)
87 (78-97)
81 (61-107)
110 (65-180)
39 (33-46)
H, H
H, H
methyl, H
benzyl, H
-(CH2)2SCH3, H
methyl, H
methyl, H
benzyl, H
-(CH2)2SCH3, H
-(CH2)3-
Ta ble 2. Inhibitory Activity of the Compounds of Series 3 against POP from Pig Brain (95% Confidence Intervals Given in
Parentheses; log P Values Are Presented If Determined)
compd
X
Y
IC50, nM
log P
-0.2
1a
1b
1c
1d
3a
3b
3c
3d
3e
3f
3g
3h
3i
3j
3k
3l
pyrrolidin-1-yl
pyrrolidin-1-yl
pyrrolidin-1-yl
pyrrolidin-1-yl
benzylamino
azepan-1-yl
piperazin-1-yl
4-Boc-piperazin-1-yl
cyclopentyl
methyl
pyrrolidin-1-yl
26 (21-32)
1.3 (1.1-1.6)
1.5 (1.1-2.2)
0.39 (0.24-0.62)
31 (27-35)
2(S)-formylpyrrolidin-1-yl
2(S)-cyanopyrrolidin-1-yl
2(S)-(hydroxyacetyl)pyrrolidin-1-yl
pyrrolidin-1-yl
pyrrolidin-1-yl
pyrrolidin-1-yl
pyrrolidin-1-yl
pyrrolidin-1-yl
pyrrolidin-1-yl
pyrrolidin-1-yl
cyclopentyl
pyrrolidin-1-yl
23 (18-28)
33 (28-39)
170 (150-200)
14 (10-20)
1.1
65 (58-74)
18 (15-22)
phenyl
methoxy
methoxy
640 (550-750)
54 (35-81)
cyclopentyl
methyl
cyclopentyl
cyclohexyl
methyl
cyclopentyl
methyl
78 (73-83)
2.7
0.8
54% inhib at 0.1 mM
1.1 (0.90-1.4)
0.72 (0.56-0.93)
4.2 (2.5-7.3)
1.6 (1.0-2.5)
1.3 (1.0-1.8)
0.61 (0.46-0.80)
1.2 (0.88-1.7)
2(S)-cyanopyrrolidin-1-yl
2(S)-cyanopyrrolidin-1-yl
2(S)-cyanopyrrolidin-1-yl
2(S)-cyanopyrrolidin-1-yl
2(S)-cyanopyrrolidin-1-yl
2(S)-(hydroxyacetyl)pyrrolidin-1-yl
2(S)-(hydroxyacetyl)pyrrolidin-1-yl
3m
3n
3o
3p
3q
3r
-0.6
0.3
1.1
isopropyl
phenyl
cyclopentyl
methyl
0.2
phthalic acid monomethyl ester, which was activated
with pivaloyl chloride and reacted with the R-aminoacyl
derivative AX. The methyl ester group was hydrolyzed.
The resulting carboxylic acid was activated with pivaloyl
chloride and reacted with L-proline methyl ester. The
methyl ester group was hydrolyzed. The resulting
carboxylic acid was activated with pivaloyl chloride and
reacted with the amine Y.
The synthesis of the unsymmetrical compounds with
a central 3,3-dimethylglutaroyl group was started by
reacting 3,3-dimethylglutaric anhydride with 2(S)-(cy-
clopentanecarbonyl)pyrrolidine. The resulting carboxylic
acid was activated with pivaloyl chloride and reacted
with L-proline methyl ester. The methyl ester group was
hydrolyzed. The resulting carboxylic acid was activated
with pivaloyl chloride and reacted with the amine Y.
In Vitr o Assa y for P OP Activity. The inhibitory
effect of the novel compounds on POP activity of pig
brain was determined according to a method described
earlier.19 Suc-Gly-Pro-7-amino-4-methylcoumarin was
used as substrate, and the formation of 7-amino-4-
methylcoumarin was determined fluorometrically with
a microplate fluorescence reader.
alanyl derivative 2c being the most potent ones having
IC50 values of 87 and 81 nM, respectively. However,
these two compounds were still less potent than the
corresponding L-prolyl derivative 1a . To study the
relevance of symmetry in this series, an unsymmetrical
compound was also made. The unsymmetrical mixed
L-prolyl and L-alanyl derivative 2e had an IC50 value of
39 nM. Although it was more potent than the sym-
metrical L-alanyl derivative 2b, it was still slightly less
potent than the symmetrical L-prolyl derivative 1a . This
series of compounds shows that replacing either one or
both L-prolyl groups at the P2 and P4 sites by acyclic
R-amino acid residues lowers the potency.
The second strategy was to make modifications on the
pyrrolidinyl groups at the P1 and P5 sites of compound
1a , resulting in a series of compounds 3 (Table 2). In
Resu lts a n d Discu ssion
The importance of the L-prolyl groups at the P2 and
P4 sites in compound 1a was studied by synthesizing a
series of compounds 2a -e (Table 1). In compounds
2a -d both L-prolyl groups were replaced symmetrically
by glycyl, L-alanyl, L-phenylalanyl, and L-methionyl
groups, respectively. All four compounds inhibited the
enzyme, the L-alanyl derivative 2b and the L-phenyl-
compounds 3a -c the P5 pyrrolidinyl group was replaced
by other alkylamino groups such as the benzylamino,
azepanyl, and piperazinyl groups, respectively, whereas
the P1 pyrrolidinyl group was not modified. Interest-
ingly, all three compounds were equipotent as compared

4546 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 21
Walle´n et al.
Ta ble 3. Inhibitory Activity of the Compounds of Series 4
against POP from Pig Brain (95% Confidence Intervals Given
in Parentheses; log P Values Are Presented If Determined)
to compound 1a , having IC50 values in the range of 23-
33 nM. The Boc-protected piperazinyl derivative 3d ,
which was an intermediate in the synthesis of 3c, was
less active with an IC50 value of 170 nM. Compounds
3a -d give strong evidence for great freedom at the P5
site. This is in agreement with an earlier study in which
the P5 site was not contributing to the binding of the
substrate to the enzyme.20
compd
X
Y
IC50, nM
log P
1e
4a
pyrrolidin-1-yl pyrrolidin-1-yl
13 (12-14)
0.57 (0.38-0.84)
cyclopentyl
2(S)-cyanopyrrol-
idin-1-yl
4b
cyclopentyl
2(S)-(hydroxyacetyl)- 0.32 (0.23-0.45) 0.7
pyrrolidin-1-yl
Focusing on increasing the log P value of compound
1a , a replacement of the nitrogen atom of the pyrroli-
dinyl group at the P5 site by a carbon atom was made,
resulting in the cyclopentyl derivative 3e. The pyrrol-
idinyl group at the P5 site was also replaced by methyl
and phenyl groups, resulting in compounds 3f and 3g,
respectively. As compared to compound 1a , the cyclo-
pentyl derivative 3e and the phenyl derivative 3g were
more potent, having IC50 values of 14 and 18 nM,
respectively, whereas the methyl derivative 3f was less
potent, having an IC50 value of 65 nM. This result shows
that although the P5 site does not contribute to the
binding of the substrate to the enzyme, it does have a
small effect on the in vitro potencies of the inhibitors
3e and 3f, where the only difference is the alkyl
substituent at the P5 site.
result from compound 3k , it is apparent that a methyl
group cannot act as a P1 group.
