Inhibition of cholinesterases with cationic phosphonyl oximes highlights distinctive properties of the charged pyridine groups of quaternary oxime reactivators

Article abstract of DOI:10.1016/S0006-2952(03)00204-1

Oxime-induced reactivation of phosphonylated cholinesterases (ChEs) produces charged phosphonyl pyridine oxime intermediates (POXs) that are most potent organophosphate (OP) inhibitors of ChEs. To understand the role of cationic pyridine oxime leaving groups in the enhanced anti-ChE activity of POXs, the bimolecular rate constants for the inhibition (k_i) of acetylcholinesterases (AChE) and butyrylcholinesterases (BChE), and the rate of decomposition (k_d) of authentic O-alkyl methylphosphonyl pyridine oximes (AlkMeP-POXs) and N,N-dimethylamidophosphoryl pyridine oximes (EDMP-POXs), were studied. Stability ranking order in aqueous solutions correlated well with the electronic features and optimized geometries that were obtained by ab initio calculations at 6-31G** basis set level. AlkMeP-POXs of the 2-pyridine oxime series were found to be 4- to 8-fold more stable (t_1/2=0.7 to 1.5min) than the homologous O,O-diethylphosphoryl (DEP) oxime. Results suggest that re-inhibition of enzyme activity by POX is less likely during the reactivation of DEP-ChEs (obtained by use of DEP-containing pesticides) by certain oximes, compared to nerve agent-inhibited ChEs. The greatest inhibition was observed for the O-cyclohexyl methylphosphonyl-2PAM derivative (4.0×10⁹M^-1min^-1; mouse AChE) and is 10-fold higher than the k_i of cyclosarin. Increasing the size of the O-alkyl substituent of AlkMeP-POXs had only a small to moderate effect on the k_i of ChEs, signifying a major role for the cationic pyridine oxime leaving group in the inhibition reaction. The shape of plots of log k_i vs. pK_a of the leaving groups for AlkMeP-PAMs and DEP-PAMs, could be used as a diagnostic tool to highlight and rationalize the unique properties of the cationic moiety of pyridine oxime reactivators.

Full text of DOI:10.1016/S0006-2952(03)00204-1

Biochemical Pharmacology 66 (2003) 191–202
Inhibition of cholinesterases with cationic phosphonyl oximes
highlights distinctive properties of the charged pyridine groups
of quaternary oxime reactivators$
Yacov Ashania,1, Apurba K. Bhattacharjeeb, Haim Leadera,1,
Ashima Saxenaa,*, Bhupendra P. Doctora
aDivision of Biochemistry, Walter Reed Army Institute of Research, 503 Robert Grant Road,
Silver Spring, MD 20910-7500, USA
bDivision of Experimental Therapeutics, Walter Reed Army Institute of Research, 503 Robert Grant Road,
Silver Spring, MD 20910-7500, USA
Received 12 July 2002; accepted 11 November 2002
Abstract
Oxime-induced reactivation of phosphonylated cholinesterases (ChEs) produces charged phosphonyl pyridine oxime intermediates
(POXs) that are most potent organophosphate (OP) inhibitors of ChEs. To understand the role of cationic pyridine oxime leaving
groups in the enhanced anti-ChE activity of POXs, the bimolecular rate constants for the inhibition (ki) of acetylcholinesterases
(AChE) and butyrylcholinesterases (BChE), and the rate of decomposition (kd) of authentic O-alkyl methylphosphonyl pyridine oximes
(AlkMeP-POXs) and N,N-dimethylamidophosphoryl pyridine oximes (EDMP-POXs), were studied. Stability ranking order in aqueous
solutions correlated well with the electronic features and optimized geometries that were obtained by ab initio calculations at 6-31Gꢀꢀ
basis set level. AlkMeP-POXs of the 2-pyridine oxime series were found to be 4- to 8-fold more stable (t1=2 ¼ 0:7 to 1.5 min) than the
homologous O,O-diethylphosphoryl (DEP) oxime. Results suggest that re-inhibition of enzyme activity by POX is less likely during the
reactivation of DEP–ChEs (obtained by use of DEP-containing pesticides) by certain oximes, compared to nerve agent-inhibited ChEs.
The greatest inhibition was observed for the O-cyclohexyl methylphosphonyl-2PAM derivative (4:0 Â 109 MÀ1 minÀ1; mouse AChE) and
is 10-fold higher than the ki of cyclosarin. Increasing the size of the O-alkyl substituent of AlkMeP-POXs had only a small to moderate
effect on the ki of ChEs, signifying a major role for the cationic pyridine oxime leaving group in the inhibition reaction. The shape of plots
of log ki vs. pKa of the leaving groups for AlkMeP-PAMs and DEP-PAMs, could be used as a diagnostic tool to highlight and rationalize
the unique properties of the cationic moiety of pyridine oxime reactivators.
# 2003 Elsevier Science Inc. All rights reserved.
Keywords: Cholinesterases; Inhibition; Decomposition; Organophosphates; Phosphonyl oximes; Ab initio
$ A preliminary communication was presented at The US Army Medical Defense Bioscience Review 2000, 4–9 June, Hunt Valley Inn, MD.
* Corresponding author. Tel.: þ1-301-319-9406; fax: þ1-301-319-9150.
E-mail address: ashima.saxena@na.amedd.army.mil (A. Saxena).
1 Recipients of the National Research Council fellowship; on sabbatical leave from Israel Institute for Biological Research, P.O. Box 19, Ness Ziona, Israel.
Abbreviations: ChE, cholinesterases; AChE, acetylcholinesterase; BChE, butyrylcholinesterase; rMoAChE, recombinant mouse AChE; FBS AChE, fetal
bovine serum AChE; HuBChE, human BChE; EqBChE, equine BChE; OP, organophosphates; 2-PAM, 1-methyl-2-pyridinium carboxaldehyde oxime; POX,
phosphonyl, phosphoryl, or dimethylamidophosphoryl pyridine oximes; AlkMeP-POXs, O-alkyl methylphosphonyl pyridine oximes; DEP, O,O-
diethylphosphate; DEPQ, 7-(O,O-diethylphosphinyloxy)-1-methylquinolinium methyl sulfate; MEPQ, 7-(O-ethyl methylphosphinyloxy)-1-methylquinoli-
nium iodide; DEP-POX, O,O-diethylphosphoryl pyridine oximes; DEP-TMP, 3-(O,O-diethylphosphinyloxy)-m-(N,N,N-trimethylammonio) phenyl iodide;
DEP-MHP, 3-(O,O-diethylphosphinyloxy)-1-methyl pyridinium iodide; EDMP-POXs, O-ethyl N,N-dimethylamidophosphoryl oximes; IMP-thiocholine, O-
isopropyl S-(N,N,N-trimethylammonio ethyl)methylphosphonothiolate; CNPY, cyanopyridinium cations; MEP, molecular electrostatic potentials.
0006-2952/03/$ – see front matter # 2003 Elsevier Science Inc. All rights reserved.
doi:10.1016/S0006-2952(03)00204-1

