Substrate selectivity of high-activity mutants of human butyrylcholinesterase

Article abstract of DOI:10.1039/c3ob41713a

Cocaine is one of the most addictive drugs, and there is still no FDA (Food and Drug Administration)-approved medication specific for cocaine abuse. A promising therapeutic strategy is to accelerate cocaine metabolism, producing biologically inactive metabolites via a route similar to the primary cocaine-metabolizing pathway, i.e. cocaine hydrolysis catalyzed by butyrylcholinesterase (BChE) in plasma. However, the native BChE has a low catalytic efficiency against the abused cocaine, i.e. (-)-cocaine. Our recently designed and discovered A199S/F227A/S287G/A328W/Y332G mutant and other mutants of human BChE have a considerably improved catalytic efficiency against (-)-cocaine. In the present study, we carried out both computational modeling and experimental kinetic analysis on the catalytic activities of these promising new BChE mutants against other known substrates, including neurotransmitter acetylcholine (ACh), acetylthiocholine (ATC), butyrylthiocholine (BTC), and (+)-cocaine, in comparison with the corresponding catalytic activity against (-)-cocaine. Both the computational modeling and kinetic analysis have consistently revealed that all the examined amino acid mutations only considerably improve the catalytic efficiency of human BChE against (-)-cocaine, without significantly improving the catalytic efficiency of the enzyme against any of the other substrates examined. In particular, all the examined BChE mutants have a slightly lower catalytic efficiency against neurotransmitter ACh compared to the wild-type BChE. This observation gives us confidence in developing an anti-cocaine enzyme therapy by using one of these BChE mutants, particularly the A199S/F227A/S287G/A328W/Y332G mutant.

Full text of DOI:10.1039/c3ob41713a

Organic &
Biomolecular Chemistry
View Article Online
View Journal
PAPER
Substrate selectivity of high-activity mutants of human
butyrylcholinesterase†
Cite this: DOI: 10.1039/c3ob41713a
Shurong Hou,‡ Liu Xue,‡ Wenchao Yang,‡ Lei Fang, Fang Zheng and
Chang-Guo Zhan*
Cocaine is one of the most addictive drugs, and there is still no FDA (Food and Drug Administration)-
approved medication specific for cocaine abuse. A promising therapeutic strategy is to accelerate cocaine
metabolism, producing biologically inactive metabolites via a route similar to the primary cocaine-meta-
bolizing pathway, i.e. cocaine hydrolysis catalyzed by butyrylcholinesterase (BChE) in plasma. However,
the native BChE has a low catalytic efficiency against the abused cocaine, i.e. (−)-cocaine. Our recently
designed and discovered A199S/F227A/S287G/A328W/Y332G mutant and other mutants of human
BChE have a considerably improved catalytic efficiency against (−)-cocaine. In the present study, we
carried out both computational modeling and experimental kinetic analysis on the catalytic activities of
these promising new BChE mutants against other known substrates, including neurotransmitter acetyl-
choline (ACh), acetylthiocholine (ATC), butyrylthiocholine (BTC), and (+)-cocaine, in comparison with the
corresponding catalytic activity against (−)-cocaine. Both the computational modeling and kinetic analy-
sis have consistently revealed that all the examined amino acid mutations only considerably improve the
catalytic efficiency of human BChE against (−)-cocaine, without significantly improving the catalytic
efficiency of the enzyme against any of the other substrates examined. In particular, all the examined
BChE mutants have a slightly lower catalytic efficiency against neurotransmitter ACh compared to the
wild-type BChE. This observation gives us confidence in developing an anti-cocaine enzyme therapy by
using one of these BChE mutants, particularly the A199S/F227A/S287G/A328W/Y332G mutant.
Received 7th June 2013,
Accepted 3rd September 2013
DOI: 10.1039/c3ob41713a
www.rsc.org/obc
As is well known, computational design of high-activity
mutants of an enzyme is extremely challenging, particularly
Introduction
Cocaine is a widely abused drug1 without an FDA-approved when the chemical reaction process is rate-determining for the
medication. The disastrous medical and social consequences enzymatic reaction.15–17 To computationally design a mutant
of cocaine abuse have made the development of a feasible enzyme with an improved catalytic activity for a given sub-
anti-cocaine medication a high priority.2,3 An ideal anti- strate, one needs to design possible amino acid mutations that
cocaine medication would be to accelerate cocaine meta- can accelerate the rate-determining step of the catalytic reac-
bolism, producing biologically inactive metabolites via a route tion process11,18,19 while other steps of the reaction are not
similar to the primary cocaine-metabolizing pathway, i.e. slowed down by the mutations. The fundamental reaction
cocaine hydrolysis catalyzed by butyrylcholinesterase (BChE) in pathway for BChE-catalyzed hydrolysis of (−)-cocaine was
plasma.4–9 Unfortunately, wild-type BChE has a low catalytic uncovered by extensive molecular dynamics (MD) simu-
activity against naturally occurring (−)-cocaine (kcat = 4.1 min−1 lations11,18 and reaction-coordinate calculations18,19 using
and KM = 4.5 μM).10–14 It is of interest for development of anti- quantum mechanics (QM) and hybrid quantum mechanics/
cocaine medication to design a human BChE mutant as a molecular mechanics (QM/MM).20–24 The computational
cocaine hydrolase (CocH) with significantly improved catalytic studies revealed11,15,18,25 that the rate-determining step of
activity against (−)-cocaine.
