ACE inhibitors hypothesis generation for selective design, synthesis and biological evaluation of 3-mercapto-2-methyl-propanoyl-pyrrolidine-3-imine derivatives as antihypertensive agents

Article abstract of DOI:10.1016/j.bmc.2009.03.008

A series of new 3-mercapto-2-methyl-propanoyl-pyrrolidine derivatives (V, VIa-e) were designed. A new validated ACE inhibitors pharmacophore model (hypothesis) was generated for the first time in this research from the biologically active (frozen) conform

Full text of DOI:10.1016/j.bmc.2009.03.008

Bioorganic & Medicinal Chemistry 17 (2009) 3739–3746
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
ACE inhibitors hypothesis generation for selective design, synthesis and
biological evaluation of 3-mercapto-2-methyl-propanoyl-pyrrolidine-3-imine
derivatives as antihypertensive agents
b
a
a
Mohamed A. H. Ismail a, , M. Nabil Aboul-Enein , Khaled A. M. Abouzid , Dalal A. Abou El Ella ,
*
Nasser S. M. Ismail a
a Pharmaceutical Chemistry Department, Faculty of Pharmacy, Ain Shams University, ElKhalifa ElMaamoon St., 11566 Abbasseya, Cairo, Egypt
b Laboratory of Pharmaceutical Sciences, National Research Centre, Dokki, Egypt
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of new 3-mercapto-2-methyl-propanoyl-pyrrolidine derivatives (V, VIa–e) were designed. A new
validated ACE inhibitors pharmacophore model (hypothesis) was generated for the first time in this
research from the biologically active (frozen) conformation of Lisinopril–Human ACE complex that was
downloaded from PDB, using stepwise technique of CATALYST modules. The molecular modeling com-
pare–fit study of the designed molecules (V, VIa–e), with such ACE inhibitors hypothesis was fulfilled,and
several compounds showed significant high simulation fit values. The compounds with high fit values
were synthesized and biologically evaluated in vivo as hypotensive agents. It appears that the in vivo
hypotensive activity of compounds V, VIa, VIb, and VIe was consistent with their molecular modeling
results, and compound VIe showed the highest activity in comparison to Captopril.
Received 4 December 2008
Revised 1 March 2009
Accepted 3 March 2009
Available online 10 March 2009
Keywords:
Molecular modeling
ACE inhibitors hypothesis generation
Hypotensive agents
3-Mercapto-2-methyl-
propanoylpyrrolidines
Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction
The design of new 3-mercapto-2-methyl-propanoyl-pyrrolidine
derivatives, as ACE inhibitors, was based on structural modification
Renin is a proteolytic enzyme produced mainly in the juxta-glo-
merular apparatus of the kidney, which acts on the circulating
R-globulin angiotensinogen produced by the liver resulting in the
formation of angiotensin I (Ang I), a decapeptide that has very little
biological activity. Angiotensin-converting enzyme (ACE) then con-
verts Ang I into the octapeptide Ang II,1,2 a potent vasoconstrictor
that plays an integral role in the pathophysiology of hyperten-
sion,3–5 through sympathetic Ang II receptor activity.6 This directed
many researchers toward designing drugs to inhibit the effects of
ACE.7
ACE inhibitors have gained wide acceptance clinically and are
commonly prescribed for the treatment of hypertension, conges-
tive heart failure,8,9 protection against hypertension-related organ
damage,10,11 The first breakthrough in the area of ACEIs was Capto-
prilÒ,12 which was launched in 1981 for hypertension manage-
ment. Furthermore new ACEI leads; 1,13 2,13 3,14 4,15 5,15 and
Lisinopril16 (Fig. 1) were also developed to have more activity as
antihypertensive agent than Captopril. Our current investigation
is based on developing new Me-Too ACEIs bearing pyrrolidine ring
system which possess potential antihypertensive activity.
of the lead compound; Captopril, aiming to get optimum activity.
These modifications were performed through the following strate-
gies: considering the SAR of the lead compounds that have ACE
inhibitor activity, prompted us to retain the sulfhydryl-side chain
linked to pyrrolidine ring system. This is because, it was reported
that SH function of the ACE inhibitor could strikingly bind to Zn
ion (present in the skeleton of the ACE).13,16 In addition, the
p-methoxybenzyl moiety as well as substituted imino functions
was introduced onto the pyrrolidine ring system, in order to get
the essential lipophilic pharmacophores in the designed molecules
(V and VIa–e).13 Meanwhile, such lipophilic features could
improvethepharmacokineticproperties andpotenciesof thesemol-
ecules.13 Furthermore, the design of the proposed molecules (V and
VIa–e) was also, based upon molecular modeling studies between a
newly generated ACEIs pharmacophore model (hypothesis) and
these test set proposed molecules (V and VIa–e). In this research,
the ACEIs hypothesis was generated for the first time from the
docked bioactive conformer of Lisinopril molecule that was com-
plexed with the Human ACE. Such complex was downloaded from
the website of PDB and the docked conformer of Lisinopril was
selected and used without any conformational changes (as a frozen
bioactive conformation), to generate the ACEIs hypothesis using
CATALYST modules adopting stepwise technique. Also, such hypo-
* Corresponding author. Tel.: +20 12 3269674; fax: +20 22 4051107.
