Inhibition of acetylcholinesterase by physostigmine analogs: Conformational mobility of cysteine loop due to the steric effect of the alkyl chain

Article abstract of DOI:10.1002/jbt.10026

The effect of a series of physostigmine analogs on acetylcholinesterase activity was investigated. The second-order rate constant k_on of the enzyme-inhibitor complex correlates with the conformational positioning of aromatic residues, especially Trp84, in the transition state complex. The van der Waals interactions are an important structural element of this conformational change. A transient mobility of the cysteine loop (Cys67-Cys94) was confined only to the presence of a significant steric effect. Even with this limitation, however, the steric effect seems to be an appropriate model for future tests on the back door hypothesis involving facilitated opening for faster product clearance.

Full text of DOI:10.1002/jbt.10026

J BIOCHEM MOLECULAR TOXICOLOGY
Volume 16, Number 2, 2002
Inhibition of Acetylcholinesterase by Physostigmine
Analogs: Conformational Mobility of Cysteine Loop
Due to the Steric Effect of the Alkyl Chain
Enrico Gavuzzo1 and Massimo Pomponi2
1Ist. di Strutturistica Chimica, CNR Montelibretti
2Ist. di Chimica, Facolta` di Medicina, UCSC, Rome, Italy; E-mail: m.pomponi@rm.unicatt.it
Received 31 July 2001; revised 27 February 2002; accepted 12 March 2002
ABSTRACT: The effect of a series of physostigmine
analogs on acetylcholinesterase activity was inves-
tigated. The second-order rate constant kon of the
enzyme–inhibitor complex correlates with the con-
formational positioning of aromatic residues, espe-
cially Trp84, in the transition state complex. The van
der Waals interactions are an important structural el-
ement of this conformational change. A transient mo-
bility of the cysteine loop (Cys67–Cys94) was confined
only to the presence of a significant steric effect. Even
with this limitation, however, the steric effect seems
to be an appropriate model for future tests on the
“back door” hypothesis involving facilitated opening
C
AChE (TcAChE) [2] revealed that its active site is deeply
buried within the molecule, in a long and narrow gorge
lined with aromatic residues. The hydrolysis of the sub-
strate (acetylcholine, ACh) is accomplished by the cat-
alytic triad consisting of Ser200, Glu327, and His440,
with Ser200 being transiently acetylated. Sussman et al.
[2] have proposed a model for cation–␲ interactions [3]
between the ammonium group of ACh and the aro-
matic amino acids (Tyr130, Trp84, and Phe330) of the
active site. All these aromatic groups provide polariz-
able ␲ electrons that might contribute to the induced,
short-range, van der Waals interactions. The differen-
tial catalytic activity of Trp86Ala (Trp84 in TcAChE) and
Tyr337Ala (F330 in TcAChE) human mutants [4] sup-
ports this hypothesis. It may also be noted that Trp84 is
part of that surface loop (Cys67–Cys94), which con-
stitutes the thin portion of the long and narrow gorge,
penetrating half-way into the enzyme and containing
for faster product clearance.
2002 Wiley Periodicals,
Inc. J Biochem Mol Toxicol 16:64–69, 2002; Published on-
line in Wiley Interscience (www.interscience.wiley.com).
DOI 10.1002/jbt.10026
KEYWORDS: Inhibition of AChE; Cysteine Loop (Cys67–
Cys94); Steric Effects
˚
the catalytic site about 4 A from its base [2].
The authors have examined, through combination
of kinetic studies and molecular modeling, both the ori-
entation of the Trp84 side chain and the conformational
adjustment of the cysteine loop. This is the first time
such an approach has been used to study these com-
pounds (Figure 1, 3–7).
INTRODUCTION
Acetylcholinesterase (AChE, EC 3.1.1.7) inhibitors
donepezil, tacrine, and physostigmine are the only
FDA-approved primary treatment for patients with
mild to moderately severe Alzheimer’s disease [1].
However, notwithstanding the abundance of data col-
lected on the extraordinary catalytic efficiency of AChE,
comprehensive characterization of AChE structure–
function relationships has only lately begun to emerge.
However, the X-ray 3-D structure of Torpedo californica
MATERIALS AND METHODS
All the reagents for synthesis of the inhibitors were
purchased from Aldrich and were of analytical grade.