(−)-cocaine hydrolysis catalyzed by the A328W/Y332A and
A328W/Y332G mutants of BChE is the first step of the chemi-
cal reaction process. Therefore, starting from the A328W/
Y332A or A328W/Y332G mutant, rational design of BChE
mutants against (−)-cocaine has been focused on decreasing
the energy barrier for the first reaction step without signifi-
cantly affecting the other steps. We have developed unique
computational strategies and protocols based on the virtual
Department of Pharmaceutical Sciences, College of Pharmacy, University of Kentucky,
789 South Limestone Street, Lexington, KY 40536, USA. E-mail: zhan@uky.edu;
Tel: +859-323-3943
†Electronic supplementary information (ESI)
10.1039/c3ob41713a
available. See DOI:
‡These authors contributed equally to the work.
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
screening of rate-determining transition states of the enzy- structure, the lengths of transition bonds (i.e. the covalent
matic reaction to design enzyme mutants with improved cata- bonds that are being broken or formed gradually during the
lytic activity.25–31 The computational design was followed by initial reaction step) were restrained while all other geometric
in vitro experiments, including site-directed mutagenesis, parameters were allowed to move. The transition-bond lengths
protein expression, and enzyme activity assays. The integrated used in our modeling of the TS1 structures for each pair of
computational–experimental studies have led to the discovery enzyme and substrate were based on our previously reported
of a series of BChE mutants with a significantly improved cata- molecular modeling and QM/MM reaction-coordinate calcu-
lytic efficiency against (−)-cocaine.25–31 The first one of our lations on BChE-catalyzed hydrolysis of (−)-cocaine or ACh or
designed and discovered high-activity mutants of human ATC.28,31,36 Specifically, the transition-bond lengths used in
BChE, i.e. the A199S/S287G/A328W/Y332G mutant,25 was vali- our modeling of the TS1 structures with (−)-cocaine or
dated by an independent group of scientists32,33 who con- (+)-cocaine were the same as those in the QM/MM-optimized
cluded that this mutant is “a true CocH with a catalytic TS1 geometry for BChE-catalyzed hydrolysis of (−)-cocaine.28
efficiency that is 1000-fold greater than wild-type BChE”. This The transition-bond lengths used in our modeling of the TS1
BChE mutant is currently in double-blind, placebo-controlled structures with ACh were the same as those in the QM/MM-
clinical trials in humans by Teva Pharmaceutical Industries optimized TS1 geometry for BChE-catalyzed hydrolysis of
Ltd for cocaine abuse treatment.34 Our recently designed and Ach.36 The transition-bond lengths used in our modeling of
discovered new mutants28,30 of human BChE are even more the TS1 structures with ATC or BTC were the same as those in
active against (−)-cocaine in vitro and in vivo, without knowing the QM/MM-optimized TS1 geometry for BChE-catalyzed
whether these mutations also considerably increase the cataly- hydrolysis of ATC.37 The amino acid mutations and the minor
tic efficiencies against other substrates.
The present study was focused on the substrate selectivity significantly change the transition-bond lengths based on our
of our discovered high-activity mutants of human BChE. We previous experience in the transition-state simu-
carried out both computational modeling and experimental lations.16,17,25,28,30,35 As discussed in our previous compu-
kinetic analysis on the catalytic activities of the promising tational studies related to the transition-state
structural difference in the substrate were not expected to
BChE mutants against other known substrates, including modeling,16,17,25,28,30,35 the computational procedures for
acetylcholine (ACh), acetylthiocholine (ATC), butyrylthiocho- modeling a TS1 structure were the same as those for modeling
line (BTC), and (+)-cocaine, in comparison with the corres- the corresponding ES structure, except for keeping the tran-
ponding catalytic activity against (−)-cocaine. In particular, sition bond lengths restrained during the energy minimization
ACh is the only known natural substrate of BChE in the body. or MD simulation on the TS1 structure. Technically, each tran-
The catalytic activities of wild-type BChE against ACh, ATC, sition-bond length in a TS1 structure was restrained by defin-
BTC, and (+)-cocaine are all much higher than that against ing a new type of covalent bond whose force constant was one
(−)-cocaine. So, we wanted to know whether the same amino half of the normal covalent bond between the two atoms. It
acid mutations designed to considerably increase the catalytic should be pointed out that the sole purpose of performing this
activity of BChE against (−)-cocaine also considerably increase type of computational modeling on a transition state was to
the catalytic activities of BChE against other substrates. The examine the hydrogen bonding interaction between the carbo-
obtained kinetic data have demonstrated that the computa- nyl group of the substrate and the oxyanion hole of the
tionally designed mutations only considerably improve the enzyme. We were only interested in the modeled structures, as
catalytic efficiency of human BChE against (−)-cocaine, the total energies calculated in this way would be meaningless.
without significantly improving the catalytic activities against The modeled structures were used to estimate the hydrogen
other substrates.
bonding energies (HBE) by using an HBE equation25 utilized
in our earlier studies.