E-mail address: mhismaeel@yahoo.com (M.A.H. Ismail).
0968-0896/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2009.03.008

3740
M. A. H. Ismail et al. / Bioorg. Med. Chem. 17 (2009) 3739–3746
F
F
F
F
S
HS
COOH
O
O
N
O
O
O
F
F
N
N
N
HS
O
Ph
N
O
O
O
O
O
H
COOH
SH
O
O
O
Captopril
RO 68-7629
RO 67-7447
2
3
1
Ph
Ph
C)4
H2N(H2
S
S
O
HOOC
COO
N
COOH
N
O
S
N
1/2 Ca++
N
H
Ph
O
Ph
HS
O
HOOC
Lisinopril
Zofenoprilat
Zofenopril
5
4
Figure 1. The structures of the lead ACE inhibitors.
thesis was perfectly validated and used for virtual studies to predict
the biological activity of database molecules as ACEIs. The compare–
fit studies between generated hypothesis and V, VIa–e were per-
formed in order to prioritize their activities. The result of virtual
study revealed that the molecules (V and VIa–e) have high fitting
values and they could be considered active hit molecules as ACEIs.
Accordingly, the designed molecules (V and VIa–e) were
synthesized following Scheme 1 and evaluated for their hypoten-
sive activity on normotensive anesthetized cats. The study
revealed that the hypotensive activities of V and VIa–e were sim-
ilar to or higher than those of Captopril and matched with the vir-
tual screening.
COOC2H5
O
O
O
+
COOH
SH
N
Thioacetic
Methacrylic
acid
acid
O
b
I
a
O
O
O
O
S
Cl
III
N
HCl
H
II
c
O
O
N
O
O
S
IV
d
R
O
N
O
O
e
N
N
O
O
HS
HS
V
VIa-e
Scheme 1. Compounds: VIa, R = OH; VIb, R = CH2CH2OH; VIc, R = NH–CS–NH2; VId, R = CH2–C6H5; VIe, R = CH(CH3)COOC2H5. Reagents and conditions: (a) 5 N HCl, reflux 5 h;
(b) (i) aqueous KOH 7%; (ii) SOCl2, reflux 1 h; (c) anhyd K2CO3, acetonitrile; (d) 5.5 N NH4OH, MeOH; (e) NH2–R, anhyd CH3COONa, EtOH, reflux 6 h.

M. A. H. Ismail et al. / Bioorg. Med. Chem. 17 (2009) 3739–3746
3741
in its presumed biologically active conformation was selected
together with adjacent binding site (about 10 Å distance) in a
new independent file of PDB format (Fig. 2B), then the frozen
bioactive conformer of Lisinopril was isolated from the remaining
protein structures. Such conformer included atom types in its basic
architecture by only single bonds. Thus the structure needed mod-
ifications in order to correct the atom types and bond orders, using
Cerius2 3D-sketcher, then the corrected bioactive conformation of
Lisinopril was saved (without changing its conformation, i.e; fro-
zen conformation) (Fig. 2C and D).
The bioactive frozen conformer of Lisinopril was transferred to
Catalyst workstation and the reported crucial five features,13,18,19
required for effective ACE inhibitory activity, were added manu-
ally, depending on the reported data, to generate the hypothesis
in stepwise manner (Fig. 3A and B). The added crucial features18
are; two HB-Acceptor features (A1 and A2), one negative ionizable
feature (N), and two hydrophobic features (H1 and H2).
2. Results and discussion
2.1. Molecular modeling. Stepwise generation of the ACEIs
hypothesis
The molecular modeling studies for constructing the hypothesis
for the ACEIs, derived from the frozen conformation of Lisinopril
docked at the human ACE were performed using both Cerius2
and CATALYST modules. The crystal structure of human ACE in
complex with the Lisinopril (1086)17 was obtained from the World
Wide Web based protein data bank (PDB). The PDB file in a simpli-
fied view contains a listing of the amino acid sequence, then a list-
ing of connectivity data that contain information about how the
amino acid sequence (the primary structure) is arranged in three
dimensional space (Fig. 2A). Using the protein tools deck in Cerius2
Modules, all the non-protein residues present in the structure can
be identified and selected as independent models. Thus, Lisinopril
Figure 2. Isolation steps for bioactive conformation of Lisinopril.