Analogues of compounds 3e-g in which the pyrroli-
dinyl group at the P1 site was replaced by 2(S)-cyano-
pyrrolidine and 2(S)-(hydroxyacetyl)pyrrolidine groups
were synthesized in order to increase the potency. In
this group of compounds the P5 site was varied further,
including compounds with cyclohexyl and isopropyl
groups at the P5 site. The resulting compounds 3l-r
were highly potent, having IC50 values in the range of
0.6-4.2 nM. The 2(S)-(hydroxyacetyl)pyrrolidine group
at the P1 site increased the potency slightly more than
the 2(S)-cyanopyrrolidine group. The methyl group at
the P5 site resulted again in the least potent compounds
with both 2(S)-cyanopyrrolidinyl and 2(S)-(hydroxy-
acetyl)pyrrolidinyl groups at the P1 site.
Because the cyclopentyl and phenyl groups can also
act as the P1 site, as reported in an earlier study
[4-phenylbutanoyl-2(S)-(cyclopentanecarbonyl)pyrroli-
dine and 4-phenylbutanoyl-2(S)-benzoylpyrrolidine had
IC50 values of 30 and 23 nM, respectively, in the same
in vitro assay],21 the high potency of compounds 3e and
3g could result from their binding in the opposite
direction with the P5 site as the P1 site to the active
site of the enzyme. To study this hypothesis the two
intermediates 3h and 3i were also tested. Compound
3h has a cyclopentyl group and compound 3i has a
pyrrolidinyl group at the P1 site. Both compounds have
an L-proline methyl ester group at the P4-P5 sites,
which is not a preferred group at the P1-P2 sites. The
L-proline methyl ester group might be hydrolyzed in the
in vitro assay. However, if this happens, the resulting
free L-proline group is also not a preferred group at the
P1-P2 sites. The cyclopentyl derivative 3h had an IC50
value of only 640 nM, whereas the pyrrolidinyl deriva-
tive 3i had an IC50 value of 54 nM. To study the
direction of binding further, both pyrrolidinyl groups at
the P1 and P5 sites of compound 1a were replaced
symmetrically by cyclopentyl and methyl groups, result-
ing in compounds 3j and 3k , respectively. The cyclo-
pentyl derivative 3j was moderately potent, having an
IC50 value of 78 nM, whereas the methyl derivative 3k
was a poor inhibitor, having only 54% inhibition at 0.1
mM.
As reported earlier for compound 1e, the central
isophthaloyl group at the P3 site could be replaced by a
3,3-dimethylglutaroyl group.16 The 3,3-dimethylglu-
taroyl analogues of compounds 3l and 3q were made,
resulting in compounds 4a and 4b, respectively (Table
3). Both compounds were highly potent, having IC50
values in the range of 0.3-0.6 nM.
The increased potency of compounds 1b-d , 3l-r , and
4a ,b is a result of the interaction between the formyl,
cyano, and hydroxyacetyl substituents of the pyrroli-
dinyl group at the P1 site and the serine residue
(Ser554) at the active site of the enzyme. The formyl
group has been reported to form a reversible hemiacetal
adduct with the serine residue at the active site of the
enzyme,22 and the hydroxyacetyl group has been re-
ported to give a tight binding type of inhibition.19 Most
probably the cyano group interacts analogously. There-
fore, the compounds 1b-d , 3l-r , and 4a ,b are transi-
tion state analogue inhibitors, whereas the other com-
pounds are substrate analogue inhibitors.
This study resulted in very potent compounds, and
the log P values for a selection of them were determined
(Tables 2 and 3). The log P values for compounds 1a ,
3e, and 3j were -0.2, 1.1, and 2.7, respectively, indicat-
ing that each replacement of a nitrogen atom of the
pyrrolidine groups at the P1 and P5 sites by a carbon
atom resulted in an increase of the log P value by >1
log unit. The effect of different P5 groups on the log P
value was studied by comparing compounds 3l, 3n , 3o,
and 3p , having log P values of 0.8, -0.6, 0.3, and 1.1,
respectively. A replacement of the cyclopentyl group at
the P5 site by a methyl or a isopropyl group lowered
the log P value, whereas a replacement by a phenyl
On the basis of the results from compounds 3h -j, it
is obvious that a cyclopentyl group can act as a P1
group. However, replacing the pyrrolidinyl group at the
P1 site by a cyclopentyl group always decreased the
potency. Therefore, it is unlikely that compound 3e
would bind in the opposite direction with the P5 site as
the P1 site to the active site of the enzyme. This
conclusion applies most probably also to compound 3g,
on the basis of the result from the earlier study in which
the cyclopentyl and the phenyl groups at the P1 site
were reported to give equipotent compounds.21 From the

Prolyl Oligopeptidase Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 21 4547
group slightly raised the log P value. A comparison
between compounds 3e, 3l, and 3q, having log P values
of 1.1, 0.8, and 0.2, respectively, showed that a replace-
ment of the pyrrolidinyl group at the P1 site by a 2(S)-
cyanopyrrolidine group lowered the log P only slightly,
whereas a replacement by a 2(S)-(hydroxyacetyl)pyrrol-
idine group lowered the log P value by almost 1 log unit.
The log P value for compound 4b was also determined
to establish the effect of the replacement of the central
isophthaloyl group at the P3 site by a 3,3-dimethylglu-
taroyl group. Compound 4b had a log P value of 0.7,
which was slightly higher than the log P value of 0.2
for compound 3q.
The trends of the measured log P values combined
with the data of the in vitro potency clearly point out
compounds 3e, 3g, 3l, 3m , 3p , 4a , and 4b as good
candidates for further pharmacological studies. The log
P values for these compounds are in the same range as
for SUAM-1221 and J TP-4819, which were reported to
be active in vivo. Compound 3j has the highest log P
value, and it might also have to be considered as a
candidate for further pharmacological studies. However,
the log P value is a first prediction of the blood-brain
barrier penetrability, and no selection of compounds for
further pharmacological studies should be made on the
basis of the log P value alone.
Exp er im en ta l Section
An a lytica l. NMR spectra were recorded on
a Bruker
Avance 500 spectrometer (500.1 MHz for 1H and 125.8 MHz
for 13C) or a Bruker AM 400 spectrometer (400.1 MHz for 1H
and 100.6 MHz for 13C), CDCl3 was used as solvent (if not
otherwise noted), and chemical shifts are expressed in parts
per million relative to tetramethylsilane as internal standard.
For compounds 3c, 3l-r , and 4a ,b the 13C NMR spectra are
not presented, due to difficulties in assigning the spectra.
