Asymmetric synthesis and molecular docking study of enantiomerically pure pyrrolidine derivatives with potential antithrombin activity

Article abstract of DOI:10.1016/j.tetasy.2013.05.023

The (2R,4R,5S)- and (2S,4S,5R)-enantiomers of 4-(tert-butyl) 2-methyl 5-(4-bromophenyl)-pyrrolidine-2,4-dicarboxylate 3 were synthesized efficiently with an ee of >90% on a gram scale using a FAM-catalytic methodology. Subsequent modification afforded ena

Full text of DOI:10.1016/j.tetasy.2013.05.023

Tetrahedron: Asymmetry 24 (2013) 838–843
Contents lists available at SciVerse ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Asymmetric synthesis and molecular docking study of
enantiomerically pure pyrrolidine derivatives with potential
antithrombin activity
b
b
b
Seylan Ayan a, Özdemir Dogan a, , Polina M. Ivantcova , Nikita G. Datsuk , Dmitry A. Shulga ,
⇑
Vladimir I. Chupakhin b, Dmitry V. Zabolotnev b, Konstantin V. Kudryavtsev b,c,
⇑
a Department of Chemistry, Middle East Technical University, 06800 Ankara, Turkey
b Department of Chemistry, M. V. Lomonosov Moscow State University, Leninskie Gory 1/3, Moscow 119991, Russia
c Institute of Physiologically Active Compounds, Russian Academy of Sciences, Chernogolovka 142432, Moscow region, Russia
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 22 April 2013
Accepted 29 May 2013
The (2R,4R,5S)- and (2S,4S,5R)-enantiomers of 4-(tert-butyl) 2-methyl 5-(4-bromophenyl)-pyrrolidine-
2,4-dicarboxylate 3 were synthesized efficiently with an ee of >90% on a gram scale using a FAM-catalytic
methodology. Subsequent modification afforded enantiopure N-((4-chlorophenyl)thio)acetyl pyrrolidine
derivatives 4, which are potential thrombin inhibitors according to comprehensive molecular docking
studies.
Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction
scale, we evaluated zinc-FAM1q and silver-POFAM1i chiral catalysts
for a 1,3-dipolar azomethine ylide cycloaddition to form enantio-
The pyrrolidine ring system is a structural element of many nat-
ural products and drugs. The enantioselective construction of poly-
substituted pyrrolidines via 1,3-dipolar cycloaddition of electron
deficient dipolarophiles with azomethine ylides has been studied
by many researchers using metal-catalysts1 and organocatalysts.2
Nevertheless, there are limitations in the application of the cata-
lytic systems which can be dipole and/or dipolarophile dependent
and required a case study for each starting material. N-acylated
derivatives of racemic 5-arylpyrrolidine-2,4-dicarboxylic acids
were synthesized by using 1,3-dipolar cycloaddition as a scaffold
forming step and, following tests, demonstrated inhibitory activity
of the some compounds towards thrombin.3 Both enantiomerically
pure forms of the discovered inhibitors3 are needed in order to elu-
cidate the relationship of absolute ligand configuration with its
biological activity. Recently we separated racemic 4-(tert-butyl)
merically pure pyrrolidine derivatives. Using zinc-FAM chiral cata-
lysts, we successfully synthesized both enantiomers of 3 with high
enantiomeric purity and then converted them into N-((4-chloro-
phenyl)thio)acetyl pyrrolidine derivatives 4. Subsequent molecular
modelling of pyrrolidine-based ligands 4 gives a rationale to the
development of thrombin inhibitors based on enantiomerically
pure well-defined 5-arylpyrrolidine-2,4-dicarboxylic acid scaffold.
2. Results and discussion
Three chiral ligands have been tested for the asymmetric induc-
tion in the 1,3-dipolar cycloaddition of tert-butyl acrylate 1 with
aldimine 2 (Fig. 1). Although POFAM1i and FAM1q ligands have
been applied to the synthesis of large sets of functionalized pyrrol-
idines, they were not used for the synthesis of orthogonally pro-
tected diesters of 5-bromophenylpyrrolidine-2,4-dicarboxylic
acid 3 in previous research.
