Interactions of chiral quinuclidin-3-yl benzoates with butyrylcholinesterase: Kinetic study and docking simulations

Article abstract of DOI:10.1002/poc.494

Both enantiomers of quinuclidin-3-yl benzoate (RQBz and SQBz) were synthesized in order to examine the stereoselectivity of the hydrolysis of these esters catalyzed by horse serum butyrylcholinesterase (BChE). The hydrolysis of benzoylcholine (BzCh) was a

Full text of DOI:10.1002/poc.494

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2002; 15: 608–614
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/poc.494
Interactions of chiral quinuclidin-3-yl benzoates with
butyrylcholinesterase: kinetic study and docking
simulations†
±
Ï
Â
Â
Ines PrimozÏic,* Tomica Hrenar, Srdanka Tomic and Zlatko Meic
Department of Chemistry, Faculty of Science, University of Zagreb, Strossmayerov trg 14, HR-10000 Zagreb, Croatia
Received 25 October 2001; revised 1 February 2002; accepted 7 February 2002
ABSTRACT: Both enantiomers of quinuclidin-3-yl benzoate (RQBz and SQBz) were synthesized in order to
epoc
examine the stereoselectivity of the hydrolysis of these esters catalyzed by horse serum butyrylcholinesterase (BChE).
The hydrolysis of benzoylcholine (BzCh) was also studied in order to determine the influence of the alcohol part of
the esters upon the kinetics. The kcat value for the substrates decreased in order BzCh > RQBz (4-fold slower) ꢀ
SQBz (76-fold slower reaction). KM values determined for quinuclidinium substrates revealed that the binding affinity
of RQBz (0.28 mM) is approximately 2-fold lower than that of SQBz (0.13 mM) towards BChE. From the ratio of the
enantiomeric kcat/KM values, an enantiomeric excess of 78% was calculated, indicating that the resolution of racemic
quinuclidin-3-yl benzoate can be achieved by hydrolysis with BChE. The orientations of all the studied benzoate
esters and butyrylcholine (BuCh) in the active site of human BChE were proposed by flexible ligand docking with
AutoDock 3.0. Analyses of the Michaelis complexes obtained revealed that there are numerous similar close contacts
in the active site. The main difference in binding of quinuclidinium and choline esters was found in the ammonium
electrostatic region which includes cation–p interaction of the ammonium moiety of substrates with the indole ring
of Trp84. The important cation–p interaction with Trp84 was lowest in the case of the S-enantiomer of QBz, which
might be the main explanation for the slowest rate of hydrolysis of that compound. Copyright  2002 John Wiley &
Sons, Ltd.
Additional material for this paper is available from the epoc website at http://www.wiley.com/epoc
KEYWORDS: (R)- and (S)-quinuclidin-3-yl benzoates; butyrylcholinesterase; hydrolysis kinetics; benzoylcholine;
butyrylcholine; docking
INTRODUCTION
metric carbon atom, many investigations have concen-
trated on the resolution of racemic compounds using
chemical3 and biocatalytic4 methods in order to provide a
simple, efficient and inexpensive procedure.
Quiniclidin-3-ol and its esters may be looked upon as
bicyclic analogues of acetylcholine, and therefore their
interaction with various receptors in a cholinergic system
is not surprising. Many esters of quinuclidin-3-ol have
been tested and recognized as pharmacological agents,
and as such they are commercially available as
therapeutics.1 Compounds containing the quinuiclidine
moiety have also been found to be potential antidotes
against poisoning by organophosphorus (anticholinester-
ase) compounds, including pesticides, insecticides and
warfare agents.2 Since 3-quiniclidinol contains an asym-
One of the enzymes tested as a biocatalyst was
butyrylcholinesterase (BChE, EC 3.1.1.8). BChE is a
serine hydrolase that has a structure similar to that of
acetylcholinesterase.5 The active site of this enzyme
consists of several major domains: (i) esteratic site,
containing active serine as a part of the catalytic triad
(Ser–His–Glu); (ii) acyl pocket, a hydrophobic region
which accommodates the acyl group of an ester; (iii)
choline subsite, consisting of tryptophan for molecular
recognition of the substrate’s quaternary ammonium
group; and (iv) oxyanion hole, formed by the main chain
N—H dipoles interacting with the carbonyl oxygen of the
substrate. The catalytic steps in ester hydrolyses involve
initial formation of an enzyme–substrate complex,
followed by an acylation step, and finally water causes
deacylation of the enzyme (Scheme 1).