Figure 3. (A and B) selected features for ACEIs.

3742
M. A. H. Ismail et al. / Bioorg. Med. Chem. 17 (2009) 3739–3746
Figure 4. Generation of ACE inhibitors hypothesis.
Figure 6. Constraint distances of ACEIs hypothesis.
O
N
O
N
N
5
3
-O
4
O
S
2
1
Du
Figure 5. Constraint distances between the active site of a reported hypothetical
bioactive lead.
The structure of Lisinopril was then deleted, and constraint
dimension (distances and angels between all the features) were
automatically determined and the corresponding hypothesis was
then generated using the stepwise technique of CATALYST mod-
ules.20 Such generated hypothesis, reported herein, in this investi-
gation, for the first time, showed all the features reported for a
potent ACE inhibitor18 (Fig. 4).
Figure 7. Constraint angles of ACEIs hypothesis.
Dammkoehler et al.18 reported hypothetical constraint
distances between the most essential groups in a biologically ac-
tive molecule as outlined in Figure 5.
2.1.1. Validation of the generated Catalyst hypothesis of ACE
inhibitors
(a) Mapping of Lisinopril and the other lead compounds such as
Captopril to the generated hypothesis showed complete
mapping of all the features with high fit values (4.9 and
3.9, respectively) and with low conformational energies
(3.7and 2.7) (Fig. 8 and Table 2).
Meanwhile, the same authors did not mention the constraint
angles between these essential groups. But, in this research we
reported the range of the constraint distances and the angles
between the essential features existed in the generated hypothesis,
as shown in Figures 6, 7 and Table 1.
Table 1
Constraint dimensions of the generated ACEIs hypothesis in comparison to Richard’s ACEI hypothetical bioactive molecule
Dimensions
Richard’s ACEI hypothetical
bioactive lead
Our catalyst ACEI hypothesis (values are recorded in ranges)
Constraint distances between
different features (Å)
1, 8.543; 2, 4.938; 3, 3.808; 4,
5.278; 5, 4.095
A1–N, 6.903–8.903; A1–A2, 4.042–6.042; A2–N, 2.203–4.203; A1–H2, 5.478–7.478; H2–N,
9.841–11.841
Constraint angles between different
features (°)
Not reported
A1–N–A2, 31.39–41.298; N–A1–H2, 92.372–102.34; A1–H2–N, 41.29–51.291; A1–A2–N,
140.90–150.90; A1–N–H2, 31.398–41.298
Figure 8. Mapping of (A) Lisinopril (B) compound V (C) hydrolyzed ester of compound VIe with the generated ACE inhibitors hypothesis.

M. A. H. Ismail et al. / Bioorg. Med. Chem. 17 (2009) 3739–3746
3743
Table 2
de-esterified VIe have high fitting values in comparison to the lead
compounds (Fig. 8 and Table 2).
Mapping of the best fit conformers of lead and test set compounds (V and VIa–e) to
the generated ACEIs hypothesis
Compd no.
No. of
conformers
Conf. energy
Fit value
(out of five)
2.1.3. Conclusions of simulation studies
(kcal molꢀ1
)
The above molecular modeling simulation studies indicated
that the pyrrolidine derivatives (V and VIa–e) have potential ACE
inhibitor activity and thereby they are considered as active antihy-
pertensive hits. It seems that the high fit values of V, VIa, VIb, and
VIe with the ACEIs hypothesis, are attributed to the hydrophobic
interaction of the p-methoxybenzyl group with the hydrophobic
feature (H2). Accordingly, we concluded that compounds V, VIa,
VIb, and VIe could be considered as promising active hits as ACE
inhibitors. Hence such active hits were synthesized (Scheme 1)
and evaluated for their hypotensive activity in normotensive adult
cats.