Amides N-terminal to prolyl groups have energetically similar
cis and trans isomers (also sometimes referred to as rotamers),
which complicates the NMR spectra. For the intermediates
3h and 3i no 13C NMR spectra were measured. Positive ion
mass spectra were acquired with ESI-MS, using a Finnegan
MAT LCQ quadropole ion trap mass spectrometer equipped
with an ESI source. Combustion analysis for CHN was
measured on an EA1110 ThermoQuest CE Instruments el-
emental analyzer. No melting points were determined as the
products were mostly hygroscopic amorphous compounds.
Syn th esis. All chemicals and solvents were of commercial
quality and were purified if necessary following standard
procedures. Boc-L-prolyl-pyrrolidine, Boc-2(S)-(cyclopentane-
carbonyl)pyrrolidine, Boc-2(S)-(cyclohexanecarbonyl)pyrroli-
dine, Boc-2(S)-isobutanoylpyrrolidine, Boc-2(S)-acetylpyrroli-
dine, Boc-2(S)-benzoylpyrrolidine, Boc-2(S)-(acetoxyacetyl)pyr-
rolidine, and Boc-2(S)-cyanopyrrolidine were synthesized ac-
cording to reported procedures.21,23,24
P r oced u r e A: Am id e F or m a tion Usin g P iva loyl Ch lo-
r id e Activa tion . A solution of pivaloyl chloride (1.0 mmol)
in dichloromethane was added to a solution of carboxylic acid
(1.0 mmol) and triethylamine (1.1 mmol) in dichloromethane
at 0 °C. The reaction mixture was stirred at 0 °C for 1 h. A
solution of triethylamine (1.1 mmol) and the amine (1.0-1.1
mmol) in dichloromethane was slowly added at 0 °C (if the
amine was in the form of a trifluoroacetic acid salt or HCl salt,
then 3.3 mmol of triethylamine was used and the triethyl-
amine was added separately before the addition of the amine).
The reaction mixture was stirred for 2-3 h or overnight at
room temperature. The dichloromethane solution was washed
with 30% citric acid, saturated NaCl, and saturated NaHCO3.
The dichloromethane phase was dried with anhydrous Na2-
SO4 and evaporated.
P r oced u r e B: Rem ova l of Boc-P r otectin g Gr ou p . Tri-
fluoroacetic acid (2-4 mL) was added to the Boc-protected
amine (1.0 mmol) in dichloromethane (5-10 mL) at 0 °C, and
the reaction was stirred at 0 °C for 2 h. The solvent was
removed, and the product was evaporated in vacuo, yielding
the corresponding amine trifluoroacetic acid salt. Alternatively,
an excess of ethyl acetate saturated with HCl was added to
the Boc-protected amine at room temperature, and the reaction
mixture was stirred at room temperature for 30 min; after-
ward, the solvent was removed and the product was evapo-
rated in vacuo, yielding the corresponding amine HCl salt.
P r oced u r e C: Bis-a m id e F or m a tion w ith Isop h th a loyl
Dich lor id e. A solution of isophthaloyl dichloride (1.0 mmol)
in diethyl ether or dichloromethane was mixed with a solution
of amine (2.0 mmol) in diethyl ether or dichloromethane in
the presence of an excess of 4 M NaOH at 0 °C with vigorous
stirring. The reaction was stirred vigorously for 1-2 h at room
temperature. Dichloromethane was added, and the phases
were separated. The organic phase was washed with saturated
NaCl, dried with anhydrous Na2SO4, and evaporated.
Con clu sion s
The structure-activity relationship for compound 1a
was studied with the focus on increasing the log P value.
Replacing one or both L-prolyl groups at the P2 and the
P4 sites with acyclic R-amino acid residues, such as
glycyl, L-alanyl, L-phenylalanyl, and L-methionyl, de-
creased the potency. Replacing the pyrrolidinyl group
at the P5 site by other alkylamino groups, such as
benzylamino, azepanyl, and piperazinyl groups, resulted
in equipotent compounds, indicating that there is a
great freedom at the P5 site. Replacing the nitrogen
atom of the pyrrolidinyl group at the P5 site by a carbon
atom slightly increased the potency. On the other hand,
replacing the nitrogen atoms of both pyrrolidinyl groups
at the P1 and P5 sites by carbon atoms slightly
decreased the potency. However, each replacement of a
nitrogen atom by a carbon atom increased the log P by
>1 log unit. Furthermore, replacing the pyrrolidinyl
group at the P5 site by methyl, isopropyl, cyclohexyl,
and phenyl groups was also allowed, whereas replacing
the pyrrolidinyl group at the P1 site by a methyl group
was not allowed. The most potent compounds resulted
from replacing the pyrrolidinyl group at the P5 site with
cyclopentyl, cyclohexyl, or phenyl combined with replac-
ing the pyrrolidinyl group at the P1 site by 2(S)-
cyanopyrrolidinyl and 2(S)-(hydroxyacetyl)pyrrolidinyl
groups. The 2(S)-cyanopyrrolidinyl group decreased the
log P value only slightly, whereas the 2(S)-(hydroxy-
acetyl)pyrrolidine group decreased the log P value by
almost 1 log unit. By replacing the central isophthaloyl
group at the P3 site by a 3,3-dimethylglutaroyl group,
the log P value of the 2(S)-(hydroxyacetyl)pyrrolidinyl
derivative could be raised slightly. In all, this study
resulted in very potent POP inhibitors with improved
log P values for penetrating the blood-brain barrier.
Further pharmacological studies for these compounds
have to be carried out next.
P r oced u r e D: Hyd r olysis of a Ca r boxylic Acid Meth yl
Ester . LiOH (1.0-1.5 mmol) was added to a solution of the
carboxylic acid methyl ester (1.0 mmol) in a small volume of
25% water in methanol. The reaction was stirred overnight
or longer at room temperature. The methanol was evaporated,
and the residue was dissolved in water. The aqueous phase
was washed with dichloromethane. The aqueous phase was
made acidic with 2-3 M HCl, and the product was extracted
with dichloromethane. The dichloromethane phase was dried
with anhydrous Na2SO4 and evaporated, yielding the product.
P r oced u r e E: Hyd r olysis of a n O-Acetyl Gr ou p . K2CO3
(1.1 mmol) was added very slowly to a solution of the acetyl-

4548 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 21
Walle´n et al.
protected compound (1.0 mmol) in 50% methanol in water (6
mL) at 0 °C. The reaction was stirred 1 h at room temperature.
The methanol was evaporated. Dichloromethane was added,
and the dichloromethane phase was washed with saturated
NaCl, dried with anhydrous Na2SO4, and evaporated.
flash chromatography. The methyl ester group was hydrolyzed
according to procedure D. Yield 0.76 g (2.0 mmol, 30%).