AChE from Electrophorus electricus (Type V-S), acetylth-
iocholine (ATC), and 5,5-dithiobis(2-nitrobenzoic acid)
(DTNB) were purchased from Sigma. Physovenine
was provided by Prof. Arnold Brossi and Prof. Nigel H.
Greig (NIH, Bethesda) [5] and 8-carbaphysostigmine
was provided by Dr. Yuhpyng L.Chen (Pfizer,
Groton, CT) [6]. The synthetic route to Phy analogs
Correspondence to: Dr. Massimo Pomponi.
Contract Grant Sponsor: Italian Ministero dell’Universita` e della
Ricerca Scientifica e Tecnologica (MURST).
Contract Grant Sponsor: Consiglio Nazionale delle Ricerche
(CNR).
c
2002 Wiley Periodicals, Inc.
64

Volume 16, Number 2, 2002
AChE: MOBILITY OF CYSTEINE LOOP
65
(pyrrolo[2,3-b]indol-5-ol 3,3a,8,8a-hexahydro-1,3a,8-
trimethyl-N-alkyl-carbamates, 3aS-cis) was similar in
many respects to the one used for the synthesis of
heptylphysostigmine [7].
presented similar UV and IR spectra. NMR spectrum
in deuterated benzene: 0.90 (t), 1.25 (s), 1.40(s), 2.00 (t),
2.50 (s), 2.80 (t), 2.90 (s), 2.95–3.30 (m), 4.30 (s), 5.00–
5.30 (m), 6.25 (dd), 6.65 (d), 6.75 (dd). Mass spectrum:
401 (13), 218 (100). Structures of the inhibitors used are
shown in Figure 1.
Preparation of Trimethyl-1,3a,8-hexahydro-
1,2,3,3a,8,8a-pyrrol[2,3-b]indole-5(3aS, 8aR)-
heptylphysostignine (6)
Enzyme Activity
Five hundered and fifty milligrams of physostig-
mine (2 mmol) and 108 mg mmol of sodium methoxide
(2 mmol) were placed in a flask (50 mL) to which a vac-
uum between 5 and 10 mm Hg was applied. Absolute
ethanol (70 mL) was then added over 2 h in small vol-
umes, under agitation at ambient temperature. When
the ethanol had been removed, benzene (50 mL) was
added (temperature range: 30 C). To this residue was
slowly added 50 mL of a benzene solution containing
4 mmol of heptyl isocyanate; the reaction mixture was
left to stand for 3 h, benzene being added when neces-
sary. During this time, eseroline gets converted to car-
bamate. Benzene was evaporated, and petroleum ether
was added to the reaction residue, followed by 10 mL of
10 3 M HCl. The vacuum was removed, and the crude
reaction product was extracted with diethyl ether af-
ter previous saturation with sodium bicarbonate. The
combined ether solutions were rapidly extracted with
50 mL of 0.05 M HCl. The aqueous acid solution, sat-
urated with sodium bicarbonate, was again extracted
with petroleum ether. The organic layer was repeatedly
washed with water before being dried over sodium sul-
fate. The solvent was finally removed under vacuum.
If necessary, the extract was chromatographed over a
silica gel column, using chloroform/methanol (98:2) as
eluent. The product was finally dissolved in benzene,
heptane was added, and the solution was cooled. The
residue slowly crystallized and had a m.p. of 83–86 C.
UV spectrum in methanol: 303 (3300), 253 (14200). IR
spectruminchloroform:3460(m), 2920(s), 2860(s), 1720
(s), 1670 (s), 1600 (s), 1490 (s), 1240 (s), 1200 (s), 1120 (m).
Structural assignment was confirmed by NMR spec-
troscopy on a 1H-(300 MHz) Varian Gemini spectrome-
ter. NMR spectrum in deuterated benzene: 0.90 (t), 1.25
(s), 1.35 (s), 1.80 (t), 2.40 (s), 2.50 (t), 2.60 (s), 3.35 (q),
4.00 (s), 5.30–5.70 (m), 6.20 (d), 7.00 (s), 7.05 (dd). Mass
spectrum (LKB Model 9000, single focusing): 359 (15),
218 (100), 174 (30), 161 (28), 160 (29).