2-methyl 5-(4-bromophenyl)pyrrolidine-2,4-dicarboxylate 3,4
a
promising thrombin inhibitor scaffold, into its individual enantio-
mers by chiral HPLC and studied the acylative kinetic resolution
of the racemate.5 Although we achieved 88.2% ee for unreacted
(2R,4R,5S)-3, the preparative isolation afforded the enantiomeri-
cally enriched product in only a low yield.5 To develop an appropri-
ate method for the synthesis of both enantiomers of 3 on a gram
Different conditions were screened for an efficient synthesis of
enantiomerically pure pyrrolidines
3 (Table 1). Experiments
showed that a silver-POFAM catalytic system produced (ꢀ)-3 in
low yields and ee (Table 1, entry 1). Therefore no further studies
were carried out with this catalyst. Preliminary studies with a
zinc-FAM catalytic system gave much better results (Table 1, entry
2). Further optimization in terms of time, temperature and
substrate amounts was performed (Table 1, entries 3–6). From
these studies, ꢀ20 °C and 6 h found to be the optimum conditions
(Table 1, entry 3). Increasing the ligand amount and extending the
⇑
Corresponding authors. Tel.: +90 3122105134; fax: +90 3122103200 (O.D.); tel.:
+7 495 939 3564; fax: +7 495 932 8846 (K.V.K.).
E-mail addresses: dogano@metu.edu.tr (Ö. Dogan), kudr@org.chem.msu.ru
(K.V. Kudryavtsev).
0957-4166/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tetasy.2013.05.023

S. Ayan et al. / Tetrahedron: Asymmetry 24 (2013) 838–843
839
identical to the racemate.3 The enantiomeric purity of enantiomers
(ꢀ)-4 and (+)-4 was determined by chiral HPLC.
OH
OH
OH
(S)
(R)
(S)
Ph
(R)
Ph
P(O)Ph2
(R)
(S)
N
(S)
Fc
N
N
Fc
Fc
Theoretical studies of the interaction of pyrrolidines (ꢀ)-4 and
(+)-4 with the active sites of thrombin and trypsin were also per-
formed in order to establish potential guiding features of en-
zyme–ligand interactions. For docking studies, we selected X-ray
structures of enzymes from Protein Data Bank: 2ZC9 for thrombin
and 1K1L for trypsin. Three different docking programs were em-
ployed to balance diverse calculation approaches and to attempt
an implicit recording of potential halogen bonding6 between the
molecules studied and serine proteases. Ligand–protein interac-
tions were calculated using AutoDock (version 4.2.3)7 and Auto-
Dock Vina (version 1.1.2),8 where each ligand binding mode
within the enzymes active site was characterized by free energy
of binding, and Fred (version 2.2.5) with Chemgauss3 scoring func-
tion.9 The structures of the proteins for modelling were prepared
according to a standard scenario: water molecules were removed
from the protein structure and hydrogen atoms were added to all
amino acid residues. The 3D molecular structures of enantiomers
(ꢀ)-4 and (+)-4 were obtained with OpenBabel10 and optimized
using the MMFF94 force field within the same software. Calcula-
tions were carried out for the conformers with a different dihedral
angle at the amide bond (0 and 180° (index inv)) of each enantio-
mer in the pair. Modelling results differentiate between enantio-
mers (ꢀ)-4 and (+)-4 at the active sites of both proteases
(Table 2), which rationalizes the development of efficient asym-
metric synthesis of a molecular scaffold for potential inhibitors.
(R)
(R)Me
Me
Et
FAM
POFAM
ent-FAM
Figure 1. Structures of ligands used for asymmetric 1,3-dipolar cycloaddition
experiments (Fc – ferrocenyl).
reaction time did not affect the yield or enantioselectivity (Table 1,
entries 4 and 5). Scaling of the reaction up to 500 mg of imine 2
was appropriate for the successful synthesis of (ꢀ)-3 (Table 1, en-
try 7). The physicochemical characteristics of enantiomerically
pure 5-bromophenylpyrrolidine-2,4-dicarboxylic acid diester (ꢀ)-
3 were in accordance with the literature.5 When ligand ent-FAM1q
was used as the catalyst pyrrolidine (+)-3 was obtained as ex-
pected in good yields and ee (Table 1, entries 8 and 9). Spectro-
scopic and physicochemical data for (+)-3 also corresponded to
the literature.5
In the next step, both enantiomers of cycloadduct 3 were con-
verted into pyrrolidine derivatives (ꢀ)-4 and (+)-4 in two steps.