Isop h th a lic Acid L-Ala n yl-p yr r olid in e L-P r olyl-p yr r o-
lid in e Am id e (2e). Isophthalic acid L-alanyl-pyrrolidine L-
proline amide (0.76 g, 2.0 mmol) and pyrrolidine (0.18 mL, 2.2
mmol) were coupled according to procedure A. Purification was
by flash chromatography. Yield 0.68 g (1.54 mmol, 79%); MS
Isop h th a lic a cid bis(glycyl-p yr r olid in e) a m id e (2a )
was prepared according to the method described for 1a .16 The
product was crystallized from dichloromethane-hexane. MS
(ESI), m/z 387 ([MH]+); 1H NMR δ 1.89 (qui, 4H, J ) 6.8 Hz),
2.01 (qui, 4H, J ) 6.8 Hz), 3.46 (t, 4H, J ) 6.8 Hz), 3.54 (t,
4H, J ) 6.8 Hz), 4.19 (d, 4H, J ) 4.3 Hz), 7.45-8.33 (m, 4H)
7.62 (br t, 2H, J ) 4.3 Hz); 13C NMR δ 24.11, 25.92, 42.55,
45.53, 46.00, 125.89, 128.71, 130.15, 134.33, 166.44, 166.54.
Anal. (C20H26N4O4‚0.5CH2Cl2) C, H, N.
Boc-L-a la n yl-p yr r olid in e. A solution of N,N′-dicyclohex-
ylcarbodiimide (10.9 g, 53 mmol) in acetonitrile (100 mL) was
added to a solution of Boc-L-alanine (10.0 g, 53 mmol) and
N-hydroxysuccinimide (6.2 g, 53 mmol) in acetonitrile (100 mL)
at -20 °C. The reaction was stirred at -20 °C for 2 h, and
then it was left without stirring at -20 °C overnight. The
formed N,N′-dicyclohexylurea was filtered off and the filtrate
evaporated. The residue was scraped with hexane, and the last
traces of hexane were evaporated in vacuo, yielding the
activated ester (15.0 g, 52 mmol). Pyrrolidine (8.7 mL, 104
mmol) was added to a solution of the activated ester in
tetrahydrofuran (200 mL) at room temperature. The reaction
mixture was stirred overnight. The solvent was evaporated,
and the residue was dissolved in dichloromethane. The dichlo-
romethane solution was washed with 30% citric acid, saturated
NaCl, and saturated NaHCO3. The dichloromethane phase was
dried with anhydrous Na2SO4 and evaporated. Purification was
by flash chromatography (yield 9.1 g, 37 mmol, 72%). Boc-L-
phenylalanyl-pyrrolidine and Boc-L-methionyl-pyrrolidine were
prepared according to the same method.
1
(ESI), m/z 441 ([MH]+); H NMR δ 1.44 (d, J ) 6.9 Hz, 3H),
1.50-2.28 (m, 12H), 2.55-3.99 (m, 10H), 4.25-4.95 (m, 2H),
7.28-8.07 (m, 4H); 13C NMR δ 18.40, 24.15, 24.20, 25.57, 26.09,
26.25, 29.03, 46.08, 46.13, 46.46, 47.42, 50.31, 58.24, 125.95,
128.58, 128.78, 130.45, 134.31, 136.94, 165.75, 168.55, 170.41,
170.83. Anal. (C24H32N4O4‚0.3H2O) C, H, N.
Isop h th a lic Acid Mon o(L-p r olyl-p yr r olid in e) Am id e.
Boc-L-prolyl-pyrrolidine (8.3 g, 30.9 mmol) was deprotected
using HCl-saturated ethyl acetate (50 mL) according to
procedure B. Isophthalic acid monomethyl ester (5.6 g, 30.9
mmol) and the L-prolyl-pyrrolidine HCl salt were coupled
according to procedure A. Purification was by flash chroma-
tography. The methyl ester group was hydrolyzed according
to procedure D. Yield 5.1 g (16.0 mmol, 52%).
Isop h th a lic Acid (L-P r olin e Meth yl Ester ) L-P r olyl-
p yr r olid in e Am id e (3i). Isophthalic acid mono(L-prolyl-
pyrrolidine) amide (5.1 g, 16.0 mmol) and L-proline methyl
ester HCl salt (2.6 g, 16.0 mmol) were coupled according to
procedure A. Purification was by flash chromatography. Yield
4.2 g (9.8 mmol, 61%); MS (ESI), m/z 428 ([MH]+); 1H NMR δ
1.30 (m, 12H), 2.67-3.98 (m, 8H), 3.78 (s, 3H), 4.27-4.85 (m,
2H), 7.38-7.77 (m, 4H). Anal. (C23H29N3O5‚0.4H2O) C, H, N.
Isoph th alic Acid L-P r olin e L-P r olyl-pyr r olidin e Am ide.
The methyl ester group of 3i was hydrolyzed according to
procedure D. Yield 3.2 g (7.9 mmol, 81%).
Isop h th a lic Acid L-P r olylben zyla m in e L-P r olyl-p yr r ol-
id in e Am id e (3a ). Isophthalic acid L-proline L-prolyl-pyrro-
lidine amide (0.65 g, 1.6 mmol) and benzylamine (0.17 mL,
1.6 mmol) were coupled according to procedure A. Purification
was by flash chromatography. Yield 0.48 g (0.95 mmol, 60%);
Isop h th a lic Acid Bis(L-a la n yl-p yr r olid in e) Am id e (2b).
Boc-L-alanyl-pyrrolidine (1.2 g, 5.0 mmol) was deprotected
according to procedure B. The L-alanyl-pyrrolidine trifluoro-
acetic acid salt was reacted according to procedure C. Purifica-
tion was by flash chromatography (yield 0.98 g, 2.4 mmol,
95%). MS (ESI), m/z 415 ([MH]+); 1H NMR δ 1.45 (d, 6H, J )
6.9 Hz), 1.86-2.05 (m, 8H), 3.42-3.75 (m, 8H), 4.95 (br qui,
2H, J ≈ 7 Hz), 7.46-8.31 (m, 4H), 7.53 (br d, 2H, J ) 7.4 Hz);
13C NMR δ 18.26, 24.16, 26.10, 46.15, 46.51, 47.41, 125.62,
128.80, 130.33, 134.46, 165.74, 171.00. Anal. (C22H30N4O4‚
0.2H2O) C, H, N.
1
MS (ESI), m/z 503 ([MH]+); H NMR δ 1.30-2.71 (m, 12H),
2.97-3.74 (m, 8H), 3.74-4.85 (m, 4H), 7.06-7.76 (m, 9H); 13
C
NMR δ 24.19, 25.43, 25.58, 26.23, 27.41, 28.97, 43.53, 46.07,
46.43, 50.30, 50.49, 58.28, 60.12, 125.98, 127.26, 127.50,
128.46, 128.64, 128.73, 129.20, 136.35, 136.80, 138.38, 168.40,
170.31, 170.49, 170.89. Anal. (C29H34N4O4‚0.3H2O) C, H, N.
Isop h th a lic a cid L-p r olyla zep a n e L-p r olyl-p yr r olid in e
a m id e (3b) was prepared according to the method described
1
for 3a . MS (ESI), m/z 495 ([MH]+); H NMR δ 1.30-2.32 (m,
Isoph th alic acid bis(L-ph en ylalan yl-pyr r olidin e) am ide
(2c) was prepared according to the method described for 2b.