AChE activities were measured in 0.1 M sodium
phosphate buffer (pH 8) containing 0.1% BSA and
0.25 mM DTNB at 20 C according to Ellman et al. [8],
using acetylthiocholine (ATC) as substrate. The reac-
tion for fast-binding inhibitors was initiated by addi-
tion of substrate. When inhibition by the slow-binding
inhibitors was measured, the reaction was started by
the addition of enzyme. The assay medium contained
0.5 mM ATC iodide in a 0.1 M phosphate buffer (pH
7), and the reaction was monitored at 412 nm by means
of a graphics software (Varian Cary 3E) at 20 C, us-
ing thermostated (Haake) cells. Typically, four inhibitor
concentrations were used. The resulting dependence
of activity on time was fit by nonlinear least-squares
kt
procedures [9] to the equation v = (v0 vss)e
+ vss,
where v is the observed velocity at any time, v0 is the
initial velocity of the reaction, vss is the steady-state ve-
locity of the reaction, and k is the pseudo-first-order
rate constant. The equation applies whether the reac-
tion is started by addition of the enzyme to the reaction
Preparation of Trimethyl-1,3a,8-hexahydro-
1,2,3,3a,8,8a-pyrrol[2,3-b]indole-5(3aS,8aR)-
decylphysostigmine (7)
FIGURE 1. Molecular formulas: m-(N, N, N-trimethylammonio)-
trifluoroacetophenone (TMTFA, 1); ACh (2); 8-carbaphysostigmine
(3, Y = CH2, X = NEt, R = CH3); physostigmine (Phy, 4, Y = NCH3,
X = NCH3, R = CH3); physovenine (5, Y = NCH3, X = O, R = CH3);
heptylphysostigmine (Phy-C7, 6, Y = NCH3, X = NCH3, R = C7H15);
decyl-Phy (Phy-C10, 7, Y = NCH3, X = NCH3, R = C10H21).
The procedure for the preparation of 6 was fol-
lowed. After evaporation of the solvent, an oily residue
remained but was not crystallized. Inhibitors 7 and 6

66
GAVUZZO AND POMPONI
Volume 16, Number 2, 2002
mixture containing substrate and inhibitor or by the ad-
dition of substrate after preincubation of enzyme and
inhibitor.
As predicted by this relation, the dependence of k1
on inhibitor concentration was linear, from which kon
=
1
(3.5 0.2) 103 M 1 s1 and koff = (1.7 0.4) 10 3 s
Building and analysis of the 3-D molecular mod-
els was performed on Silicon Graphics work sta-
tions, Indigo2 and O2, using the SYBYL modeling (in
vacuo) software (TRIPOS, Inc. St. Louis, MO). The
X-ray coordinates of m-(N,N,N-trimethylammonio)-
trifluoroacetophenone–acetylcholinesterase (TMTFA–
AChE) complex were used as templates. Specifically,
the ligand was positioned to make a tetrahedral bond
with the O␥ of Ser200. Positioning of the ligand was
also guided by interactions of quaternary nitrogen (N1)
with residue Trp84; the quaternary group was placed
(Figure 2A). The apparent association constant of the E–
Inhibitor complex was (2.0 0.5) 106 M 1. At high
concentrations of 4 Eq. (1) reduced to k1 = (kon/(1 +
[S]/Km))[Inhibitor]. For inhibitor concentrations in the
range 0.06–0.10 mM, the dependence of k1 on 4 was lin-
ear (plot not shown). The value kon = (1.5 0.3) 103
1
M
1 s calculated from the slope of this plot was
in good agreement with the value determined from
Figure 2A. The same procedure was utilized for all the
inhibitors used.
It is noteworthy, from the analysis of the numeri-
cal values of kon (Table 1) for charged and uncharged
analogs, that charge does not confer a particular advan-
tage for interaction in the active site. In order to ver-
ify the presence of any structural correlation, the tetra-
hedral intermediates generated during the inhibition
step were structurally aligned on the crystallographic
coordinates of theTcAChE complex (Brookhaven Pro-
tein Data, ID Codes 1AMN) with the transition state
analog inhibitor TMTFA [11], and then subjected to
˚
within van der Waals distance ( 3.5 A).
