First gallamine-tacrine hybrid: Design and characterization at cholinesterases and the M2 muscarinic receptor

Article abstract of DOI:10.1021/jm070859s

Gallamine and tacrine are allosteric antagonists at muscarinic M ₂ acetylcholine receptors and inhibitors of acetylcholinesterase. At both acetylcholine-binding proteins, gallamine and tacrine are known to occupy two different binding sites: in M₂ receptors within the allosteric binding area and in acetylcholinesterase at its catalytic and its peripheral site. To find new ligands of both targets, we designed a gallamine-tacrine dimer and several derived hybrid compounds to address the two binding sites. Their M₂ receptor allosteric and acetylcholinesterase inhibitory potential was determined. The hybrid compounds revealed an allosteric potency in the low nanomolar range exceeding the allosteric potency of gallamine and tacrine by factors of 100 and 4800, respectively. Cholinesterase inhibition was augmented by hybrid formation, and all compounds exhibited IC₅₀ values in the lower nanomolar range. Thus, gallamine-tacrine hybrid formation is a valuable approach toward high affinity ligands concurrently targeting these acetylcholine-binding proteins.

Full text of DOI:10.1021/jm070859s

J. Med. Chem. 2007, 50, 5685-5695
5685
First Gallamine-Tacrine Hybrid: Design and Characterization at Cholinesterases and the M2
Muscarinic Receptor
Paul W. Elsinghorst,† Julia S. Cieslik,‡ Klaus Mohr,‡ Christian Tra¨nkle,‡ and Michael Gu¨tschow*,†
Pharmaceutical Institute, Pharmaceutical Chemistry I, UniVersity of Bonn, An der Immenburg 4, 53121 Bonn, Germany, and
Pharmaceutical Institute, Pharmacology and Toxicology, UniVersity of Bonn, Gerhard-Domagk-Str. 3, 53121 Bonn, Germany
ReceiVed July 16, 2007
Gallamine and tacrine are allosteric antagonists at muscarinic M2 acetylcholine receptors and inhibitors of
acetylcholinesterase. At both acetylcholine-binding proteins, gallamine and tacrine are known to occupy
two different binding sites: in M2 receptors within the allosteric binding area and in acetylcholinesterase at
its catalytic and its peripheral site. To find new ligands of both targets, we designed a gallamine-tacrine
dimer and several derived hybrid compounds to address the two binding sites. Their M2 receptor allosteric
and acetylcholinesterase inhibitory potential was determined. The hybrid compounds revealed an allosteric
potency in the low nanomolar range exceeding the allosteric potency of gallamine and tacrine by factors of
100 and 4800, respectively. Cholinesterase inhibition was augmented by hybrid formation, and all compounds
exhibited IC50 values in the lower nanomolar range. Thus, gallamine-tacrine hybrid formation is a valuable
approach toward high affinity ligands concurrently targeting these acetylcholine-binding proteins.
Chart 1. Two Building Blocks, Gallamine and Tacrine, To Be
Linked by a Suitable Spacer
Introduction
Acetylcholine is the linchpin of neurotransmission in para-
sympathetically innervated target cells, autonomic ganglia, the
neuromuscular junction and the CNS as it carries the nervous
signal across the synaptic cleft and to the final effector cells.
The level of acetylcholine inside the synaptic cleft is therefore
subject to sensitive regulation mechanisms. Released at the
presynaptic terminal, acetylcholine diffuses toward the postsy-
naptic side. There it may either bind to muscarinic receptors
that are G-protein coupled or to nicotinic acetylcholine receptors,
representing a group of ion channels. To terminate the resulting
signal, the synaptic cleft needs to be freed of acetylcholine, a
task that is mainly accomplished by acetylcholinesterase, a serine
hydrolase of therapeutic importance.1-3 Tacrine is a prototypal
cholinesterase inhibitor, but it is also an allosteric modulator
of muscarinic receptors. Gallamine is an archetypal muscarinic
allosteric agent but at high concentrations it also inhibits
cholinesterases.
At both targets, tacrine and gallamine can bind at topographi-
cally distinct places. A previous study on acetylcholinesterase
inhibition by tacrine-trimethoxybenzene heterodimers4 was
based on the dual binding site model, that involves the peripheral
and the active site.5,6 While tacrine commonly binds to the active
site of acetylcholinesterase, the trisquaternary phenolic ether
gallamine is a long known ligand of the peripheral site.