The initial structures of BChE and the mutants used in the
molecular modeling were prepared on the basis of our pre-
vious MD simulation11,25,31 on the enzyme–substrate complex
for wild-type BChE binding with (−)-cocaine. Our previous MD
Methods
Molecular modeling
Various substrates interacting with human BChE and its simulations on the enzyme–substrate complexes started from
mutants were modeled for their enzyme–substrate binding the X-ray crystal structure38 deposited in the Protein Data Bank
complexes (denoted as ES) and transition states for the initial (pdb code: 1P0P). The general procedure for carrying out the
reaction step (denoted as TS1) by using the same modeling MD simulations on the enzyme–substrate interactions in water
strategy and approach that we used to study (−)-cocaine inter- was essentially the same as that used in our previously
acting with the enzymes.25–31 The general strategy of perform- reported computational studies on other complexes.17,25,30,31
ing an energy minimization or molecular dynamics (MD) Each starting structure was neutralized by adding a counter
simulation on a transition-state structure of an enzymatic reac- ion (chloride ion) and was solvated in an orthorhombic box of
tion using a classical force field (molecular mechanics) has TIP3P water molecules with a minimum solute–wall distance
been described and justified in our recent reports.25,28,30,31,35 of 10 Å (which means that the shortest distance between an
During the energy minimization or MD simulation on the TS1 atom of the enzyme–substrate complex and the boundary of
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
the box is longer than 10 Å). The obtained box sizes of the sol- benzoic acid (a product of (−)-cocaine hydrolysis) was extracted
vated systems were about 99 Å × 92 Å × 87 Å. All the energy by 1 ml of toluene and measured by scintillation counting.
minimizations and MD simulations (using Newton’s equations Finally, the measured (−)-cocaine concentration-dependent
of motion) were performed by using the Sander module of the radiometric data were analyzed by using the standard Michaelis–
Amber11 package.39 The solvated systems were carefully equili- Menten kinetics so that the catalytic parameters were deter-
brated and fully energy-minimized. First, the solvent mole- mined along with the use of a well-established standard
cules were energy-minimized for 5000 steps (including 2500 enzyme-linked immunosorbent assay (ELISA) protocol.27 The
steps using the steepest descent method and 2500 steps using enzyme activity assay with [3H]ACh was similar to the assay
the conjugate gradient method) with the ligand and enzyme with [3H](−)-cocaine. The primary difference is that the enzy-
restrained. Second, the solvent, ligand, and side chains of the matic reaction was stopped by the addition of 200 µl of 0.2 M
enzyme were energy-minimized for 1000 steps (including 500 HCl containing 2 M NaCl and that the product was [3H]acetic
steps using the steepest descent method and 500 steps using acid for the ACh hydrolysis. All measurements were performed
the conjugate gradient method) with the backbone of enzyme at room temperature.
restrained. Finally the whole system was energy-minimized for
5000 steps (including 2500 steps using the steepest descent and (+)-cocaine were determined by UV-Vis spectrophotometric
method and 2500 steps using the conjugate gradient method). assays using GENios Pro Microplate Reader (TECAN,
The catalytic activities of the enzymes against ATC, BTC,
a
These systems were gradually heated from T = 10 K to T = Research Triangle Park, NC) with the XFluor software.
298.15 K in 30 ps before running the MD simulation at T = (+)-Cocaine was dissolved in water to make a 0.1 M stock solu-
298.15 K for 1 ns or longer, making sure that we obtained a tion containing 34 mg ml−1. Aliquots were frozen at −20 °C,
stable MD trajectory for each of the simulated structures. The thawed once, and discarded. The reaction rate of (+)-cocaine
time step used for the MD simulations was 2 fs. Periodic hydrolysis was measured at 25 °C by recording the time-depen-
boundary conditions in the NPT ensemble at T = 298.15 K with dent absorption at 230 nm. ATC and BTC stock solutions of
Berendsen temperature coupling and P = 1 atm with isotropic 0.2 M were prepared in water and frozen at −20 °C. The reaction
molecule-based scaling were applied. The SHAKE algorithm rates were measured at 25 °C by recording the time-dependent
was used to fix all covalent bonds containing hydrogen atoms. absorption at 450 nm in the presence of 1 mM dithiobisnitro-
The non-bonded pair list was updated every 10 steps. The par- benzoic acid, in 0.1 M potassium phosphate, pH 7.2.
ticle mesh Ewald (PME) method was used to treat long-range
electrostatic interactions. A residue-based cutoff of 10 Å was
utilized for the non-covalent interactions. The final snapshot
of the stable MD trajectory was fully energy-minimized using
the steepest descent method for 5000 steps and then the con-
Results and discussion
Insights from molecular modeling
jugate gradient method until the convergence criterion for an Molecular modeling enabled us to understand how human
energy gradient of 0.1 kcal mol−1
Å
−1 was achieved.