192
Y. Ashani et al. / Biochemical Pharmacology 66 (2003) 191–202
1. Introduction
rate by which it displaces the OP residue from its covalent
attachment to AChE. For this reason, the rapid decomposi-
tion of POX, accumulated during oxime treatment of OP
poisoning, is of considerable importance for the therapeutic
management of nerve agent and pesticide poisoning.
Inhibition of AChE (EC 3.1.1.7) and BChE (EC 3.1.1.8)
by OPs yield covalent conjugates with the active-site serine
located at the bottom of a deep narrow gorge [1,2]. In
contrast to cationic aliphatic OPs, the potential for charge
dispersion on a planar cationic aromatic system such as
POX may provide a large unsymmetrically charged sur-
face capable of interactions with negative electrostatic
potential sites at the entrance and along the catalytic gorge.
Phosphonylated oximes (POXs, Fig. 1) are intermediates
formed during oxime-induced reactivation of OP-inhibited
ChEs (Fig. 2a). The POXs listed in Table 1 are expected to
accumulate during the reactivation of AChEs inhibited with
the G- and V-type nerve-agents and with diethylphosphate-
based pesticides (Fig. 1), by a variety of quaternary pyridine
oximes. The re-inhibition of enzymatic activity by POX
slows down the regeneration of AChE activity and compli-
cates the interpretation of results from kinetic studies both in
vitro and in vivo. The antidotal potency of an oxime reacti-
vator against OP toxicity depends mostly on the apparent
Fig. 1. Structures of organophosphate inhibitors and pyridine oxime reactivators. The syn isomers of POXs and PAMs are shown. When R0 ¼ CH3 and R ¼ Et,
iPr, pinacolyl, cyclohexyl, and 2-methylcyclohexyl, the POXs represent the parent OP inhibitors VX, sarin, soman, cyclosarin, and 2-methylcyclosarin,
respectively. When R0 ¼ ðCH3Þ2N and R ¼ OEt, the parent inhibitor is tabun.

Y. Ashani et al. / Biochemical Pharmacology 66 (2003) 191–202
193
Fig. 2. Reaction pathways associated with POXs. (a) Reactivation and re-inhibition; (b) Decomposition in accordance with the b-elimination reaction;
(c) Resonance stabilization of the dispersed positive charge on the pyridine ring.
Table 1
Rate constants for the decomposition of POXs (t1,2) and the inhibition of ChEs by POXs (ki)a
10À6 kic (MÀ1 minÀ1
)
b
t1/2 (min)
POX
R
R0
rMoAChE
FBS AChE
HuBChE
EqBChE
EMP-2PAM
IMP-2PAM
C2H5
CH3
0.70
1.05
1.50
1.30
0.79
1.10
0.18
588
>600
139
62
1940
1420
1570
4020
3400
11
3590
2170
1070
2800
2740
14
1630
632
236
672
840
3.7
280
42
CH(CH3)2
CH(CH3)C(CH3)3
Cyclohexyl
2-Methylcyclohexyl
C2H5
CH3
PMP-2PAM
CMP-2PAM
MCMP-2PAM
EDMP-2PAM
DEP-2PAMd
EMP-3 PAM
CMP-3 PAM
DEP-3 PAM
EMP-4PAM
IMP-4PAM
CH3
CH3
139
211
244
3.0
140
1.2
44
CH3
(CH3)2N
C2H5
CH3
C2H5
210
2.7
140
1.9
2030
34
C2H5
Cyclohexyl
C2H5
C2H5
CH3
195
0.16
382
110
237
1800
596
2.5
89
401
66
C2H5
CH3
0.084
188
60
0.92
69
296
73
CH(CH3)2
CH(CH3)C(CH3)3
Cyclohexyl
2-Methylcyclohexyl
C2H5
CH3
75
12
PMP-4PAM
CMP-4PAM
MCMP-4PAM
EDMP-4PAM
DEP-4PAMd
CH3
112
84
159
1580
251
1.4
363
1860
823
2.2
29
CH3
CH3
265
137
0.44
40
100
63
(CH3)2N
C2H5
C2H5
16
6.0
1.0
710
MEPQ
1980
660
7.0
1380
4420
7.7
290
532
2.1
DEPQ
Echothiophate
a For structural details see Fig. 1. EMP-, IMP-, PMP-, CMP-, MCMP-, EDMP- and DEP-POXs are expected during reactivation of ChEs inhibited with
VX, sarin, soman, cyclosarin, 2-methylcyclosarin, tabun, and paraoxon, respectively.
b Mean of 3–6 determinations ÆSD < 20%.
c Mean of 4–9 determinations, ÆSD < 20%, except for DEP-4PAM vs. EqBChE (ÆSD ¼ 35%).
d From [9].