Since its discovery7 in 1947, gallamine has been used as a
muscle relaxant (Flaxedil) and was additionally found to
allosterically modulate muscarinic receptors.8 Among the five
muscarinic receptor subtypes classified until today, gallamine
exhibits a selectivity for the M2 subtype.9 Gallamine belongs
to the group of typical allosteric modulators that bind in the
core region of the allosteric site.10,11 Tacrine is an atypical
muscarinic allosteric agent.12,13 Recently, the atypical molecular
interaction of tacrine at M2 receptors was elucidated by applying
a tacrine homodimer.14 The homodimer had a higher allosteric
affinity than the monomer. This finding suggested that the
allosteric site has room to accommodate two tacrine molecules
simultaneously, one in the core region of the allosteric site and
one in its periphery.14
Inspired by this analogy of two distinct binding sites for
gallamine and tacrine at both, acetylcholinesterase and the
allosteric site of the M2 muscarinic receptor, we envisaged
hybrids of tacrine and gallamine or gallamine-related moieties
(Chart 1) held together by a previously established linker.4 The
commonly utilized diaminoalkane linker of seven or eight
methylene groups5 was not suitable to carry the gallamine unit.
However, our linker was chosen to stretch a comparable
distance, bringing the gallamine moiety close to the peripheral
binding site. These novel hybrid compounds were thought to
exhibit increased affinity toward the M2 muscarinic receptor
and acetylcholinesterase by simultaneously addressing the
respective tacrine and gallamine binding sites.
All compounds were evaluated as inhibitors of several
acetylcholinesterases with IC50 values in the nanomolar con-
centration range. Comparable potency was observed for human
butyrylcholinesterase, whose inhibition is expected to provide
additional benefits in Alzheimer’s disease.3,15 The allosteric
potential of the hybrids was assessed by their ability to bind to
M2 receptors whose acetylcholine binding site was blocked by
the radiolabeled antagonist [3H]N-methyl-scopolamine ([3H]-
NMS). In comparison to their building blocks, gallamine and
* To whom correspondence should be addressed. Phone: +49 228
732317. Fax +49 228 732567. E-mail: guetschow@uni-bonn.de.
† Pharmaceutical Chemistry I.
‡ Pharmacology and Toxicology.
10.1021/jm070859s CCC: $37.00 © 2007 American Chemical Society
Published on Web 10/18/2007

5686 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 23
Elsinghorst et al.
Scheme 1. Synthesis of the Mono- and Bisquaternary Gallamine-Tacrine Hybrids 13-16a
a Reagents and conditions: (a) Ac2O, Py, RT, 1 h; (b) (1) (COCl)2, CH2Cl2, RT, 2 h; (2) H2N(CH2)5COOEt‚HCl, DIEA, CH2Cl2, RT, 1 h; (3) K2CO3,
MeOH, RT, 12 h; (c) K2CO3, 18-crown-6, (CH2Br)2, 80 °C, 36 h; (d) Et3N, MeNO2, 60 °C, 24 h; (e) N2H4, EtOH, 80 °C, 72 h; (f) 6,9-dichloro-/9-chloro-
1,2,3,4-tetrahydroacridine, EtOH, 140 °C, 24 h.
tacrine, the hybrids behaved as M2 receptor modulators of
remarkably high potency.
valuable method to obtain these quaternary intermediates as well
as the final compounds as solid substances. The mono- and
bisquaternary carboxylic esters 9 and 10 were then converted
into their corresponding carboxylic hydrazides 11 and 12 by
reaction with aqueous hydrazine. The reaction conditions
established in the course of previous investigations4 had to be
adjusted to achieve an almost complete reaction of 9 and 10.
Only complete conversion to 11 and 12 allows their collection
by precipitation from ethyl acetate without further purification.
The final seventh step was the introduction of a tacrine moiety
by treatment with either 9-chloro- or 6,9-dichloro-1,2,3,4-
tetrahydroacridine in sealed reaction vessels. The desired
compounds 13-16 were obtained as amorphous yellow powders
of notable hygroscopicity in overall yields between 20 and 36%.