BChE and its mutants interact with ACh, ATC, BTC,
(+)-cocaine, and (−)-cocaine in the ES and TS1 structures.
According to the modeling, for (−)-cocaine interacting with
Experimental methods
Both wild-type and mutants of human BChE were expressed wild-type BChE, there is only one hydrogen bond (H-bond)
and their enzyme activities against various substrates were between the carbonyl oxygen of (−)-cocaine and the oxyanion
assayed at the same time under the same experimental con- hole (G116, G117, and A199) in the ES structure, and there are
ditions; the wild-type was used as a standard reference and two H-bonds in the TS1 structure, as seen in Fig. 1A and 1B;
validated according to the catalytic parameters reported for the more detailed computational data are provided as ESI.† With
wild-type in the literature. The proteins (wild-type and mutants the A199S/F227A/S287G/A328W/Y332G mutant (E12-7), there
of BChE) were expressed in human embryonic kidney 293F are two H-bonds between the carbonyl oxygen of (−)-cocaine
cells. Cells at a density of ∼1 × 106 cells per ml were transfected and the oxyanion hole (G116, G117, and S199) in the ES struc-
by 293fectin reagent–DNA complexes in the ratio of 2 μl : 1 μg ture and three H-bonds in the TS1 structure, as seen in Fig. 1C
ml−1 of the cells. Cells were cultured for five more days. The and 1D. The extra H-bond in both the ES and TS1 structures is
culture medium was harvested for the BChE activity assays.
with the hydroxyl group of S199 after the A199S mutation. The
For determining the catalytic activity of the enzymes modeled ES and TS1 structures suggest that this mutant
against (−)-cocaine, we used a sensitive radiometric assay should have a significantly higher catalytic activity against
based on toluene extraction of [3H](−)-cocaine labeled on its (−)-cocaine compared to the wild-type BChE.
benzene ring.40 In brief, to initiate the enzymatic reaction,
For (−)-cocaine interacting with other mutants examined
100 nCi of [3H](−)-cocaine along with (−)-cocaine was mixed (i.e. the A199S/A328W/Y332G, A199S/F227A/A328W/Y332G,
with the culture medium. The enzymatic reactions proceeded A199S/S287G/A328W/Y332G, and A199S/F227A/S287G/A328W/
at room temperature (25 °C) with varying concentrations of E441D mutants), the modeled ES and TS1 structures are all
(−)-cocaine. The reactions were stopped by adding 200 μl of qualitatively similar to the corresponding ES and TS1 struc-
0.1 M HCl, which neutralized the liberated benzoic acid while tures with the A199S/F227A/S287G/A328W/Y332G mutant in
ensuring a positive charge on the residual (−)-cocaine. [3H] terms of the number of H-bonds. Thus, we only depict the ES
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 2 The energy-minimized ES and TS1 structures for (+)-cocaine interacting
with wild-type human BChE and its A199S/F227A/S287G/A328W/Y332G
mutant (E12-7).
Fig. 1 The energy-minimized ES and TS1 structures for (−)-cocaine interacting
with wild-type human BChE and its A199S/F227A/S287G/A328W/Y332G
mutant (E12-7).
and TS1 structures with the A199S/F227A/S287G/A328W/Y332G
mutant in Fig. 1 as a typical example of the five mutants exam-
ined. The modeled ES and TS1 structures qualitatively suggest
that each of these mutants should have a considerably higher
catalytic activity against (−)-cocaine compared to the wild-type
BChE.
For other substrates (including ACh, ATC, BTC, and
(+)-cocaine) interacting with the wild-type BChE or any of the
aforementioned mutants, there are always two H-bonds
between the carbonyl oxygen of the substrate and the oxyanion
hole (G116, G117, and A/S199) in the ES structure and three
H-bonds in the TS1 structure. Depicted in Fig. 2–5 are the
modeled ES and TS1 structures with wild-type BChE and a
representative mutant (A199S/F227A/S287G/A328W/Y332G).
More detailed computational data are provided in the ESI.†
The hydroxyl group of S199 in the mutant does not form an
extra H-bond in the ES or TS1 structure for the mutant inter-
acting with any substrate other than (−)-cocaine, suggesting
that these BChE mutants should not have dramatically
improved catalytic activities against ACh, ATC, BTC, and
(+)-cocaine compared to the wild-type BChE.
Fig. 3 The energy-minimized ES and TS1 structures for ACh interacting with
wild-type human BChE and its A199S/F227A/S287G/A328W/Y332G mutant
(E12-7).