Initial optimization included restriction of the dis-
tances between the carbonyl oxygen of the amide ni-
trogen atoms of residues Gly118 and Gly119, which
are two potential hydrogen bond donors. This restric-
tion was relieved in the subsequent refinement. The
POWELL minimizer was selected. The TRYPOS force
field consisted of both van der Waals and electrostatic
(Geisteiger–Marsili) energies. The cutoff criterion was
˚
8 A radius for all interactions. The energy convergence
criterionwas 0.1kcal mol 1. Thecomplexeshavebeen
superimposed using the C␣ of ␣-helix 384–411 residues.
This ␣-chain results the most invariant secondary struc-
ture for these intermediates.
RESULTS
Physostigmine (4), caused reversible inhibition.
Theconcentrationvalueoftheinhibitorintheassaywas
comparable to the inhibition constant, Ki (=koff/kon), of
the enzyme–inhibitor complex. These observations and
the additional experiments are consistent with the fol-
lowing inhibition scheme (Scheme 1).
In the presence of inhibitor, the AChE activity
dropped in a time-dependent manner that is well de-
scribed by first-order kinetics. All the inhibitors pro-
duced complete inhibition of the enzyme. In accord
with Scheme 1, the pseudo-first-order rate constant de-
termined from the fit of the progress curve is described
by the following equation [9]:
k1 = (kon/(1 + [S]/Km))[Inhibitor] + koff
(1)
The Michaelis constant, Km, was 0.11 0.01 mM.
FIGURE 2 (A) Dependence of the observed first-order inhibition
rate constant on inhibitior concentration. The rate constants kon and
koff were calculated from the slope (corrected for the presence of sub-
strate) and intercept, respectively, of the displayed linear fit. Inhibi-
tion at 20 C. AChE activity was determined by a slight modification
of the method used by Ellman et al. [8]. (B) Linear plot of kon vs
the mean of the distances from indole ring of Trp84 (ligand–AChE
complex). The plot allows us to derive the kon of ACh [10].
SCHEME 1

Volume 16, Number 2, 2002
AChE: MOBILITY OF CYSTEINE LOOP
67
TABLE 1. Second-Order Rate Constant kon for Phy Analogs
Inhibitors of Acetylcholinesterase
(acceptor) of the inhibitor carbonyl group seemed to be
stabilized in the tetrahedral intermediate by potential
hydrogen bonds from the amide nitrogen of Gly118 and
Gly119.
The relationship between the binding rate constant,
kon, and the mean of the relative distances (computed
as reported in Figure 4) of indole rings of the side chain
of Trp84 of each inhibitor-AChE complex is reported
in Figure 2B. The result suggests a direct relationship
between the relative distance from the Trp84 residue
and the value of binding rate constant, kon.
Inhibitor
ln kon
TMTFA (1)
ACh (2)
11.6a
9.6b
8.0
8.15
9.4
3
4
5
6
7
6.1
2.9
Enzyme inhibition was determined as described in Materials and Methods
section. Values represent mean of triplicate determinations with standard de-
viation not exceeding 20%.
Phy (4) and 8-carbaphysostigmine (3) possess the
amine group N1, and at physiological pH, exist as
cations. Since the indole nitrogen lone pair increases
the electron density and partial negative charge of the
aromatic system, the ammonium of an advancing lig-
and would induce electronic perturbations in the aro-
matic system, thus increasing the van der Waals inter-
actions. At any rate, the catalytic machinery of AChE
also seemed to utilize a second way of interaction typ-
ical of charged ligands. The coulombic effect of the
positiveammoniumgroupseemedtostabilizethetetra-
hedralintermediateandtransitionstatewithdirectcon-
sequences on the value of kon. If aromatic residues, par-
ticularly Trp84, are crucial for molecular recognition,
the above observation may also apply to substrates.
The case is different for inhibitors 6 and 7, where
a specific steric effect seems to influence negatively
the tetrahedral intermediate (Figure 3), which by a
Hammond effect [13], produces a proportional destabi-
lization of the transition state. In particular, compound
7, which possesses the longer alkyl chain, caused a tran-
sient shift in the cysteine loop, Cys67–Cys94 (Figure 5).
Apparently, thisbehaviorwasconfinedonlytothepres-
ence of a significant steric effect.