Lisinopril
Captopril
V
VIa
VIb
VIc
VId
210
131
125
85
126
175
149
220
3.7
2.7
6.4
5.8
5.6
13
4.9
3.9
3.95
3.33
3.47
2.9
12.3
9
3.1
3.87
De-esterified VIe
Table 3
Mapping of Catalyst database molecules with ACE inhibitors hypothesis
2.2. Synthesis
Database
Total numbers
of molecules
Number of the retrieved
(mapped) hits
% of retrieved
(mapped) hits
The intermediates (I,22 III23) were synthesized according to
reported methods. Compound I was deethoxycarboxylated and
N-deacetylated in one step through refluxing in 5 N HCl to afford
II. Acylation of II with equivalent amount of III, afforded the 2-
(4-methoxybenzyl)-1-acetylthio-2-methylpropanoyl) pyrrolidine-
3-one (IV) Hydrolysis of thioester function of (IV) using 5.5 N
methanolic ammonia gave the target compound; 2-(4-methoxy-
benzyl)-1-(30-mercapto-20-methyl propanoyl) pyrrolidine-3-one
(V). A new series of 30-mercapto-20-methyl-propanoyl-pyrroli-
dine-3-imines (VIa–e) were prepared by condensation of the
2-(4-methoxybenzyl)-1-(30-mercapto-20-methyl propanoyl) pyr-
rolidine-3-one (V) with the respective primary amines (viz;
hydroxylamine HCl, 2-ethanolamine, thiosemicarbazide, benzyl-
amine, and 2-aminopropionic acid ethyl ester) in the presence of
anhydrous sodium acetate and absolute ethanol (Scheme 1).
NCI 2000
238,819
55,273
2000
291
9
0
0.122
0.016
0
Maybridge 2001
MiniMaybridge
Total
296,092
300
0.101
(b) The simulated fitting values of such hypothesis with the
hypotensive test set compounds (V and VIe) were consistent
with the experimental results, as it will be mentioned later,
in the biological evaluation results (Table 2).
(c) Additional validation of our ACEIs hypothesis was achieved
by databases search study with the molecular structures of
the provided Catalyst databases (Maybridge 2001, MiniMay-
bridge 2001 and NCI 2000). The results showed that 0.101%
of the databases molecules was retrieved among 296,092
molecules (Table 3). The low percentage of the retrieved
molecules of the databases proved the high selectivity of
such hypothesis.21
2.3. Biological screening. Assessment of the hypotensive
activity in normotensive cats
In addition, it was found that the ACEI molecules; 3, 4, and 5
(Fig. 1), that exist among the used databases molecules, were
among the recognized mapped molecules in this search.
The above finding indicates the validity of the generated ACE
inhibitors hypothesis (Fig. 4).
Preliminary pharmacological testing of selected compounds (V,
VIa, VIb, and VIe), which have high compare/fit scoring value, was
carried out in anesthetized normotensive adult cats. All tested
compounds demonstrated a hypotensive activity on adult normo-
tensive cats (Table 4 and Graph 1).
Compounds V and VIe showed higher hypotensive activity in
comparison to Captopril at doses, 0.0014 and 0.0012 mmol/kg,
respectively. Whereas compounds VIa and VIb showed moderate
hypotensive activity in comparison to the same reference
compound at doses 0.0031 and 0.00285 mmol/kg, respectively.
The hypotensive effects of the different congeners of this group
were arranged in the following decreasing order: VIe > V > VIb > -
VIa. This indicated that the required SAR for these series of com-
pounds should involve 3-one and 3-(1-carbethoxy)ethylimino
moieties to exert higher hypotensive activity than Captopril, while
2.1.2. Molecular modeling simulation study of the generated ACE
inhibitors hypothesis with the test set molecules; 3-mercapto-2-
methylpropanoyl-pyrrolidine derivatives (V and VIa–e)
Examination of the mapping of the test set compounds (V and
VIa–e) to the generated ACE inhibitors hypothesis for predicting
their activity was performed using compare/fit (Best Fit) process.
Different mappings for all conformers of each compound of the test
set to the generated hypothesis were visualized. Such simulation
study revealed that pyrrolidine compounds V, VIa, VIb, and the
Table 4
Mean percentage reduction in the systolic and diastolic blood pressure of normotensive adult cats in comparison with their fitting affinities with ACEIs hypothesis
Compd
no.
Fit value (out of
five)
Conf. energy
Dose (mmol/
kg)
Systolic BP (mmHg)
mean SE
Mean change
%
Diastolic BP (mmHg)
mean SE
Mean change
%
(kcal molꢀ1
)
Control!
Captopril
V
—
3.9
3.95
3.33
3.47
3.87
—
134.1 2.01
105.9 1.4
104.9 23.65
121.9 0.89
106 2.76
—
125 1.35
97.6 1.56
—
2.7
6.4
5.8
5.6
9
0.00246
0.00246
0.00246
0.00246
0.00246
ꢀ20.91%
ꢀ21.67%
ꢀ9%
ꢀ21.89%
115
0
ꢀ8%
VIa
109.7 0.75
113.7 2.32
84 0.89
ꢀ12.22%
ꢀ9%
VIb
ꢀ20.83%
ꢀ30.37%
VIea
93.3 1.6
ꢀ32.8%
!
Statistically significant difference from the control group at p < 0.05.