*Correspondence to: I. Primozicˇ, Department of Chemistry, Faculty of
Science, University of Zagreb, Strossmayerov trg 14, HR-10000
Zagreb, Croatia.
ˇ
E-mail: ines@chem.pmf.hr
†Presented at the 8th European Symposium on Organic Reactivity
(ESOR-8), Cavtat (Dubrovnik), Croatia, September 2001.
Contract/grant sponsor: Ministry of Science and Technology of the
Republic of Croatia; Contract/grant number: 119401; Contract/grant
number: 119410.
BChE has been used previously for the resolution of
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 608–614

CHIRAL QUINUCLIDIN-3-YL BENZOATES–BUTYRYLCHOLINESTERASE
609
Scheme 1
Figure 2. The ligands chosen for docking, with rotatable
bonds highlighted with arrows
racemic 3-quinuclidinyl butyrate. The stereoselectivity of
that hydrolysis was in favour of the R-enantiomer.4a
However, the difference in hydrolysis rates was not
sufficient to achieve complete resolution of enantiomers.
In the course of our work with quinuclidinium esters, a
problem of resolution of racemic mixtures arose.2
Therefore, we chose to determine the kinetics of BChE-
catalysed hydrolysis of benzoate esters: (R)-quinuclidin-
3-yl benzoate (RQBz) and (S)-quinuclidin-3-yl benzoate
(SQBz) in comparison with benzoylcholine (BzCh) (Fig.
1).
on a Varian Cary 3 UV–Visible spectrophotometer.
HPLC analyses (Thermo Separation Prod-ucts, Spectra-
SYSTEM 2000) were performed on an RP-18 column
(Waters, SymmetryShield, 5 mm, 150 Â 3.9 mm i.d.) at
40°C. The mobile phase used was water–methanol–
acetonitrile–acetic e.g. acid–triethylamine (60:25:15:
0.33:0.2) at a flow-rate of 0.7 ml minÀ1. The reactions
were carried out in a Heidolph UNIMAX 1100 shaker.
Furthermore, in order to investigate the nature of the
interaction of benzoates with the enzyme and to predict
possible structural changes for obtaining better resolu-
tion, docking simulations were also performed on
butyrylcholine (BuCh) (Fig. 2).
Preparation of substrates
(R)- and (S)-3-quinuclidinol were resolved according to
the published procedure using L- and D-tartaric acid.3a
(R)-Quinuclidin-3-yl benzoate (RQBz) was prepared
by dissolving (R)-3-quinuclidinol (500 mg, 3.9 mmol) in
benzoic anhydride6 (3.7 ml, 20 mmol) and stirring the
mixture for 1 h at 130°C. The reaction mixture was then
cooled, made alkaline with saturated aqueous K2CO3 and
extracted with chloroform. The extract was dried over
Na2SO4 and evaporated under reduced pressure. The
residue was purified by column chromatography [alumi-
nium oxide 90 active neutral (70–230 mesh ASTM);
Merck, Darmstadt, Germany] with dichloromethane–
methanol–ammonia solution (8:2:0.35) as the eluent and
distilled under reduced pressure to give the title
EXPERIMENTAL
Materials and instrumentation
BChE (EC 3.1.1.8), type IV-S lyophilized powder from
horse serum (Sigma Chemical), benzoylcholine chloride
(BDH) and (Æ)-3-quinuclidinol (Aldrich Chemical) were
used as received. Optical rotations were measured on an
Optical Activity AA-10 automatic polarimeter at ambient
1
temperature in ethanol. H and 13C, 1D and 2D NMR
spectra were recorded on a Varian XL-GEM 300
spectrometer. Chemical shifts are given in ppm downfield
from TMS as internal standard. UV spectra were recorded
compound as a colourless oil, yield 860 mg (95%),
26
‰ꢀŠ ‡ 12:3ꢁ(c = 2.0, EtOH). ESMS: m/z calculated for
D
‡
1
C14H18NO2 , 232.1; found, 232.0. H NMR (300 MHz,
DMSO-d6), ꢁ (ppm): 1.33–1.77 (m, 4H, H5 and H8);
1.99–2.07 (m, 1H, H4); 2.47–2.88 (m, 5H, H6, H7 and
2
3
H2); 3.20 (dd, 1H, J = 14.7 and J = 8.2 Hz, H2); 4.88–
4.96 (m, 1H; H3); 7.46–7.54 (m, 2H, H3 and H5 Ph);
7.60–7.69 (m, 1H, H4 Ph); 7.97 (d, 2H, J = 7.93 Hz, H2
and H6 Ph). 13C NMR (300 MHz, DMSO-d6), ꢁ (ppm):
19.44 (C5); 24.16 (C8); 25.19 (C4); 46.01 (C6); 46.98
(C7); 55.24 (C2); 71.94 (C3); 129.03 (C2 and C6 Ph);
129.39 (C3 and C5 Ph); 130.26 (C1 Ph); 133.57 (C4 Ph);
165.79 (C
=O). UV (EtOH), lmax (nm): 238.4 (" = 4766
dm3 molÀ1 cmÀ1).