An observation that can be added at this point is
that the amine group (N1) present in Phy (4) may not
be essential for its anti-AChE activity. In fact, this was
the noted in physostigmine analogs bearing oxygen at
position 1 (physovenine, 5). One reason for the activity
of compound 5 may be that the aromatic electrons of
indole ring of Trp84 modulate their charge density on
the electronic environment of the ligand. However, a
different contribution is also possible in this case as the
O1 oxygen can form a potential hydrogen bond with
the hydroxy moiety of Tyr130.
a Ref. [11].
1
b This value of kon
(
1.4 104
M
s
1) is similar to the value of kcat 1.4
104
M
s
1, reported by Rosenberry [10].
1
field-fit optimization. TMTFA, a potent inhibitor of
AChE, forms a covalent complex with the enzyme that
it structurally mimics the theoretical acylation tetrahe-
dral intermediate [11,12] of acetylcholine (ACh). How-
ever, carbamates (for example, physostigmine) also
react with AChE to produce a covalent, tetrahedral in-
termediate very similar to the complex formed during
ACh hydrolysis. These analogies justify the similar ap-
proach that has been used in this study for TMTFA,
Ach, and Phy analogs. In all cases the resultant com-
plex corresponds to the tetrahedral geometry of an in-
termediate acylating/carbamoylating step of catalysis.
The minimization has been successively performed us-
ing decreasing values of force constant from 100 to
2
˚
0, kcal/((mole) (A) ; for this last value (without any
constraint) the minimization has also been calculated
in presence of charges. Notably, with or without con-
straints, there is no difference in the van der Waals dis-
tance of inhibitor from Trp84. Figure 3 shows the rela-
tive position of this residue for TMTFA (␣-helix 384–411
was used for this superimposition).
In particular, the results, relative to the “cation–␲
interaction subsite” Trp84, were consistent with stack-
ing of the positively charged ligand against the ␲ elec-
tron system of residue at position 84, as suggested
by the crystal structure [12]. In this context, Figure 4
shows the distances for the closer, positive atom of
˚
TMTFA. The mean of these distances (3.64 0.01 A)
correlate with the binding rate constant kon. The same
approach was also valid for ACh and physostigmine
analogs. The most striking observation of the modeling
study was a remarkable overall similarity between the
orientation of the TMTFA molecule and the physostig-
mine analogs, i.e., the principal interactions appeared
to be shared. In particular, the quaternary nitrogen (N1
for physostigmine analogs) of the inhibitor was posi-
DISCUSSION
In this paper we compare the structure of TMTFA–
TcAChE complex [12] with those of selected inhibitor–
TcAChE complexes, with the goal of identifying partic-
ular topographical features that might be indicative of
the inhibition mechanism. The results described herein
˚
tioned within van der Waals distance (3.5 A) of Trp84;
the O␥ of Ser200 lay within hydrogen-bonding distance
of His440; Glu327 was also near His440. The oxygen

68
GAVUZZO AND POMPONI
Volume 16, Number 2, 2002
FIGURE 5 Displacement of the loop (Cys67–Cys94) structure in
AChE. Superposition of the loop trace of tetrahedral ACh complex
and that of the modeled structure of decyl-Phy analog. Same condi-
tions as for Figure 3. RMS 0.25. The Trp84 residue is linked to the
helical turn structure (Ser79-Met83).
FIGURE 3 Superimposition of TMTFA, ACh, Phy, and decyl-Phy
(PhyC10) minimized structures showing the Trp84 shift. ␣-helix 384–
411 has been used for this superimposition. The coordinates for
TMTFA–AChE complex were obtained from Brookhaven Protein
Data Bank (entry 1AMN). This molecular structure was used as the
starting geometry for the minimization study. The carbamoyl (or
carbonyl) group was positioned to make a tetrahedral bond with
O␥ of Ser200. The quaternary nitrogen of the inhibitor is positioned
from the indolic ring of Trp84, computed as mean of
these relative distances (Figures 4 and 2B).