Following this procedure, efforts were undertaken to acquire
the envisaged gallamine-tacrine heterodimers 24 and 25
(Scheme 2). Preliminary investigations concerning the alkylation
of methyl 3,4,5-trihydroxybenzoate with 1,2-dibromoethane
unfortunately revealed the almost quantitative formation of a
benzo[b][1,4]dioxine (Chart 2). To avoid this ring closure, the
para-hydroxy group had to be selectively protected, and a two-
step alkylation process was employed. A benzyl protection was
considered the most practical solution as its removal conditions
are suitable for quaternary compounds. Thus, starting from 3,4,5-
trihydroxybenzoic acid (17), the first two steps were again the
acetyl protection of all three phenolic groups to give 18 and
Results and Discussion
Synthesis. To gain a straightforward access toward com-
pounds 24 and 25 shown in Scheme 2 and to study the
importance of the substituents in the gallamine scaffold, the
synthetic route was initially established for the simplified
compounds 13-16 (Scheme 1). Starting from hydroxy-
substituted benzoic acids 1 and 2, the anticipated linker was
introduced at first. A prerequisite was the protection of the
phenolic groups to allow activation of the carboxylic acid for
coupling with ethyl 6-aminohexanoate. Thus, 4-hydroxybenzoic
acid (1) and 3,5-dihydroxybenzoic acid (2) were acetyl-protected
by treatment with acetic anhydride in pyridine. Both 4-acetoxy-
benzoic acid (3) and 3,5-bis(acetoxy)benzoic acid (4) were
subsequently reacted with ethyl 6-aminohexanoate by classical
acyl chloride activation. Selective deprotection was achieved
using potassium carbonate in methanolic solution to obtain
intermediates 5 and 6. During the fourth step, the phenolic
groups were alkylated with 1,2-dibromoethane following a
procedure of Sarkar et al.16 to yield the 2-bromoethoxy-
substituted compounds 7 and 8. These were quaternized with
triethylamine, and bromide salts 9 and 10 precipitated when a
concentrated solution in anhydrous ethanol was added to ethyl
acetate. The precipitation from ethyl acetate proved to be a

Gallamine-Tacrine Hybrid
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 23 5687
Scheme 2. Synthesis of the Trisquaternary Gallamine-Tacrine Hybrids 24 and 25a
a Reagents and conditions: (a) Ac2O, Py, RT, 1 h; (b) (1) (COCl)2, CH2Cl2, RT, 2 h; (2) H2N(CH2)5COOEt‚HCl, DIEA, CH2Cl2, RT, 1 h; (3) BnCl,
K2CO3, KI, Me2CO, reflux, 18 h; (4) K2CO3, EtOAc, MeOH, H2O, reflux, 1 h; (c) K2CO3, 18-crown-6, (CH2Br)2, 80 °C, 36 h; (d) (1) Et3N, MeNO2,
60 °C, 24 h; (2) H2, Pd/C, MeOH, 6 h; (3) K2CO3, 18-crown-6, (CH2Br)2, MeCN, 80 °C, 36 h; (e) Et3N, MeCN, 100 °C, 48 h; (f) N2H4, EtOH, 80 °C, 72
h; (g) 6,9-dichloro-/9-chloro-1,2,3,4-tetrahydroacridine, EtOH, 140 °C, 24 h.
Chart 2. Protection of the para-Hydroxy Group is Essential
Because Direct Alkylation of Gallic Acid Derivatives with
1,2-Dibromoethane Leads Benzo[b][1,4]dioxins
cooling. The following hydrazide formation and coupling of
23 with 6-chloro- or 6,9-dichloro-1,2,3,4-tetrahydroacridine
proceeded smoothly, and the final compounds 24 and 25 were
obtained after 11 steps in 15% overall yield each.
Molecular Docking and Cholinesterase Inhibition. Possible
binding modes of the quaternary target compounds 13, 15 and
24 were investigated using the AutoDock software suite.18 The
1ZGB19 crystal structure of acetylcholinesterase from Torpedo
californica in complex with a tacrine heterodimer was used as
the docking template, because it already accommodates a hybrid
inhibitor that stretches the entire gorge, binding simultaneously
to both binding sites. The tacrine moiety was found to bind to
the active site as in the crystal structures 1ACJ20 (TcAChE-
tacrine) or 1ZGB (TcAChE-tacrine-hupyridone), while interac-
tions of the hydrazide linker with Asp72 and Tyr121 were
observed as similarly reported before21 (residue numbering
throughout this report will refer to the organism of the complex
discussed). The main emphasis was put on possible interaction
partners at the utmost gorge entrance. The propidium-Mm-
AChE (PDB 1N5R22) complex showed, that the quaternary side
chain is pointing out of the gorge in two distinct orientations.