Generally speaking, structural information obtained from
MD simulations on a protein–ligand complex may be used to stabilized by hydrogen bonding between the carbonyl oxygen
analyze the interaction energies.41,42 Based on the MD-simu- of the substrate and the oxyanion hole. With the structure of
lated H⋯O distances between the carbonyl oxygen of the sub- the reaction system changing from ES to TS1, the calculated
strate and the oxyanion hole, we estimated the total HBE tHBE value decreases by only about 1 kcal mol−1 for wild-type
(tHBE) values in the ES and TS1 structures for the wild-type BChE-catalyzed hydrolysis of (−)-cocaine (or we have Δ(tHBE)
BChE- and E12-7-catalyzed hydrolysis of the five substrates (see ≈ −1 kcal mol−1), and about 13–15 kcal mol−1 for E12-7-cata-
the ESI† for the data). The calculated tHBE values indicate lyzed hydrolysis of (−)-cocaine (or we have Δ(tHBE)
≈
that, for a given enzymatic hydrolysis, the tHBE value in the −13–15 kcal mol−1). In other words, the Δ(tHBE) value
TS1 structure is always lower than that in the corresponding decreases by about 12–14 kcal mol−1 from the wild-type BChE
ES structure; the lower tHBE value (minus value) means stron- to E12-7 for (−)-cocaine. So, the amino acid mutations in E12-7
ger overall hydrogen bonding. So, the TS1 structure is always have dramatically enhanced the overall hydrogen bonding
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
considerably improving the catalytic efficiency of human BChE
against ACh, ATC, BTC, or (+)-cocaine.
Kinetic parameters
Based on the computational insights, we carried out in vitro
experimental tests, including the protein expression and
enzyme activity assays, on the A199S/A328W/Y332G, A199S/
F227A/A328W/Y332G, A199S/S287G/A328W/Y332G, A199S/
F227A/S287G/A328W/E441D, and A199S/F227A/S287G/A328W/
Y332G mutants of human BChE. To minimize the possible sys-
tematic experimental errors of the kinetic data, we also
expressed the wild-type enzyme and performed the kinetic ana-
lysis along with the mutants under the same experimental con-
ditions, and compared the catalytic efficiencies of the mutants
to the corresponding catalytic efficiencies of the wild-type
enzyme against various substrates. Depicted in Fig. 6–10 are
the measured kinetic data. Summarized in Table 1 are the
determined kinetic parameters.
Based on the kinetic parameters summarized in Table 1, all
five BChE mutants examined in this study have a considerably
improved catalytic efficiency (kcat/KM) against (−)-cocaine, with
the improvement ranging from 121- to 2020-fold. The same
mutations do not dramatically improve the catalytic efficien-
cies of human BChE against the other substrates. Within the
five mutants examined, the values of the relative catalytic
efficiency (RCE), i.e. the ratio of the catalytic efficiency of the
mutant to that of the wild-type BChE, range from 0.4 to 1.68
for (+)-cocaine, 0.066 to 0.95 for ACh, 0.18 to 1.99 for ATC, and
0.39 to 3.91 for BTC. The largest RCE value is associated with
Fig. 4 The energy-minimized ES and TS1 structures for ATC interacting with
wild-type human BChE and its A199S/F227A/S287G/A328W/Y332G mutant
(E12-7).
Fig. 5 The energy-minimized ES and TS1 structures for BTC interacting with
wild-type human BChE and its A199S/F227A/S287G/A328W/Y332G mutant
(E12-7).
between the carbonyl oxygen of the substrate and the oxyanion
hole in the TS1 structure for the (−)-cocaine hydrolysis. In
comparison, the changes in the Δ(tHBE) values calculated for
(+)-cocaine, ACh, ATC, and BTC are all much smaller than the
change of the Δ(tHBE) value calculated for (−)-cocaine.
Based on the modeling results, the computationally exam-
ined five sets of mutations (A199S/A328W/Y332G, A199S/
F227A/A328W/Y332G, A199S/S287G/A328W/Y332G, A199S/
F227A/S287G/A328W/E441D, and A199S/F227A/S287G/A328W/
Y332G) are all expected to considerably improve the catalytic
efficiency of human BChE against (−)-cocaine, without
Fig. 6 Kinetic data for (−)-cocaine hydrolysis catalyzed by wild-type human
BChE and various BChE mutants.
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 7 Kinetic data for (+)-cocaine hydrolysis catalyzed by wild-type human
Fig. 9 Kinetic data for ATC hydrolysis catalyzed by wild-type human BChE and
BChE and various BChE mutants.
various BChE mutants.
Fig. 10 Kinetic data for BTC hydrolysis catalyzed by wild-type human BChE and
Fig. 8 Kinetic data for ACh hydrolysis catalyzed by wild-type human BChE and
various BChE mutants.
various BChE mutants.
For ACh (which is the only known endogenous substrate of
the A199S/F227A/S287G/A328W/E441D mutant for both BChE in the body), all the BChE mutants actually have a
(+)-cocaine (1.68-fold) and ATC (1.99-fold), and the A199S/ slightly lower catalytic efficiency compared to the wild-type
F227A/A328W/Y332G mutant for BTC (3.91-fold).