194
Y. Ashani et al. / Biochemical Pharmacology 66 (2003) 191–202
For example, the bimolecular rate constants (ki) for the
inhibition of AChEs from various sources by the cationic
aromatic ligands DEPQ and MEPQ (Fig. 1) were found to
be 1À5 Â 108 MÀ1 minÀ1 [3–5], while those of homolo-
gous cationic aliphatic OPs such as echothiophate or one of
its methylphosphonate analogs, IMP-thiocholine (Fig. 1),
were reported at 0.06 to 0:16 Â 108 MÀ1 minÀ1 [6–8]. The
remarkably high ki values of DEPQ and MEPQ relative to
those of echothiophate and IMP-thiocholine may be
explained by the >100-fold increase in acidity of the 1-
methyl 7-hydroxyquinolinium leaving group (pKa ¼ 5:7;
[4]) compared to that of thiocholine (pKa ꢁ 7:7; [9]).
However, results from recent experiments generated
ki values of 91À130 Â 108 MÀ1 minÀ1 for certain bis-
quaternary aromatic OPs of the phosphonyl oxime family
with leaving groups exhibiting pKa values near 8.2 [3].
The three phosphonylated oxime series depicted in Fig. 1
and listed in Table 1 are structurally related to the most
potent OP inhibitors of ChE, namely, bis-quaternary
phosphonyl oximes [3]. Hence, they offer suitable candi-
dates for the systematic evaluation of the effect of charge
delocalization and molecular geometry on the rate of
ChE inhibition, and on the stability of POXs in aqueous
solutions.
Previous studies reported on the in situ characterization
of AlkMeP-POX [3,10–14] and DEP-POX [11,12]. Only
recently have authentic DEP-POXs been synthesized and
characterized as anti-ChEs [15]. We report here on the
experimental and theoretical studies of 12 newly synthe-
sized AlkMeP-POXs and two EDMP-POXs. It is suggested
that the enhanced anti-ChE activity of POXs and their
stability ranking order arise from the unique properties of
the positive electrostatic potential of the charged aromatic
leaving group. Results also highlight the contribution
of an O-cyclohexyl substituent to the increased anti-ChE
potency of OPs.
pyridinium iodide (DEP-MHP) were prepared by reacting
the tertiary amines with O,O-diethylphosphoryl chloride
followed by quaternization with methyl iodide. The com-
pletion of all reactions and the purity of the new compounds
were established by TLC and 31P, 1H, and 13C NMR
spectroscopy. MEPQ and DEPQ were prepared as
described earlier ([5] and [16], respectively), and echothio-
phate was obtained from Ayerst Laboratories.
Purified rMoAChE was provided by Professor P. Taylor
´
and Dr. Z. Radic. FBS AChE, HuBChE, and EqBChE were
purified by procainamide-Sepharose 4B gel-affinity chro-
matography [17,18].
2.2. Enzyme assays
The activities of ChEs were determined by the method of
Ellman et al. [19] using 1 mM acetylthiocholine iodide and
butyrylthiocholine iodide as substrates for AChE and
BChE, respectively. Assays were carried out in 50 mM
phosphate buffer, pH 8.0, at 258.
2.3. Determination of kd and ki
Fresh aqueous stock solutions of POXs (approximately
10 mM) adjusted to pH 4.5 by adding a few drops of 0.1 M
acetate buffer, were kept in an ice-water bath. The con-
centrations of stock solutions were confirmed further by
titration of a known concentration of HuBChE pre-deter-
mined by use of DEPQ [16]. The decomposition rate
constants (kd) for various POXs, and the rate constants
for the inhibition of AChEs and BChEs by POXs (ki) were
determined essentially as described [15], using 10 mM
HEPES buffer, pH 7.8, at 298.
2.3.1. Decomposition of POXs
The decomposition of 2- and 4-POXs were determined
by monitoring decreases in absorbance at 278 and 254 nm,
respectively. The decomposition of the 3-POX series was
based on a decrease in absorbance at 273 and an increase at
267 nm. Rate constants were calculated using the follow-
ing equation:
2. Materials and methods
2.1. Materials
ODt À OD1 ¼ ðOD0 À OD1Þ eÀk t
(1)
d
AlkMeP-POXs and EDMP-POXs were prepared using
procedures described earlier for the synthesis of DEP-POXs
[15]. Briefly, tertiary POXs of 2-, 3- or 4-pyridine carbox-
aldehyde oxime were obtained by reacting the relatively
non-toxic O-alkyl methylphosphonochloridates or O-ethyl
N,N-dimethylamidophosphorochloridate with the corres-
ponding pyridine carboxaldehyde oxime (Aldrich Chemical
Co). The tertiary POXs were purified by column chromato-
graphy and converted to quaternary compounds in the
presence of excess methyl iodide or dimethyl sulfate in
CH3CN at room temperature. DEP-3PAM was prepared
using the procedure previously described for DEP-4PAM
[15]. The DEP esters of the cations m-(N,N,N-trimethylam-
monio) phenol iodide (DEP-TMP) and 1-methyl 3-hydroxy
where, OD0 and ODt are the absorbances at time zero and
at time t, respectively. OD is the value assigned when no
1
changes in absorbance were observed. In all cases, the
decomposition products were identified by UV spectra.
Results were compared to standard solutions of authentic
2-, 3-, and 4-CNPY prepared as previously described
[15]. In several cases, these kd values were validated by
determining the rates at which the POXs lost their anti-ChE
activity.
2.3.2. Inhibition of ChEs
At t ¼ 0, an aliquot of a stock solution of POX was
diluted into 10 mM HEPES buffer, pH 7.8, containing

Y. Ashani et al. / Biochemical Pharmacology 66 (2003) 191–202
195
ChE and 0.1% BSA, and pre-incubated at 298. The final
concentrations of the inhibitors were sufficiently high to
establish pseudo-first-order reaction conditions. For 3- and
4-POXs, curve-fitting of the data points was carried out by
nonlinear regression analysis using the mono-exponential
decay equation:
and anti pyridine aldoximes and their quaternary salts,2
it is proposed that the OP moieties of POXs listed in
Table 1, are syn to the carboxaldehyde hydrogen,
PÀOÀN=C(H)ÀCpyr. This assignment implies that the
syn geometry of the starting materials (i.e. the tertiary
oximes) was conserved. As was demonstrated previously
with DEP-POXs [15], the anti-parallel orientation of the
bulky OP and pyridine moieties seems to be the preferred
geometry in all POXs and is consistent with the X-ray
crystal structure of IMP-4PAM [20].
obst
Et ¼ E0 eÀk
(2)
where, Et and E0 are enzyme activities at time t and zero,
respectively. ki was calculated from the slope of the straight
line obtained by the plot of kobs vs. [POX]. Due to the
rapid decomposition of all 2-POXs, ki was determined as
previously described for DEP-2PAM [15]. Briefly, as time
increases, the ratio of the residual enzyme to its initial
activity approaches a limiting value as shown in Eq. (3):
3.2. Decomposition of POXs
Based on spectral changes in aqueous solutions, and
a comparison of the UV spectra of the decomposition
products with authentic samples, the major decomposition
products were identified as the corresponding CNPY
cations (Fig. 2b). This observation is consistent with pre-
vious reports ([15], and references cited therein).
ꢀ
ꢁ
ꢀ ꢁ
E1
E0
ki
ln
¼ À½2-POXŠ
(3)
0
kd
where E is the enzyme activity that does not change over
The t1/2 values of AlkMeP-2PAMs and EDMP-2PAM
were 0.7 to 1.5 min (Table 1), while those for the 4-POXs
series were 57- to 126-fold higher, approaching a value of
112 min for PMP-4PAM. When the OP-oxime moiety was
at the meta-position of the pyridine ring (3-POXs), a
further increase in stability was observed (t1=2 > 580 min).
A similar dependency on the position of the OP-oxime
moiety was observed for the decomposition rates for DEP-
POXs. However, DEP-POXs decomposed at rates that were
4- to 8-fold higher than those of AlkMeP-POX and EDMP-
POXs. In the 2- and 4-POXs, the change from an ethyl
(EMP-POXs) to a pinacolyl group (PMP-POXs) resulted in
a ꢁ2-fold increase in stability, which can be attributed to a
combined effect of steric factors and an increased electron
donation by the pinacolyl group.
1
15 min, and E0 is the activity at time zero. [2-POX]0 is the
initial concentration of the tested POX. ki was calculated
from Eq. (3) using the value of kd obtained as described
above.
2.4. Hydrolysis of DEPQ
The bimolecular rate constant for the hydroxide-induced
hydrolysis of DEPQ was determined by monitoring the
increase in absorbance of the leaving group 1-methyl
7-hydroxyquinolinium at 406 nm, in the presence of
0.01 to 0.1 NaOH.
2.5. Quantum chemical calculations
The geometry and electronic features of the POXs were
calculated using quantum chemical AM1 and ab initio
methods. Several low energy conformers with varying
population densities were generated using the systematic
search technique via AM1 method as implemented
in SPARTAN version 5.1.2 (Wavefunction, Inc.). The
geometry of the minimum energy and most abundant
conformer was considered for further optimization and
calculation of electronic properties. Geometry optimiza-
tion and energetics calculations were performed using the
Gaussian 94 version (Revision A.1).
3.3. Inhibition of ChEs
AlkMeP-POXs and EDMP-POXs are racemic mixtures
of 2 or 4 enantiomers. Since replacement of thiocholine
with the pyridine oxime function does not change the
definition of absolute configuration, it is assumed that
the enantiomers with the Sp assigned configuration [7]
are also the active isomers of POXs. The enantioselectivity
ratio Sp/Rp for AChEs could be appreciated only with the
relatively stable 4-POX series. Based on the titration of
pre-determined concentrations of rMoAChE, a significant
enantioselectivity was displayed by all 4-POXs with
the exception of EMP-4PAM, which showed a reduced
enantioselectivity (not shown).3 This observation is
consistent with the behavior of homologous O-ethyl
methylphosphonates with different leaving groups, such
3. Results
3.1. Structures of POXs
2 For literature citations see [15]. The syn configuration of the active
geometrical form of 2-, 3- and 4-PAM are depicted in Fig. 1.
3 A 1:2 molar ratio of AChE to racemic POX was required to inhibit
100% of enzyme activity; less than 2 mol were required to produce
complete inhibition with EMP-4PAM.
All NMR signals were in agreement with the proposed
structures and suggest that only one kind of geometrical
isomer was isolated. On the basis of the published NMR
chemical shifts and coupling constants of well-defined syn