It is either bound to an outer surface loop, comprised by
Leu289-Ser293, or it interacts with His287. These orientations
coupling with ethyl 6-aminohexanoate. Following a method
adapted from Pearson et al.,17 the para-acetoxy moiety was
exchanged by a benzyl group and the two remaining meta-
acetoxy groups were cleaved to obtain 19, which was subse-
quently treated with 1,2-dibromoethane to yield intermediate
20. Reaction with triethylamine produced a bisquaternary
carboxylic ester that was deprotected by hydrogenolysis and
again alkylated with 1,2-dibromoethane. This second alkylation
was found extremely sensitive to oxidation and had to be
carefully protected by argon to achieve 21 free from any
degradation products. The subsequent quaternization toward the
trisquaternary carboxylic ester 22 was another crucial step as
the use of nitromethane led to constant impurities that could
not be removed. However, acetonitrile dissolved both educt and
product at higher temperatures, and pure 22 precipitated upon

5688 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 23
Elsinghorst et al.
Figure 2. Cholinesterase selectivity: a chlorine at C6 fits perfectly
into hAChE (a), while hBChE offers less space and impairs binding
(b). Graphics were created using PyMOL.40
believed that the binding to the outer gorge lip plays a dominant
role in ligand association. The second ammonium side chain
of compound 15 (Figure 1b) is directed toward the gorge middle,
where it is found on an axis connecting Tyr70 and Tyr334. This
location allows the excellent accommodation of the ethyl
substitutents, each of them pointing into a separate surface
pocket. Compound 24 (Figure 1c) carrying three quaternary side
chains was the most complex structure to dock with a high level
of torsional constraints. Again, the triplet of Phe284, Asp285,
and Ser286 is addressed by one ammonium group, this time in
para position, while one of the meta side chains points into the
gorge and the other toward the entrance. It is the inward pointing
side chain that finds a perfect fit at the gorge center and forces
the opposite side chain outward.
Figure 1. Compounds 13, 15, and 24 docked into acetylcholinesterase
(PDB 1ZGB19) using AutoDock.18 Graphics were created using
PyMOL40 and amino acid residue numbering refers to TcAChE.
The inhibition of acetylcholinesterase was investigated for
enzymes from Electrophorus electricus, Torpedo californica,
and Homo sapiens, while butyrylcholinesterase inhibition was
determined using the human enzyme. IC50 values derived from
duplicate experiments are listed in Table 1, also including the
selectivity between the two human cholinesterases. Data ob-
tained for tacrine and gallamine in the presence of 6%
acetonitrile differed notably from our results (Table 1) and
literature reports24-26 in the absence of acetonitrile. For two
acetylcholinesterases, higher IC50 values were observed in the
presence of acetonitrile, whereas the opposite applies to human
butyrylcholinesterase inhibited by tacrine. These findings will
be covered in more detail in a future report.
result from two possible binding modes of the phenylphenan-
thridinium system, that forms a π-π stacking with Trp286 and
may be flipped by 180 degrees. For our compounds, a π-π
stacking of the benzamide unit with the corresponding Trp279
was not observed because the amide bond forms hydrogen bonds
to mid-gorge amino acid residues, as was already found with a
series of tacrine-trimethoxybenzene heterodimers.23
In all three compounds 13, 15, and 24, one of the quaternary
side chains is located at the same, fixed position. The backbone
carbonyl oxygens of Phe284 and Ser286, and the side chain
oxygens of Asp285 and Ser286 were found at the base of a
pyramid, with the ammonium nitrogen at its top (Figure 1a).
As this interaction is observed in all three docking sets, it is
It was found that increasing substitution of the gallamine-
derived moiety tends to reduce the inhibitory potency. In the

Gallamine-Tacrine Hybrid
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 23 5689
Table 1. Inhibition of Acetylcholinesterase from Electrophorus electricus (EeAChE) and Torpedo californica (TcAChE), Human Acetylcholinesterase
(hAChE), and Human Butyrylcholinesterase (hBChE) by Compounds 13-16, 24, and 25
IC50 ( SEM (*nM, **µM)a
selectivity
(IC50 hAChE/
cmpd
13*
14*
15*
16*
24*
25*
tacrine*
gallamine**
tacrine*
gallamine**
EeAChE
TcAChE
hAChE
hBChE
IC50 hBChE)
0.467 ( 0.032
26.0 ( 1.9
1.28 ( 0.05
48.0 ( 3.4
28.9 ( 2.4
90.0 ( 3.4
300 ( 13
3150 ( 110
42.3 ( 2.0
1070 ( 60
17.6 ( 1.6
127 ( 3
8.59 ( 0.27
301 ( 10
27.7 ( 1.6
334 ( 41
n.d.