Acylation of (ꢀ)-3 and (+)-3 with chloroacetyl chloride followed
by reaction with 4-chlorophenylthiol led to (ꢀ)-4 and (+)-4 in
65% and 72% yields, respectively, (Scheme 1). The synthetic meth-
odology of these steps is analogous to the synthesis of racemic pyr-
rolidine 4;3 the NMR characteristics of enantiomers of 4 were
Table 1
Catalytic asymmetric synthesis of pyrrolidine enantiomers 3a
COOBut
COOBut
ButOOC
1
FAM · Zn(II)
ent-
POFAM · Ag(I)
+
MeOOC
N
H
COOMe
FAM · Zn(II)
or
N
H
Br
Br
MeOOC
N
3
(-)-
2
Br
3
(+)-
Entry
Ligand
Ligand (mol %)
Metal
Metal (mol %)
Temperature (°C)
Time (h)
Yield (%)
ee (%)
1
2
3
4
POFAM
FAM
FAM
FAM
FAM
FAM
FAM
6
AgOAc
1.5
10
10
10
10
10
10
10
10
ꢀ20
ꢀ20
ꢀ20
ꢀ20
ꢀ20
ꢀ40
ꢀ20
ꢀ20
ꢀ20
48
14
6
10
10
48
6
16
76
88
87
87
0
87
88
87
10
83
88
87
87
—
90
87
93
11.5
11.5
11.5
20
11.5
11.5
11.5
11.5
Zn(OTf)2
Zn(OTf)2
Zn(OTf)2
Zn(OTf)2
Zn(OTf)2
Zn(OTf)2
Zn(OTf)2
Zn(OTf)2
5
6
7b
8
ent-FAM
ent-FAM
6
6
9b
a
For entries 1–7 the product is (ꢀ)-3; for entries 8 and 9 the product is (+)-3.
The reaction was performed with 500 mg of imine 2.
b
ButOOC
COOBut
1) ClCH2COCl, Et3N, THF
2) 4-ClPhSH, DMF, K2CO3
1) ClCH2COCl, Et3N, THF
COOMe
N
MeOOC
N
(+)-3
(-)-3
Br
O
Br
O
2) 4-ClPhSH, DMF, K2CO3
S
S
Cl
Cl
(-)-4, 65%, ee 98%
4
(+)- , 72%, ee 99%
Scheme 1. Synthesis of enantiomerically pure potential thrombin inhibitors (ꢀ)-4 and (+)-4.

840
S. Ayan et al. / Tetrahedron: Asymmetry 24 (2013) 838–843
Table 2
Energies and binding constants of ligands (+)-4 and (ꢀ)-4 with thrombin (PDB ID: 2ZC9) and trypsin (PDB ID: 1K1L), calculated by the scoring functions of AutoDock 4.2 (AD),
AutoDock Vina (ADV) and Fred Chemgauss3 (ChG3)
ButOOC
ButOOC
COOBut
COOBut
COOMe
COOMe
MeOOC
MeOOC
N
N
N
N
Br
Br
Cl
O
Br
O
Br
O
O
Structure/
conformation
S
S
S
S
Cl
Cl
Cl
(+)-4
(+)-4_inv
(ꢀ)-4
(ꢀ)-4_inv
E, kcal/mola (Ki,
l
M)
E, kcal/mola (Ki,
lM)
E, kcal/mola (Ki,
lM)
E, kcal/mola (Ki,
lM)
ADthrombin
ADVthrombin
ChG3thrombin
ADtrypsin
ꢀ6.31 (23.9)
ꢀ7.50 (3.7)
ꢀ65.56
ꢀ6.47 (18.0)
ꢀ8.20 (1.0)
ꢀ62.09
ꢀ6.34 (22.5)
ꢀ8.10 (1.4)
ꢀ70.82
ꢀ6.09 (34.6)
ꢀ7.10 (7.3)
ꢀ75.31
ꢀ6.17 (30.1)
ꢀ6.80 (12.0)
ꢀ53.94
ꢀ6.29 (24.4)
ꢀ6.40 (23.3)
ꢀ38.19
ꢀ6.43 (19.3)
ꢀ7.00 (8.6)
ꢀ40.85
ꢀ6.86 (9.4)
ꢀ6.80 (12.0)
ꢀ39.63
ADVtrypsin
ChG3trypsin
b
ADselectivity
1.02 (1.27)
1.10 (3.24)
1.22
1.03 (1.36)
1.3 (23.3)
1.63
0.99 (0.86)
1.16 (6.14)
1.73
0.89 (0.27)
1.04 (1.64)
1.90
b
ADVselectivity
b
ChG3selectivity
a
Ki Estimations are available for scoring functions of AutoDock and AutoDock Vina, but not for Chemgauss3. For the last scoring function, energy binding parameters are
dimensionless.
b
The selectivity is reported as a thrombin/trypsin binding energy (scoring function) ratio and Ki (trypsin)/Ki(thrombin) ratio in parentheses, where applicable.