196
Y. Ashani et al. / Biochemical Pharmacology 66 (2003) 191–202
as MEPQ and VX [5,21]. As rationalized before for
O-alkyl methylphsphonothiolates [7], the reduced Sp/Rp
ratio of EMP-4PAM can be attributed to a partial accom-
modation of the relatively small ethoxy group of the less
potent stereoisomer in the acyl pocket of AChE.
cyclohexyl ring was associated with an increase in ki for
the inhibition of ChEs. An analogous increase in ki was
observed in a series of O-alkyl methylphosphonofluori-
dates such as sarin, soman, and cyclosarin [22,24]. Intro-
ducing a methyl group on the cyclohexyl ring (MCMP-
POXs) resulted in a slight to moderate decrease in ki that is
attributable to steric constraints. The small but significant
differences in ki values for rMoAChE and FBS AChE,
point to subtle structural differences in the catalytic gorge
of these two enzymes.
Inhibition ranking order of POXs for BChEs differed
significantly from that observed with AChEs. For example,
all three DEP-POXs were found to be more potent inhi-
bitors of HuBChE compared to AChEs, suggesting a better
accommodation of DEP-POXs in the enlarged gorge [25].
The significantly slower rate of inhibition of EqBChE by
POXs compared to HuBChE can be attributed to structural
variations between the two BChEs.
Replacement of one of the ethoxy groups in DEP-POXs
with the bulkier dimethylamino residue provides the ex-
pected tabun-type derivatives, EDMP-2PAM and EDMP-
4PAM. These compounds are poor inhibitors of both AChEs
and BChEs, an observation that is attributed to a combina-
tion of steric factors and a reduction in the electro-positivity
of the phosphorus atom. Interestingly, the ki of EDMP-
2PAM was 3-fold lower for HuBChE compared to AChE.
The bimolecular rate constants for the inhibition of
rMoAChE, FBS AChE, HuBChE, and EqBChE are sum-
marized in Table 1. For rMoAChE and FBS AChE, all
AlkMeP-2PAMs were 7- to 100-fold more potent than
the parent OPs [22–24]. For example, replacement of F
or SCH2CH2N(iPr)2 in sarin and VX with 2-PAM,
increased the ki for rMoAChE from 0.014 and 0:12 Â
109 MÀ1 minÀ1 to 1:42 Â 109 MÀ1 minÀ1 (IMP-2PAM)
and 1:94 Â 109 MÀ1 minÀ1 (EMP-2PAM), respectively.
Introducing 4-PAM as the leaving group also resulted in
an increase in ki relative to the parent compounds, however,
to a lesser extent. AlkMeP-3PAMs were the least potent
inhibitors of AChEs. In fact, the ki values of EMP-3PAM
and CMP-3PAM were reduced 3- to 5-fold relative to those
of VX and cyclosarin, respectively.
Notably, increasing the size of the O-alkyl substituent of
AlkMeP-POXs has only a small effect on the ki for either
enzyme, with the exception of the O-cyclohexyl ring. Of all
the POXs tested, the maximal rate of inhibition was noted
for CMP-2PAM (4:02 Â 109 MÀ1 minÀ1). Regardless of
the oxime moiety, within a given POX series, the rigid
Table 2
Semi-empirical calculations on POXs
POX
R
R0
Molecular electrostatic potentials (kcal/mol)
Dihedral angle (8)
O(1)
P(2)
H(6)
4–5–7–8
1–2–3–4
1–2–OÀR
2-POXs
EMP-2PAM
C2H5
CH3
CH3
CH3
CH3
CH3
C2H5
À55.5
À57.0
À58.2
À59.7
À58.1
À59.6
142.3
145.0
134.9
140.0
135.2
146.9
94.2
96.5
88.9
92.0
89.3
96.6
133
À135
134
40
À41
40
29
1.7
IMP-2PAM
PMP-2PAM
CMP-2PAM
MCMP-2PAM
DEP-2PAM
CH(CH3)2
CH(CH3)C(CH3)3
Cyclohexyl
2-Methylcyclohexyl
C2H5
20
À132
À146
À132
À43
À44
59
À4.5
À22
20
3-POXs
EMP-3PAM
CMP-3PAM
DEP-3PAM
C2H5
CH3
CH3
C2H5
À65.3
À65.7
À60.7
137.8
134.0
128.8
89.2
86.5
82.7
À155
À157
À161
À39
À39
À49
3.9
2.2
Cyclohexyl
C2H5
À37
4-POXs
EMP-4PAM
IMP-4PAM
PMP-4PAM
CMP-4PAM
MCMP-4PAM
DEP-4PAM
C2H5
CH3
CH3
CH3
CH3
CH3
C2H5
À68.0
À68.3
À68.3
À68.4
À67.8
À63.8
132.7
133.4
128.5
134.9
132.1
131.8
84.3
85.0
82.4
86.7
84.3
84.9
166
167
177
167
166
171
41
41
30
19
CH(CH3)2
CH(CH3)C(CH3)3
Cyclohexyl
2-Methylcyclohexyl
C2H5
À43
39
À26
21
37
22
53
39