7.61 ( 0.41
5.44 ( 0.48
15.8 ( 0.8
6.75 ( 0.36
23.2 ( 1.2
6.54 ( 0.44
926 ( 30
2110 ( 70
136 ( 4
1480 ( 50
1.50 ( 0.08
8.55 ( 0.42
1.42 ( 0.07
16.7 ( 1.9
2.40 ( 0.14
26.5 ( 0.9
10.2 ( 0.3
2390 ( 100
28.9 ( 0.8
235 ( 10
5.07
0.64
11.1
0.4
9.7
0.3
91
0.9
4.7
6.3
n.d.
n.d.
n.d.
a IC50 values were determined in duplicate with at least four different inhibitor concentrations. Assays were performed in 0.1 M sodium phosphate buffer,
pH 7.3, 0.1 M NaCl, 6% MeCN. The last two lines refer to measurements in the absence of MeCN. n.d. ) not determined. For details, see Experimental
Section.
Table 2. Parameters Describing the Allosteric Inhibition of [3H]NMS
Dissociation from Porcine Muscarinic M2 Receptors by the Indicated
Test Compounds
a
cmpd
13
14
15
16
24
25
tacrine
gallamine
pEC50,diss
EC50,diss (nM)
nH
8.88 ( 0.06b
8.31 ( 0.03
9.07 ( 0.09b
8.82 ( 0.06b
8.86 ( 0.04b
8.99 ( 0.02b
5.20 ( 0.03
6.88 ( 0.10b
1.3
5.0
0.9
1.5
1.4
-1.08 ( 0.16
-1.21 ( 0.10c
-0.97 ( 0.18
-0.88 ( 0.13
-0.88 ( 0.08
-0.91 ( 0.04
-1.60 ( 0.23c
-0.82 ( 0.21
1.0
6300
130
Figure 3. Interaction of the indicated test compounds with muscarinic
M2 receptors whose acetylcholine binding site is blocked by [3H]NMS.
Ordinate: apparent rate constant k-1 of [3H]NMS-dissociation as a
percentage of the control value determined in the absence of test
compound. Curve inflection points of the hybrids showed no significant
differences (one-way ANOVA, p > 0.05). Shown are means (
SEM derived from 3-12 dissociation experiments with duplicated
values.
a pEC50: -log concentrations inducing a half-maximal reduction of the
apparent rate constant k-1 of [3H]NMS dissociation as a measure of binding
affinity. For details, see Experimental Section. b Slope factor is not different
from unity and was fixed to nH ) 1 for pEC50 evaluation. c Slope factor is
different from unity (F-test, p < 0.05).
case of acetylcholinesterase from Electrophorus electricus, the
affinity decreased by 2 orders of magnitude when comparing
compounds 13, 15, and 24. This loss in inhibitory potency due
to increasing substitution was only 2- or 3-fold for the other
three cholinesterases, as it was also observed in the case of the
chloro-substituted derivatives 14, 16, and 25 with all studied
cholinesterases. This might indicate the overall increase in
energy consumption due to an increase of torsional constraints.
In case of a chloro substitution at position 6 of the tacrine moiety
(14, 16, 25), a significant loss in inhibitory potency toward
acetylcholinesterase from Torpedo californica and a switch in
selectivity from human butyrylcholinesterase to acetylcholin-
esterase was observed. Both phenomena are tightly linked to
Pro446 in hAChE and its counterpart Met437 in hBChE or
Ile439 in TcAChE. While the proline offers enough space to
accommodate a chlorine at position 6, neither the methionine
nor the isoleucine do so and thus impair the binding of a chloro-
substituted tacrine to the active site (Figure 2). The 5-fold
selectivity toward acetylcholinesterase achieved by the introduc-
tion of chlorine is in accordance with previous reports.27,28
indicated by a reduction of the apparent rate constant k-1 (Figure
3). pEC50,diss values, that is, minus log concentrations for a half-
maximum effect, correspond to test compound binding affinity
for the allosteric site (Table 2). The pEC50,diss values found for
gallamine and tacrine were in agreement with previous re-
ports.13,29 Hybrid formation increased the allosteric potency by
factors of 100 relative to gallamine and 4800 relative to tacrine.
This gain in potency is of particular interest since homodimer-
ization of tacrine only resulted in a 20 fold increase in allosteric
potency.14 Remarkably, five hybrid substances exhibited EC50,diss
values around 1 nM, only for 14 a slightly increased value of
5 nM was observed. Neither the introduction of chlorine (14,
16, 25) into the tacrine ring system, nor the successive
simplification of the gallamine moiety had a significant effect
on allosteric potency. For example, the removal of one para-
(15 vs 24) or two meta-triethylammonioethoxy substituents (13
vs 24) retained the inhibition of [3H]NMS dissociation. Thus,
the number and position of the triethylammonioethoxy sub-
stituents is not decisive for the pronounced gain in potency.