MS (ESI), m/z 567 ([MH]+); 1H NMR δ 1.52-1.82 (m, 8H),
2.61-2.68 (m, 2H), 3.08-3.21 (m, 4H), 3.30-3.36 (m, 2H),
3.43-3.52 (m, 4H), 5.11-5.15 (m, 2H), 7.21-7.30 (m, 10H),
7.43-8.29 (m, 4H), 7.60 (br d, 2H, J ) 8.0 Hz); 13C NMR δ
24.04, 25.78, 39.55, 45.87, 46.47, 53.09, 125.58, 127.03, 128.46,
128.81, 129.50, 130.54, 134.20, 136.46, 165.89, 169.86. Anal.
(C34H38N4O4‚1.5H2O) C, H, N.
Isop h th a lic a cid bis(L-m eth ion yl-p yr r olid in e) a m id e
(2d ) was prepared according to the method described for 2b.
MS (ESI), m/z 535 ([MH]+); 1H NMR δ 1.85-2.14 (m, 6H), 2.12
(s, 3H), 2.58-2.68 (m, 2H), 3.41-3.48 (m, 1H), 3.56-3.65 (m,
2H), 3.79-3.85 (m, 1H), 5.08-5.14 (m, 2H), 7.33-8.22 (m, 4H),
7.80 (br d, J ) 8.1 Hz); 13C NMR δ 15.75, 24.21, 26.14, 30.56,
32.11, 46.19, 46.64, 50.67, 125.83, 128.56, 130.27, 133.93,
166.10, 170.25. Anal. (C26H38N4O4S2‚0.4H2O) C, H, N.
Isop h th a lic Acid Mon o(L-a la n yl-p yr r olid in e) Am id e.
Boc-L-alanyl-pyrrolidine (2.4 g, 10 mmol) was deprotected
according to procedure B. Isophthalic acid monomethyl ester
(1.8 g, 10 mmol) and the L-alanyl-pyrrolidine trifluoroacetic
acid salt were coupled according to procedure A. The methyl
ester group was hydrolyzed according to procedure D. Yield
1.9 g (6.6 mmol, 66%).
20H), 2.70-4.00 (m, 12H), 4.27-4.52 (m, 2H), 7.35-7.81 (m,
4H); 13C NMR δ 24.20, 25.56, 26.25, 26.77, 26.93, 26.95, 27.58,
29.01, 29.13, 29.58, 46.03, 46.42, 46.53, 47.95, 50.29, 50.33,
56.83, 58.21, 126.04, 128.20, 129.02, 129.07, 136.68, 136.78,
168.54, 168.67, 170.36, 171.59. Anal. (C28H38N4O4‚0.9H2O) C,
H, N.
Mon o-Boc-p ip er a zin e. A solution of di-tert-butyl dicar-
bonate (2.2 g, 10 mmol) in dichloromethane (20 mL) was added
slowly to a solution of piperazine (1.7 g, 20 mmol) in dichlo-
romethane at 0 °C. The reaction mixture was stirred at room
temperature overnight. The dichloromethane solution was
washed with saturated NaHCO3 and water. Diethyl ether (60
mL) was added, and the organic phase was washed with water.
The organic phase was dried with anhydrous Na2SO4 and
evaporated. Purification was by flash chromatography. Yield
1.0 g (5.4 mmol, 54%).
Isop h th a lic a cid L-p r olyl-(Boc-p ip er a zin e) L-p r olyl-
p yr r olid in e a m id e (3d ) was prepared according to the
method described for 3a . MS (ESI), m/z 582 ([MH]+); 1H NMR
δ 1.44 (s, 1.4H), 1.45 (s, 0.7H), 1.47 (s, 4.8H), 1.48 (s, 1.9H),
1.60-2.30 (m, 12H), 2.66-4.98 (m, 16H), 4.23-5.06 (m, 2H),
7.31-7.79 (m, 4H); 13C NMR δ 24.19, 25.57, 26.24, 28.39, 28.98,
29.39, 30.80, 42.17, 45.72, 46.03, 46.40, 47.14, 50.18, 50.30,
56.35, 58.23, 60.34, 80.21, 126.06, 128.27, 128.99, 129.15,
136.46, 136.70, 154.62, 168.54, 168.65, 170.30, 170.58. Anal.
(C31H43N5O6‚1.2H2O) C, H, N.
Isoph th alic Acid L-Alan yl-pyr r olidin e L-P r olin e Am ide.
Isophthalic acid mono(L-alanyl-pyrrolidine) amide (1.9 g, 6.6
mmol) and L-proline methyl ester HCl salt (1.1 g, 6.6 mmol)
were coupled according to procedure A. Purification was by
Isop h t h a lic Acid 2(S)-L-P r olyl-p ip er a zin e L-P r olyl-
p yr r olid in e Am id e (3c). Compound 3d (0.21 g, 0.36 mmol)

Prolyl Oligopeptidase Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 21 4549
was deprotected using ethyl acetate saturated with HCl
according to procedure B (methanol had to added to increase
the solubility, and it was reacted overnight). Purification was
by flash chromatography. The product was crystallized from
dichloromethane and ethyl acetate saturated with HCl. Yield
0.15 g (0.29 mmol, 80%); MS (ESI), m/z 482 ([MH]+); 1H NMR
(DMSO-d6) δ 1.44-2.32 (m, 12H), 2.58-4.01 (m, 16H), 4.34-
5.02 (m, 2H), 7.14-7.62 (m, 4H), 9.10-9.80 (m, 2H). Anal.
(C26H35N5O4‚HCl‚1.8H2O) C, H, N.
flash chromatography. Yield 1.19 g (2.8 mmol, 68%); MS (ESI),
m/z 427 ([MH]+); 1H NMR δ 1.30-2.36 (m, 16H), 2.60 (qui,
0.1H, J ) 8.0 Hz), 3.12 (qui, 0.9H, J ) 8.0 Hz), 3.47-3.67 (m,
4H), 3.78 (br s, 3H), 4.29-4.88 (m, 2H), 7.37-7.77 (m, 4H).
Anal. (C24H30N2O5) C, H, N.
Isop h th a lic Acid 2(S)-(Cyclop en ta n eca r bon yl)p yr r o-
lid in e L-P r olin e Am id e. The methyl ester group of 3h (0.89
g, 2.1 mmol) was hydrolyzed according to procedure D. Yield
0.87 g (2.1 mmol, 100%). Isophthalic acid 2(S)-(cyclohexane-
carbonyl)pyrrolidine L-proline amide, isophthalic acid 2(S)-
isobutanoylpyrrolidine L-proline amide, isophthalic acid 2(S)-
acetylpyrrolidine L-proline amide, and isophthalic acid 2(S)-
benzoylpyrrolidine L-proline amide were prepared according
to the same method.