Y. Ashani et al. / Biochemical Pharmacology 66 (2003) 191–202
197
Fig. 3. Space-filled models for the 6-31Gꢀꢀ optimized geometries of EMP-POXs. Color codes: Yellow, phosphorus; Red, oxygen; Blue, nitrogen; White,
carbon; Green, hydrogen.
3.4. Molecular electrostatic features and low
energy conformers
The low-energy-high-abundance conformers of the
2- and 4-POXs reveal some differences in the structure
of the optimized geometry (Fig. 4, 2-POXs). CMP-, and
MCMP-2PAM displayed similar O(1)ÀP(2)ÀO(3)ÀN(4)
(À43 and À448) dihedral angles, however, they differ more
in the N(4)ÀC(5)ÀC(7)ÀC(8) (À132 and À1468) and
O(1)À2(P)ÀOÀR (À4.5 to À228) dihedral angles. Yet,
the ki values of CMP-2PAM and MCMP-2PAM were
about the same. On the other hand, despite differences in
the optimized structures of EMP-2PAM, IMP-2PAM,
and PMP-2-PAM, the ki values were similar. This lack of
correlation of the anti-ChE potency with a specific geo-
metry was further pronounced with the 4-POX series.
To substantiate some of the experimental conclusions,
the electronic and geometric features of POXs were cal-
culated using AM1 and ab initio quantum chemical meth-
ods using the 6-31Gꢀꢀ basis set. The optimized geometry of
IMP-4PAM was highly similar to its reported X-ray crystal
structure [20]. For example, in the crystal structure, the
OÀN=C(H)ÀCpyr plane was reported to be 14.28 to the
best plane through the pyridine ring, an angle that is
close to the calculated value of 138 (180ꢂ À 1678) for the
N(4)ÀC(5)ÀC(7)ÀC(8) dihedral angle (Table 2). Further,
the acidity rank order of 2-, 3-, and 4-PAM (pKa 8.0, 9.2, and
8.6, respectively; [26]) correlated well with the rank order of
the positive potential values of the acidic hydrogen of the
hydroximino function of 2-, 3-, and 4-PAM (103.5, 96.5, and
98.2 kcal/mol, respectively). These values are a measure of
the intrinsic acidity estimated by scanning the MEP sur-
faces. Thus, theexistingexperimentaldatavalidatethesteric
and electronic properties derived from these calculations.
The MEP values of O(1), P(2) and H(6) atoms for the
three POX series are summarized in Table 2. The most
positive electrostatic potential was found to be on the
phosphorus atom and the most negative potential was
associated with the P=O oxygen O(1). The 2-POX series
displayed the greatest positive potential of the 6(H) atom,
and thus, the strongest acidity. The acidity of the 6(H) atom
in 3- and 4-POXs was similar. The AM1 and the ab initio
calculations gave similar results. This discrepancy, can be
reconciled by taking into account the possible steric hin-
drance that can affect the interaction between the positive
charge of the pyridine ring and the P=O oxygen atom as
illustrated for EMP-POXs in Fig. 3.
4. Discussion
4.1. Decomposition of POXs
The salient characteristic of the stability study is the
correlation between the calculated relative acidity of the
6(H) atom, PÀOÀN=C(H)ÀCpyr of 2- and 4-POXs, and
kd. Thus, the more acidic the hydrogen, the less stable the
POX. The acidity is influenced by through-space repulsion
forces, which depend on the distance between the 6(H)
atom and the positive charge of the pyridine ring. The
optimized geometries shown in Fig. 3 indicate that the N-
methyl group in EMP-2PAM interferes with any possible
through-space interaction between the positive charge on
the pyridine nitrogen and the most negative potential of the
P=O oxygen. This enables the molecule to dissipate more
of its positive charge and makes the 6(H) atom more acidic
than in the other two POXs. In addition, resonance stabi-
lization of the charge is expected to increase the acidity of

198
Y. Ashani et al. / Biochemical Pharmacology 66 (2003) 191–202
Fig. 4. Molecular electrostatic potential maps of the 6-31Gꢀꢀ optimized geometries. Colors show the most positive potential (deepest blue) and the most
negative potential (deepest red) regions of the 2-POX series.
6(H) in 2- and 4-POXs, while no such mechanism is
possible in the 3-POXs (Fig. 2c). The average acidity of
6(H) in 2- and 4-POX series could be well correlated with
their rate of decomposition, namely, the higher the acidity
the less stable the POX. However, the 3-POXs that were
similar in acidity to that of the 4-POXs were found to be at
least 5-fold more stable than the 4-POXs. This discrepancy
can be explained by the observation that the greatest
interaction between the charged nitrogen and the negative
potential on the P=O oxygen atom occurs in the least
hindered EMP-3PAM (Fig. 3). This interaction is expected
to decrease the attraction forces between the charged
nitrogen and the base acceptor, which may slow down
proton abstraction (Fig. 2b) and make the 3-POX series the
most stable.
In each series, the decreased stability of DEP-POXs
compared to the AlkMeP-POXs and EDMP-POXs, can be
attributed to the pKa of R0P(O)(OR)OH (Fig. 2b). Thus,
O,O-diethylphosphoric acid (R0 ¼ OC2H5; R ¼ C2H5;
pKa ¼ 1:04) has a 10-fold higher acidity than the corre-
sponding O-alkyl methylphosphonic acids (R0 ¼ CH3;
R ¼ alkyl) [27]. It appears that the greater resonance
stabilization of the negative charge on the conjugate base
(C2H5O)2P(O)OÀ compared to CH3P(O)(OR)OÀ, is res-
ponsible for the increased acidity, and is likely to accel-
erate decomposition of DEP-POXs. Also, replacement of
CH3ÀP with (CH3)2NÀP is not expected to change the
poor pp–dp overlap between phosphorus and these sub-
stituents [28], and therefore, the acidities of CH3P(O)-
(OC2H5)OH, and (CH3)2NP(O)(OC2H5)OH are likely to
be similar. Direct pKa measurements, however, will be
required to validate this suggestion.
In summary, it is likely that AlkMeP-POXs decompose
via b-elimination reactions with rates that are strongly
influenced by the structure of the PAM moiety and less by
the nature of the O-alkyl susbstituent.
4.2. Inhibition of ChEs with POXs
Once the POXs are transferred from the aqueous solu-
tion to ChEs, the reaction center shifts from the 6(H) atom
to the phosphorus atom and produces a substantial increase
in the SN2(P) reaction. This is supported by the facts that
DEP–ChE conjugates were obtained [15], and the ki values
for the inhibition of the four ChEs by AlkMeP-POXs and
DEP-POXs increased with increasing acidity of the leaving
groups.4
With the exception of the cyclohexyl derivatives,
increasing the size of the O-alkyl substituent in
AlkMeP-POXs had in general a small effect on the ki
for ChEs. These results differ from those reported for
cationic aliphatic OPs [7], where 20- to 50-fold increases
in ki values for AChE and BChE were reported, respec-
tively, when the O-isopropyl group was replaced with the
O-3,3-dimethylbutyl group. The diminished dependency
of ki on the size of the O-alkyl substituent suggests that the
4 pKa values (2-PAM, 8.0; 4-PAM, 8.6; 3-PAM, 9.2) were reported by
Ginsburg and Wilson [26] for a series of biologically reactive isomers that
were later assigned the syn configuration ([29,30]; Fig. 1).