Allosteric M2 Receptor Modulation. The interaction of the
building blocks and the hybrids 13-16, 24, and 25 with M2
receptors was measured in receptors whose acetylcholine-
binding site was blocked by the radioligand [3H]NMS. As a
consequence of ligand binding to the allosteric site, which is
located in the extracellular entrance of the receptor’s binding
cleft, the dissociation of [3H]NMS was inhibited. All compounds
retarded [3H]NMS dissociation concentration-dependently, as
Whereas the concentration-effect curve of tacrine was steep
(Table 2), as expected for an atypical allosteric modulator, most
hybrids showed curve slopes not different from unity, which is
in agreement with a typical mode of allosteric action. Again,
14 was an exception in that its curve slope was slightly steeper
than unity (Table 2).

5690 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 23
Elsinghorst et al.
formamide. Oxalyl chloride (34.4 mmol, 3.0 mL) was added while
stirring, and once the gas evolution had ceased, it was evaporated
to dryness. The residue was again dissolved in dichloromethane
(20 mL), and ethyl 6-aminohexanoate hydrochloride (30.0 mmol,
5.88 g) was suspended in the solution. After dropwise addition of
N-ethyl-N,N-diisopropylamine (60.0 mmol, 10.5 mL) over 1 h,
washing with water and a saturated solution of sodium hydrogen
carbonate, drying with anhydrous sodium sulfate, and evaporation
in vacuo yielded a white solid that was dissolved in anhydrous
methanol (60 mL). After adding potassium carbonate (3.2 mmol,
0.44 g) and stirring for 12 h, the solution was poured into water
(200 mL) containing 2 M hydrochloric acid (4 mL). The resulting
suspension was extracted three times using ethyl acetate, and the
combined organic phases were dried using anhydrous sodium
sulfate. Compound 5 was obtained by evaporation in vacuo as a
Conclusion
The successful synthesis of the gallamine-tacrine hybrid 24
was accomplished through a multistep convergent synthesis
based on a concise protection-deprotection strategy and the
previously established linker chemistry. All compounds exhib-
ited a profound inhibitory potential toward acetylcholinesterases
as well as butyrylcholinesterase. The simplified derivative 13
was found to be a very potent inhibitor of acetylcholinesterase
from Electrophorus electricus with an IC50 value of about 500
pM and showed a three times increased activity at human
acetylcholinesterase compared to the trisquaternary compound
24. The anticipated influence of a 6-chloro substitution of the
tacrine moiety on cholinesterase selectivity was observed, though
all compounds tend to be rather unselective inhibitors of both
human cholinesterases. The allosteric potency of the gallamine-
tacrine hybrids at M2 receptors is strongly increased compared
to their building blocks, gallamine and tacrine. As the orthosteric
site was occupied by the radioligand in these experiments, the
gain in potency is explained by simultaneous hybrid binding to
the core region and the peripheral region of the allosteric site.
The detailed elucidation of the allosteric modulation of mus-
carinic receptors, including the question of whether orthosteri-
cally free receptors might allow for an allosteric-orthosteric
hybrid binding topography, will be the subject of further studies
and reported in due course.
1
white powder (7.71 g, 92%), mp 75 °C. H NMR (DMSO-d6) δ
1.15 (t, 3 H, J ) 7.1 Hz), 1.28 (tt, 2 H, J ) 7.6 Hz), 1.48 (tt, 2 H,
J ) 7.4 Hz), 1.53 (tt, 2 H, J ) 7.5 Hz), 2.26 (t, 2 H, J ) 7.4 Hz),
3.19 (dt, 2 H, J ) 6.5 Hz), 4.03 (q, 2 H, J ) 7.1 Hz), 6.76 (ddd,
2 H, J ) 2.4, 2.9, 9.5 Hz), 7.68 (ddd, 2 H, J ) 2.4, 2.9, 9.5 Hz),
8.12 (t, 1 H, J ) 5.7 Hz), 9.86 (s, 1 H); 13C NMR (DMSO-d6) δ
14.24, 24.37, 26.07, 29.04, 33.62, 38.98, 59.75, 114.81, 125.60,
129.11, 160.04, 165.92, 172.97. Anal. (C15H21NO4) C, H, N.
Ethyl 6-((3,5-Dihydroxybenzoyl)amino)hexanoate (6). 3,5-Bis-
(acetyloxy)benzoic acid (4; 20.0 mmol, 4.76 g) was reacted as
4-acetyloxybenzoic acid (3), and 6 was obtained as a white
precipitate by suction filtration (4.76 g, 81%), mp 78 °C. 1H NMR
(DMSO-d6) δ 1.16 (t, 3 H, J ) 7.1 Hz), 1.27 (tt, 2 H, J ) 7.7 Hz),
1.47 (tt, 2 H, J ) 7.5 Hz), 1.53 (tt, 2 H, J ) 7.6 Hz), 2.26 (t, 2 H,
J ) 7.3 Hz), 3.16 (dt, 2 H, J ) 6.6 Hz), 4.03 (q, 2 H, J ) 7.2 Hz),
6.32 (t, 1 H, J ) 2.2 Hz), 6.63 (d, 2 H, J ) 2.2 Hz), 8.15 (t, 1 H,
J ) 5.7 Hz), 9.35 (s, 2 H); 13C NMR (DMSO-d6) δ 14.26, 24.35,
26.03, 28.87, 33.63, 39.00, 59.77, 105.00, 105.53, 137.19, 158.31,
166.49, 172.98. Anal. (C15H21NO5‚H2O) C, H, N.
Experimental Section
General Methods and Materials. Melting points were deter-
mined on a Boe¨tius melting point apparatus and are uncorrected.
Several compounds were prepared using the Bu¨chi Glas Uster
autoclave “TinyClave”. Thin-layer chromatography was performed
on Merck aluminum sheets, silica gel 60 F254. Preparative column
chromatography was performed on Merck silica gel 60, 70-230
mesh. 13C NMR (125 MHz) and 1H NMR spectra (500 MHz) were
recorded on a Bruker Avance DRX 500 spectrometer; 13C NMR
signals were assigned on the basis of 13C/1H correlation experiments
(HSQC, HMBC). Acetylcholinesterase was purchased from Fluka
(Deisenhofen, Germany; Electrophorus electricus) and Sigma
(Steinheim, Germany; Homo sapiens), and butyrylcholinesterase
was purchased from Lee Biosolutions (St. Louis, U.S.A.; Homo
sapiens). Acetylcholinesterase from Torpedo californica was a kind
gift from the Institute for Physiological Chemistry, University of
Bonn, Germany. Elemental analyses were performed with a Vario
EL apparatus.
Ethyl 6-((4-(2-Bromoethoxy)benzoyl)amino)hexanoate (7).
Ethyl 6-((4-hydroxybenzoyl)amino)hexanoate (5; 25.0 mmol, 6.98
g), potassium carbonate (150.0 mmol, 20.73 g), and 18-crown-6
(3.8 mmol, 1.00 g) were suspended in 1,2-dibromoethane (120 mL)
and heated to 80 °C for 36 h while stirring. Subsequent removal of
the inorganic salts by suction filtration, washing with dichlo-
romethane, and evaporation in vacuo produced a crude residue that
was subjected to column chromatography using petrol ether and
ethyl acetate in a ratio of 1:2. Compound 7 was achieved from the
1
first fractions as a white solid (8.74 g, 91%), mp 82 °C. H NMR
(DMSO-d6) δ 1.15 (t, 3 H, J ) 7.1 Hz), 1.29 (tt, 2 H, J ) 7.6 Hz),
1.50 (tt, 2 H, J ) 7.3 Hz), 1.54 (tt, 2 H, J ) 7.6 Hz), 2.27 (t, 2 H,
J ) 7.4 Hz), 3.21 (dt, 2 H, J ) 6.6 Hz), 3.80 (t, 2 H, J ) 5.5 Hz),
4.03 (q, 2 H, J ) 7.1 Hz), 4.37 (t, 2 H, J ) 5.4 Hz), 7.00 (ddd, 2
H, J ) 2.5, 2.8, 9.8. Hz), 7.80 (ddd, 2 H, J ) 2.5, 2.8, 9.8 Hz),
8.26 (t, 1 H, J ) 5.7 Hz); 13C NMR (DMSO-d6) δ 14.24, 24.36,
26.05, 28.97, 31.37, 33.60, 39.05, 59.75, 67.99, 114.19, 127.58,
129.09, 160.16, 165.58, 172.95. Anal. (C17H24BrNO4) C, H, N.