Fit value is measured using the hydrolyzed ester form.
a

3744
M. A. H. Ismail et al. / Bioorg. Med. Chem. 17 (2009) 3739–3746
Systolic P
Diastolic P
35.00%
30.00%
25.00%
20.00%
15.00%
10.00%
5.00%
Figure 10. Atomic charge distribution on the de-esterified; VIe.
0.00%
V
VIa
VIb
VIe
Captopril
Graph 1. Percentage reduction in systolic and diastolic BP compared to the
reference compound (Captopril).
3. Experimental
All reagents were supplied by Aldrich, Merck, and Acros Organ-
ics, and were used with no further purification. TLC was performed
on Silica Gel 60 F254 plates (Merck). Melting points were deter-
mined on Stuart Scientific apparatus and are uncorrected. FT-IR
spectra were recorded on a Perkin–Elmer FT spectrophotometer.
1H NMR spectra were measured in d scale on a Perkin–Elmer
300 MHz spectrometer. EIMS spectra were recorded on Finnigan
Mat SSQ 7000 (70 eV) mass spectrometer. Elemental analyses were
performed at the Microanalytical Center, Cairo University, Egypt.
Preparation of 1-acetyl-4-ethoxycarbonyl-2-(4-methoxybenzy-
l)pyrrolidin-3-one (I) and 3-acetylthio-2-methyl-propionic acid
incorporation of 3-oxime or 3(2-hydroxyethylimine moieties will
lead to lower hypotensive activity.
2.4. Correlation between molecular modeling simulation
studies and the in vivo hypotensive activity
The in vivo hypotensive activity of the designed compounds (V
and VIa–e) was found to matching with the fit values. However,
compound VIe having the in vivo hypotensive activity was found
to be the most active molecule, in spite of V having higher fit va-
lue (3.95) than the fit value of VIe (3.87). A similar disparity was
explained in one of our previous reported researches,21 where it
was found that the existence of an electron-rich edge on the mol-
ecule may interfere in the interaction forces between the com-
pound and its binding site. Accordingly, we decided to
determine the atomic charge distribution for compounds VIe
and V, using Cerious2 software, version 4.7 Accelrys, in order to
predict the effect of charge distribution of these compounds on
the binding affinity with their binding sites. The result is given
in (Figs. 9 and 10). We could find that compound V has low elec-
tron-rich edge around the pyrrolidin-3-one ring system, which
may weakly interact with the complementary positive ionizable
feature of the receptor side, and hence it has lower hypotensive
activity. The opposite is true in compound VIe, where it showed
high electron-rich edge around the pyrrolidine-3-alkyl carboxyl-
ate moiety, and hence it could easily combine with the positive
ionizable feature of the binding site; that is, it will have higher
in vivo hypotensive activity than V.
chloride (III) was performed adopting reported procedures.22
A
solution of I (5 g, 0.084 mol) and 5 N hydrochloric acid (25 mL)
was refluxed for 5 h and evaporated under vacuum and the residue
was crystallized from isopropanol to give 4 g (82%) of II, mp: 225–
227 °C. IR 3340 and 1719 cmꢀ1 1H NMR (CDCl3) 2.8–3.2 (m, 6H,
.
C6H4CH2, N–CH2CH2C@O), 3.75 (s, 3H, OCH3), 4.4 (s, 1H, NCHC@O)
and 6.7–7 (dd, J = 8.4, 8.6 Hz, 4H, Ar-H).
3.1. 1-(3-Acetylthio-2-methylpropanoyl)-2-(4-methoxybenzyl)
pyrrolidin-3-one (IV)
A mixture of II (3.27 g, 17.5 mmol), 3-acetylthio-2-methyl-pro-
pionyl chloride (III) (8.75 g, 19.3 mmol), and potassium carbonate
(4.83 g, 35 mmol) in acetonitrile (100 mL) was stirred at 25 °C for
24 h. The reaction mixture was filtered and the filtrate was evapo-
rated under vacuum. The residue was diluted with water and the
resulting mixture was then extracted with ethyl acetate. The com-
bined organic phases were washed with water, dried over anhy-
drous magnesium sulfate, and evaporated under vacuum. The
appropriate products as crude oil residue were purified by column
chromatography using n-hexane/chloroform (3:1) to produce the
titled compound IV as pale yellow oil (4.9 g, 92%). IR
m (neat,
cmꢀ1): 3010, 2293, 1756, 1689 and 1642. 1H NMR (DMSO-d6) d
ppm 1.7 (d, J = 7.1 Hz, 3H, –CH–CH3), 2.1–3.6 (m, 12H, CH3C@O,
S–CH2–CH–, N–CH2–CH2, –CH2-ph), 3.9 (s, 3H, CH3O–), 4.22 (t,
J = 7.49 Hz, 1H, CH–N) and 6.66–7.2 (dd, J = 8.34, 8.61 Hz, 4H, Ar-
H); MS (EI+): m/z: 349.12 [MÅ]+; Microanalysis calcd C, H, N for
C18H23NO4S: 61.87, 6.63, 4.01, found 61.69, 6.54, 3.92.