Y. Ashani et al. / Biochemical Pharmacology 66 (2003) 191–202
199
increased reactivity of POXs is mainly due to the interac-
tion between the pyridinium ring and conserved residues
such as D74 and W86 in AChE (D70 and W82 in HuBChE)
[31]. Thus, it can be generalized that the cationic aromatic
moieties that serve as leaving groups, contribute sufficient
stabilization energy for the productive alignment of the
inhibitor in the catalytic gorge.
DEP-containing inhibitors echothiophate, IMP-thiocho-
line, and its O-3,3-dimethylbutyl homologue [7] were
similar for rMoAChE and HuBChE highlight the
unique contribution of the cationic aromatic moiety to
the enhanced inhibition of ChEs. The increase in inhibition
potency of DEP-POXs toward HuBChE seems to be
controlled by two factors: (1) the enlargement of the
acyl pocket and choline binding site in HuBChE can
accommodate DEPs and other bulky O-alkyl substituents
[7,32,33] more comfortably than AChE; and (2) the
enlarged gorge seems to provide more favorable intera-
ctions for charged aromatic DEPs (e.g. DEPQ, DEP-2,
3-, and 4-PAM), with functionally conserved residues such
as D70 and W82 in HuBChE.
Remarkable differences in the reactivity of AlkMeP-
POXs and DEP-POXs were observed between AChEs and
BChEs. Up to a 6-fold increase in the ki value for the
inhibition of rMoAChE compared to HuBChE was
recorded with AlkMeP-2PAMs. However, the ki values
of AlkMeP-4PAM vary only slightly between the two
enzymes, while EMP-3PAM and CMP-3PAM inhibited
HuBChE 2- and 12-fold faster than rMoAChE. It seems
that the location of the charged nitrogen in the pyridine ring
determines the relative potency of POXs due to the dif-
ferent projections of the leaving groups in the gorge. One
may argue that for the 2-PAM series, the enlargement of the
acyl pocket, and near-by sites in HuBChE permits several
low-level POX-enzyme adducts that eventually will slow
down inhibition because of non-productive binding.
Removing the hydroximinomethyl function away from
the charged nitrogen apparently favors an orientation that
increases the probability of expulsion of the leaving group
in HuBChE and consequently the rate of inhibition. The
latter is maximized with the AlkMeP-3PAMs, a finding
that is consistent with the pyridinium oxime moiety being
in a different orientation in its optimized structure com-
pared to AlkMeP-2PAMs and AlkMeP-4PAMs (Fig. 3).
While AlkMeP-2PAMs were more potent inhibitors
of AChEs compared to BChEs, all DEP-POXs-inhibited
HuBChE 10- to 412-fold faster than the two AChEs. The
fact that ki values of cationic aliphatic OPs such as the
4.3. Analysis of linear-free energy relationships
The analysis of linear-free energy relationships was
applied as a diagnostic tool for the further interpretation
of the results of the inhibition studies.
4.3.1. Anti-ChE activity of cyclohexyl-containing OPs
The relationship between log ki and the acidity of 2-,
3-, and 4-PAM was used to underscore the increased rate
of inhibition by cyclohexyl-containing AlkMeP-POXs
(Fig. 5). Despite the limited number of data points
¨
(N ¼ 3 for each ChE) the use of Bronsted plots for the
qualitative comparison of CMP-POXs with EMP- and
DEP-POXs is valid because essentially similar shapes were
obtained for all four ChEs. The lines in Fig. 5 clearly show
that of all the POXs examined, the most potent CMP-POX
series displayed the smallest dependency on the acidity of
the leaving group. The decreased dependency of log ki on
pKa can be considered as an indication of a change in the
¨
Fig. 5. Bronsted plots (log ki ¼ bpKa þ C) for the inhibition of ChEs by POXs. Panel A, CMP-POX; panel B, EMP-POX; and panel C, DEP-POXs (the leaving
groups for each panel are 2-, 3-, and 4-PAM; pKa taken from [26]. Symbols represent HuBChE (*), FBS AChE (&), EqBChE (~), and rMoAChE (*).

200
Y. Ashani et al. / Biochemical Pharmacology 66 (2003) 191–202
rate-determining step. A possible kinetic scheme that could
¨
account for the visual change in the Bronsted slopes is
shown in Eq. (4). It involves the rapid formation of a
reversible enzyme-POX complex (KI) that follows a slow
conformational change (kc) prior to the unimolecular
phosphonylation step (kp). That OPs can induce consider-
able reversible conformational changes in AChE has been
recently confirmed by X-ray crystallography [1,2]:
KI
EOH þ POX@EÀOH Á POX
kc
kp
@EÀOHꢀPOX!EÀOÀP þ HX
(4)
k
Àc
The second order rate constants for the inhibition of
enzyme (EOH) activity were approximated as follows [34]:
kc
kp > kÀc
kp < kÀc
;
ki ¼
ki ¼
(5)
(6)
KI
kckp
;
KIkÀc
We assume that constants that reflect binding of POXs to
AChE (i.e. KI, kc, kÀc) are relatively independent of the pKa
of the leaving group, while kp must be highly dependent on
it. The dependency of ki on kp will only be evident when
kp < kÀc (Eq. (6)) which is probably the case with EMP-,
and DEP-POXs. For the CMP-POX series it appears that
kp > kÀc, and therefore, ki is significantly less dependent
on the acidity of the leaving group (Eq. (5)). The greater
activity of CMP-POXs is attributed to the cyclohexyl ring
that is part of a pharmacophore that offers simultaneous
multiple hydrophobic interactions with residues that are
involved in the accommodation of OP ligands in the active-
site gorge. The prevalence of kp > kÀc for CMP-POXs is
reasonably explained on the basis of the tight binding to
AChE that slows down kÀc. Of particular interest is the 72-
fold increase in ki of CMP-3PAM against rMoAChE
relative to EMP-3PAM. Here, the contribution of the cyclic
ring is greatly pronounced and appears to suppress unfa-
vorable factors such as the reduced acidity of the leaving
group, 3-PAM.
¨
Fig. 6. Bronsted plot for the inhibition of rMoAChE by O,O-diethylpho-
sphates. The numbers refer to the following DEPs: 1, DEP-MHP; 2,
DEPQ; 3, DEP-2PAM; 4, DEP-4PAM; 5, DEP-3PAM; 6, echothiophate; 7,
DEP-TMP. Symbols represent cationic aromatic DEPs (&) and charged
aliphatic DEPs (&). For structural details see Fig. 1.
DEP-2 and DEP-4PAM), is not likely to differ sterically,
and yet, the ki values of DEP-MHP (pKa 4.96; [35]) and
DEPQ (that contains a different leaving group) leveled off
¨
in the Bronsted plot (Fig. 6). This break in the curve is
indicative of a change in the rate-limiting step, the mechan-
ism of which is most likely similar to that proposed for the
scheme in Eq. (4). A sharp break at ꢁpKa 8.0 recently was
reported for phosphatase-induced hydrolysis of phenyl-
containing substrates with relatively small structural
alterations in the leaving group [36]. These observations,
together with a slope >1 for the descending portion of the
line, seem to indicate that bond breaking is advanced in the
transition state [37,38]. It is possible that within a certain
pKa range, presumably below 8, the electron withdrawal
capacity of the leaving group is an important driving force
for the inhibition reaction by DEP-POXs.
Fig. 6 also shows that the ki values of echothiophate and
DEP-TMP for rMoAChE are 45- and 520-fold smaller,
respectively, than would be predicted from their pKa
values. It is proposed that the greater anti-ChE activity
of cationic aromatic DEPs is due to a combination of
increased acidity of the leaving group and better compli-
mentarity with the catalytic gorge that arises from the
coupling of a positive charge to an aromatic system. The
continuity of positive electrostatic potential on a planar
pyridine ring with a large surface area is expected to
provide more efficient contact with the p face of several
aromatic planes in the gorge [39–46]. The importance of
cationic aromatic systems in methylphosphonates is also
evident from the fact that the ki values of IMP-2PAM and
4.3.2. Reactivity of cationic aromatic vs. charged
aliphatic OPs
Sharp curvatures in the plots of log ki of enzymatic
reactions vs. the pKa of structurally unrelated leaving
groups previously were observed around pKa values of
7–8, for the inhibition of AChE by non-charged DEPs
[4,34]. The hydroxide-catalyzed hydrolysis of DEPs
including DEPQ,5 and of other series of OPs, displayed
linear log kOH vs. pKa plots over a wide range of pKa values
of leaving groups [34]. Therefore, the curvature observed
with rMoAChE is peculiar to the enzymatic inhibition
reaction. The nucleophilic attack by AChE at the phos-
phorous atom of DEP-MHP and DEP-3PAM (and probably
5 Present study, data not shown.