BChE. So, all these mutants only have a considerably improved
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
Table 1 Kinetic parameters determined for (−)-cocaine, (+)-cocaine, ACh, ATC, and BTC hydrolyses catalyzed by wild-type BChE and its mutants
Substrate
Enzymea
KM (µM)
kcat (min−1
)
kcat/KM (M−1 min−1
)
RCEb
(−)-Cocaine
Wild-type BChEc
A199S/A328W/Y332G
A199S/F227A/A328W/Y332G
A199S/S287G/A328W/Y332G
A199S/F227A/S287G/A328W/E441D
A199S/F227A/S287G/A328W/Y332G
Wild-type BChE
4.5
5.1
4.4
3.1
1.1
3.1
4.7
5.0
4.4
6.3
4.7
4.6
148
156
189
36
27
37
33
31
41
21
12
20
4.1
560
1560
3060
1730
5700
6420
2820
6060
5620
10 800
8990
61 200
4190
7430
9.1 × 105
1.1 × 108
3.6 × 108
9.9 × 108
1.6 × 109
1.8 × 109
1.4 × 109
5.6 × 108
1.4 × 109
8.9 × 108
2.3 × 109
2.0 × 109
4.1 × 108
2.7 × 107
3.9 × 107
1.5 × 108
3.9 × 108
3.2 × 108
6.1 × 108
1.1 × 108
1.6 × 108
3.7 × 108
1.2 × 109
7.2 × 108
1.7 × 109
6.8 × 108
6.8 × 109
2.7 × 109
2.0 × 109
2.2 × 109
1
121
396
1080
1730
2020
1
0.40
1.01
0.65
1.68
1.43
1
0.066
0.095
0.37
0.95
0.78
1
0.18
0.26
0.60
1.99
1.19
1
(+)-Cocaine
ACh
A199S/A328W/Y332G
A199S/F227A/A328W/Y332G
A199S/S287G/A328W/Y332G
A199S/F227A/S287G/A328W/E441D
A199S/F227A/S287G/A328W/Y332G
Wild-type BChEd
A199S/A328W/Y332G
A199S/F227A/A328W/Y332G
A199S/S287G/A328W/Y332G
A199S/F227A/S287G/A328W/E441D
A199S/F227A/S287G/A328W/Y332G
Wild-type BChEe
5320
10 400
11 900
20 200
3410
6870
7880
14 000
14 800
29 500
6100
74 700
14 400
17 800
28 000
ATC
A199S/A328W/Y332G
A199S/F227A/A328W/Y332G
A199S/S287G/A328W/Y332G
A199S/F227A/S287G/A328W/E441D
A199S/F227A/S287G/A328W/Y332G
Wild-type BChEf
BTC
17
8.9
11
5.3
8.9
13
A199S/A328W/Y332G
0.39
3.91
1.57
1.15
1.24
A199S/F227A/A328W/Y332G
A199S/S287G/A328W/Y332G
A199S/F227A/S287G/A328W/E441D
A199S/F227A/S287G/A328W/Y332G
a Unless indicated otherwise, all kinetic parameters listed in this table have been determined in the present study. b RCE refers to the relative
catalytic efficiency (kcat/KM), i.e. the ratio of the kcat/KM value of the mutant to that of wild-type BChE against the same substrate. c Data for wild-
type BChE from ref. 10. d The kcat value for wild-type BChE was reported in ref. 51. e The kcat value for wild-type BChE from ref. 52. f The kcat value
for wild-type BChE was reported in ref. 52.
catalytic efficiency against (−)-cocaine without an improve- reasons, even high levels of plasma BChE activity are unlikely
ment in the catalytic efficiency of BChE against ACh. As a to affect motor transmission. The current observation that
result, the catalytic efficiencies of these BChE mutants against none of these high-activity mutants of human BChE has an
(−)-cocaine are all higher than the corresponding catalytic improved catalytic efficiency against ACh compared to the
efficiencies against ACh, as seen in Table 1.
wild-type BChE gives additional confidence in developing an
The observed substrate selectivity of these mutants may be enzyme therapy by using one of these BChE mutants, particu-
used to address a potential question concerning whether the larly the A199S/F227A/S287G/A328W/Y332G mutant with the
enzyme therapy using a high-activity mutant of human BChE highest catalytic efficiency against (−)-cocaine.
would significantly affect the cholinergic transmission and,
thus, produce adverse effects. In fact, previous studies evaluat-
ing wild-type human BChE as a prophylaxis against chemical
warfare nerve agents found no autonomic or motor impair-
Conclusion
ment in rats, guinea pigs, or primates, even with the high
doses raising the plasma enzyme levels by 50- to 100-fold.43–48
This is not surprising due to several factors. First, the molar
concentrations of acetylcholinesterase (AChE) and BChE in
blood are roughly similar49 and BChE has a lower catalytic
efficiency against ACh compared to AChE. Second, cholinergic
synapses in the brain are insulated from plasma enzymes by
the blood–brain barrier and, thus, the exogenous enzymes in
plasma would not reach the brain. In addition, peripheral
cholinergic synapses are densely packed with AChE. It has
been known that the mouse neuromuscular junction has 5 ×
1019 catalytic AChE subunits per cc, i.e. ∼0.1 mM,50 whereas
mouse plasma BChE levels are below 0.1 μM.49 For these
Both the computational modeling and experimental kinetic
analysis have consistently revealed that all five BChE mutants
examined in this study only have a considerably improved cata-
lytic efficiency against (−)-cocaine, without dramatic improve-
ment in the catalytic efficiency against any of the other
substrates examined compared to the wild-type BChE. In par-
ticular, all these BChE mutants have a slightly lower catalytic
efficiency against ACh compared to the wild-type BChE.