Isop h th a lic Acid 2(S)-(Cyclop en ta n eca r bon yl)p yr r ol-
id in e L-P r olyl-2(S)-cya n op yr r olid in e Am id e (3l). Boc-2(S)-
cyanopyrrolidine (0.42 g, 2.1 mmol) was deprotected according
to procedure B. Isophthalic acid 2(S)-(cyclopentanecarbonyl)-
pyrrolidine L-proline amide (0.87 g, 2.1 mmol) and the 2(S)-
cyanopyrrolidine trifluoroacetic acid salt were coupled accord-
ing to procedure A. Purification was by flash chromatography.
Yield 0.23 g (0.47 mmol, 22%); MS (ESI), m/z 491 ([MH]+); 1H
NMR δ 1.33-2.32 (m, 20H), 2.55-4.03 (m, 7H), 4.15-4.88 (m,
3H), 7.35-7.75 (m, 4H). Anal. (C28H34N4O4‚0.1H2O) C, H, N.
Isop h th a lic a cid 2(S)-(cycloh exa n eca r bon yl)p yr r oli-
d in e L-p r olyl-2(S)-cya n op yr r olid in e a m id e (3m ) was pre-
pared according to the method described for 3l. MS (ESI), m/z
505 ([MH]+); 1H NMR δ 1.18-2.32 (m, 22H), 2.56-2.73 (m,
1H), 3.16-4.03 (m, 6H), 4.16-4.90 (m, 3H), 7.31-7.75 (m, 4H).
Anal. (C29H36N4O4‚0.3H2O) C, H, N.
Isop h th a lic a cid 2(S)-a cetylp yr r olid in e L-p r olyl-2(S)-
cya n op yr r olid in e a m id e (3n ) was prepared according to the
method described for 3l. MS (ESI), m/z 437 ([MH]+); 1H NMR
δ 1.71-2.33 (m, 12H), 2.29 (s, 3H), 2.67-4.02 (m, 6H), 4.20-
4.88 (m, 3H), 7.38-7.76 (m, 4H). Anal. (C24H28N4O4‚0.2H2O)
C, H, N.
Isop h th a lic a cid 2(S)-isobu ta n oylp yr r olid in e L-p r olyl-
2(S)-cya n op yr r olid in e a m id e (3o) was prepared according
to the method described for 3l. MS (ESI), m/z 465 ([MH]+); 1H
NMR δ 1.17 (d, 3H, J ) 6.9 Hz), 1.24 (d, 3H, J ) 6.9 Hz),
1.76-2.32 (m, 12H), 2.62-4.03 (m, 7H), 4.15-4.91 (m, 3H),
7.32-7.75 (m, 4H). Anal. (C26H32N4O4‚0.5H2O) C, H, N.
Isop h th a lic a cid 2(S)-ben zoylp yr r olid in e L-p r olyl-2(S)-
cya n op yr r olid in e a m id e (3p ) was prepared according to the
method described for 3l. MS (ESI), m/z 499 ([MH]+); 1H NMR
δ 1.70-2.46 (m, 12H), 2.60-4.02 (m, 6H), 4.14-4.91 (m, 2H),
5.10-5.71 (m, 1H), 7.12-8.06 (m, 9H). Anal. (C29H30N4O4‚
0.7H2O) C, H, N.
Isop h th a lic Acid 2(S)-(Cyclop en ta n eca r bon yl)p yr r ol-
id in e L-P r olyl-2(S)-(h yd r oxya cet yl)p yr r olid in e Am id e
(3q). Boc-2(S)-(acetoxyacetyl)pyrrolidine (0.63 g, 2.3 mmol)
was deprotected according to procedure B. Isophthalic acid
2(S)-(cyclopentanecarbonyl)pyrrolidine L-proline amide (0.90
g, 2.2 mmol) and the 2(S)-(acetoxyacetyl)pyrrolidine trifluo-
roacetic acid salt were coupled according to procedure A.
Purification was by flash chromatography. The O-acetyl group
was hydrolyzed according to procedure E. Purification was by
flash chromatography. Yield 0.23 g (0.44 mmol, 19%); MS
(ESI), m/z 524 ([MH]+); 1H NMR δ 1.30-2.33 (m, 20H), 2.54-
4.06 (m, 7H), 4.16-4.87 (m, 5H), 7.34-7.75 (m, 4H). Anal.
(C29H37N3O6‚0.4H2O) C, H, N.
Isop h th a lic Acid 2(S)-(Cyclop en ta n eca r bon yl)p yr r ol-
id in e L-P r olyl-p yr r olid in e Am id e (3e). Boc-2(S)-(cyclopen-
tanecarbonyl)pyrrolidine (0.40 g, 1.5 mmol) was deprotected
according to procedure B. Isophthalic acid mono(L-prolyl-
pyrrolidine) amide (0.49 g, 1.5 mmol) and the 2(S)-(cyclopen-
tanecarbonyl)pyrrolidine trifluoroacetic acid salt were coupled
according to procedure A. Purification was by flash chroma-
tography. Yield 0.46 g (0.99 mmol, 66%); MS (ESI), m/z 466
1
([MH]+); H NMR δ 1.29-2.31 (m, 20H), 2.52-3.98 (m, 8H),
4.28-4.87 (m, 2H), 7.35-7.78 (m, 4H); 13C NMR δ 24.19, 25.48,
25.58, 26.06, 26.18, 26.24, 28.61, 28.70, 29.00, 29.60, 46.06,
46.42, 49.19, 50.30, 50.33, 58.27, 64.46, 125.99, 128.36, 129.00,
129.15, 136.44, 136.65, 168.60, 168.64, 170.36, 210.92. Anal.
(C27H35N3O4‚0.2H2O) C, H, N.
Isop h th a lic a cid 2(S)-a cetylp yr r olid in e L-p r olyl-p yr -
r olid in e a m id e (3f) was prepared according to the method
described for 3e. MS (ESI), m/z 412 ([MH]+); 1H NMR δ 1.60-
2.32 (m, 12 H), 2.29 (s, 3H), 2.67-3.99 (m, 8H), 4.29-4.87 (m,
2H), 7.36-7.80 (m, 4H); 13C NMR δ 24.18, 25.39, 25.60, 26.23,
27.06, 28.21, 28.98, 46.04, 46.40, 50.24, 50.33, 58.26, 65.43,
125.99, 128.42, 129.01, 129.25, 136.21, 136.63, 168.46, 168.93,
170.28, 206.35. Anal. (C23H29N3O4‚0.7H2O) C, H, N.
Isop h th a lic a cid 2(S)-ben zoylp yr r olid in e L-p r olyl-p yr -
r olid in e a m id e (3g) was prepared according to the method
described for 3e. MS (ESI), m/z 474 ([MH]+); 1H NMR δ 1.50-
2.44 (12H), 2.16-3.98 (m, 8H), 4.27-4.86 (m, 1H), 5.16-5.71
(m, 1H), 7.15-8.08 (m, 9H); 13C NMR δ 24.20, 25.39, 25.61,
26.25, 29.00, 29.50, 46.05, 46.43, 50.15, 50.34, 58.28, 61.51,
126.04, 128.11, 128.37, 128.58, 128.69, 129.09, 129.16, 133.30,
135.36, 136.51, 168.55, 168.60, 170.37, 197.41. Anal. (C28H31
-
N3O4‚0.7H2O) C, H, N.