˚
within van der Waals distance (3.5 A); the O␥ of Ser200 lies within
However, the catalytic power of enzymes has long
been attributed to specific interactions with substrates
in the transition state. Since transition states have es-
sentially zero lifetimes, it is impossible to observe them
directly and information about their geometries must
be acquired indirectly. The X-ray of the complex of
TcAChE with the transition state analog TMTFA is an
example of how this data can be obtained. In addition,
the manifest overall correspondence between the align-
ment of the TMTFA molecule and the physostigmine
analogs provides a useful way of organizing inhibitors
3–6 (Figure 1) in the molecular modeling of tetrahe-
dral intermediates. In addition, the results described
herein (Figure 2B) show that charged and uncharged
inhibitors follow a common correlation. This effect es-
tablishes that all compounds occupy a common site
and share the principal interactions. The catalytic ma-
chinery of AChE also utilizes coulombic interactions, as
suggested by data obtained for charged ligands. A con-
sequence of this study is the feasibility extending this
approach to substrates. In fact, the value of the ACh
kon constant, calculated from the computational rela-
hydrogen-bonding distance of His440; Glu327 is also near His440. No
constraints were used. Procedure: ANNEAL (SYBYL): interesting re-
gion 18 A; hot region 16 A. TMTFA (1); ACh (2), RMS 0.36; Phy (3),
RMS 0.25; decyl-Phy (PhyC10, 7), RMS 0.22.
˚
˚
indicate that within a structural congeneric series (in
this case analogs of the alkaloid Phy), in the inhibitor–
AChE complex is observed a relationship between the
bimolecular rate constant kon and the relative distances
1
1
tive distance (3.46 0.11 A), was 1.4 104 M
s
˚
(vs. 1.4 104, reported by Rosenberry [10]).
According to this relationship the expected value
1
kon for 8-carbaphysostigmine, (2.9 0.2) 103 M 1 s
,
is in a good agreement with the value reported in
Table 1. Moreover, an additional comment is necessary
for physovenine (5) and decyl-Phy (7). The kon value for
physovenine may be due to the hard lone pairs of O1,
which generate a greater negative charge density while
interacting with the partial negative charge of ␲ system
of indole ring of the Trp84 residue. On the other hand,
FIGURE 4 3-D structure of the “cation–␲ interaction subsite” Trp84
and inhibitor TMTFA, in the tetrahedral intermediate. The mean of
the distances is considered to be proportional to the distance from
the geometric center of the aromatic ring. Conditions as in Figure 3.

Volume 16, Number 2, 2002
AChE: MOBILITY OF CYSTEINE LOOP
69
3. Dougherty DA, Stauffer DA. Acetylcholine binding by
a synthetic receptor: Implications for biological recogni-
tion. Science 1990;250:1558–1560.
the behavior of decyl-Phy may be due to the unusual
steric effect of the long alkyl chain. The refined and su-
perimposed structures ACh (2) and decyl-Phy (7) reveal
aconformationaldistortionofthecysteineloop(Cys67–
Cys94) (Figure 5). From a structural point of view, other
studies on AChE complexes with reversible inhibitors
have shown very similar results [14].
The mobility of Trp84 and the dynamic behavior
of the cysteine loop (Cys67–Cys94) bearing Trp84 has
been a field of study that has provided very conflict-
ing results [15,16]. In particular, molecular dynamics
[15] suggests the existence of a molecular device (the
shutter-like back door) that opens to admit or remove
molecules, although site-directed mutagenesis studies
failed to provide experimental support for the back
door model [17]. In the present study, the adjustment
of the loop was observed for the longer N-carbamyl
chains (7). However, at the same time, the steric effects
seemed to interfere with the validity of the present ap-
proach: when bulky inhibitors were squeezed into the
binding pocket, the modeling showed that the cysteine
loop moved away (Figure 5).
4. Ordentlinch A, Barak D, Kronman C, Ariel N, Segall Y,
Velan B, Shafferman A. Contribution of aromatic moieties
of tyrosine 133 and of the anionic subsite tryptophan 86
to catalytic efficiency and allosteric modulation of acetyl-
cholinesterase. J Biol Chem 1995;270(5):2082–2091.
5. Yu QS, Liu C, Brzostowska M, Chrisey L, Brossi A,
Greig NH, Atack JR, Soncrant TT, Rapoport SI, Radunz
HE. Physovenines: Efficient synthesis of ( )- and (+)-
physovenine and synthesis of carbamate analogs of ( )-
physovenine. Helv Chem Acta 1991;74:761–766.