3.2. 1-(3-Mercapto-2-methylpropanoyl)-2-(4-methoxybenzyl)
pyrrolidin-3-one (V)
Compound VI (0.85 g) was dissolved in methanolic ammonium
hydroxide (7.8 mL). The solution was stirred for 2 h at room tem-
perature.24 The reaction mixture then concentrated under vacuum
Figure 9. Atomic charge distribution on compound V.

M. A. H. Ismail et al. / Bioorg. Med. Chem. 17 (2009) 3739–3746
3745
and the residue then washed with water and extracted with ethyl
acetate, dried over anhydrous magnesium sulfate, and evaporated
under vacuum to give the oily residue which was purified by col-
umn chromatography using n-hexane/chloroform (3:1) to produce
(br, 1H, SH); and 6.6–8.3 (m, 9H, Ar-H); MS (EI+): m/z: 396.1
[MÅ]+. Microanalysis calcd C, H, N for C23H28N2O2S: 69.66, 7.12,
7.06, found 69.37, 7.28, 6.89.
0.6 g (92%) of the titled compound V as pale yellow oil. IR
m
(neat,
3.8. Ethyl-2-(1-(3-mercapto-2-methylpropanoyl)-2-(4-
methoxybenzyl)pyrrolidin-3-ylidene amino)propanoate (VIe)
cmꢀ1): 3020, 22933–2836, 1732 and 165. 1H NMR (DMSO-d6) d
ppm 1.7 (d, J = 7.45 Hz, 3H, CH3–CH–CO), 2.3–3.4 (m, 9H, S–CH2–
CH–, N–CH2–CH2, –CH2-ph), 3.7 (s, 3H, CH3O–), 4.2 (t, J = 7.51, 1H,
ph-CH2–CH–N), 6.7–7.2 (dd, J = 8.74, 8.8 Hz, 4H, Ar-H); MS (EI+):
m/z: 307.1 [MÅ]+; Microanalysis calcd C, H, N for C18H23NO4S:
61.87, 6.63, 4.01, found 61.67, 6.45, 4.18.
Yield 75%, mp 135–136 °C; IR
m
(neat, cmꢀ1): 1731 (C@O, ester),
1665 (C@O, amid); 1H NMR (DMSO-d6) d ppm 1.2 (d, J = 8 Hz, 3H,
CH3–CH–CO); 1.35 (t, J = 7.3 Hz, 3H, CH3–CH2–O); 1.9–3.5 (m, 9H,
N–CH2–CH2–C@N, S CH2CH–CO, ph-CH2–); 3.75 (s, 3H, CH3O–);
4.2 (q, J = 6.3 Hz, 2H, CH3CH2–O); 4.4 (t, J = 7 Hz, 1H, CH–N) and
6.9–7.2 (dd, J = 8.41, 8.62 Hz, 4H, Ar-H); MS (EI+): m/z: 406.2
[MÅ]+. Microanalysis calcd C, H, N for C21H30N2O4S: 62.04, 7.44,
6.89, found 61.85, 7.23, 7.11.
3.3. 2-(4-Methoxybenzyl)-1-(3-mercapto-2-methylpropanoyl)-
3-substituted imino pyrrolidine (VIa–e) (cf.22
)
A mixture of V (1.5, 0.012 mmol), primary amine (viz; hydrox-
ylamine HCl, 2-ethanolamine, thiosemicarbazide, benzylamine,
and 2-aminopropionic acid ethyl ester) (0.02 mmol), and anhy-
drous sodium acetate (1.4 g, 0.03 mmol) in absolute ethanol
(25 mL) was refluxed under stirring for 6 h. After cooling, the sol-
vent was evaporated under vacuum and the residual solid was
treated with ice cold water and filtered. The precipitate was crys-
tallized from ethanol to produce the titled compounds (VIa–e).