Isop h th a lic Acid Bis-2(S)-(cyclop en ta n eca r bon yl)p yr -
r olid in e Am id e (3j). Boc-2(S)-(cyclopentanecarbonyl)pyrrol-
idine (0.61 g, 2.28 mmol) was deprotected according to
procedure B. The 2(S)-(cyclopentanecarbonyl)pyrrolidine tri-
fluoroacetic acid salt in dichloromethane was reacted according
to procedure C. Purification was by flash chromatography.
Yield 0.45 g (0.98 mmol, 86%); MS (ESI), m/z 465 ([MH]+); 1H
NMR δ 1.29-2.05 (m, 22H), 2.20-2.32 (m, 2H), 2.53-2.63 (m,
0.4H), 3.07-3.15 (m, 1.6H), 3.43-3.83 (m, 4H), 4.36-4.57 (m,
0.4H), 4.81-4.88 (m, 1.6H), 7.23-7.75 (m, 4H); 13C NMR δ
26.05, 26.05, 26.19, 28.63, 28.65, 29.59, 49.27, 50.31, 64.42,
125.90, 128.43, 129.02, 136.54, 168.57, 210.86. Anal. (C28H36
-
N2O4‚0.2H2O) C, H, N.
Isop h th a lic a cid bis-2(S)-a cetylp yr r olid in e a m id e (3k )
was prepared according to the method described for 3j. MS
(ESI), m/z 357 ([MH]+); 1H NMR δ 1.84-2.28 (m, 8H), 2.29
(bs, 6H), 3.45-3.81 (m, 4H), 4.40-4.47 (m, 0.4H), 4.71-4.77
(m, 1.6H), 7.14-7.78 (m, 4H); 13C NMR δ 25.38, 27.20, 28.20,
50.22, 65.38, 126.05, 128.49, 128.56, 129.07, 136.29, 136.35,
168.75, 206.25. Anal. (C20H24N2O4‚0.4H2O) C, H, N: calcd, 7.70;
found, 7.16.
Isop h th a lic Acid Mon o-2(S)-(cyclop en ta n eca r bon yl)-
p yr r olid in e Am id e. Boc-2(S)-(cyclopentanecarbonyl)pyrroli-
dine (1.3 g, 4.9 mmol) was deprotected according to procedure
B. Isophthalic acid monomethyl ester (0.88 g, 4.9 mmol) and
the 2(S)-(cyclopentanecarbonyl)pyrrolidine trifluoroacetic acid
salt were coupled according to procedure A. The methyl ester
group was hydrolyzed according to procedure D. Yield 1.28 g
(4.1 mmol, 84%).
Isop h th a lic Acid 2(S)-(Cyclop en ta n eca r bon yl)p yr r ol-
id in e (L-P r olin e Meth yl Ester ) Am id e (3h ). Isophthalic acid
mono-2(S)-(cyclopentanecarbonyl)pyrrolidine amide (1.28 g, 4.1
mmol) and L-proline methyl ester HCl salt (0.75 g, 4.5 mmol)
were coupled according to procedure A. Purification was by
Isop h th a lic a cid 2(S)-a cetylp yr r olid in e L-p r olyl-2(S)-
(h yd r oxya cetyl)p yr r olid in e a m id e (3r ) was prepared ac-
cording to the method described for 3q. MS (ESI), m/z 470
1
([MH]+); H NMR δ 1.34-2.34 (m, 12H), 2.29 (s, 3H), 2.68-
4.06 (m, 6H), 4.22-4.84 (m, 5H), 7.37-7.77 (m, 4H). Anal.
(C25H31N3O6‚0.2H2O) C, H, N.
3,3-Dim eth yl Glu ta r ic Acid Mon o-2(S)-(cyclop en ta n e-
ca r bon yl)p yr r olid in e Am id e. Boc-2(S)-(cyclopentanecar-
bonyl)pyrrolidine (1.3 g, 5.0 mmol) was deprotected according
to procedure B. The 2(S)-(cyclopentanecarbonyl)pyrrolidine
trifluoroacetic acid salt and triethylamine (2.3 mL, 16.5 mmol)
were dissolved in tetrahydrofuran (50 mL). A solution of 3,3-
dimethyl glutaric acid anhydride (0.71 g, 5.0 mmol) in tet-

4550 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 21
Walle´n et al.
rahydrofuran (20 mL) was added at 0 °C. The reaction was
stirred overnight at room temperature. The solvent was
evaporated, and the residue was dissolved in water and ethyl
acetate. The phases were separated, and the ethyl acetate
phase was washed with 0.1 M HCl. The ethyl acetate phase
was dried with anhydrous Na2SO4 and evaporated. Yield
1.8 g.
interface module (D-7000), a pump (L-7100), an autosampler
(L-7250), and a Purospher RP-C18e column (125 × 4 mm,
5 µm).
Ack n ow led gm en t. This research was supported by
Finncovery Ltd. and the National Technology Agency
in Finland (TEKES). We thank Tiina Koivunen, Helly
Rissanen, and Pa¨ivi Sutinen for outstanding technical
assistance, Prof. Dr. J ouko Vepsa¨la¨inen for help in
assigning the NMR spectra, Dr. Seppo Auriola and Unni
Tengvall, M.Sc., for performing the ESI-MS analysis,
and Dr. Pekka J arho for guidance in the log P measure-
ments.
3,3-Dim eth yl Glu ta r ic Acid 2(S)-(Cyclop en ta n eca r bon -
yl)p yr r olid in e L-P r olin e Am id e. 3,3-Dimethyl glutaric acid
mono-2(S)-(cyclopentanecarbonyl)pyrrolidine amide (0.93 g,
3.0 mmol) and proline methyl ester HCl salt (0.50 g, 3.0 mmol)
were coupled according to procedure A. The methyl ester
group was hydrolyzed according to procedure D. Yield 1.2 g
(3.0 mmol, 100%).
3,3-Dim eth ylglu ta r ic 2(S)-(cyclop en ta n eca r bon yl)p yr -
r olid in e L-p r olyl-2(S)-cya n op yr r olid in e a m id e (4a ) was
prepared according to the method described for 3l. MS (ESI),
m/z 485 ([MH]+); 1H NMR δ 1.17-1.21 (m, 6H), 1.52-2.61
(m, 24H), 2.95-3.09 (m, 1H), 3.47-3.88 (m, 6H), 4.16-4.82
(m, 3H). Anal. (C27H40N4O4‚0.4H2O) C, H, N.
Refer en ces
(1) Polga´r, L. The prolyl oligopeptidase family. Cell. Mol. Life Sci.
2002, 59, 349-362.
(2) Wilk, S. Prolyl endopeptidase. Life Sci. 1983, 33, 2149-2157.
(3) Mentlein, R. Proline residues in the maturation and degradation
of peptide hormones and neuropeptides. FEBS Lett. 1988, 234,
251-256.
(4) Cunningham, D. F.; O’Connor, B. Proline specific peptidases.