Figure 1. Substrates for enzymatic hydrolysis
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 608–614

ˇ
ˇ
I. PRIMOZIC ET AL.
610
(S)-Quinuclidin-3-yl benzoate (SQBz) was prepared
a precalculated electrostatic grid map for each atom type
present in the substrate molecule. This was achieved by
the AutoGrid part of the suite. All electrostatic maps were
26
by the same method, ‰ꢀŠ À 11:8ꢁ(c = 2.0, EtOH).
D
˚
calculated with 0.2 A spacing between grid points and in
Kinetic procedure
all cases the centre of the grid was attached to the g-
oxygen atom of catalytic Ser200. The dimensions of the
˚
The hydrolysis of benzoate esters catalysed by BChE was
monitored by following the production of benzoic acid at
238 nm by a UV detector (HPLC). The reactions were
conducted at 30°C in 0.1 M sodium phosphate buffer (pH
7.4). All experiments were performed in a total volume of
active site box were set at 10 Â 20 Â 10 A, which ensured
an appropriate size of the ligand-accessible space. The
consistencies of the maps were ascertained by checking
maximum and minimum van der Waals energies and
electrostatic potentials for each calculated grid map.
Flexible ligand docking was performed for choline
(BuCh and BzCh) and two enantiomeric quinuclidinium
(RQBz and SQBz) esters. Docking calculations were
carried out using the Lamarckian genetic algorithm
(LGA) and all parameters were the same for each
docking. We used initially a population of random
individuals (population size: 400), a maximum number of
2 Â 106 energy evaluations, a maximum number of
generations of 200 000, an elitism value of 1 and a
mutation rate of 0.08. For the local search, the pseudo-
Solis and Wets method was used: maximum of 300
iterations per local search, the probability of performing a
local search on an individual in the same population was
0.06, the maximum number of consecutive successes or
failures before doubling or halving the local search step
size was 4 in both cases and the termination criterion for
the local search was 0.01. At the end of a docking
procedure (50 docking runs), the resulting positions were
1.0 ml and an enzyme concentration of 1.5 Â 10À9
M
(0.016 mg mlÀ1). The enzymatic reaction was stopped by
addition of aliquots (20 ml) of the reaction mixture to the
HPLC mobile phase (200 ml). Two to four measurements
were made with each substrate concentration (0.08–
0.5 mM). Rates were corrected to account for non-
enzymatic hydrolysis. The concentration of active sites
of BChE was determined by titration with a high-affinity
phosphorylating agent 7-(diethoxyphosphoryl)oxy-1-
methylquinolinium iodide (DEPQ), and ascertaining the
minimal concentration that produced complete inhibi-
tion.7 KM and kcat values were obtained by non-linear
regression of the experimental data to the Michaelis–
Menten equation.
Structures of BChE and substrates
˚
A theoretical 3D structure of human BChE was derived
by homology modelling on the basis of the known x-ray
structure of Torpedo acetylcholinesterase.5 Using the
AutoDock 3.0 suite of programs,11 we added polar
hydrogen atoms to amino acid residues. Gasteiger–
Marsili8 atomic partial charges were assigned to all
atoms of the enzyme. The charge of the non-polar
hydrogens was added to the atom to which that hydrogen
was bonded. Then, atomic solvation parameters and
fragmental volumes were assigned to the protein atoms
using AutoDock’s AddSol utility.