Y. Ashani et al. / Biochemical Pharmacology 66 (2003) 191–202
201
[4] Kitz RJ, Ginsburg S, Wilson IB. The reaction of acetylcholinesterase
with diethylphosphoryl esters of quaternary and tertiary aminophe-
nols. Mol Pharmacol 1967;3:225–32.
IMP-4PAM are at least 89- and 7-fold greater than those
reported for the inhibition of rMoAChE by the aliphatic
inhibitor IMP-thiocholine.6 This rationalization for the
increased activity of charged aromatic DEPs appears to
also hold for AlkMeP-POXs and MEPQ.
[5] Levy D, Ashani Y. Synthesis and in vitro properties of a powerful
quaternary methylphosphonate inhibitor of acetylcholinesterase. A
new marker in blood–brain barrier research. Biochem Pharmacol
1986;35:1079–85.
[6] Maglothin JA, Wilson IB. Equilibrium constants for the phosphoryla-
tion of acetylcholinesterase by some diethyl phosphorothiolates and
phosphates. Biochemistry 1974;13:3520–7.
5. Conclusion
[7] Hosea NA, Berman HA, Taylor PM. Specificity and orientation of
trigonal carboxyl esters and tetrahedral alkylphosphonyl esters in
cholinesterases. Biochemistry 1995;34:11528–36.
The experimental and theoretical study of POXs reveals
distinct types of OP inhibitors of ChEs, and underscores the
unique contribution of the charged pyridine moiety and the
cyclohexyl group to the inhibition of ChEs. The orientation
of the large aromatic cation that serves as the leaving
group seems to be a result of tight interactions with con-
served residues in AChE and BChE that also accelerate the
inhibition reaction. A second generalization is that oxime-
induced reactivation of AChE inhibited with O,O-diethyl-
based phosphates (e.g. commercial pesticides), should be
less complicated than that inhibited with nerve-agents. This
is due to the rapid decomposition of DEP-POXs compared
to AlkMeP-POXs and EDMP-POXs.
Finally, it was reported that soman- and cyclosarin-
inhibited AChEs were reactivated more readily with 2-
¨
PAM derived oxime antidotes (HI-6, HS-6 and HLO7) than
with the 4-PAM-based analogs (TMB4 and obidoxime)
[24,47]. Results of this study suggest that this could be due
in part, to the formation of stable and potent PMP- and
CMP-4PAM based POXs with ki values for ChEs that are
[8] Radic Z, Quinn DM, Vellom DC, Camp S, Taylor P. Allosteric control
of acetylcholinesterase catalysis by fasciculin.
J Biol Chem
1995;270:20391–9.
[9] Heilbronn E. Hydrolysis of carboxylic acid esters of thiocholine and
its analogues. 2. alkaline hydrolysis. Acta Chem Scand 1958;12:1492–
506.
[10] HackleyBE,OwensOO.PreparationofO-(isopropylmethylphosphono)-
4-formyl-1-methylpyridinium iodide oxime. J Org Chem 1959;21:1120.
[11] De Jong LPA, Ceulen DI. Anticholinesterase activity and rate of
decomposition of some phosphylated oximes. Biochem Pharmacol
1978;27:857–63.
[12] Harvey B, Scott RP, Sellers DJ, Watts P. In vitro studies on the
reactivation by oximes of phosphylated acetylcholinesterase-II. On
the formation of O,O-diethyl phosphorylated AChE and O-ethyl
methylphosphonylated AChE and their reactivation by PS2. Biochem
Pharmacol 1986;35:745–51.
[13] Lamb JC, Steinberg GM, Solomon S, Hackley BE. Isopropyl methyl-
phosphonylated bisquaternary oximes; powerful inhibitors of choli-
nesterases. Biochim Biophys Acta 1964;89:174–6.
[14] Nenner M. Phosphonylated aldoximes. Antagonism of acetylcholi-
nesterase and hydrolytic decomposition. Biochem Pharmacol
1974;23:1255–62.
likely to exceed 1 Â 1010
M
À1 minÀ1 [3]. Thus, the proper-
ties of POXs should be taken into consideration in the
evaluation of kinetic studies on the reactivation and aging
of OP-ChE conjugates.
[15] Leader H, Vincze A, Manisterski B, Rothchild N, Dosoretz C, Ashani
Y. Characterization of O,O-diethylphosphoryl oximes as inhibitors of
cholinesterases and substrates of phosphotriesterases. Biochem Phar-
macol 1999;58:503–15.
[16] Gordon MA, Carpenter DE, Barret HW, Wilson IB. Determination of
the normality of cholinesterase solutions. Anal Biochem 1978;85:
519–27.
[17] De La Hoz D, Doctor BP, Ralston JC, Rush RS, Wolfe AD. A
simplified procedure for the purification of large quantities of fetal
bovine serum acetylcholinesterase. Life Sci 1986;39:195–9.
[18] Grunwald J, Marcus D, Papier Y, Raveh L, Pittel Z, Ashani Y. Large-
scale purification and long-term stability of human butyrylcholines-
terase: a potential bioscavenger drug. J Biochem Biophys Methods
1997;34:123–35.
Acknowledgments
The authors wish to thank SGT Connie Hinrichs for the
excellent technical assistance.
References
[19] Ellman GL, Courtney KD, Andres V, Featherstone RM. A new and
rapid colorimetric determination of acetylcholinesterase activity.
Biochem Pharmacol 1961;7:88–95.
[1] Millard CB, Koellner G, Ordentlich A, Shafferman A, Silman I,
Sussman JL. Reaction products of acetylcholinesterase and VX reveal
a mobile histidine in the catalytic triad. J Am Chem Soc 1999;121:
9883–4.
[20] Portmann R, Neiederhauser A, Hofman W, Frey A. Synthesis of 4-
({[isopropyloxy]methylphosphonyloxy}iminomethyl)-1-methylpyridi-
nium iodide and its characterization. Helv Chim Acta 1991;74:331–5.
[21] HallCR, InchTD, InnsRH, MuirAW, Sellers DJ, Smith AP. Differences
between some biological properties of enantiomers of alkyl S-alkyl
methylphosphonothioates. J Pharm Pharmacol 1977;29:574–6.
[22] Gray PJ, Dawson RM. Kinetic constants for the inhibition of eel and
rabbit brain acetylcholinesterase by some organophosphates and
carbamates of military significance. Toxicol Appl Pharmacol
1987;91:140–4.
[2] Millard CB, Kryger G, Ordentlich A, Greenblat HM, Harel M, Raves
ML, Segall Y, Barak D, Shafferman A, Silman I, Sussman JL. Crystal
structures of aged phosphonylated acetylcholinesterase: nerve agent
reaction products at the atomic level. Biochemistry 1999;38:7032–9.
[3] Luo C, Saxena A, Smith M, Garcia G, Radic Z, Taylor P, Doctor BP.
Phosphoryl oxime inhibition of acetylcholinesterase during oxime
reactivation is prevented by edrophonium. Biochemistry 1999;38:
9937–47.
[23] Raveh L, Grunwald J, Marcus D, Papier Y, Cohen E, Ashani Y. Human
butyrylcholinesterase as a general prophylactic antidote for nerve
agent toxicity. In vitro and in vivo quantitative characterization.
Biochem Pharmacol 1993;22:2465–74.
6 Values are for the Sp enantiomer of IMP-thiocholine [7], and should be
divided by 2 to normalize the ki to that of the racemic mixtures used in this
work.