Biochim. Biophys. Acta 1997, 1343, 160-186.
(5) Huston, J . P.; Hasenohrl, R. U. The role of neuropeptides in
learning: focus on the neurokinin substance P. Behav. Brain
Res. 1995, 66, 117-127.
(6) Kowall, N. W.; Beal, M. F.; Busgiglio, J .; Duffy, L. K.; Yankner,
B. A. An in vivo model for the neurodegenerative effects of beta
amyloid and protection by substance P. Proc. Natl. Acad. Sci.
U.S.A. 1991, 88, 7247-7251.
(7) Laitinen, K. S. M.; van Groen, T.; Tanila, H.; Vena¨la¨inen, J .;
Ma¨nnisto¨, P. T.; Alafuzoff, I. Brain prolyl oligopeptidase activity
is associated with neuronal damage rather than â-amyloid
accumulation. Neuroreport 2001, 12, 3309-3312.
(8) Rampon, C.; J iang, 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.
(9) J iang, 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.
(10) Toide, K.; Fujiwara, T.; Iwamoto, Y.; Shinoda, M.; Okamiya, K.;
Kato, T. Effect of a novel prolyl endopeptidase inhibitor, J TP-
4819, on neuropeptide metabolism in the rat brain. Naunyn
Schmied. Arch. Pharmacol. 1996, 353, 355-362.
(11) Yoshimoto, T.; Kado, K.; Matsubara, F.; Koriyama, N.; Kaneto,
H.; Tsuru, D. Specific inhibitors for prolyl endopeptidase and
their anti-amnesic effect. J . Pharmacobio-Dyn. 1987, 10, 730-
735.
(12) Atack, J . R.; Suman-Chauhan, N.; Dawson, G.; Kulagowski, J .
J . In vitro and in vivo inhibition of prolyl endopeptidase. Eur.
J . Pharmacol. 1991, 205, 157-163.
(13) Miura, N.; Shibata, S.; Watanabe, S. Increase in the septal
vasopressin content by prolyl endopeptidase inhibitors in rats.
Neurosci. Lett. 1995, 196, 128-130.
(14) 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.
(15) De Nanteuil, G.; Portevin, B.; Lepagnol, J . Prolyl endopeptidase
inhibitors: a new class of memory enhancing drugs. Drugs
Future 1998, 23, 167-179.
(16) 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.
3,3-Dim eth ylglu ta r ic a cid 2(S)-(cyclop en ta n eca r bon -
yl)p yr r olid in e L-p r olyl-2(S)-(h yd r oxya cetyl)p yr r olid in e
a m id e (4b) was prepared according to the method described
1
for 3q. MS (ESI), m/z 518 ([MH]+); H NMR δ 1.17-1.20 (m,
6H), 1.52-2.59 (m, 24H), 2.95-3.08 (m, 1H), 3.16 (br s, 1H),
3.52-3.68 (m, 5H), 3.87-3.94 (m, 1H), 4.28-4.43 (m, 2H),
4.63-4.70 (m, 3H). Anal. (C28H43N3O6‚0.2H2O) C, H, N.
In Vitr o Assa y for P OP Activity. The whole pig brains,
excluding cerebellum and most of the brain stem, of three pigs
were frozen in liquid nitrogen within 30 min from slaughtering
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 superna-
tants were pooled and stored in small aliquots at -80 °C until
used. The supernatant was thawed in ice and diluted (1:2) with
homogenization buffer. In the microplate assay procedure, 10
µL of the enzyme preparation (protein concentration ) 4.3 mg/
mL) was preincubated with 460 µL of 0.1 M sodium-potas-
sium 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 (final DMSO
concentration was <0.1%). 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-amino-4-methylcoumarin dis-
solved 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). Formation of 7-amino-4-methylcou-
marin was determined fluorometrically with microplate fluo-
rescence reader (excitation at 360 nm and emission at 460 nm).
In the uninhibited reaction, the specific POP activity was 3.8
nmol/min/mg of protein. The final concentration of the com-
pounds in the assay mixture varied from 10-12 to 10-5 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 GraphPad Prism
3.02 software.
Deter m in a tion of log P Va lu es. A known concentration
of an inhibitor in phosphate buffer (saturated with 1-octanol,
pH 7.4) was shaken with a suitable volume of 1-octanol for 60
min at room temperature. The phases were separated by
centrifugation for 5 min at 2000 rpm, and the aqueous phase
was analyzed. The partition coefficient was calculated in
relation to a control that was treated in the same way as the
samples but did not contain 1-octanol. Each partition coef-
ficient was determined at least in triplicate. For each HPLC
method 20 mM KH2PO4 (pH 7) was used as the aqueous phase.
As the organic phase, methanol was used for compound 4b
and 90% acetonitrile for the other compounds. The HPLC
methods were tested for linearity and repeatability. The Merck
Hitachi HPLC system consisted of a UV detector (L-7400), an
(17) Saito, M.; Hashimoto, M.; Kawaguchi, N.: Shibata, H.; Fukami,
H.; Tanaka, T.; Higuchi, N. Synthesis and inhibitory activity of
acyl-peptidyl-pyrrolidine derivatives toward post-proline cleaving
enzyme; a study of subsite specificity. J . Enzyme Inhib. 1991,
5, 51-75.
(18) Toide, K.; Iwamoto, Y.; Fujiwara, T.; Abe, H. J TP-4819: a novel
prolyl endopeptidase inhibitor with potential as a cognitive
enhancer. J . Pharmacol. Exp. Ther. 1995, 274, 1370-1378.
(19) Vena¨la¨inen, J . I.; J uvonen, R. O.; Forsberg, M. M.; Garcia-
Horsman, A.; Poso, A.; Wallen, E. A. A.; Gynther, J .; Ma¨nnisto¨,
P. T. Substrate dependent, non-hyperbolic kinetics of pig brain
prolyl oligopeptidase and its tight binding inhibition by J TP-
4819. Biochem. Pharmacol. 2002, 64, 463-471.
(20) Fu¨lo¨p, V.; Szeltner, Z.; Renner, V.; Polgar, L. Structures of prolyl
oligopeptidase substrate/inhibitor complexes. J . Biol. Chem.
2001, 276, 1262-1266.

Prolyl Oligopeptidase Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 21 4551
(21) 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.
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.
(24) Houssin, R.; Bernier, J .-L.; He´nichart, J .-P. A convenient and
general method for the preparation of tert-butoxycarbonylami-
noalkanenitriles and their conversion to mono-tert-butoxycar-
bonylalkanediamines. Synthesis 1988, 259-261.
(22) Fu¨lo¨p, V.; Bo¨cskei, Z.; Polga´r, L. Prolyl oligopeptidase: an
unusual â-propeller domain regulates proteolysis. Cell 1998, 94,
161-170.
(23) Arai, H.; Nishioka, H.; Niwa, S.; Yamanaka, T.; Tanaka, Y.;
Yoshinaga, K.; Kobayashi, N.; Miura, N.; Ikeda, Y. Synthesis of
J M030811O