The observation of the substrate selectivity gives us additional
confidence in developing an enzyme therapy by using one of
these BChE mutants, particularly the A199S/F227A/S287G/
A328W/Y332G mutant with the highest catalytic efficiency
against (−)-cocaine.
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
22 J. Gao and D. G. Truhlar, Annu. Rev. Phys. Chem., 2002, 53,
467–505.
Acknowledgements
This work was supported by the NIH grants R01 DA035552, 23 S. Marti, M. Roca, J. Andres, V. Moliner, E. Silla, I. Tunon
R01 DA032910, R01 DA013930, and R01 DA025100 to Zhan. and J. Bertran, Chem. Soc. Rev., 2004, 33, 98–107.
The authors also acknowledge the Computer Center at the Uni- 24 M. W. van der Kamp and A. J. Mulholland, Biochemistry,
versity of Kentucky for supercomputing time on a Dell X-series
Cluster with 384 nodes or 4768 processors.
2013, 52, 2708–2728.
25 Y. Pan, D. Gao, W. Yang, H. Cho, G. Yang, H.-H. Tai and
C.-G. Zhan, Proc. Natl. Acad. Sci. U. S. A., 2005, 102, 16656–
16661.
26 Y. Pan, D. Gao, W. Yang, H. Cho and C.-G. Zhan, J. Am.
Chem. Soc., 2007, 129, 13537–13543.
References
1 UNODC, World Drug Report 2010 (United Nations Publi- 27 Y. Pan, D. Gao and C.-G. Zhan, J. Am. Chem. Soc., 2008, 130,
cation, Sales No. E.10.XI.13), 10. 5140–5149.
2 L. Karila, D. Gorelick, A. Weinstein, F. Noble, 28 F. Zheng, W. Yang, M.-C. Ko, J. Liu, H. Cho, D. Gao,
A. Benyamina, S. Coscas, L. Blecha, W. Lowenstein,
J. L. Martinot, M. Reynaud and J. P. Lépine,
Int. J. Neuropsychopharmacol., 2008, 11, 425–438.
3 Z.-X. Xi and L. E. Gardner, Curr. Drug Abuse Rev., 2008, 1,
303–327.
M. Tong, H.-H. Tai, J. H. Woods and C.-G. Zhan, J. Am.
Chem. Soc., 2008, 130, 12148–12155.
29 W. Yang, Y. Pan, F. Zheng, H. Cho, H.-H. Tai and
C.-G. Zhan, Biophys. J., 2009, 96, 1931–1938.
30 F. Zheng, W. Yang, L. Xue, S. Hou, J. Liu and C.-G. Zhan,
Biochemistry, 2010, 49, 9113–9119.
4 M. M. Meijler, G. F. Kaufmann, L. Qi, J. M. Mee,
A. R. Coyle, J. A. Moss, P. Wirsching, M. Matsushita and 31 L. Xue, M.-C. Ko, M. Tong, W. Yang, S. Hou, L. Fang, J. Liu,
K. D. Janda, J. Am. Chem. Soc., 2005, 127, 2477–2484.
5 M. R. A. Carrera, G. F. Kaufmann, J. M. Mee, M. M. Meijler,
F. Zheng, J. H. Woods, H.-H. Tai and C.-G. Zhan, Mol. Phar-
macol., 2011, 79, 290–297.
G. F. Koob and K. D. Janda, Proc. Natl. Acad. Sci. U. S. A., 32 S. Brimijoin, Y. Gao, J. J. Anker, L. A. Gliddon, D. LaFleur,
2004, 101, 10416–10421.
R. Shah, Q. Zhao, M. Singh and M. E. Carroll, Neuropsycho-
pharmacology, 2008, 33, 2715–2725.
33 J. J. Anker, S. Brimijoin, Y. Gao, L. Geng, N. E. Zlebnik,
R. J. Parks and M. E. Carroll, Biol. Psychiatry, 2012, 71,
700–705.
6 D. W. Landry, K. Zhao, G. X. Yang, M. Glickman and
T. M. Georgiadis, Science, 1993, 259, 1899–1901.
7 C.-G. Zhan, S. X. Deng, J. G. Skiba, B. A. Hayes,
S. M. Tschampel, G. C. Shields and D. W. Landry,
J. Comput. Chem., 2005, 26, 980–986.
34 F. Zheng and C.-G. Zhan, Future Med. Chem., 2012, 4, 125–128.
8 L. M. Kamendulis, M. R. Brzezinski, E. V. Pindel, 35 D. Gao and C.-G. Zhan, Proteins: Struct., Funct., Bioinf.,
W. F. Bosron and R. A. Dean, J. Pharmacol. Exp. Ther., 1996,
279, 713–717.
9 D. A. Gorelick, Expert Rev. Clin. Pharmacol., 2008, 1, 277–290.
2006, 62, 99–110.