Geometry optimizations of all substrates were carried
out using the Gaussian 94 program.9 Structures were
optimized by the density functional theory (DFT) method
using the B3LYP functional and 6–31G* basis set.10 The
structure of SQBz was simply prepared by the mirroring
of the coordinates obtained for RQBz. Further prepara-
tion of substrates included addition of Mulliken’s atomic
charges, removal of hydrogen atoms and addition of their
atomic charges to skeleton atoms, and finally, assignment
of proper atomic types. AutoTors was then used to define
the rotatable bonds in the ligands (Fig. 2).
clustered according to an r.m.s. criterion of 0.5 A and the
orientations obtained were analysed visually. To ensure
the validity of the results, the docking procedure was
repeated 10 times for each substrate.
RESULTS AND DISCUSSION
Reactivity of BChE towards substrates
According to Scheme 1, the Michaelis constant (KM)
corresponds to (kÀ1/k1 ‡ k2/k1)[k3/(k2 ‡ k3)]. The appar-
ent catalytic first-order rate constant (kcat) contains the
first-order rate constants for acylation (k2) and deacyla-
tion (k3) and is equal to k2k3/(k2 ‡ k3). The ratio kcat/KM is
often used as a measure of catalytic efficiency and it
reflects the initial steps leading to the formation of the
acylated enzyme, k1k2/(k ‡ k2).12 Using non-linear
À1
Table 1. Michaelis±Menten parameters for BChE-catalyzed
hydrolysis of choline esters
kcat
(kcÀat1/KM)
Substrate
KM (mM)
(103 minÀ1
)
(107 M minÀ1
)
Docking simulations
BzCh
RQBz
SQBz
0.17 Æ 0.01
0.28 Æ 0.02
0.13 Æ 0.02
19 Æ 1
11
1.6
0.2
4.4 Æ 0.1
All docking studies were performed by using the
AutoDock 3.0 suite of programs.11 This program requires
0.25 Æ 0.01
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 608–614

CHIRAL QUINUCLIDIN-3-YL BENZOATES–BUTYRYLCHOLINESTERASE
611
regression, KM, kcat and the apparent bimolecular rate
constants (kcat/KM) were calculated for the hydrolysis of
benzoate esters by BChE. Kinetic constants and standard
errors are displayed in Table 1.
Despite the marked difference in experimentally
observed kcat, the difference in KM values between
substrates was relatively small. Because of the relatively
small k2, KM can be used as a measure of the binding
affinity of SQBz and RQBz. Actual KM values determined
for the quinuclidinium substrates indicate that the binding
affinity RQBz toward BChE is approximately 2-fold lower
than that of SQBz. The greater affinity of the S-enantiomer
towards the active site indicates that the binding of that
substrate does not always lead to the tetrahedral
intermediate and therefore hydrolysis proceeds at a much
slower rate than that of the R-enantiomer or BzCh.
A change of the alcohol part from the charged
quaternary ammonium choline moiety in BzCh to
quinuclidinium resulted in a decreased catalytic effi-
ciency of BChE for both enantiomers. The activity trend
for both kcat and kcat/KM was BzCh > RQBz ꢀ SQBz.
The combined effects on KM and kcat resulted in a 7-fold
reduction for RQBz and a 53-fold reduction in the
apparent bimolecular rate constant for SQBz compared
with BzCh. A comparison of the substrate acylation rates
for the quinuclidinium derivatives showed that despite
the lower affinity towards the enzyme, the R-enantiomer
acylates the serine in the active site 8-fold faster. The
catalytic first-order rate constant (kcat) for quinuclidinium
derivatives reveals an 18-fold slower reaction for the S-
enantiomer. Since k3 is the same for all substrates, the
difference in the kcat values indicates that the major
reason for the 4-fold slower reaction of RQBz and the 76-
fold slower reaction of SQBz compared with BzCh lies in
the acylation step (Scheme 1).
Figure 3. Two orientations of BuCh obtained by ¯exible
ligand docking with AutoDock 3.0in the active site of BChE:
cluster 1 (À7.21 kcal molÀ1), the most energetically favour-
able orientation with the carbonyl group pointed towards
catalytic triad; cluster 2 (À7.13 kcal molÀ1), carbonyl group
oriented towards oxyanion hole. Two clusters obtained for
each benzoate ester (BzCh, SQBz and RQBz) are related in a
similar way to those of BuCh. This ®gure is available in colour
at www.interscience.wiley.com/epoc
docked into the active site of BChE by utilizing the
AutoDock program. We chose to explore the binding at
the active site of the enzyme to see whether differences in
interaction with those residues can explain the rates of
hydrolyses. Furthermore, observed differences in binding
in the ground states are probably also important for the
stability of the tetrahedral intermediates.
According to the energy values and substrate orienta-
tions in the BChE–ester complexes, AutoDock generated
two clusters of binding orientations for all substrates in
the active site. Based on the knowledge of the mechanism
of hydrolysis of AChE and BChE,12 we chose the most
probable orientation obtained for BuCh which can result
in the most stable tetrahedral intermediate during
hydrolysis (Fig. 3).
That orientation was then compared with all other
similar orientations obtained for the benzoate esters
(Table 2). However, AutoDock is not able to deal with
conformational flexibility of the enzyme, so that should
be taken into account when evaluating these results.
The results obtained from flexible ligand docking
revealed that all esters have similar interactions in the
active site. There are three energetically favourable
regions for the interaction of the tested esters with the
active site of BChE: (1) the hydrophobic region, which
The tested quinuclidinium benzoate esters are poorer
substrates for BChE than butyrylcholine, which is a good
substrate.13 Hence it can be assumed that catalysis is
limited by substrate orientation or a possible unimolecu-
lar isomerization step, as proposed for some other
substrates.14 Possible alternative binding modes in the
active gorge can compete with that which leads to the
tetrahedral intermediate and hydrolysis.
From the difference in free energy, DDG‡, one can
determine the selectivity of the reaction and the optical
purity of the product when performing the resolution of
enantiomers.15 DDG‡ was calculated from the ratio of
‘specificity constants’ (kcat/KM) obtained for RQBz and
SQBz. The value of 1.25 kcal molÀ1 corresponds to an
enantiomeric excess of 78% (50% conversion) in favor of
the R-enantiomer.
Table 2. Final intermolecular energy (EFIE), interaction energy
contributed by van der Waals interactions (EvdW) and
interaction energy contributed by electrostatic interactions
(EElect) obtained by AutoDock for the second, energetically
favourable orientation (cluster 2) for all substratesa
EFIE
EvdW
EElect
Substrate
(kcal molÀ1
)
(kcal molÀ1
)
(kcal molÀ1
)
BuCh
BzCh
RQBz
SQBz
À7.13
À8.81
À8.72
À8.71
À6.62
À8.31
À8.77
À8.87
À0.51
À0.50
0.05
Substrate-BChE complexes derived from docking
studies
0.16
a
Three different benzoate esters and butyrylcholine were
1 kcal = 4.184 kJ.
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 608–614

ˇ
ˇ
I. PRIMOZIC ET AL.
612
interaction of the ammonium moiety of substrate with
the indole ring of Trp84 and close contacts with
carboxylate of Glu199 and aromatic ring of Tyr130
(Scheme 2).
BuCh was docked to test whether AutoDock can
reproduce the binding of the ‘natural’ substrate of BChE.
The docking results obtained for this ligand showed two
main ways of binding which differed only in the position
of the carbonyl group of the substrate (Fig. 3). The
orientation with slightly lower energy was the one with a
direct H-bond between the carbonyl group and the
hydroxyl group of Ser200, which showed the strongest
tendency to form a hydrogen bond. The second com-
plex obtained differed only slightly in energy
(0.08 kcal molÀ1) but described nicely the expected
mode of binding (Fig. 4). The distances of the carbonyl
oxygen atom from the amide nitrogen atom of Gly118 and
Scheme 2
includes the propyl group and the benzene ring interact-
ing with the amino acids of the acyl pocket, (Leu285
,
Gly119 were 4.0 and 2.9 A, respectively, and the distance
˚
Val287, Trp233, Phe400, Phe331); (2) the carbonyl electro-
static region, which includes partial hydrogen bonds
formed between the carbonyl oxygen of substrates and
the amide backbone hydrogen of Gly119 or hydroxyl
hydrogen of Ser200, and the electrostatic interaction
between the electrophylic carbonyl carbon of substrate
and g-oxygen of catalytic Ser200; and (3) the ammonium
electrostatic region, which includes strong cation–p
of the carbonyl carbon atom from the g-oxygen atom of
the catalytic Ser200 was 3.2 A. Hydrophobic interactions
˚
in the acyl pocket are between the propyl chain of the
substrate and Leu288, Val290, Phe400, Phe331 and Trp233
.
The cation–p interactions between the quaternary
ammonium group and Trp84 are very important in
binding of the charged choline moiety.16 The quaternary
ammonium group of BuCh is oriented in such a way that
Figure 4. Proposed orientations for BuCh (grey), BzCh (yellow), RQBz (orange) and SQBz (blue) in the active
site of human BChE (cluster 2) obtained by ¯exible ligand docking with AutoDock 3.0, showing the residues
involved in the interactions. Hydrogens atoms are omitted for clarity. The nitrogen atom of the substrate is
represented with a ball model. This ®gure is available in colour at www.interscience.wiley.com/epoc
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 608–614

CHIRAL QUINUCLIDIN-3-YL BENZOATES–BUTYRYLCHOLINESTERASE
199
613
˚
˚
(2.7–3.5 A). Therefore,
5.8 A), but is closer to Glu
cation–p stabilization of the alcohol part of RQBz and
especially of SQBz is lower than in the case of choline
esters, and that is probably the main contributor to the
lower reaction rates.
Overlay of carbonyl groups of all obtained structures
(cluster 2) by flexible ligand docking showed that all
substrates ended in a similar orientation regarding the
position of the quaternary nitrogen atom and the carbonyl
group, except for SQBz (Fig. 5).
The quinuclidinium part of SQBz is properly posi-
tioned in the complex with the active site when its
carbonyl oxygen is turned toward His440 and makes a
hydrogen bond with Ser200. Such an orientation of SQBz,
which is unsuitable for hydrolysis, may contribute
significantly to its binding to the enzyme, explaining
why the S-enantiomer of QBz is a poor substrate for
BChE.
Figure 5. Overlay of carbonyl groups of structures obtained
by ¯exible ligand docking with AutoDock 3.0in the active
site of human BChE (cluster 2) showing the similarity of
obtained orientations for BuCh (grey), BzCh (yellow) and
RQBz (orange) and difference in the position of the
quaternary ammonium nitrogen of SQBz (blue). This ®gure
is available in colour at www.interscience.wiley.com/epoc
the cation–p interactions with the indole ring of Trp84 are
This knowledge should facilitate the search for and
identification of other structural analogues likely to
exhibit improved stereoselectivity towards BChE.
optimal, the distance to the centre of the indole ring being
199
˚
˚
3.7–4.5 A. This group also interacts with Glu (ꢂ3.0 A
‡
from the N atom and one methyl group to the OE2
oxygen of glutamic acid), but it was proved by mutation
studies that Trp84 is more important for efficient
catalysis.17
Acknowledgements
We thank Professor Israel Silman of the Weizmann
Institute, Rehovot, Israel for providing coordinates of the
theoretical structure of human BChE. We are very
grateful to Dr Elsa Reiner and Dr Vera Simeon, Institute
for Medicinal Research and Occupational Health,
Zagreb, Croatia, for their kind help with chemicals and
advice regarding the kinetic study. We are grateful to Dr
Haim Leader, Dr Yacov Ashani and Dr Bhupendra P.
Doctor for the gift of DEPQ. We also thank Dr Vesna
Benzoate esters. In many respects the Michaelis
complexes obtained with flexible ligand docking display
interactions with the active site residues that are very
similar to those obtained for BuCh. For all benzoates, the
most energetically favourable orientation was that with
the H-bond with the catalytic Ser200, while the orientation
in which the carbonyl group is pointed toward the
oxyanion hole was slightly higher in energy as obtained
for BuCh (Fig. 3). The modelling of the benzoate esters in
the acyl pocket indicated that the acyl pocket provides a
tight fit for the phenyl ring. The van der Waals energy is
ꢂ2 kcal molÀ1 higher in the case of the phenyl ring
compared with the propyl group of BuCh. It is worth
noting that the phenyl ring of all substrates has almost
exactly the same location in the acyl pocket making a T-
shaped aromatic complex with Phe331, Phe400 and Trp233
and having close contacts with Leu288 and Val290. This
strong interactions and lower flexibility of the phenyl ring
resulted in a longer distance of the carbonyl carbon atom
Gabelica, PLIVA Pharmaceutical Industries, for the mass
–
spectra and Dr Drazen Vikic´-Topic´, Ruder Bosˇkovic´
ˇ
Institute, for NMR spectra. This work was supported by
the Ministry of Science and Technology of the Republic
of Croatia, Project Nos 119401 and 119410.