202
Y. Ashani et al. / Biochemical Pharmacology 66 (2003) 191–202
[24] Worek F, Eyer P, Szinicz L. Inhibition reactivation and aging kinetics
of cyclohexylmethylphosphonofluoridate-inhibited human cholines-
terases. Arch Toxicol 1998;72:580–7.
[37] Kirby AJ, Varvoglis AG. The reactivity of phosphate esters. Monoester
hydrolysis. J Am Chem Soc 1967;89:415–23.
[38] Andres GO, Granados AM, de Rossi RH. Kinetic study of the
hydrolysis of phthalic anhydride and aryl hydrogen phthalates. J
Chem Soc [Perkin 1] 2001;66:7653–7.
[25] Saxena A, Redman AMG, Jiang X, Lockridge O, Doctor BP. Differ-
ences in active-site gorge dimensions of cholinesterases revealed by
binding of inhibitors to human butyrylcholinesterase. Biochemistry
1997;36:14642–51.
[39] Ma JC, Dougherty DA. The cation–p interaction. Chem Rev
1997;97:1303–24.
[26] Ginsburg S, Wilson IB. Oximes of the pyridine series. J Am Chem Soc
1957;79:481–5.
[40] Zoltewicz JA, Maier NM. p-Cation and p-dipole-stabilizing interac-
tions in a simple model system with cofacial aromatic rings. J Org
Chem 1998;63:4985–90.
[27] Van Hooidonk C, Kraaij GW, Ginjaar L. On the reactivity of orga-
nophosphorus compounds. Part IV: The alkaline hydrolysis of some
O-phosphylated 2-pyridine oximes. Rec Trav Chem 1968;87:673–86.
[28] Kirby AJ, Warren SG. The organic chemistry of phosphorus.
Amsterdam, London, New York: Elsevier; 1967.
[41] Sussman JL, Harel M, Frolow F, Oefner C, Goldman A, Toker L,
Silman I. Atomic structure of acetylcholinesterase from Torpedo
californica:
a prototypic acetylcholine-binding protein. Science
1991;253:872–9.
[29] Poziomek EJ, Kramer DN, Mosher WA, Michel HO. Configurational
analysis of 4-formyl-1-methylpyridinium iodide oxime and its rela-
tionship to a molecular complementarity theory on the reactivation of
inhibited acetylcholinesterase. J Am Chem Soc 1961;83:3916–7.
[30] Carlstrom D. A crystallographic study of N-methyl-pyridine-2-aldox-
ime (2-PAM) halides. Acta Chem Scand 1966;20:1240–6.
[31] Harel M, Sussman JL, Krecji E, Bon S, Chanal P, Massoulie J, Silman
I. Conversion of acetylcholinesterase to butyrylcholinesterase: mod-
eling and mutagenesis. Proc Natl Acad Sci USA 1992;89:10827–31.
[32] Vellom DC, Radic Z, Li Y, Pickering NA, Camp S, Taylor P. Amino
acid residues controlling acetylcholinesterase and butyrylcholinester-
ase specificity. Biochemistry 1993;32:12–7.
[42] Harel M, Schalk I, Ehert-Sabatier L, Bout F, Goeldner M, Hirth C,
Axelson PH, Silman I, Sussman JL. Quaternary ligand binding to
aromatic residues in the active-site gorge of acetylcholinesterase. Proc
Natl Acad Sci USA 1993;90:9031–5.
[43] Radic Z, Pickering NA, Vellom DC, Camp S, Taylor P. Three distinct
domains in the cholinesterase molecule confer selectivity for acetyl-
and butyrylcholinesterase inhibitors. Biochemistry 1993;32:12074–
84.
[44] Ordentlich A, Barak D, Kronman C, Flashner Y, Leitner M, Segall Y,
Ariel N, Cohen S, Velan B, Shafferman A. Dissection of the human
acetylcholinesterase active center determinants of substrate specifi-
city. Identification of residues constituting the anionic site, the hydro-
phobic site. and the acyl pocket. J Biol Chem 1993;268:17083–95.
[45] Quinn DM, Feaster SR, Nair HK, Baker NA, Radic Z, Taylor P.
Delineation and decomposition of energies involved in quaternary
ammonium binding in the active site of acetylcholinesterase. J Am
Chem Soc 2000;122:2975–80.
[33] Ordentlich A, Barak D, Kromman C, Ariel N, Segall Y, Velan B,
Shafferman A. The architecture of human acetylcholinesterase active
center probed by interactions with selected organophosphate inhibi-
tors. J Biol Chem 1996;271:11953–62.
[34] Ashani Y, Snyder SL, Wilson IB. Linear free energy relationships in
the hydrolysis of some inhibitors of acetylcholinesterase. J Med Chem
1973;16:446–50.
[46] Hosea NA, Radic Z, Tsigelny I, Berman HA, Quinn DM, Taylor PM.
Aspartate 74 as a primary determinant in acetylcholinesterase govern-
ing specificity to cationic organophosphonates. Biochemistry
1996;35:10995–1004.
[35] Albert A, Phillips JN. Ionization constants of heterocyclic substances.
Part II. Hydroxy-derivatives of nitrogenous six-membered ring com-
pounds. J Chem Soc [Perkin 1] 1956;Part II:1294–304.
[47] De Jong LPA, Wolring GZ. Stereospecific reactivation by some
Hagedorn-oximes of acetylcholinesterases from various species
including man, inhibited by soman. Biochem Pharmacol 1984;33:
1119–25.
[36] McCain DF, Cartina IE, Hengge AC, Zhang Z-Y. The catalytic
mechanism of Cdc25A phosphatase. J Biol Chem 2002;277:11190–
200.