36 X. Chen, L. Fang, J. Liu and C.-G. Zhan, J. Phys. Chem. B,
2011, 115, 1315–1322.
10 H. Sun, Y.-P. Pang, O. Lockridge and S. Brimijoin, Mol. 37 X. Chen, L. Fang, J. Liu and C.-G. Zhan, Biochemistry, 2012,
Pharmacol., 2002, 62, 220–224. 51, 1297–1305.
11 A. Hamza, H. Cho, H.-H. Tai and C.-G. Zhan, J. Phys. Chem. 38 O. L. Y. Nicolet, P. Masson, J. C. Fontecilla-Camps and
B, 2005, 109, 4776–4782.
F. J. Nachon, J. Biol. Chem., 2003, 278, 41141–41147.
39 D. A. Case, T. A. Darden, T. E. Cheatham, C. L. Simmerling,
J. Wang, R. E. Duke, R. Luo, K. M. Merz, B. Wang,
D. A. Pearlman, M. Crowley, S. Brozell, V. Tsui, H. Gohlke,
J. Mongan, V. Hornak, G. Cui, P. Beroza, C. Schafmeister,
J. W. Caldwell, W. S. Ross and P. A. Kollman, AMBER11,
University of California, San Francisco, 2010.
12 S. J. Gatley, Biochem. Pharmacol., 1991, 41, 1249–1254.
13 S. Darvesh, D. A. Hopkins and C. Geula, Nat. Rev. Neurosci.,
2003, 4, 131–138.
14 E. Giacobini, Butyrylcholinesterase: Its Function and Inhibi-
tors, Martin Dunitz, an imprint of the Taylor and Francis
Group plc, London, 2003.
15 D. Gao and C. G. Zhan, Proteins: Struct., Funct., Bioinf., 40 H. Sun, M. L. Shen, Y.-P. Pang, O. Lockridge and
2006, 62, 99–110. S. Brimijoin, J. Pharmacol. Exp. Ther., 2002, 302, 710–716.
16 D. Gao and C.-G. Zhan, J. Phys. Chem. B, 2005, 109, 23070– 41 M. P. Frushicheva, J. Cao, Z. T. Chu and A. Warshel, Proc.
23076. Natl. Acad. Sci. U. S. A., 2010, 107, 16869–16874.
17 D. Gao, H. Cho, W. Yang, Y. Pan, G. Yang, H. H. Tai and 42 G. Hou and Q. Cui, J. Am. Chem. Soc., 2011, 134, 229–246.
C.-G. Zhan, Angew. Chem., Int. Ed., 2006, 45, 653–657.
18 C.-G. Zhan, F. Zheng and D. W. Landry, J. Am. Chem. Soc.,
2003, 125, 2462–2474.
19 C.-G. Zhan and D. Gao, Biophys. J., 2005, 89, 3863–3872.
20 A. Warshel and M. Levitt, J. Mol. Biol., 1976, 103, 227–249.
21 U. C. Singh and P. A. Kollman, J. Comput. Chem., 1986, 7,
718–730.
43 T. M. Myers, W. Sun, R. S. Naik, M. G. Clark, B. P. Doctor
and A. Saxena, Neurotoxicol. Teratol., 2012, 34, 323–330.
44 A. Saxena, W. Sun, J. M. Fedorko, I. Koplovitz and
B. P. Doctor, Biochem. Pharmacol., 2011, 81, 164–169.
45 Y. J. Rosenberg, A. Saxena, W. Sun, X. Jiang, N. Chilukuri,
C. Luo, B. P. Doctor and K. D. Lee, Chem. Biol. Interact.,
2010, 187, 279–286.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
46 R. F. Genovese, W. Sun, C. C. Johnson, R. C. Ditargiani, 49 B. Li, J. A. Stribley, A. Ticu, W. Xie, L. M. Schopfer,
B. P. Doctor and A. Saxena, Basic Clin. Pharmacol. Toxicol.,
2010, 106, 428–434.
P. Hammond, S. Brimijoin, S. H. Hinrichs and
O. Lockridge, J. Neurochem., 2000, 75, 1320–1331.
47 A. Saxena, W. Sun, C. Luo and B. P. Doctor, Chem. Biol. 50 L. Anglister, J. Eichler, M. Szabo, B. Haesaert and
Interact., 2005, 157–158, 199–203. M. M. Salpeter, J. Neurosci. Methods, 1998, 81, 63–71.
48 A. Weber, H. Butterweck, U. Mais-Paul, W. Teschner, 51 Y. Gao, D. LaFleur, R. Shah, Q. Zhao, M. Singh and
L. Lei, E. M. Muchitsch, D. Kolarich, F. Altmann, S. Brimijoin, Chem. Biol. Interact., 2008, 175, 83–87.
H. J. Ehrlich and H. P. Schwarz, Vox Sang., 2011, 100, 52 A. T. Boeck, L. M. Schopfer and O. Lockridge, Biochem.
285–297.
Pharmacol., 2002, 63, 2101–2110.
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem.