Novel anticholinesterases based on the molecular skeletons of furobenzofuran and methanobenzodioxepine

Article abstract of DOI:10.1021/jm049309+

Reductive cyclization of 5-hydroxy-3-methyl-3- methoxycarbonylmethylenebenzofuran-2(3H)-one (4) gave 5-hydroxy-3a-methyl-2,3, 3a,8a-tatrahydrofuro[2,3-6]benzofuran (5) and the rearrangement product 7-hydroxy-4,5-dihydro-2,5-methano-1,3-benzodioxepine (6).

Full text of DOI:10.1021/jm049309+

9
86
J. Med. Chem. 2005, 48, 986-994
Novel Anticholinesterases Based on the Molecular Skeletons of Furobenzofuran
∇
and Methanobenzodioxepine
Weiming Luo,†,# Qian-sheng Yu, Ming Zhan, Damon Parrish, Jeffrey R. Deschamps, Santosh S. Kulkarni,
†,#
‡
§
§
|
†
⊥
⊥
[
,†
Harold W. Holloway, George M. Alley, Debomoy K. Lahiri, Arnold Brossi, and Nigel H. Greig*
Drug Design & Development Section, Laboratory of Neurosciences, and Bioinformatics Unit, Research Resources Branch,
Gerontology Research Center, National Institute on Aging, National Institutes of Health, 5600 Nathan Shock Drive,
Baltimore, Maryland 21224, Laboratory for the Structure of Matter, Department of the Navy, Naval Research Laboratory,
Washington, D.C. 20375, Medicinal Chemistry Section, Intramural Research Program, National Institute on Drug Abuse,
National Institutes of Health, 5500 Nathan Shock Drive, Baltimore, Maryland 21224, Indiana University School of Medicine,
Indianapolis, Indiana 46202, and School of Pharmacy, University of North Carolina at Chapel Hill,
North Carolina 27599-7361.
Received August 20, 2004
Reductive cyclization of 5-hydroxy-3-methyl-3-methoxycarbonylmethylenebenzofuran-2(3H)-
one (4) gave 5-hydroxy-3a-methyl-2,3,3a,8a-tatrahydrofuro[2,3-b]benzofuran (5) and the rear-
rangement product 7-hydroxy-4,5-dihydro-2,5-methano-1,3-benzodioxepine (6). Reaction of
compounds 5 and 6 with different isocyanates provided two series novel carbamates (7-12)
whose structures were confirmed by X-ray crystallography. These were assessed for anti-
cholinesterase action against freshly prepared human enzyme and proved to be potent inhibitors
of either acetyl- (AChE) or butyrylcholinesterase (BChE) with specific compounds exhibiting
remarkable selectivity. Because the two series of carbamates (7-12) differ in their phenolic
moieties, their respective potency and selectivity for AChE versus BChE was governed by their
N-substituted groups. This same characteristic was also present in a series of physovenine
analogues (1, 13, 15, 17) and physostigmine analogues (2, 14, 16, 18). These structure-activity
relations proved valuable in elucidating the mechanisms underpinning the interaction between
carbamate-based cholinesterase inhibitors and their enzyme target. In addition, because
physostigmine analogues have demonstrated activity in lowering the Alzheimer’s disease
protein, amyloid precursor protein (APP), examples of the two new series of carbamates were
characterized in culture assays of quantifying cell viability and synthesis of APP.
Introduction
Physovenine (1), with an ether linkage in ring-C of
natural physostigmine (2) instead of a N-Me group
(
Figure 1), is an alkaloid present in Physostigma
venenosum whose chemistry and biological properties
have been extensively reviewed. The almost identical
1
properties of 1 and 2 in inhibiting human AChE and
2
,3
BChE aroused our interest in further exploring the
mechanisms underpinning the action of physostigmine
(2) and its analogues toward the discovery of potent and
novel anticholinesterase because this class of compounds
represents the backbone of current Alzheimer’s disease
(
AD) treatment.4
Physostigmine (2) has served as a template in the
Figure 1. Chemical structures of physovenine (1) and phy-
sostigmine (2) based carbamates.
development of several agents for AD, including a slow-
release formulation of the parent compound (synapton,
Forest Laboratories, St Louis, MO), its heptylcarbamate
from clinical development because of efficacy and toxic-
ity issues, and its phenylcarbamate (phenserine, Na-
tional Institute on Aging, Baltimore, MD, and Axonyx,
New York, NY) that currently is in phase 3 clinical
assessment in mild to moderately afflicted patients. The
pharmacodynamics that underscores the action of this
class of drugs is the elevation of brain acetylcholine
(
eptylstigmine, Mediolanum, Italy), both withdrawn
∇
*
Historically known occasionally as furochroman.
To whom correspondence should be addressed. Phone: 410-558-
8
278. Fax: 410-558-8323. E-mail: Greign@grc.nia.nih.gov.
†
Laboratory of Neurosciences, National Institute on Aging.
Equal contribution and joint first authorship by Drs. Weiming Luo
#
(
ACh) levels that is achieved by inhibition of its hydro-
and Qian-sheng Yu.
4
‡
lyzing enzymes, AChE and BChE. A hallmark of AD
is synaptic loss, particularly of the forebrain cholinergic
Gerontology Research Center, National Institute on Aging.
§
Naval Research Laboratory.
|
National Institute on Drug Abuse.
⊥
system that is pivotal in higher brain functions associ-
Indiana University School of Medicine.
[
ated with learning, memory, and behavior.4
University of North Carolina at Chapel Hill.
1
0.1021/jm049309+ CCC: $30.25 © 2005 American Chemical Society
Published on Web 01/12/2005

Anticholinesterases
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 4 987
Scheme 1a
Extensive structural information is available on the
interaction of AChE and BChE with ACh and related
substrates because of extensive X-ray crystallography
and mutagenesis studies.5-7 The catalytic subunit of
both enzymes has a molecular mass of 60-80 kDa,
depending on its level of glycosylation and the species
6
from which it is derived. The active site for ACh
hydrolysis is buried within a primarily hydrophobic 20
7
Å deep gorge. In its interaction and inhibition of AChE
and BChE, it is generally believed that under physi-
1
ological conditions the basic N (CH3) group of physos-
8
+
tigmine (pKa ) 8.46) and analogues gains a H to form
a quaternary ammonium group. In a manner similar
to ACh, positively charged physostigmine is drawn into
the enzyme by electrostatic field forces. Thereafter, it
interacts with specific binding domains to form an
inhibitor-enzyme complex. Whereas the quaternary
ammonium group plays an important role in this
process, physovenine (1), which bears a neutral ring-C
oxygen instead of basic ring-C nitrogen, appears to be
equipotent as an anticholinesterase, as is the ring-C thia
analogue of 1, thiaphysovenine.9
a
Reagents: (i) 160-170 °C, 3 h; (ii) ClCH2COOCH3/DMF, NaH,
This phenomenon can be potentially explained by the
common capacity of the ether oxygen and amino nitro-
gen of 1 and 2 to form H-bonds with amino acids,
providing the basis of the interaction between the
compounds and enzyme. However, the common ring-
0
°C, 1 hr, then room temp, overnight; (iii) (a) LAH/THF, room
temp, 1 h; (b) oxalic acid, 0.5 h; (iv) Na, EtNCO/Et O, room temp,
1 h; (v) Na, o-tolyl isocyanate/Et2O, room temp, 0.5 h; (vi) Na,
p-isopropylphenyl isocyanate/Et2O, room temp, 45 min.
2
8
It has been reported that the oxindole corresponding
B-N (CH3) in both provides a weaker basic group (pKa
8
to compound 4, 3-methoxycarbonylmethylene oxyindole-
)
3.7) and hence reintroduces the potential involve-
2
-one, is able to convert to physovenol.17 On the basis
ment of charge into the compound-enzyme interaction.
To totally exclude the possibility of forming a positively
charged molecule, we synthesized analogues of 2 with
the nitrogen atoms of the tricyclic moiety replaced by
ether oxygen and assessed the activity of the resulting
novel compounds as anticholinesterases. The resulting
tetrahydrofuro[2,3-b]benzofuran is an important struc-
tural component of aflatoxins, mold metabolites studied
of this, we conceived that compound 4 would be ame-
nable to reductive cyclization to provide the desired
tricyclic target compound 5. Applying normal reductive
conditions in practice, however, we gained a mixture of
highly polar compounds. The departure of our obtained
products from those predicted was likely derived from
the following. First, in lieu of an amide in ring-B, as
occurs in the oxyindole,17 compound 4 has a lactone.
This may be easily reduced to form tetrol, 19, which
contains both phenol and alcohol groups. After inter-
molecular dehydration and cyclization, 19 became a
complex mixture of different products. Second, desired
compound 5 is an acetal that under acidic conditions
exists in an equilibrium between acetal 5 and aldehyde
20 (Scheme 2). The 1,4-hydroquinone in compound 20
is not a stable moiety, being easily oxidized to form
complex products. To avoid the equilibrium moving
toward the formation of aldehyde 20, strong aqueous
acid was avoided in the synthesis. Hence, after reduction
1
0
11
12
extensively by B u¨ chi, Rapoport, Townsend, and
others. However, rather than creating the tricyclic
moiety by chemistry published by these authors, we
started with 3-methyl-5-hydroxybenzofuran-2-one (3),
readily available from 1,4-cyclohexandione and pyruvic
1
3
acid, and its further application. This is reminiscent
of chemistry employed earlier by Harley-Mason in his
14
synthesis of physostigmine (2). We herein describe the
anticholinesterase activity against human enzyme for
this novel series of carbamates, and because physostig-
mine analogues have demonstrated activity to lower the
levels of the AD protein amyloid precursor protein
15
16
of compound 4 by LiAIH in THF at room temperature
for 1 h, solid oxalic acid, rather than dilute HCl, was
(
APP) that generates the neurotoxic peptide, Aâ , we
4
assessed the action of specific examples on the cellular
viability and production of secreted APP of immortal
human cells in culture.
added to the reaction mixture as a hydrolysis reagent.
This replacement provided us compound 5 together with
the unexpected compound 6 in a ratio of 1:1. In the same
manner as proposed for 5, 6, in an aqueous acidic
condition, likely is in equilibrium with aldehyde 21. The
reaction mechanism remains to be fully elucidated and
is a focus of current investigation.
Separation of compounds 5 and 6 by chromatography
proved to be difficult because of their similar polarities.
As a consequence, a mixture of 5 and 6 was reacted with
different isocyanates to provide three pairs of carbam-
ates 7 and 8, 9 and 10, and 11 and 12. Each pair of
compounds was separated by preparative TLC to gener-
ate the final products in high purity. In summary, this
Results
Chemistry. 5-Hydroxybenzofuran-2-one (3) was ob-
tained by condensation of 1,4-cyclohexandione with
pyruvic acid according to published methodology.13 The
crude product, 3, was crystallized directly from the
reaction mixture without chromatography. Its recrys-
tallization from ethanol gave pure crystalline 3 in a 40%
yield. C-3 alkylation of compound 3 with methyl chloro-
acetate, using 1 equiv of sodium hydride as a base, gave
compound 4, which also was purified by simple crystal-
lization without using chromatography (Scheme 1).

9
88 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 4
Luo et al.
Scheme 2
four-step total synthesis provided a simple means to
obtain compounds containing the molecular skeletons
of tetrahydrofurobenzofuran (5) and 4,5-dihydro-2,5-
methano-1,3-benzodioxepine (6). The reductive cycliza-
tion of compound 4 requires additional optimization to
generate a higher yield.
phenyl ring is at 67.94° to the O-phenyl (ring-A). The
1,3-dioxepine is folded at the bridgehead C2-C5. The
angle of C10-C2-O1 is 113.17°, the angle of C10-C2-
O3 is 117.8°, and the angle of O1-C2-O3 is 108.68°.
Biological Evaluation. Table 1 illustrates the bio-
logical activity of compounds 7-12 against freshly
prepared human AChE and BChE, in comparison to the
corresponding carbamates of physovenol (1, 13, 15, 16)
and eseroline (2, 14, 17, 18), whose measured values
were similarly obtained.18 Because we have previ-
ously demonstrated that physostigmine analogues with
straight-chain N-alkyl (e.g., methyl, butyl, and octyl)
Because this represents the first description of these
types of carbamates and the intromolecular rearrange-
ment that occurs during the reductive cyclization of
compound 4, the structures of 7-12 were characterized
1
13
not only by routine MS, H NMR and C NMR but
1
additionally by H NMR COSY 45. The latter proved
useful in defining the bridged structure of compound 8.
Specifically, a correlation between bridged methylene
C10-H2 (δ 2.19-1.91 ppm) and the C2-H (δ 5.69) on
the bridgehead position was present in 8 and contrasts
with the C8a-H (δ5.77 ppm) of 7, where there is no
correlation of methylenes, C2-H2 (4.06-3.56 ppm) and
C3-H2 (2.13-1.91 ppm) (see Supporting Information).
Finally, single crystals of 11 and 12 were obtained and
analyzed by X-ray crystallography to confirm their
structures (Figure 2).
For comparison, the ethylcarbamate of physovenol
3
(
13) was synthesized using a published method. The
ethylcarbamate of eseroline (14) was also synthesized
1
5
according to known procedure.
X-ray Crystallography. The results of the X-ray
studies are illustrated in Figure 2.
For compound 11, the four atoms of the carbamoyl
moiety -N-CO-O- compose a plane. The angle be-
tween planes N-phenyl and carbamoyl is 10.95°. The
angle between the carbamoyl and O-phenyl (ring-A) is
9
5.75°. These two aromatic rings are almost perpen-
dicular to each other (84.80°). The terminal five-
membered ring-C has an envelope conformation with
C3 being out of plane atoms. The central five-membered
ring-B is coplanar with its adjoining phenyl ring. The
fused ring system is folded at the bond in common with
the two five-membered rings (C3a-C8a). The angle of
O8-C8a-O1 is 108.32°.
For compound 12, the four atoms of carbamoyl -N-
CO-O- also are in a plane. The angle between planes
N-phenyl and carbamoyl is 4.33°. The angle between
carbamoyl and O-phenyl (ring-A) is 72.27°. The N-
Figure 2. X-ray crystallographic picture and corresponding
chemical structures of 11 and 12.

Anticholinesterases
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 4 989
Table 1. 50% Inhibitory Concentration of Compounds versus Human Erythrocyte AChE and Plasma BChEIC50
IC50 ( SEM (nM)
compound
AChE
BChE
selectivity
Physovenine Seriesa
27 ( 1
1
1
1
1
(-)-methylcarbamate
4 ( 1
7-fold BChE
3
5
7
(-)-ethylcarbamate
82( 4
2 ( 0.1
40-fold BChE
120-fold AChE
227-fold BChE
(-)-2′-methylphenylcarbamate
(-)-4′-isopropylphenylcarbamate
13 ( 1
3860 ( 970
1560 ( 120
17 ( 2
Physostigmine Seriesa
28 ( 2
2
1
1
1
(-)-methylcarbamate
16 ( 3
2-fold BChE
23-fold BChE
195-fold AChE
15-fold BChE
4
6
8
(-)-ethylcarbamate
94 ( 12
4 ( 0.1
(-)-2′-methylphenylcarbamate
(-)-4′-isopropylphenylcarbamate
10 ( 2
760 ( 20
1950 ( 240
50 ( 1
Tetrahydrofurobenzofuran Series
7
9
1
(()-ethylcarbamate
360 ( 22
23 ( 4
6 ( 2
>30,000
27 ( 4
60-fold BChE
>1300-fold AChE
100-fold BChE
(()-2′-methylphenylcarbamate
1
(()-4′-isopropylphenyl carbamate
2650 ( 400
Dihydrobenzodioxepine Series
8
1
1
(()-ethylcarbamate
830 ( 86
56 ( 1
24 ( 5
>30000
22 ( 5
35-fold BChE
>535-fold AChE
195-fold BChE
0
2
(()-2′-methylphenylcarbamate
(()-4′-isopropylphenylcarbamate
4300 ( 350
a
The IC50 data of compounds 1, 2, 15, 16, 18 are from ref 14, and the data of compound 17 are from ref 3. Examples of these were
reanalyzed and closely matched the provided values.
Table 2. Comparison of the IC50 Values of Anticholinesterases of Clinical Interest against Human Erythrocyte AChE and Plasma
BChE
IC50 a ( SEM (nM)
compound
AChE
BChE
selectivity
tacrine (cognex)
190 ( 40
22 ( 8
47.0 ( 10
4150 ( 1700
7300 ( 830
37 ( 5
1560 ( 45
1700 ( 385
5.0 ( 0.1
>10000
4-fold BChE
188-fold AChE
9-fold AChE
donepezil (aricept)
galanthamine (reminyl)
rivastigmine (exelon)
phenserine
800 ( 60
4150 ( 160
22 ( 1.4
125 ( 23
22 ( 2
b
122-fold BChE
70-fold AChE
14-fold AChE
4-fold BChE
2′-ethylphenylgeneserine N-oxide HCL (CHF2819)
heptylphysostigmine (eptastigmine)
huperzine A
47 ( 22
>212-fold AChE
a
b
IC50 values were determined in duplicate on a minimum of four different occasions. Rivastigmine is atypical in that its activity
4
0
against brain-derived AChE is far more potent than against red blood cell derived enzyme. The measured value may therefore noticeably
underestimate its activity in brain, and the agent has been reported to be nonselective between AChE and BChE.
carbamoyl moieties have similar anticholinesterase
in comparison to many of the clinically available anti-
cholinesterases (Table 2), and the enzyme subtype
selectivities achieved by 10 and 12 were superior to
those of the more classical physostigmine series.
Actions on Cellular Viability and Toxicity and
Secretion of Amyloid-â Precursor Protein Levels
in a Neuronal Cell Line. Because phenserine has
demonstrated the ability to lower APP, the precursor
to the putative Alzheimer toxin amyloid-â peptide (Aâ)
in neuronal cell lines,15 representative carbamates 9 and
10 were assessed for actions on APP in human neuro-
blastoma SK-N-SH cells in culture. Cell viability and
toxicity were additionally quantified by MTT assay and
measurement of lactate dehydrogenase levels, respec-
tively, at 48 h. Neither compound altered secreted levels
of APP at the concentrations assessed (10 and 30 µM),
which were well tolerated. As illustrated in Figure 3,
however, there was a significant increase in cell viability
induced by 10 at the lower concentration. This was
validated by quantifying viable cell number by trypan
blue cell count (Figure 3).
1
9-21
activities,
we synthesized the N-ethylcarbamates
(
7, 8, 13, and 14) as the representative of this class.
Ethyl isocyanate is commercially available and safer
than the methyl analogue for transport and handling.
All compounds (7-12) are racemic.
Tetrahydrofurobenzofuran Series. The tetra-
hydrofurobenzofuran carbamates (7, 9, 11) proved to be
potent anticholinesterases. Like physovenine (1), the
ethylcarbamate (7) possessed potent BChE inhibitory
action but surprisingly low AChE activity to provide it
a BChE selectivity of 60-fold. The 2′-methylphenylcar-
bamate, which for both the physostigmine and physo-
venine series (16, 15) provided AChE potency and
selectivity, similarly provided the tetrahydrofuroben-
zofuran series (9) with high AChE potency. Interest-
ingly, the compound was devoid of BChE action at the
highest concentration assessed, 30 µM, providing it a
remarkable and unexpected AChE selectivity. In con-
trast, the 4′-isopropylphenylcarbamate, which for both
the physostigmine and physovenine series (18, 17)
reverses the potency and selectivity to favor BChE,
correspondingly provided the tetrahydrofurobenzofuran
series (11) a high BChE potency.
Discussion
Novel Potent AChE and BChE Inhibitors. Our
previous studies have demonstrated that phenylcar-
bamates in the physostigmine series are long-acting and
well-tolerated cholinesterases inhibitors, with a selectiv-
ity for AChE or BChE that depends on the position of
Dihydrobenzodioxepine Series. The dihydroben-
zodioxepine carbamates (8, 10, 12) mirrored the selec-
tivity described for the tetrahydrofurobenzofurans but
with slightly less potency. This potency was favorable

9
90 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 4
Luo et al.
Figure 4. Superimpositon of compounds 15, 16, 9, and 10.
group of tolserine (16) protrudes away from the rest.
This may explain the enantiomeric selectivity found
with the physostigmine series, with the (+)-enantiomers
being cholinergically inactive, whereas both enanti-
Figure 3. Action of carbamates 9 and 10 on cell viability
upper) as assessed by MTT assay and cell number (lower) as
(
assessed by trypan blue cell count (mean ( standard error of
the mean). 10 significantly elevated cell viability and cell count
1
omers are active for the physosvenine as well as N -
norphysostigmine series.3
,23,26
Because the two novel
(
Dunnett’s t-test).
1
racemic carbamates (7-12) lack a N -methyl group but
are potent anticholinesterases, we predict that their
enantiomeric selectivities should be lower than those
of (-)-physostigmine (2) and analogues. Synthesis of
optically pure enantiomers is ongoing to test this
hypothesis.
phenyl ring substitution.22,23 In this regard, the experi-
mental drug phenserine and it 2′-substituted analogue,
tolserine (16), have demonstrated safety and efficacy in
2
2
clinical and preclinical studies, respectively. On the
basis of the same pharmacophore, the described novel
tetrohydrofurobenzofuran (7, 9, 11) and dihydrobenzo-
dioxepine (8, 10, 12) carbamates possess potent and
clinically relevant activity (Tables 1 and 2). This is
combined with unexpectedly high enzyme subtype se-
lectivity for the substituted phenylcarbamates from both
series and an economy of synthesis for the racemic
compounds that makes them highly attractive agents
for further development as drug candidates.
Chemical Interactions underpinning Activity.
The kinetic analysis of cationic ligands has shown that
the active site of AChE has a large negative charge,27
thus creating a strong dipole along the active site
gorge.28 An electronic steering mechanism has been
29
proposed for ligand binding. However, the crystal
structure of AChE from Torpedo californica has shown
high aromatic content of the active site, suggesting an
7
Enzyme-Inhibitor Binding Interactions. On the
basis of extensive X-ray crystallography and mutagen-
esis studies, AChE appears to have three binding
domains that interact separately or in a combined
aromatic guidance mechanism for ligand binding. The
novel tetrahydrofurobenzofuran (7, 9, 11) and dihy-
drobenzodioxepine (8, 10, 12) carbamates, unlike phy-
sostigmine (2), lack a ring-C amino group to capture a
proton and bear a positive charge. Thus, this novel set
of compounds has a different dipole moment and vector
of the dipole than does physostigmine and related
carbamates.
To assess H-bonding and other binding interaction
between these compounds and AChE, compound 9 and
Torpedo californica AChE (TcAChE, EC 3.1.1.7.) was
modeled (Figure 5). Specifically, the crystal structure
of TcAChE, carbamoylated by the physostigmine ana-
logue, 8-(cis-2,6-dimethylmorpholino)octyleseroline (MF
268), was utilized for this purpose.30 The carbamoyl
moiety was removed from this complex to allow com-
pound 9 to be docked in the active site of AChE (Figure
5).
2
4,25
manner with divergent enzyme inhibitors.
Their
binding affinities are dependent, in part, on their three-
dimensional fit within the substrate gorge of the en-
zyme, alignment with one or more of the binding
domains, and the chemical basis underpinning the
interaction between inhibitor and binding domain.23
Physostigmine analogues access the same two binding
domains as does the natural substrate ACh.6
Table 1 illustrates that the potency and AChE versus
BChE selectivity of the four series of carbamates depend
on the N-substituted group of the carbamoyl moiety.
Specifically, the N-alkyl and N-4′-isopropylphenylcar-
bamates are BChE-selective, whereas the N-2′-methyl-
phenylcarbamate is AChE-selective, irrespective of the
phenol: physovenines (1, 13, 15, 17), physostigmines
As illustrated and similar to studies of ACh,5,31 the
carbonyl group of compound 9 is in the vicinity of the
“oxyanion hole”, the backbone N-H of Gly 118, Gly 119,
and Ala 201. The carbamate N-H interacts with the
His 440 of the catalytic triad (Figure 5). With two points
(N-H and CdO) fixed to the backbone of the binding
domain, the degree of freedom of the N-phenyl group is
tremendously decreased. In summary, this part of each
compound is relatively fixed and tolerates little struc-
tural change without a loss of binding and resulting
inhibitory action.
(
2, 14, 16, 18), tetrahydrofurobenzofuran (7, 9, 11), or
bridged benzodioxepine (8, 10, 12). This is illustrated
in Figure 4 in which the four AChE active compounds,
15, 16, 9 (in a 3a-S configuration), and 10 (in a 5-S
configuration), are superimposed in an optimized con-
formation.
The carbamoyl moieties perfectly overlap, suggesting
a similar inhibitory mechanism and in accord with
similar inhibitory activities, but slight deviations exist
1
in the overlay of the phenols. Specifically the N -methyl

Anticholinesterases
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 4 991
however, is not cholinergically mediated because its (+)-
enantiomer, which lacks anticholinesterase activity,
lowers APP and, as a consequence, Aâ levels.15 Fur-
thermore, the inhibition of cholinergically mediated
biochemical pathways, specifically those involving MAP
kinase and PI kinase (whose activation are associated
with anticholinesterase action and muscarinic receptor
stimulation16 without the loss of action on APP) con-
firmed the disassociation between phenserine’s action
as an anticholinesterase and as an APP inhibitor.15 The
APP lowering activity appears to be post-transcription-
ally mediated via the 5′-untranslated region of APP
mRNA15 through a specific region containing a CAGA
34
box that is conserved in mammals that deposit Aâ .
Like 9 and 10, (-)-physotigmine (2) also lacks this APP
action,35 and the structure-activity relationship under-
pinning it is the focus of ongoing studies.
In contrast, we have never previously observed an
increase in cell viability, as induced by 10, despite the
fact that we routinely perform MTT assays to discrimi-
nate between drug-induced biochemical changes and
toxicity. The mechanism underpinning this activity is
being investigated, and because neither 2 nor prior
carbamates have demonstrated this action, it likely is
noncholinergically mediated.
In summary, we have synthesized and characterized
novel tetrahydrofurobenzofuran (7, 9, 11) and dihydro-
benzodioxepine (8, 10, 12) carbamates that possess
potent and clinically relevant anticholinesterase activ-
ity. These compounds not only are interesting candi-
dates as experimental agents for AD, head trauma,
glaucoma, and nerve gas prophylaxis, for which anti-
cholinesterases have previously demonstrated clinical
utility, but have proven valuable as tools to characterize
the molecular interactions that underpin the classic
cholinesterase inhibitory action of (-)-physostigmine (2),
its ring-B and ring-C congeners, and their clinically
relevant carbamates.
Figure 5. Docking model of compound 9 in the binding
domain of AChE. The amino acid residues are numbered on
the basis of the T. Californica sequence.
The noncationic tricyclic phenol portion is oriented
in the “anionic” site, lined with hydrophobic residues,
Trp 84 and Tyr 330, gaining the binding energy by
hydrophobic interactions. Recently, a X-ray crystal of a
neutral molecule within AChE demonstrated the bind-
ing of a noncationic group within the anionic site, with
cation-π mimicking C-H‚‚‚π interactions.32 This indi-
cates that the structural modification of our tricyclic
structure can be tolerated and will allow specific inter-
actions to differentiate the four series of analogues.2
The o-tolyl group of compound 9 binds in the active site
gorge near the hydrophobic residues Phe 288 and Phe
3,33
2
90, forming stacking interactions. These residues have
been implicated for the substrate selectivity of AChE
and BChE,
2
4,25
thus explaining the observed selectivity
of the compound 9 toward AChE.
The new dioxygen compounds 7-12 are, in reality,
spiroacetals. As such, they can be more readily hydro-
lyzed by acid than can ethers but are extremely resist-
ant to hydrolysis by bases. Their described ex vivo
biological evaluation was undertaken in a pH 8.0 buffer
because this represents the optimal working environ-
Experimental Section
Chemistry. Melting points (uncorrected) were measured
1
13
with a Fisher-Johns apparatus. H NMR and C NMR were
recorded on a Bruker (Bellevica, MA) AC-300 spectrometer.
MS (m/z) data were recorded on a Hewlett-Packard 5973 GC-
MS (CI). HRMS experiments were performed by the UCR Mass
Spectrometry Facility, Department of Chemistry, University
of California. All reactions involving nonaqueous solutions
were performed under an inert atmosphere.
6
,20,22-24
ment for AChE and BChE.
Hence, compounds
7-12 are predicted to be stable prior to interacting with
either cholinesterase enzyme. Their similar anticho-
linesterase activities and selectivities compared to those
of other tricyclic compounds, as exemplified by the
physovenine and physostigmine series (Table 1), support
this hypothesis. However, as spiroacetals, they may be
more supportive of a hydrogen-bonding interaction
within the enzyme. If this occurred, such interaction
could potentially lead to a charge delocalization through
the acetal moiety, a greater ease of protonation of one
of the acetal oxygens, and hydrolysis of the acetal
system to form ring open compounds. Thus, the char-
acter of the spiroacetal may represent a fruitful area
for further characterization in future studies.
(
-)-(3aS)-3a,8-Dimethyl-2,3,3a,8a-tetrahydrofuro[2,3-
b]indol-5-yl N-Ethylcarbamate (13). It was made according
3
to a known procedure for making (-)-physovenine. In this
case, the mole ratio (physovenol/ethyl isocyanate) was 1:3 and
the reaction time was increased to about 3 h (monitored by
TLC). The product was obtained quantitatively as a gum: [R]
97.4° (c 0.2, CHCl
1
-
3
); H NMR (CDCl
3
) δ 6.77 (m, 2H, C4-H
and C6-H), 6.22 (d,1H, C7-H), 5.10 (s, 1H, C8a-H), 4.85 (s,
br, 1H, N-H), 3.90 (m, 1H, C2-H ), 3.40 (m, 1H, C2-H ), 3.30
-), 2.88 (s, 3H, N -CH ), 2.00 (m, 2H, C3-H ),
.40 (s, 3H, C3a-CH ), 1.20 (t, 3H, -CH ); CI-MS (CH ), m/z
277 (MH ). HR-MS m/z calcd for C15 277.1552.
O) C, H, N.
-)-(3aS)-1,3a,8-Trimethyl-1,2,3,3a,8,8a-hexahydro-
a
b
8
(
q, 2H, -CH
2
3
2
1
3
3
4
+
Action on Cellular Viability and Amyloid-â Pre-
cursor Protein Levels. Human SK-N-SH cells in
culture express cholinergic markers, synthesize and
secrete APP, and have proven to be useful in defining
the action of experimental compounds on AD amyloido-
21 2 3
H N O :
H N O ‚ / H
20 2 3 4 2
1
Found: 277.1563. Anal. (C15
(
pyrrolo[2,3,b]indol-5-yl N-Ethylcarbamate (14). It was
1
9
made according to a known procedure by using ethyl iso-
cyanate as alkyl isocyanate. The product was obtained quan-
1
5,16
geneis.
In contrast to (-)-phenserine (10 and 30 µM),
titatively as crystals: mp 92-93 °C; [R] -60.9° (c 0.1, CHCl
3
);
1
5
which lowers secreted APP levels by 50%, 9 and 10
lacked APP action. Such action for (-)-phenserine,
1
H NMR (CDCl ) δ 6.72 (m, 2H, C4-H and C6-H), 6.29 (d,
3
1H, C7-H), 4.85 (s, br, 1H, N-H), 4.10 (s, 1H, C8a-H), 3.00

9
92 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 4
Luo et al.
8
(
q, 2H, -CH
2
-), 2.90 (s, 3H, N -CH
3
), 2.60 (m, 2H, C2-H
2
),
H), 1.47 (s, 3H, C3a-CH
3
), and 1.14 (t, 3H, CH
3
of EtNH-)
1
13
2
.52 (s, 3H, N -CH
3
), 1.90 (m, 2H, C3-H
); CI-MS (CH ), m/z 290 (MH ). HR-
: 290.1869. Found: 290.1865.
2
), 1.45 (s, 3H, C3a-
3
ppm; C NMR (CDCl ) δ 155.2, 153.9, 144.4, 131.9, 120.6,
+
CH
3
), 1.20 (t, 3H, -CH
3
4
115.9, 115.8, 108.4, 67.6, 52.9, 40.1, 35.1, 23.2, and 14.1 ppm;
+
MS m/z calcd for C16
Anal. (C16 ) C, H, N.
-Hydroxy-3-methylbenzofuran-2(3H)-one (3). Pyruvic
H
24
3
N O
2
CI-MS (CH
4
), m/z 264 (MH ), 193, 175, and 72; HR-MS m/z
H
23
3
N O
2
calcd for C14
H
17NO
4
:
263.1158. Found: 263.1162. Anal.
(
14
C H
17N O
4
) C, H, N.
5
1
acid (8.81 g, 0.10 mol) was added to 1,4-cyclohexanedione (11.2
g, 0.10 mol). The reaction mixture was heated to 160-170 °C
and then stirred for 3 h. The mixture was maintained at room
temperature to allow crystallization. Filtration gave a crude
product that then was recrystallized from ethanol to provide
Product 8: H NMR (CDCl
5.69 (d, J ) 1.80 Hz, 1H, C2-H), 4.88 (s, br, 1H, -NH-), 4.17,
3
) δ 6.82-6.67 (m, 3H, Ar-H),
3.70 (AB, Jgem ) 7.20 Hz, 2H, C4-H
EtNH-), 2.19-1.91 (m, 2H, C10-H ), 1.41 (s, 3H, C5-CH
and 1.16 (t, 3H, CH of EtNH-) ppm; C NMR (CDCl
2
), 3.22 (q, 2H, CH
2
of
3
) δ
2
),
1
3
3
3
product 3 as pale-yellow crystals (6.50 g, 40%): mp 172.0-
155.3, 149.3, 145.0, 132.7, 121.7, 117.3, 117.0, 100.8, 84.8, 40.0,
1
+
1
73.0 °C; H NMR (CDCl
3
) δ 6.80-6.51 (m, 3 H, Ph-H), 4.69
36.5, 21.1, 17.2, and 15.5 ppm; CI-MS (CH
4
), m/z 264 (MH ),
(
s, 1H, OH), 3.55 (q, 1H, C3-H), and 1.38 (d, 3H, C3-CH
3
)
246, 220, 193, 175, 149, and 72. HR-MS m/z calcd for C14H -
17
+
1
ppm; CI-MS (CH
4
), m/z 165 (MH ) and 137.
NO
4 4 8 2
: 263.1158. Found: 263.1164. Anal. (C14H17N O ‚ / H O)
C, H, N.
5
-Hydroxy-3-methyl-3-methoxycarbonylmethyleneben-
zofuran-2(3H)-one (4). Under a nitrogen atmosphere, sodium
3a-Methyl-2,3,3a,8a-tetrahydrofuro[2,3-b]benzofuran-
5yl N-(2′-Methylphenyl)carbamate (9) and 4,5-Dihydro-
2,5-methano-5-methyl-1,3-benzodioxepin-7-yl N-(2′-Meth-
ylphenyl)carbamate (10). Under a nitrogen atmosphere, two
small pieces of sodium (∼1 mg per piece) were added to a
solution of 5 and 6 (29.5 mg, 0.153 mmol) in 5 mL of anhydrous
ether at room temperature. The mixture was stirred for 2 min,
and then o-tolyl isocyanate (20.4 mg, 0.153 mmol) was added
in one portion. The reaction was continued for 30 min at room
temperature. Thereafter, 5 mL of water was added and the
ether layer was separated. After the sample was dried over
sodium sulfate and filtered, solvent was removed from the
filtrate by evaporation. The residue was chromatographed on
a preparative TLC silica gel plate (EtOAc/hexane ) 1/3) to
afford products 9 (18.6 mg, 74.9%) and 10 (16.1 mg, 64.4%) as
a colorless gum.
hydride (0.24 g, 0.01 mol) was added to a solution of compound
3
(1.64 g, 0.01 mol) and methyl chloroacetate (1.19 g, 0.01mol)
in 6 mL of dry DMF at 0 °C in portions over 1 h. The mixture
was stirred for another hour at 0 °C and then overnight at
room temperature. The reaction mixture was poured over 25
g of ice and extracted with ether (20 mL × 3). The extracted
solution thereafter was washed with brine (20 mL × 2) and
dried over magnesium sulfate. The final concentrated solution
was maintained at room temperature and crystallized to
provide a crude product that then was recrystallized from
EtOAc to give product 4 as white needlelike crystals (1.09 g,
1
4
6.1%): mp 153.5-155.0 °C; H NMR (CDCl
3
) δ 6.92-6.61 (m,
), 3.02, 2.84
CO), and 1.41 (s, 3H, C3-
) δ 179.7, 170.0, 154.0, 144.9, 132.1,
3
H, Ph-H), 5.07 (s, 1H, OH), 3.44 (s, 3H, OCH
3
(
AB, Jgem ) 18.0 Hz, 2H, C3-CH
2
13
CH
3
) ppm; C NMR (CDCl
3
1
1
14.6, 110.6, 110.2, 51.5, 44.5, 40.9, and 24.7 ppm; CI-MS
Product 9: H NMR (CDCl ) δ 7.84 (s, br, 1H, -NH-),
3
+
(
CH
12
4
), m/z 237 (MH ), 205, and 177; HR-MS m/z calcd for
: 236.0685. Found: 236.0673.
7.32-6.70 (m, 7H, Ar-H), 5.89 (s, 1H, C8a-H), 4.18-3.68 (m,
2H, C2-H ), 2.39-2.05 (m, 2H, C3-H ), 2.31 (s, 3H, Ar-CH ),
C
H
12 5
O
2
2
3
+
5
-Hydroxy-3a-methyl-2,3,3a,8a-tatrahydrofuro[2,3-b]-
benzofuran (5) and 7-Hydroxy-4,5-dihydro-2,5-methano-
and 1.56 (s, 3H, C3a-CH
3
4
) ppm; CI-MS (CH ), m/z 326 (MH ),
4
193 and 134. HR-MS m/z calcd for C19H20NO : 326.1392.
3
5
-methyl-1,3-benzodioxepine (6). Under a nitrogen atmo-
4 8 2
Found: 326.1402. Anal. (C19H19N O ‚ / H O) C, H. N: calcd
sphere, a solution of compound 4 (1.04 g, 4.40 mmol) in 20
4.31; found 3.71.
1
mL of THF was added dropwise to lithium aluminum hydride
Product 10: H NMR (CDCl ) δ 7.87 (s, br, 1H, -NH-),
3
(
0.33 g, 8.70 mmol) in THF (20 mL) at room temperature. The
7.33-6.70 (m, 7H, Ar-H), 5.79 (d, J ) 1.80 Hz, 1H, C2-H),
mixture was stirred for 1 h at room temperature, and then
oxalic acid (1.97 g, 21.90 mmol) was added and the mixture
was stirred for another 0.5 h at room temperature. The
mixture was filtered, and a solid material was partitioned
between the ether and aqueous solution of oxalic acid. There-
after, the combined THF and ether solution was evaporated
to remove solvents. The resulting residue was chromato-
graphed on silica gel (EtOAc/hexane ) 1/3) to provide a
4.22, 3.79 (AB, J
gem 2
) 6.84 Hz, 2H, C4-H ), 2.37-2.01 (m,
2H, C10-H ), 2.31 (s, 3H, Ar-CH ), and 1.51 (s, 3H, C5-CH )
2
3
+
3
ppm; CI-MS(CH ), m/z 326 (MH ) 193 and 134. HR-MS m/z
4
calcd for C H NO : 326.1392. Found: 326.1387. Anal.
1
9
20
4
(C H N O ) C, H. N: calcd 4.31; found 3.78.
1
9
19
4
3
a-Methyl-2,3,3a,8a-tetrahydrofuro[2,3-b]benzofuran-
5
yl N-(4′-Isopropylphenyl)carbamate (11) and 4,5-Di-
hydro-2,5-methano-5-methyl-1,3-benzodioxepin-7-yl N-(4′-
Isopropylphenyl)carbamate (12). Under a nitrogen atmo-
sphere, two small pieces of sodium (∼1 mg per piece) were
added into a solution of compounds 5 and 6 (18.0 mg, 0.094
mmol) in 5 mL of dry ether at room temperature. The mixture
was stirred for 2 min, and then p-isopropylphenyl isocyanate
1
mixture of 5 and 6 as a gum (84.8 mg, 10.0%): H NMR
(
5
CH
3
CDCl ) δ 6.75-6.60 (m, 6 H, Ph-H), 5.85 (s, 1H, C8a-H of
), 5.78(d, J ) 1.80 Hz, 1H, C2-H of 6), 1.55 (s, 3H, C3a-
3
of 5), 1.50 (s, 3H, C5-CH
3
of 6) ppm; GC-MS (CI): two
+
peaks (1:1), m/z 193 (MH ), respectively. It was directly used
as a reactant in the following reactions without separation.
(
15.1 mg, 0.094 mmol) was added in a single portion. The
3
a-Methyl-2,3,3a,8a-tetrahydrofuro[2,3-b]benzofuran-
reaction was continued for a further 45 min at room temper-
ature, and then 3 mL of water was added and the ether layer
was separated. After drying over sodium sulfate and filtering,
the filtrate was subjected to evaporation to remove solvent.
The resulting residue was chromatographed on a preparative
TLC silica gel plate (EtOAc/hexane ) 1/3) to afford separate
products 11 (11.6 mg, 70.0%) and 12 (10.0 mg, 60.0%) as
colorless gums.
5
5
-yl N-Ethylcarbamate (7) and 4,5-Dihydro-2,5-methano-
-methyl-1,3-benzodioxepin-7-yl N-Ethylcarbamate (8).
Under a nitrogen atmosphere, two small pieces of sodium (∼1
mg per piece) were added to a solution of compounds 5 and 6
(
32.0 mg, 0.166 mmol) in 5 mL of dry ether at room temper-
ature. The mixture was stirred for 2 min, and then ethyl
isocyanate (24.8 mg, 0.349 mmol) was added in one portion.
The reaction was continued for another 1 h at room temper-
ature, then 5 mL of water was added, and the ether layer was
separated. After the sample was dried over sodium sulfate and
filtered, the filtrate was evaporated to provide a residue that
then was chromatographed (preparative TLC, silica gel,
EtOAc/hexane ) 1/3). This latter procedure was repeated on
three occasions to afford products 7 (11.3 mg, 51.7%) and 8
Product 11. The gumlike material then was crystallized
from the mixed solvents of EtOAc and hexane to provide cubic
1
transparent crystals: mp 156-158 °C; H NMR (CDCl
3
) δ
7
4
.31-6.71 (m, 8H, Ar-H and -HN-), 5.80 (s, 1H, C8a-H),
.09-3.59 (m, 2H, C2-H
CHMe ), 2.19-1.96 (m, 2H, C3-H
and 1.18 (d, J ) 5.24 Hz, 6H, -CHMe
2
), 2.82 (septet, J ) 5.24 Hz, 1H,
-
2
2
), 1.49 (s, 3H, C3a-CH
) ppm; CI-MS(CH
m/z 354 (MH ). HR-MS m/z: calcd for C21
3
),
),
2
4
(
13.2 mg, 60.4%) as colorless gums.
+
H
24NO
4
: 354.1705.
1
Product 7: H NMR (CDCl
3
) δ 6.87-6.60 (m, 3H, Ar-H),
Found: 354.1716. Anal. (C21H23N O ) C. H: calcd 6.56; found
4
5
.77 (s, 1H, C8a-H), 4.89 (s, br, 1H, NH), 4.06-3.56 (m, 2H,
7.05. N: calcd 3.96; found 4.56. Product 12. The gumlike
C2-H), 3.23 (q, 2H, CH
2
of EtNH-), 2.13-1.91 (m, 2H, C3-
material was crystallized from mixed solvents of EtOAc and

Anticholinesterases
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 4 993
-
5
hexane to afford needle-like transparent crystals: mp 149-
(by the addition of 1 × 10 M physostigmine (2)), and the
associated alteration in absorbance was subtracted from that
observed through the concentration range of each test com-
pound. Each agent was analyzed on four separate occasions
and assayed alongside physostigmine (2) as a control and
1
1
50 °C; H NMR (CDCl
3
) δ 7.30-6.72 (m, 8H, Ar-H and
-
)
HN-), 5.71 (d, J ) 1.80 Hz, 1H, C2-H), 4.17, 3.70 (AB, Jgem
5.40 Hz, 2H, C4-H ), 2.81 (septet, J ) 5.22 Hz, 1H, CHMe ),
), 1.48 (s, 3H, C5-CH ), and 1.19
CCH ) ppm; CI-MS (CH ), m/z 354
MH ); HR-MS m/z calcd for C21 : 354.1705. Found:
O) C, H, N.
X-ray Crystallography. Single-Crystal X-ray Analysis
of 11 and 12. A colorless crystal of 11 with dimensions 0.22
2
2
2
.19-1.94 (m, 2H, C10-H
2
3
(
(
d, J ) 5.22 Hz, 6H, CH
3
3
4
external standard whose activity we have previously re-
+
3,7,16-19,23
H
2
24NO
4
ported.
The enzyme activity at each concentration of
3
354.1695. Anal. (C21
H
23N O
4
‚ /
4
H
test compound was expressed as a percent of the activity in
the absence of compound. This was transformed into a logit
format (logit ) ln{(% activity)/[100 - (% activity)} and then
was plotted as a function of its log concentration. Inhibitory
activity was calculated as an IC50, defined as the concentration
of compound (nM) required to inhibit 50% of enzymatic activity
and determined from a correlation between log concentration
and logit activity. Only results obtained from correlation
36b
mm × 0.26 mm × 0.26 mm was mounted on glass fiber using
a small amount of epoxy. A colorless crystal of 12 was mounted
in an identical manner and had the dimensions of 0.18 mm ×
3
6b
0
.21 mm × 0.40 mm. Data for compounds 11 and 12 were
collected on a Bruker three-circle platform diffractometer
equipped with a SMART 6000 CCD detector. Both crystals
were irradiated using a rotating anode Cu KR source (λ )
2
coefficients of r > -0.98 were considered acceptable. Studies
that did not obtain this threshold were repeated.
1
.541 78) with incident beam G o¨ bel mirrors. Data collection
Quantitation of Proliferation and Action on â-Amyl-
oid Precursor Protein (APP) Levels. Human neuroblas-
toma SK-N-SH cells were seeded onto 60 mm culture plates
supplemented with 10% FBS (Gibco/InVitrogen) in MEM Eagle
was performed, and the unit cell was initially refined by using
3
6a
SMART [version 5.625].
Compounds 11 and 12 were both nonmerohedral twins with
the approximate ratios of 70:30 and 75:25, respectively. This
required an orientation matrix to be found for each component.
This was done, and the data reduction for both components
(
Mediatech/Cellgro) and grown to 80% of confluence. Com-
pounds 9 and 10 were dissolved in 100% DMSO to a concen-
tration of 0.01 M and were further diluted with media to obtain
final treatment levels of 10 and 30 µM. A final sample size of
3
6b
was performed using SAINT [verson 6.36A]
and XPREP
36c
[
verson 6.12]. Corrections were applied for Lorentz polariza-
2
0 plates and a treatment/control number of 4 per group were
utilized as a study design: vehicle control (0.3% DMSO) and
and 10 both at 10 and 30µM. At the initiation of the study,
3
6d
tion and absorption effects using TWINABS [version 1.05].
Each structure was solved and refined with the aid of the
9
3
6e
SHELXTL-plus [verson 6.10] system of programs. The full-
low serum media (0.05% FBS in MEM Eagle) and compound
or control vehicle was added to cells. At 48 h thereafter, the
conditioned media and cell pellet were collected for analysis.
Specifically, (i) a MTT assay was performed on live cells to
measure cell viability at the 48 h harvest time (n ) 4). (ii)
Levels of lactate dehydrogenase (LDH) were assayed in the
2
matrix least-squares refinement on F included atomic coor-
dinates and anisotropic thermal parameters for all non-H
atoms. The H atoms were included using a riding model. The
final refinements included the structure factors from both twin
components.
The two racemates of compound 11 occupy the same
crystallographic site in the lattice. This appears as a super-
position of the two molecules resulting in a disorder around
C3A and C8A. The disorder was modeled with a 50:50 ratio of
occupancies between the racemates.
conditioned media for assessment of cellular toxicity (n )
37,38
4).
(iii) A small aliquot of cells from each group was
removed and stored on ice for cell counting by the trypan blue
method (n ) 1) as a secondary measure of cellular viability,
and (iv) conditioned media samples were subjected to Western
blot analysis using a primary antibody against total secreted
APP (sAPP). This was undertaken on a 10% SDS-PAGE gel,
and samples (n ) 4 per treatment or control group) were
loaded at equal protein (30 µg), based on the Bradford
microassay protein estimation. Primary antibodies were 22C11
Quantitation of Anticholinesterase Activity. The action
of compounds 7-14 to inhibit the ability of freshly prepared
human AChE and BChE to enzymatically degrade their
respective specific substrates, acetyl(â-methyl)thiocholine and
s-butyrylthiocholine (0.5 mmol/L) (Sigma Chemical Co., St.
3
,7,16-19,23
Louis, MO), was quantified.
Samples of AChE and
(
APP) at 1:500 dilution and B-actin at 1:2000 dilution, both
probed for 3 h at room temperature.
BChE were derived from whole red blood cells and plasma,
respectively. Compounds were dissolved in Tween-80/EtOH 3:1
Computer-Aided Molecular Modeling. Computer-aided
(
v/v; <150 µL total volume) and were diluted in 0.1 M
molecular modeling was undertaken using Sybyl [version 6.9
Na PO buffer (pH 8.0) in half-log concentrations to provide a
3
4
(
Tripos Inc.)]. The atomic coordinates of the carbamoylated
final concentration range that spanned 0.3 nM to 30 µM.
Tween-80/EtOH was diluted to in excess of 1 in 5000, and no
inhibitory action on either AChE or BChE was detected in
separate prior experiments.
AChE, obtained from the Protein Data Bank (PDB entry:
3
0
1
oce), were used for docking studies. All hydrogens were
added with the biopolymer module after removing the crystal-
lographic water and the carbamoylated part, and atomic
positions were refined by energy minimization. The active site
residue Phe 330 was replaced with Tyr with similar conforma-
For the preparation of BChE, freshly collected blood was
centrifuged (10000g, 10 min, 4 °C) and plasma was removed
3 4
and diluted 1:125 with 0.1 M Na PO buffer (pH 7.4). For
3
9
tion to match the active site of human AChE. Thereafter,
compound 9 was manually docked with Ser 200 and residues
of the oxyanion hole as the guide. The active site (residues
within 10 Å radius of compound 9) of the complex was energy-
minimized by the conjugate gradient method with a distance-
dependent dielectric constant of 4.
AChE preparation, whole red blood cells were washed five
times in isotonic saline, lysed in 9 volumes of 0.1 M Na PO
3
4
buffer (pH 7.4) containing 0.5% Triton-X (Sigma) and then
were diluted with an additional 19 volumes of buffer to a final
dilution of 1:200.
Analysis of anticholinesterase activity was undertaken by
utilizing a 25 µL sample of each enzyme preparation and was
undertaken at their optimal working pH, 8.0, in 0.1 M
Na PO buffer (0.75 mL total volume). Compounds were pre-
3 4
incubated with enzymes (30 min at room temperature) and
then were incubated with their respective substrates and with
Acknowledgment. The authors are grateful to the
Medicinal Chemistry Section, National Institute on
Drug Abuse, NIH, for use of chemical characterization
and molecular modeling equipment. S.S.K. was sup-
ported by a National Institutes of Health Visiting
Fellowship.
5
,5′-dithiobis-2-nitrobenzoic acid (25 min, 37° C). Substrate-
enzyme interaction was immediately halted by the addition
-
5
of excess enzyme inhibitor (physostigmine 1 × 10 M), and
production of a yellow thionitrobenzoate anion was then
measured by spectrophotometer at 412 nm wavelength. To
correct for nonspecific substrate hydrolysis, aliquots were
co-incubated under conditions of absolute enzyme inhibition
Supporting Information Available: Elemental analysis,
NMR, and crystallographic data. This material is available free
of charge via the Internet at http://pubs.acs.org.

9
94 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 4
Luo et al.
References
(23) Yu, Q. S.; Holloway, H. W.; Flippen-Anderson, J. L.; Hoffman,
B.; Brossi, A.; Greig, N. H. Methyl analogues of the experimental
Alzheimer drug phenserine: synthesis and structure/activity
relationship for acetyl- and butyrylcholinesterase inhibitory
action. J. Med. Chem. 2001, 44, 4062-4071.
(
1) (a) Marion, L. Alkaloids 1952, 2, 438. (b) Coxworth, E. Alkaloids
1
965, 8, 27. (c) Robinson, B. Alkaloids 1967, 10, 383. (d)
Robinson, B. Alkaloids 1971, 13, 213. (e) Takano, S.; Ogasawara,
K. Alkaloids 1989, 36, 889.
(
24) Giacobini, E. Cholinesterase and Cholinesterase Inhibitors;
(
2) Dale, F. J.; Robinson, B. The synthesis and anti-acetylcholinest-
erase activities of (+)-physostigmine and (+)-physovenine. J.
Pharm. Pharmacol. 1970, 22, 889-896.
Martin Dunitz: London, 2000.
(25) Giacobini, E. Butyrylcholinesterase, Its Function and Inhibitors;
(
3) Yu, Q. S.; Liu, C.; Brzostowska, M.; Chrisey, L.; Brossi, A.; Greig,
N. H.; et al. Physovenines: Efficient Synthesis of (-)- and (+)-
Physovenine and Synthesis of Carbamate Analogues of (-)-
Physovenine. Anticholinesterase Activity and Analgesic Prop-
erties of Optically Active Physovenines. Helv. Chim. Acta 1991,
Martin Dunitz: London, 2003.
26) Yu, Q. S.; Luo, W.; Holloway, H. W.; Utsuki, T.; Perry, T. A.;
(
1
Lahiri, D. K.; Greig, N. H.; Brossi, A. Racemic N -norphenserine
and its enatiomers: unpredicted inhibition of human acetyl- and
butylylcholinesterase and â-amyloid precursor protein in vitro.
Heterocycles 2003, 619, 529-539.
74, 761-766.
(
4) Lahiri, D. K.; Farlow, M. R.; Sambamurti, K.; Greig, N. H.;
Giacobini, E.; Schneider, L. S. A critical analysis of new
molecular targets and strategies for drug developments in
Alzheimer’s disease. Curr. Drug Targets 2003, 4, 97-112.
5) Kaplan, D.; Barak, D.; Ordentlich, A.; Kronman, C.; Velan, B.;
Shafferman, A. Is aromaticity essential for trapping the catalytic
histidine 447 in human acetylcholinesterase? Biochemistry 2004,
(27) Nolte, H. J.; Rosenberry, T. L.; Neumann, E. Effective Charge
on Acetylcholinesterase Active Sites Determined from the Ionic
Strength Dependence of Association Rate Constants with Cat-
ionic Ligands. Biochemistry 1980, 19, 3705-3711.
(
(
28) Ripoll, D. R.; Faerman, C. H.; Axelsen, P. H.; Silman, I.;
Sussman, J. L. An Electrostatic Mechanism for Substrate
Guidance down the Aromatic Gorge of Acetylcholinesterase.
Proc. Natl. Acad. Sci. U.S.A. 1993, 90, 5128-5132.
43, 3129-36.
(
6) Soreq, H.; Zakut, H. Human Cholinesterases and Anticholinest-
erases; Academic Press: New York, 1993.
7) Sussman, J. L.; Harel, M.; Frolow, F.; Oefner, C.; Goldman, A.;
Toker, L.; Silman, I. Atomic Structure of Acetylcholineesterase
from Torpedo californica: A Prototypic Acetylcholine-Binding
Protein. Science 1991, 253, 872-879.
(
(29) Tan, R. C.; Truong, T. N.; McCammon, J. A.; Sussman, J. L.
Acetylcholinesterase: Electrostatic Steering Increases the Rate
of Ligand Binding. Biochemistry 1993, 32, 401-403.
(
(
(
(
(
30) Bartolucci, C.; Perola, E.; Cellai, L.; Brufani, M.; Lamba, D.
Back Door” Opening Implied by the Crystal Structure of
Carbamoylated Acetylcholinesterase. Biochemistry 1999, 38,
714-5719.
31) Gilson, M. K.; Straatsma, J. A.; McCammon, J. A.; Ripoll, D.
R.; Faerman, C. H.; Axelsen, P. H.; Silman, I.; Sussman, J. L.
Open “Back Door” in a Molecular Dynamics Simulation of
Acetylcholinesterase. Science 1994, 263, 1276-1278.
32) Koellner, G.; Steiner, T.; Millard, C. B.; Silman, I.; Sussman, J.
L. A Neutral Molecule in a Cation-Binding Site: Specific Binding
of a PEG-SH to Acetylcholinesterase from Torpedo californica.
J. Mol. Biol. 2002, 320, 721-725.
33) Pei, X. F.; Yu, Q. S.; Holloway, H. W.; Brossi, A.; Greig, N. H.
Syntheses and biological evaluation of ring-C opened analogues
of the cholinesterase inhibitors physostigmine, phenserine and
cymserine. Med. Chem. Res. 1999, 9, 50-60.
34) Lahiri, D. K.; Alley, M.; Farlow, H.; Rogers, J. T.; Greig, N. H.
Physostigmine and its analogue phenserine have different effects
of Alzheimer’s amyloid-beta precursor protein. In Progress in
Alzheimer’s and Parkinson’s Disease; Hanin, I., Fisher, A., Eds.,
in press.
(
8) Data were calculated by using software Pallas 3.0 produced by
CompuDrug International Inc., 705 Grandniew Drive, South San
Francisco, CA 94708.
“
5
(
9) He, X. S.; Greig, N. H.; Broosi, A. Thiaphysovenine and carbam-
ate analogues: a new class of potent inhibitors of cholinesterase.
A. Med. Chem. Res. 1992, 2, 229-237.
(
10) B u¨ chi, G.; Weinreb, M. Total syntheses of aflatoxin M
1
and G
. J. Am. Chem. Soc.
1
and an improved synthesis of aflatoxin B
1
1
971, 93, 746-752.
(
11) Civitello, E. R.; Rapoport, H. Synthesis of the Enantiomeric
Furobenzofurans, Late Precursors for the Synthesis of (+)- and
(
-)-Aflatoxins B
1 2 1 2
, B , G and G . J. Org. Chem. 1994, 59, 3775-
3
782.
(
12) Graybill, T. L.; Casillas, E. G.; Pal, K.; Townsend, C. A. Silyl
Triflate-Mediated Ring-Closure and Rearrangement in the
Synthesis of Potential Bisfuran-Containing Intermediates of
Aflatoxin Biosynthesis. J. Am.Chem. Soc. 1999, 121, 7729-7746.
13) Ozaki, Y.; Quan, Z. S.; Watabe, K. (n e´ e Okamura); Kim, S. W.
Synthesis of 5-hydroxyoxaindan-2-ones and indol-5-ols from 1,4-
cyclohexanedione. Heterocycles 1999, 51, 727-731.
(
(
(
14) Harley-Mason, J.; Jackson, A. H. Hydroxytryptamine. Part II.
A New Synthesis of Physostigmine. J. Chem. Soc. 1954, 3651-
3654.
15) Shaw, K. T. Y.; Rogers, J.; Utsuki, T.; Yu, Q. S.; Sambamurti,
K.; Brossi, A.; Ge, Y. W.; Lahiri, D. K.; Greig, N. H. Phenserine
regulates traslation of â-amyloid precursor protein mRNA by a
putative interleukin-1 responsive element: a novel target for
drug development. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 7605-
(35) Maloney, B.; Ge, Y.; Greig, N. H.; Lahiri, D. K. Characterization
of the 5′-untranslated region of APP mRNA: implications in drug
development for Alzheimer’s disease. FASEB J. 2004, 18, 1288-
1
290.
(
36) Computer software used for X-ray crystallography: (a) . SMART,
version 5.625; Bruker AXS Inc.: Madison, WI, 2001. (b) SAINT,
version 6.36A; Bruker AXS Inc.: Madison, WI, 2002. (c) XPREP,
version 6.12; Bruker AXS Inc.: Madison, WI, 2001. (d) TWIN-
ABS, version 1.05; Bruker AXS Inc.: Madison, WI, 2003. (e)
SHELXTL, version 6.10; Bruker AXS Inc.; Madison, WI, 2000.
7610.
(
(
(
16) Sambamurti, K.; Greig, N. H.; Lahiri, D. K. Advances in the
cellular and molecular biology of the beta-amyloid protein in
Alzheimer’s disease. NeuroMol. Med. 2002, 1, 1-31.
17) Pei, X. F.; Bi, S.; Brossi, A. Reduction of oxindoles with sodium
bis(2-methoxyethoxy) aluminum hydride, a novel reducing agent.
Heterocycles 1994, 39, 357-360.
(37) Lahiri, D. K.; Farlow, M. R.; Sambamurti, K. The Secretion of
Amyloid-beta Peptides Is Inhibited in Tacrine-Treated Human
Neuroblastoma Cells. Mol. Brain Res. 1998, 62, 131-140.
18) Brzostowska, M.; He, X. S.; Greig, N. H.; Rapoport, S.; Brossi,
A. Phenylcarbamates of (-)-Eceroline and (-)-Physovenol: Se-
lective Inhibition of Acetyl and/or Butyrylcholinesterases by
Phenylcarbamates. Med. Chem. Res. 1992, 2, 238-246.
19) Yu, Q. S.; Atack, J. A.; Rapoport, S.; Brossi, A. Carbamate
analogues of (-)-physostigmine: in vitro inhibition of acetyl- and
butyrylcholinesterase. FEBS Lett. 1988, 234, 127-130.
20) Atack, J. R.; Yu, Q. S.; Soncrant, T. T.; Brossi, A.; Rapoport, S.
Comparative Inhibitory Effects if Variou Physostigmine Analogs
against Acetyl- and Butyrylcholinesterases. J. Pharmacol. Exp.
Ther. 1989, 249, 194-249.
(
38) Zhu, X.; Yu, Q. S.; Cutler, R. G.; Culmsee, C.; Holloway, H. W.;
Mattson, M. P.; Greig, N. H. Novel p53 inhibitors with neuro-
protective action: syntheses and pharmacological evaluation of
2-imino-2,3,4,5,6,7-hexahydrobenzothiazoleand2-imino-2,3,4,5,6,7-
hexahydrobenzoxazole derivatives. J. Med. Chem. 2002, 45,
5090-5097.
(
(
(39) Yamamoto, Y.; Ishihara, Y.; Kuntz, I. D. Docking Analysis of a
Series of Benzylamino Acetylcholinesterase Inhibitors with a
Phthalimide, Benzoyl, or Indanone Moiety. J. Med. Chem. 1994,
(
21) Pei, X. F.; Greig, N. N.; Bi, S.; Broosi, A.; Toome, V. Inhibition
of human acetylcholinesterase by unnatural (+)-(3aR)-N1-nor-
physostigmine and arylcarbamate analogues. Med. Chem. Res.
3
7, 3141-3153.
(
40) Enz, A.; Amstutz, R.; Boddeke, H.; Gmelin, G.; Malanowski, J.
Brain selective inhibition of acetylcholinesterase: a novel ap-
proach to therapy in Alzheimer’s disease. Prog. Brain Res. 1993,
98, 431-438.
1
995, 5, 265-270.
(
22) Greig, N. H.; DeMicheli, E.; Utsuki, T.; Holloway, H. W.; Yu, Q.
S.; Perry, T. A.; Brossi, A.; Deutsch, J.; Ingram, D. K.; Lahiri,
D. K.; Soncrant, T. The experimental Alzheimer drug
phenserine: pharmacodynamics and kinetics in the rat. Acta
Neurol. Scand. 2000, 102, 74-84.
JM049309+