The thrombin active site consists of three main pockets. The S1
pocket includes the key residues Asp189 involved in the formation
of the salt bridge with charged thrombin inhibitors and Tyr228
which has become a subject of intense research since the discovery
of attractive halogen–p
interactions11 of non-basic ligands contain-
ing aryl-halogenated structural fragments. The volume of S1 pock-
et of thrombin and trypsin differs hence it also determines ligand
selectivity. The hydrophobic S2 pocket of thrombin, formed by
Tyr60A and Trp60D, is absent in trypsin, thus being the main target
for thrombin/trypsin selectivity optimization. The hydrophobic S3/
S4 pocket is comprised of Trp215, Ile174, Leu99 and Glu217. In
addition to the pockets, the main chain of Gly216 and Gly219 res-
idues also influences the affinity and selectivity of thrombin li-
gands via hydrogen bond formation with the ligand.
According to AutoDock predictions, the most favourable inter-
actions with the thrombin active site are formed in the case of
the (+)-4_inv conformer as the ligand. The best (+)-4_inv binding
mode is characterized with ꢀ6.47 kcal/mol binding energy (Table 2,
0.18 kcal/mol profitable to trypsin complex) and includes nearby
ligand–enzyme interactions (Fig. 2): (a) immersion of a p-Cl–phe-
nyl moiety in the S1 thrombin pocket, with the chlorine atom
pointing to Tyr228 at an approximate distance of 4 Å (presumably
forming favourable halogen–p interaction); (b) complementary
filling of the S3 thrombin pocket with a p-Br–phenyl substituent;
(c) the tert-butoxycarboxylic group near the S2 pocket with possi-
ble OH(Tyr60A)ꢁ ꢁ ꢁO@C hydrogen bond formation (2.7 Å) and (d)
two hydrogen bonds with a backbone of Gly216 (2.7 Å) and
Gly219 (2.0 Å). In our study AutoDock generally tends to place
the p-Cl–phenyl moiety in S1 pocket of thrombin, which is an in-
tended binding pattern used in the design of the whole inhibitor
class.3
The most stable thrombin/pyrrolidine 4 complex predicted with
AutoDock Vina is also for the (+)-4_inv conformer. The calculated
energy of the complex is ꢀ8.20 kcal/mol that differentiates it sig-
nificantly from all of the possible trypsin/pyrrolidine 4 complexes
(Table 2). The complex also contains favourable ligand–enzyme
interactions (Fig. 3): (a) complementary filling of the S3 thrombin
pocket with a tert-butoxycarboxylic group with possible hydrogen
bond (1.7 Å) of the ligand carboxylic group and a Tyr60A hydroxyl
Figure 2. AD predicted most favourable binding mode of (+)-4_inv conformer
(green) in the thrombin active site.
group; (b) insertion of the methoxycarboxylic fragment in the S2
thrombin pocket and hydrogen bond (2.0 Å) between the carbox-
ylic group and Gly219; (c) hydrogen bond (2.7 Å) between the pyr-
rolidine nitrogen atom of the inhibitor and the hydrogen of the
main chain of Gly216. The p-Cl–phenyl ligand fragment is im-
mersed into the S1 thrombin pocket but considerably less deeply
compared to AutoDock predictions and the position of the m-Cl–
phenyl group in the original 2ZC9 ligand. According to AutoDock
Vina predictions, the binding mode with p-Cl–phenyl moiety occu-

S. Ayan et al. / Tetrahedron: Asymmetry 24 (2013) 838–843
841
between His57 and the oxygen atom of the ligand carboxymethyl
group (2.8 Å) and the hydrogen bond between Gly216 and oxygen
atom of the ligand carboxy-tert-butyl group (2.7 Å); (b) the p-Cl–
phenyl ligand moiety is tightly bound near the S2 thrombin pocket
and forms a favourable cation–p interaction with Lys60F (3.0 Å)
and hydrophobic interactions with Phe60H (not shown); (c) the
gorge of the S1 thrombin pocket is filled with the p-Br–phenyl frag-
ment with the bromine atom pointing to hydrophobic subpocket,
formed by Val213 and Ala190. The halogen–p interaction is less
probable in the last case because of the larger distance (6.7 Å) be-
tween Tyr228 and the halogen atom.
Overall, both the p-Cl–phenyl and p-Br–phenyl substituents of
N-((4-chlorophenyl)thio)acetyl-5-(4-bromophenyl)pyrrolidine-
2,4-dicarboxylate (ꢀ)-4 and its enantiomer (+)-4 are frequently
found in S3 and S1 thrombin pockets according to all of the docking
methods studied. The constrained conformations of the pyrrolidine
ring system of 4 arrange well-defined interactions with thrombin
active site pockets, where the enantiomers differ in binding modes.
The p-Cl–phenyl substituent of enantiomers (ꢀ)-4 and (+)-4 in
different conformations does not occupy the S1 pocket of trypsin
according to the calculations performed. The evident reason for
the thrombin versus trypsin preferences of pyrrolidines 4 is due
to the smaller volume and higher hydrophobicity of the S1 trypsin
pocket compared to thrombin. Instead, the p-Cl–phenyl moiety of
pyrrolidines (ꢀ)-4 and (+)-4 is frequently posed in the S3 trypsin
pocket forming stacking interactions with Trp215 at the pocket
bottom.
Figure 3. ADV predicted most favourable binding mode of (+)-4_inv (green)
conformer in the thrombin active site.
Thrombin/trypsin selectivity is an important issue for the devel-
opment of oral anticoagulants based on direct thrombin inhibi-
tors.12 In this regard, AutoDock predictions for N-((4-
chlorophenyl)thio)acetyl pyrrolidine derivatives 4 are very promis-
ing, since the most favourable binding modes are those that can
pying S1 pocket is only typical for the (+)-4_inv conformer and
does not occur in any of the top 10 positions for the other three
structural conformers studied.
Fred with Chemgauss3 scoring function predicts the best
thrombin binder to be (ꢀ)-4_inv conformer which is 1.5–2 times
more efficient than interactions of all 4 conformers with trypsin
(Table 2). The (ꢀ)-4_inv binding mode is characterized by several
ligand–enzyme interactions (Fig. 4): (a) the hydrogen bond
have an additional 2 kcal/mol stabilization due to the halogen–
p
interaction.11a The latter should lead to high nanomolar activity
of the desired thrombin inhibitors.11 From the other side, Auto-
Dock Vina and Fred/Chemgauss3 docking results also give reason-
able alternative binding modes of (ꢀ)-4 and (+)-4 in the active site
of both proteases with interesting ligand–enzyme interactions.
3. Conclusion
Finding enantiomerically pure small molecules with well-
defined structures is desired for the development of effective
drug-like ligands for natural biomacromolecules. Herein we have
performed an efficient asymmetric synthesis of both enantiomeric
forms of functionalized trisubstituted pyrrolidine scaffold 3 and
transformed them by a structure-based driven approach to poten-
tial chiral ligands of serine proteases 4. The elucidation of ligand–
enzyme interactions between well-defined pyrrolidines 4 and the
thrombin/trypsin pair by bioinformatic methods points to the util-
ity of the studied compound class for the development of selective
thrombin inhibitors. The appropriate biological testing results will
be reported elsewhere.
4. Experimental
4.1. General
Aziridinyl alcohol ligands POFAM1i and FAM1q were synthe-
sized according to known procedures. Flash column chromatogra-
phy was performed using Silica gel 60 (particle size
0.040ꢀ0.063 mm) (Alfa Aesar, UK). Melting points were deter-
mined in an open capillary and are uncorrected. Optical rotations
were measured on a Jasco DIP-360 polarimeter. The 1H NMR spec-
tra were recorded on Bruker Avance (400 MHz) spectrometer with
Figure 4. OpenEye Fred (Chemgauss3) predicted most favourable binding mode of
(ꢀ)-4_inv conformer (green) in the thrombin active site.

842
S. Ayan et al. / Tetrahedron: Asymmetry 24 (2013) 838–843
residual solvent signals as a reference. Elemental analysis was per-
formed using Carlo Erba analyser. Analytical HPLC of compounds 3
and 4 was performed using a Chiralpak OD-H column, detection at
254 nm, 0.7–0.9 mL/min flow rate, n-hexane–i-PrOH 7:3 mixture
as eluting system.
4.3.1. (ꢀ)-4-(tert-Butyl) 2-methyl (2R,4R,5S)-5-(4-bromophenyl)-
1-(2-((4-chlorophenyl)thio)acetyl)pyrrolidine-2,4-dicarboxylate
(ꢀ)-4
Colourless powder (740 mg, 65%): mp 65 °C. 98.4% ee (HPLC,
Chiralpak OD-H, tR 21.5 min). ½a D20
ꢃ
¼ ꢀ73:3 (c 0.15, C6H6). 1H
NMR (400 MHz, CDCl3): d 1.14 (s, 9H, COOtBu); 2.14-2.24 (m, 1H,
H-3A); 2.36 (ddd, 1H, J 12.8, 6.9 and 6.6 Hz, H-3B); 3.24 (d, 1H, J
16.0 Hz, CH2S); 3.61 (ddd, 1H, J 12.8, 8.9 and 6.6 Hz, H-4); 3.71
(s, 3H, COOMe); 4.00 (d, 1H, J 16.0 Hz, CH2S); 4.33 (dd, 1H, J 11.3
and 6.9 Hz, H-2); 5.44 (d, 1H, J 8.9 Hz, H-5); 7.08–7.12 (m, 2H,
4.2. Catalytic asymmetric cycloaddition of tert-butyl acrylate 1
with iminoester 2 (optimized procedure, Table 1, entries 7 and
9)
H
Ar); 7.27–7.31 (m, 2H, HAr); 7.56–7.61 (m, 4H, HAr). Anal. Calcd
At first, Zn(OTf)2 (71 mg, 0.195 mmol) was weighed in a glove-
bag under nitrogen and transferred to a reaction flask, which was
flame dried and flashed with nitrogen three times. To this flask
was added FAM (or ent-FAM) ligand (77.5 mg, 0.215 mmol, ben-
zene azeotroped prior to use) and dry DCM (5.0 mL). The flask
was stirred for 1 h at room temperature, then it was cooled to
for C25H27BrClNO5S (%): C, 52.78; H, 4.78; N, 2.46. Found (%): C,
52.87; H, 4.80; N, 2.40.
4.3.2. (+)-4-(tert-Butyl) 2-methyl (2S,4S,5R)-5-(4-bromophenyl)-
1-(2-((4-chlorophenyl)thio)acetyl)pyrrolidine-2,4-dicarboxylate
(+)-4
ꢀ20 °C and imine
2
(500 mg, 1.95 mmol), Et3N (27.5
lL,
Colourless powder (819 mg, 72%): mp 65 °C. >99.0% ee (HPLC,
0.195 mmol) and tert-butyl acrylate 1 (313
lL, 2.13 mmol) were
Chiralpak OD-H, tR 10.3 min). ½a D20
ꢃ
¼ þ100:0 (c 0.15, C6H6). 1H
added, respectively. After stirring the mixture for 6 h, TLC showed
a very concentrated spot of the product. The reaction was stopped
and the reaction mixture was concentrated by a rotary evaporator.
The crude product was purified by flash column chromatography
on silca gel by eluting with hexane–ethyl acetate (2:1).
NMR spectrum was identical to the spectrum of (ꢀ)-(2R,4R,5S)-4.
Anal. Calcd for C25H27BrClNO5S (%): C, 52.78; H, 4.78; N, 2.46.
Found (%): C, 52.88; H, 4.80; N, 2.50.
4.4. Molecular modelling
4.2.1. (ꢀ)-4-(tert-Butyl) 2-methyl (2R,4R,5S)-5-(4-bromophenyl)-
pyrrolidine-2,4-dicarboxylate (ꢀ)-35
Enzyme structures for thrombin (2ZC9) and trypsin (1K1L) were
taken from Protein Data Bank. The modelling software used were
AutoDock (version 4.2.3),7 AutoDock Vina (version 1.1.2)8 and
OpenEye Fred (version 2.2.5) with Chemgauss39 scoring function.
Water molecules were removed from the protein structure and
hydrogen atoms were added to all of the amino acid residues in
the protein models. The 3D ligand structures were obtained with
OpenBabel10 and optimized using the MMFF94 force field within
the same software. Conformations of the compounds studied for
docking with Fred were generated with Omega2 (v. 2.4.6).
Colourless powder (653 mg, 87%): mp 84 °C. 90% ee (HPLC, Chir-
alpak OD-H, tR 8.7 min). Single recrystallization from hexane in-
creased ee to 99%. 1H NMR (400 MHz, CDCl3): d 1.07 (s, 9H,
COOtBu); 2.31 (ddd, 1H, J 13.3, 8.2 and 6.3 Hz, H-3B); 2.43 (dt,
1H, J 13.3 and 8.2 Hz, H-3A); 3.22-3.27 (m, 1H, H-4); 3.81 (s, 3H,
COOMe); 3.94 (t, 1H, J 8.4 Hz, H-2); 4.43 (d, 1H, J 7.8 Hz, H-5);
7.26 (d, 1H, J 8.6 Hz, HAr); 7.45 (d, 1H, J 8.6 Hz, HAr).
4.2.2. (+)-4-(tert-Butyl) 2-methyl (2S,4S,5R)-5-(4-bromophenyl)-
pyrrolidine-2,4-dicarboxylate [(+)-3]5
Acknowledgments
Colourless powder (653 mg, 87%): mp 84 °C. 93% ee (HPLC, Chir-
alpak OD-H, tR 7.0 min). Single recrystallization from hexane in-
creased ee to >99%. 1H NMR spectrum was identical to the
spectrum of (ꢀ)-(2R,4R,5S)-3.
The study was supported by the Russian Foundation for Basic
Research (project Nos. 11-03-00630-a, 11-03-91375-ST_a and 12-
03-92005-NNS_a), the Scientific and Technological Research Coun-
cil of Turkey (TUBITAK, project No 110T075), the State Contract
11.519.11.2032 (the Russian Ministry of Education and Science)
and the RAS Programme OXHM 9 ‘‘Medicinal Chemistry’’. OpenEye
Scientific Software, Inc. (http://www.eyesopen.com) is acknowl-
edged for an academic license to FRED and OMEGA software.
4.3. Synthesis of N-((4-chlorophenyl)thio)acetyl derivatives 4.
General procedure3
To a solution of pyrrolidine-2,4-dicarboxylic acid diester 3
(0.768 g, 2 mmol) in THF (5 mL), Et3N (0.263 g, 2.6 mmol,
0.363 mL) was added under an argon atmosphere. The reaction
References
mixture was cooled and
a solution of chloroacetyl chloride
1. For selected copper catalyzed recent examples see: (a) Llamas, T.; Arrayás, R.
G.; Carretero, J. C. Org. Lett. 2006, 8, 1795–1798; (b) Yan, X.-X.; Peng, Q.; Zhang,
Y.; Zhang, K.; Hong, W.; Hou, X.-L.; Wu, Y. D. Angew. Chem., Int. Ed. 2006, 45,
1979–1983; (c) Fukuzawa, S. I.; Oki, H. Org. Lett. 2008, 10, 1747–1750; (d)
López-Pérez, A.; Adrio, J.; Carretero, J. C. J. Am. Chem. Soc. 2008, 130, 10084–
10085; (e) Wang, C.-J.; Xue, Z.-Y.; Liang, G.; Lu, Z. Chem. Commun. 2009, 2905–
2907; (f) Wang, C.-J.; Liang, G.; Xue, Z.-Y.; Gao, F. J. Am. Chem. Soc. 2008, 130,
17250–17251; (g) Padilla, S.; Tejero, R.; Adrio, J.; Carretero, J. C. Org. Lett. 2010,
12, 5608–5611; For selected silver catalyzed recent examples see: (h) Han, M.-
L.; Wang, D-Yo; Zeng, P.-W.; Zheng, Z.; Hub, X.-P. Tetrahedron Asymmetry 2012,
23, 306–312; (i) Eröksüz, S.; Dogan, O.; Garner, P. P. Tetrahedron: Asymmetry
2010, 21, 2535–2541; (j) Oura, I.; Shimizu, K.; Ogata, K.; Fukuzawa, S. Org. Lett.
2010, 12, 1752–1755; (k) Wang, C.-J.; Xue, Z.-Y.; Liang, G.; Zhou, L. Chem.
Commun. 2009, 2905–2907; (l) Najera, C.; Retamosa, M. G.; Sansano, J. M.;
Cozar, A.; Cossio, F. P. Tetrahedron: Asymmetry 2008, 19, 2913–2923; (m)
Najera, C.; Retamosa, M. G.; Sansano, J. M. Angew. Chem., Int. Ed. 2008, 47, 6055–
6058; (n) Najera, C.; Retamosa, M. G.; Martin-Rodriguez, M.; Sansano, J. M.;
Cozar, A.; Cossio, F. P. Eur. J. Org. Chem. 2009, 5622–5634; (o) Martin-Rodriguez,
M.; Najera, C.; Sansano, J. M.; Cozar, A.; Cossio, F. P. Beilstein J. Org. Chem. 2011,
7, 988–996; (p) Zeng, W.; Chen, G.-Y.; Zhou, Y.-G.; Li, Y.-X. J. Am. Chem. Soc.
2007, 129, 750–751; For zinc catalyzed recent examples see: (q) Dogan, O.;
Koyuncu, H.; Garner, P.; Bulut, A.; Youngs, W. J.; Panzner, M. Org. Lett. 2006, 8,
(0.271 g, 2.4 mmol, 0.191 mL) in THF (2 mL) was added with stir-
ring at a rate that the temperature did not exceed +5 °C. The mix-
ture was stirred for 5 h at room temperature after which the
Et3NꢁHCl residue was filtered off. The filtrate was concentrated
in vacuo with a rotary evaporator. The residue (N-chloroacetyl
derivative of compound 3) was dissolved in DMF (12 mL) under
an argon atmosphere. To the solution, K2CO3 (0.276 g, 2.0 mmol)
was added and
a solution of 4-chlorothiophenol (0.290 g,
2.0 mmol) in DMF (4 mL) was added with stirring. The reaction
mixture was stirred for 24 h at rt; then, DMF was evaporated in
vacuo. Water (6 mL) and 1 M NaOH (6 mL) were added and ex-
tracted with EtOAc (3 ꢂ 10 mL). The combined organic extracts
were washed with saturated NaCl solution (5 mL) and dried over
Na2SO4. The drying agent was filtered off and the filtrate was con-
centrated in vacuo with a rotary evaporator. The residue was
chromatographed on silica gel using CHCl3ꢀMeOH (20:1) as an
eluent.