3.9. Molecular modeling experiments
All molecular modeling studies were performed using a Silicon
Graphics desktop (SGI) Fuel workstation under an IRIX 6.8 operating
system, at the Faculty of Pharmacy, Ain Shams University, Cairo,
Egypt. The bioactive conformer of Lisinopril was isolated from the
remaining protein structures using Cerius2 module (Fig. 2). The gen-
eration of the pharmacophore model for ACE inhibitors hypothesis
was accomplished using Accelrys CATALYST, version 4.8. Molecules
were built within CATALYST modules and conformational models
for each compound were generated automatically using the poling
algorithm. This emphasizes representative coverage over a 20 kcal/
mol energy range above the estimated global energy minimum
and the best quality generation technique was chosen. The bioactive
conformer of Lisinopril was used for stepwise hypothesis generation
using the stepwise technique. In this study, hydrogen bond Accep-
tors, hydrophobic features, and negative ionizable points were used
as the chemical features, which were reported to be crucial for the
ACE inhibitors activity.16,18
3.4. 1-(3-(Hydroxyimino)-2-(4-methoxybenzyl)pyrrolidin-1-
yl)-3-mercapto-2-methyl propan-1-one (VIa)
Yield 75%, mp 185–186 °C; IR
m
(neat, cmꢀ1): 3520 (br, OH),
1664 (C@O, amid); 1H NMR (DMSO-d6) d ppm 1.3 (d, J = 7.4 Hz,
3H, CH3–CH–CO); 2.1–3.4 (m, 9H, N–CH2–CH2–C@N, S CH2CH–
CO, ph-CH2–); 3.75 (s, 3H, CH3O–); 4.2 (t, J = 7.49 Hz, 1H, CH–N);
4.9 (br, 1H, SH); 5.7 (br, 1H, OH) and 6.6–7.2 (dd, J = 8.61,
8.66 Hz, 4H, Ar-H); MS (EI+): m/z: 322.2 [MÅ]+. Microanalysis calcd
C, H, N for C16H22N2O3S: 59.60, 6.88, 8.69, found 59.55, 6.77, 8.51.
3.5. 1-(3-(2-Hydroxyethylimino)-2-(4-methoxybenzyl)pyrrolidin-
1-yl)-3-mercapto-2-methylpropan-1-one (VIb)
References and notes
Yield 77%, mp 174–175 °C; IR
m
(neat, cmꢀ1): 3560–3200 (br,
1. Jackson, E. K.; Garrison, J. C. Renin and Angiotensin. In Goodman and Gilman’s:
The Pharmacological Basis of Therapeutics; Hardman, J. G., Limbird, L. E.,
Molinoff, P. B., Ruddon, R. W., Gilman, A. G., Eds., 9th ed.; McGraw-Hill: New
York, 1996; pp 733–758.
2. Brunner, H. R.; Nussberger, J.; Wäber, B. Angiotensin Antagonists. In The Renin–
Angiotensin System; Robertson, J. I. S., Nicholls, M. G., Eds.; Gower Medical
Publishing: London, 1993; pp 86.1–86.14.
3. Smith, R. D.; Timmermans, P. B. M. W. M. Pharmacol. Res. 1995, 31, 289.
4. Goodfriend, T. L.; Elliott, M. E.; Catt, K. N. Eng. J. Med. 1996, 334, 1649.
5. Veelken, R.; Hilgers, K. F.; Scrogin, K. E.; Mann, J. F. E.; Schmieder, R. E. Br. J.
Pharmacol. 1998, 125, 1761.
OH), 1660 (C@O, amid); 1H NMR (DMSO-d6) d ppm 1.25 (d,
J = 7.1 Hz, 3H, CH3–CH–CO); 1.8–3.4 (m, 13H, N–CH2–CH2–C@N, S
CH2CH–CO, ph-CH2–, N@CH2–CH2–); 3.75 (s, 3H, CH3O–); 4.4 (t,
J = 7.29 Hz, 1H, CH–N); 4.7–5.4 (br, 2H, OH and SH) and 6.8–7.2
(dd, J = 8.71, 8.41 Hz, 4H, Ar-H); MS (EI+): m/z: 350.1 [MÅ]+. Micro-
analysis calcd C, H, N for C16H22N2O3S: 61.69, 7.48, 7.9, found
61.37, 7.28, 8.01.
6. Bottari, S. P.; de Gasparo, M.; Steckelings, U. M.; Levens, N. R. Front.
Neuroendocrin. 1993, 14, 123.
7. Raasch, W.; Bartels, T.; Gieselberg, A.; Dendorfer, A.; Dominiak, P. J. Pharmacol.
Exp. Ther. 2002, 300, 428.
3.6. 2-(1-(3-Mercapto-2-methylpropanoyl)-2-(4-methoxybenzyl)
pyrrolidin-3-ylidene) hydrazinecarbothioamide (VIc)
m
(neat, cmꢀ1): 3320–3150 (br,
8. Palkowitz, A. D.; Steinberg, M. I.; Thrasher, K. J.; Reel, J. K.; Hauser, K. L.;
Zimmerman, K. M.; Wiest, S. A.; Whitesitt, C. A.; Simon, R. L.; Heifer, W.; Lifer, S.