REFERENCES
1. Mashkovsky MD, Yakhontov LN, Kaminka ME, Mikhlina EE.
Prog. Drug Res. 1983; 27: 9–61.
ˇ
ˇ
from the g-oxygen of the catalytic Ser200, 4.0 A for BzCh,
2. Reiner E, Skrinjaric´-Spoljar M, Dunaj S, Simeon-Rudolf V,
˚
Primozicˇ I, Tomic´ S. Chem.-Biol. Interact. 1999; 119–120: 173–
ˇ
˚
˚
4.5 A for RQBz and 4.7 A for SQBz. Furthermore, the
distances of the carbonyl oxygen atom and the amide
nitrogen atom of Gly119 were extended to 3.8, 4.1 and
181.
3. (a) Ringdahl B, Jope RS, Jenden DJ. Biochem. Pharmacol. 1984;
33: 2819–2822; (b) Sternbach LH, Kaiser S. J. Am. Chem, Soc.
1952; 74: 2215–2218; (c) Kalir A, Sali E, Shirin E. Isr. J. Chem.
1971; 9: 267–268.
˚
4.7 A for BzCh, RQBz and SQBz, respectively (Fig. 4).
The quaternary ammonium group of BzCh is located at
the same position as that of BuCh, while the positions of
the quinuclidiniums are in a less favourable position. The
nitrogen and adjacent methylenic groups of (R)-quinu-
4. (a) Rehavi M, Maayani S, Sokolovsky M. Life Sci. 1977; 21: 1293–
1302; (b) Muchmore DC. US Patent 5215918, 1993.
5. Harel M, Sussman JL, Krejci E, Bon S, Chanal P, Massoulie J,
Silman I. Proc. Natl. Acad. Sci. USA 1992; 89: 10827–10831.
6. Allen CFH, Kibler CJ, McLachlin DM, Wilson CV. Org. Synth.
Coll. Vol. 1995; 3: 28–30.
7. Levy D, Ashani Y. Biochem. Pharmacol. 1986; 35: 1079–1085.
8. Gasteiger J, Marsili M. Tetrahedron 1980; 36: 3219–3288.
9. Frisch MJ, Trucks GW, Schlegel HB, Gill PMW, Johnson BG,
˚
clidinium are 3.8–5.0 A from the centre of the indole ring
of Trp84. The position of the same atoms in (S)-
quinuclidinium is even more distant from Trp84 (4.9–
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 608–614

ˇ
ˇ
I. PRIMOZIC ET AL.
614
Robb MA, Cheeseman JR, Keith T, Petersson GA, Montgomery
13. Taylor P, Radic Z, Hosea NA, Camp S, Marchot P, Berman HA.
Toxicol. Lett. 1995; 82/83: 453–458.
14. Rosenberry TL. Adv. Enzymol. Relat. Areas Mol. Biol. 1975; 43:
103–218.
15. Faber K. Biotransformations in Organic Chemistry (2nd edn).
Springer: Berlin, 1995; 17.
16. Ma JC, Dougherty DA. Chem. Rev. 1997; 97: 1303–1324.; Quinn
DM, Feaster SR, Nair HK, Baker NA, Radic Z, Taylor P. J. Am.
Chem. Soc. 2000; 122: 2975–2980.
JA, Raghavachari K, Al-Laham MA, Zakrzewski VG, Ortiz JV,
Foresman JB, Cioslowski J, Stefanov BB, Nanayakkara A,
Challacombe M, Peng CY, Ayala PY, Chen W, Wong MW,
Andres JL, Replogle ES, Gomperts R, Martin RL, Fox DJ, Binkley
JS, Defrees DJ, Baker J, Stewart JP, Head-Gordon M, Gonzalez C,
Pople JA. Gaussian 94, Revision D.1. Gaussian: Pittsburgh, PA,
1995.
10. Becke AD. J. Chem. Phys. 1993; 98: 5648–5652.
11. Morris GM, Goodsell DS, Halliday RS, Huey R, Hart WE, Belew
RK, Olson AJ. J. Comput. Chem. 1998; 19: 1639–1662.
12. Quinn DM. Chem. Rev. 1987; 87: 955–979.
17. Ordentlich A, Barak D, Kronman C, Ariel N, Segall Y, Velan B,
Schafferman A. J. Biol. Chem. 1995; 270: 2082–91.
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 608–614