S. Ayan et al. / Tetrahedron: Asymmetry 24 (2013) 838–843
843
4687–4690; For calcium catalyzed examples see: (r) Shi, J.-W.; Zhao, M.-X.; Lei,
Z.-Y.; Shi, M. J. Org. Chem. 2008, 73, 305–308; For gold catalyzed examples see:
(s) Saito, S.; Tsubogo, T.; Kobayashi, S. J. Am. Chem. Soc. 2007, 129, 5364–5365.
see also Ref.10
4. Kudryavtsev, K. V. Russ. J. Org. Chem. 2010, 46, 372–379.
5. Chulakov, E. N.; Gruzdev, D. A.; Levit, G. L.; Kudryavtsev, K. V.; Krasnov, V. P.
Tetrahedron: Asymmetry 2012, 23, 1683–1688.
6. (a) Lu, Y.; Shi, T.; Wang, Y.; Yang, H.; Yan, X.; Luo, X.; Jiang, H.; Zhu, W. J. Med.
Chem. 2009, 52, 2854–2862; (b) Lu, Y.; Liu, Y.; Xu, Z.; Li, H.; Liu, H.; Zhu, W.
Expert Opin. Drug Discov. 2012, 7, 375–383.
7. Morris, G. M.; Huey, R.; Lindstrom, W.; Sanner, M. F.; Belew, R. K.; Goodsell, D.
S.; Olson, A. J. J. Comput. Chem. 2009, 30, 2785–2791.
8. Trott, O.; Olson, A. J. J. Comput. Chem. 2010, 31, 455–461.
9. McGann, M. J. Chem. Inf. Model. 2011, 51, 578–596.
10. O’Boyle, N. M.; Banck, M.; James, C. A.; Morley, C.; Vandermeersch, T.;
Hutchison, G. R. J. Chem. Inf. 2011, 3, 33.
11. (a) Imai, Y. N.; Inoue, Y.; Nakanishi, I.; Kitaura, K. Protein Sci. 2008, 17, 1129–
1137; (b) Baum, B.; Mohamed, M.; Zayed, M.; Gerlach, C.; Heine, A.; Hangauer,
D.; Klebe, G. J. Mol. Biol. 2009, 390, 56–69.
2. For selected examples, see: (a) He, L.; Chen, X.-H.; Wang, D.-N.; Luo, S.-W.;
Zhang, W.-Q.; Yu, J.; Ren, L.; Gong, L.-Z. J. Am. Chem. Soc. 2011, 133, 13504–
13518; (b) Chen, X.-H.; Zhang, W.-Q.; Gong, L.-Z. J. Am. Chem. Soc. 2008, 130,
5652–5653; (c) Guo, C.; Xue, M.-X.; Zhu, M.-K.; Gong, L.-Z. Angew. Chem., Int. Ed.
2008, 47, 3414–3417; (d) Kudryavtsev, K. V.; Zagulyaeva, A. A. Russ. J. Org.
Chem. 2008, 44, 378–387; (e) Liu, Y.-K.; Liu, H.; Du, W.; Yue, L.; Chen, Y.-C.
Chem. Eur. J. 2008, 14, 9873–9877; (f) Xue, M.-X.; Zhang, X.-M.; Gong, L.-Z.
Synlett 2008, 691–694; (g) Nakano, M.; Terada, M. Synlett 2009, 1670–1674; (h)
Xie, J.-W.; Yoshida, K.; Takasu, K.; Takemoto, Y. Tetrahedron Lett. 2008, 49,
6910–6913; (i) Vicario, J. L.; Reboredo, S.; Badia, D.; Carrillo, L. Angew. Chem.,
Int. Ed. 2007, 46, 5168–5170; (j) Ibrahem, I.; Rios, R.; Vesely, J.; Cordova, A.
Tetrahedron Lett. 2007, 48, 6252–6257.
12. Wu, E. L.; Han, K.; Zhang, J. Z. H. Chem. Eur. J. 2008, 14, 8704–8714.
3. Kudryavtsev, K. V.; Shulga, D. A.; Chupakhin, V. I.; Churakov, A. V.; Datsuk, N.
G.; Zabolotnev, D. V.; Zefirov, N. S. Russ. Chem. Bull. 2011, 60, 685–693.