L.; Boyd, D. B.; Barnett, C. J.; Wilson, T. M.; Deeter, J. B.; Takeuchi, K.; Riley, R. E.;
Miller, W. D.; Marshall, W. S. J. Med. Chem. 1994, 32, 4508.
Yield 80%, mp 155–156 °C; IR
NH), 1683 (C@O, amid); 1H NMR (DMSO-d6) d ppm 1.3 (d,
J = 81 Hz, 3H, CH3–CH–CO); 2.3–3.5 (m, 9H, N–CH2–CH2–C@N, S
CH2CH–CO, ph-CH2–); 3.9 (s, 3H, CH3O–); 4.2 (t, J = 7.9 Hz, 1H,
CH–N); 5 (br, 1H, SH) and 6.7–7.3 (dd, J = 8.9, 8,71 Hz, 4H, Ar-H),
8.1–8.5 (br, 3H, NHCSNH2); MS (EI+): m/z: 380.1 [MÅ]+. Microanaly-
sis calcd C, H, N for C17H24N4O2S2: 53.66, 6.36, 14.72, found 53.69,
4.92, 15.00.
9. Guidelines Committee J. Hypertens. 2003, 21, 1011.
10. Kirch, W.; Horn, B.; Schweizer, J. Eur. J. Clin. Invest. 2001, 31, 698.
11. Dahlof, B.; Devereux, R. B.; Kjeldsen, S. E.; Julius, S.; Beevers, G.; Faire, U.;
Fyhrquist, F.; Ibsen, H.; Kristainsson, K.; Lederbalte-Pedersen, O.; Lindholm, L.
H.; Nieminen, M. S.; Omvik, P.; Oparil, S.; Wedel, H. Lancet 2002, 359, 995.
12. Gordon, E. M.; Godfrey, J. D.; Weller, H. N.; Natarajan, S.; Pluscec, J.; Rom, M. B.;
Niemela, K.; Sabo, E. F.; Cushman, D. W. Bioorganic Chem 1986, 14, 148.
13. Wei, W.; ShengRong, S.; FengQin, F.; GuoQing, H.; ZhanLi, W. Sci. China, Ser. B—
Chem. 2008, 51, 786.
3.7. 1-(3-(Benzylimino)-2-(4-methoxybenzyl)pyrrolidin-1-yl)-
3-mercapto-2-methylpropan-1-one (VId)
14. Zaluski, M. C.; Coric, P.; Thery, V.; Gonzallez, W.; Turcaud, S.; Michel, J.; Roques,
B. J. Med. Chem. 1996, 39, 2594.
15. Subissi, A.; Evangelista, S.; Giachetti, A. Cardiovasc. Drug Rev. 1999, 17, 115.
16. Tzakos, A. G.; Gerothanassis, I. P. ChemBioChem 2005, 6, 1089.
17. Crystal structure of Human ACE in complex with lisinoprol (1086), Natesh, R.;
Shwager, S. L. U.; Sturrock, E. D.; Achorya, K. R. Nature 2003, 421, 551.
18. Dammkoehler, R. A.; Karasek, S. K.; Shands, E. F. B.; Marshall, G. R. J. Comput.
Aided Mol. Des. 1989, 3, 3.
Yield 65%, bright yellow oil; IR
m
(neat, cmꢀ1): 1675 (C@O,
amid); 1H NMR (DMSO-d6) d ppm 1.2 (d, J = 8 Hz, 3H, CH3–CH–
CO); 2.3–3.4 (m, 11H, N–CH2–CH2–C@N, S CH2CH–CO, ph-CH2–,
N–CH2ph); 3.8 (s, 3H, CH3O–); 4.3 (t, J = 7.19 Hz, 1H, CH–N); 5

3746
M. A. H. Ismail et al. / Bioorg. Med. Chem. 17 (2009) 3739–3746
19. Kustera, D. J.; Marshalla, G. R. J. Comput. Aided Mol. Des. 2005, 19, 609.
20. Ismail, N. S. M.; Salah El Dine, R.; Hattori, M.; Takahashi, K.; Ihara, M. Bioorg.
Med. Chem. 2008, 16, 7877.
21. Ismail, M. A. H.; Todd, M. H.; Barker, S.; Abou El Ella, D. A.; Abouzid, K. A. M.;
Toubar, R. A. J. Med. Chem. 2006, 49, 1526.
22. Nabil Aboul-Enein, M.; Khalifa, M.; El-Difrawy, S. M. Pharm. Acta Helv. 1973, 48,
405.
23. Mc Kinney, L. L.; Uhing, W. H.; Stezdorm, E. A.; Cown, J. C. J. Am. Chem. Soc.
1950, 72, 2599.
24. Johnson, A. L. J. Org. Chem. 1976, 41, 1320.