6. Chen YL, Nielsen J, Hedberg K, Dunaiskis A, Jones
S, Russo L, Ives J, Liston D. Syntheses, resolu-
tion, and structure-activity relationships of potent
acetyl-cholinesterase inhibitors: 8-Carbaphysostigmine
analogs. J Med Chem 1992;35:1429–1434.
7. Marta M, Gatta F, Pomponi M. Physostigmine analogs
anticholin-esterases: Effects of the lengthening of the N-
carbamic chain on the inhibition kinetics. Biochim Bio-
phys Acta 1992;1120(3):262–266.
8. Ellman GL, Courtney KD, Andres V Jr, Featherstone
M. A new and rapid colorimetric determination of
acetylcholinesterase activity. Biochem Pharmac 1961;7:
88–95.
9. Williams JW, Morrison JF. The kinetics of reversible tight-
binding inhibition. Methods Enzymol 1979;63:437–467.
10. (a) Rosenberry TL. Acetylcholinesterase. Adv Enzymol
Relat Areas Mol Biol 1975;43:103–218; (b) Rosenberry TL,
Barbett P, Mays C. Acetylcholinesterase. Methods Enzy-
mol 1982;82(Pt A):325–339.
11. Nair HK, Seravalli J, Arbuckle T, Quinn DM. Molec-
ular recognition in acetylcholinesterase catalysis: Free-
energy correlations for substrates turnover and inhibition
by trifluoro ketone transition-state analogs. Biochemistry
1994;33:8533–8576.
In conclusion, the authors hope that this procedure,
and future studies based on steric effects, might help in
discoveringsomefeaturesofthetrafficofproductsfrom
the active site of AChE. This traffic might be governed
by the mobility of the cysteine loop and this dynamic
behavior may be part of a more general mechanism of
AChE catalysis.
12. Harel M, Quinn DM, Nair HK, Silman I, Sussman JL.
The X-ray structure of a transition state analog complex
reveals the molecular origins of the catalytic power and
substrate specificity of acetylcholinesterase. J Am Chem
Soc 1996;118:2340–2346.
13. March J. The hammond postulate. In Advanced Organic
Chemistry. London: McGraw-Hill; 1977. pp. 194–195.
14. Pomponi M, Marta M, Sacchi S, Colella A, Patamia M,
Gatta F, Capone F, Oliverio A, Pavone F. Studies on a new
series of THA analogs: Effects of the aromatic residues
that line the gorge of AChE. FEBS Lett 1997;409(2):155–
160.
ACKNOWLEDGMENTS
The authors thank Professor Arnold Brossi and
Prof. Nigel H. Greig (NIH, Bethesda), and Dr. Yuhpyng
L. Chen(Pfizer, Groton, CT)forthekindgiftofphysove-
nine and 8-carbaphysostigmine, respectively.
REFERENCES
15. Gilson MK, Straatsma TP, McCammon JA, Ripoll DR,
Faerman CH, Axelsen PH, Silman I, Sussman JL. Open
”back door” in a molecular dynamics simulation of
acetylcholinesterase. Science 1994;263:1276–1278.
16. Kronman C, Ordentlich A, Barak D, Velan B, Shafferman
A. The ”back door” hypothesis for product clearance in
acetylcholinesterase challenged by site-directed mutage-
nesis. J Biol Chem 1994;269(45):27819–27822.
17. Faerman C, Ripoll D, Bon S, Le Feuvre Y, Morel N,
Massoulie J, Sussman JL, Silman I. Site-directed mu-
tants designed to test back-door hypotheses of acetyl-
cholinesterase function. FEBS Lett 1996;386(1):65–71.
1. (a) Taylor P. Development of acetylcholinesterase in-
hibitors in the therapy of Alzheimer’s disease. Neurol-
ogy 1998;51(1, Suppl 1):S30–S35, S65–S67; (b) Knopman
DS. Current pharmacotherapies for Alzheimer’s disease.
Geriatrics 1998;53:S31–S34.
2. Sussman JL, Harel M, Frolow F, Oefner C, Goldman
A, Toker L, Silman I. Atomic structure of acetyl-
cholinesterase from Torpedo californica: A prototypic
acetylcholine-binding protein. Science 1991;253:872–879.
Entry PDB 2ACE; authors: Harel M, Raves ML, Silman I,
& Sussman JL (23 June 1996).