Ethyl 6-((3,5-Bis(2-bromoethoxy)benzoyl)amino)hexanoate
(8). Ethyl 6-((3,5-dihydroxybenzoyl)amino)hexanoate (6; 40.0
mmol, 11.81 g) was reacted as ethyl 6-((4-hydroxybenzoyl)amino)-
hexanoate (5) to obtain 8 as a white powder (3.10 g, 64%), mp
4-Acetyloxybenzoic Acid (3). 4-Hydroxybenzoic acid (1; 0.15
mol, 20.7 g) was dissolved in a mixture of pyridine (50 mL) and
acetic anhydride (50 mL). After stirring at room temperature for 1
h, the reaction mixture was poured into water (500 mL) and the
pH was adjusted to about 2 using concentrated hydrochloric acid.
The solution was extracted three times with ethyl acetate (500 mL),
the combined organic extracts were dried using anhydrous sodium
sulfate, and the solvent was evaporated in vacuo. The remaining
crude product was washed with petrol ether and recrystallized from
ethyl acetate to yield 3 (19.5 g, 72%) as white needles, mp
1
98 °C. H NMR (DMSO-d6) δ 1.15 (t, 3 H, J ) 7.3 Hz), 1.29 (tt,
195 °C, lit.30 187-192 °C. H NMR (DMSO-d6) δ 2.28 (s, 3 H),
1
2 H, J ) 7.6 Hz), 1.50 (tt, 2 H, J ) 7.6 Hz), 1.54 (tt, 2 H, J ) 7.6
Hz), 2.27 (t, 2 H, J ) 7.3 Hz), 3.21 (dt, 2 H, J ) 6.5 Hz), 3.79 (t,
4 H, J ) 5.4 Hz), 4.03 (q, 2 H, J ) 7.2 Hz), 4.35 (t, 4 H, J ) 5.4
Hz), 6.69 (t, 1 H, J ) 2.2 Hz), 7.03 (d, 2 H, J ) 2.2 Hz), 8.39 (t,
1 H, J ) 5.5 Hz); 13C NMR (DMSO-d6) δ 14.24, 24.34, 26.03,
28.82, 31.44, 33.59, 39.17, 59.75, 68.19, 104.15, 106.52, 137.05,
159.08, 165.39, 172.95. Anal. (C19H27Br2NO5) C, H, N.
7.25 (ddd, 2 H, J ) 2.4, 2.5, 8.9 Hz), 7.98 (ddd, 2 H, J ) 2.2, 2.5,
8.9 Hz), 12.96 (bs, 1 H); 13C NMR (DMSO-d6) δ 21.01, 122.18,
128.47, 130.99, 154.09, 166.74, 168.97. Anal. (C9H8O4) C, H, N.
3,5-Bis(acetyloxy)benzoic Acid (4). 3,5-Dihydroxybenzoic acid
(2; 0.15 mol, 23.1 g) was reacted as 4-hydroxybenzoic acid (1) to
yield 4 (30.0 g, 84%) as a white powder, mp 164 °C, lit.31 161 °C.
1H NMR (DMSO-d6) δ 2.28 (s, 6 H), 7.26 (t, 1 H, J ) 2.2 Hz),
7.57 (d, 2 H, J ) 2.2 Hz), 13.34 (s, 1 H); 13C NMR (DMSO-d6) δ
20.92, 120.29, 120.50, 133.05, 151.06, 165.89, 169.05. Anal.
(C11H10O6) C, H, N.
2-(4-(((6-Ethoxy-6-oxohexyl)amino)carbonyl)phenoxy)-N,N,N-
triethylethanammonium Bromide (9). Ethyl 6-((4-(2-bromo-
ethoxy)benzoyl)amino)hexanoate (7; 20.0 mmol, 7.73 g) was heated
to 60 °C for 24 h in a solution of triethylamine (200 mmol, 27.9
mL) and anhydrous nitromethane (100 mL). After cooling to room
temperature, the solvent and excess amine were evaporated in vacuo
and the oily residue was taken up in absolute ethanol (10 mL).
Ethyl 6-((4-Hydroxybenzoyl)amino)hexanoate (5). 4-Acetyl-
oxybenzoic acid (3; 30.0 mmol, 5.40 g) was dissolved in anhydrous
dichloromethane (50 mL) with catalytic amounts of N,N-dimethyl-

Gallamine-Tacrine Hybrid
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 23 5691
Pouring the alcoholic solution into ethyl acetate (1000 mL) produced
the final product 9 as a white and hygroscopic precipitate (7.54 g,
77%). 1H NMR (DMSO-d6) δ 1.15 (t, 3 H, J ) 7.1 Hz), 1.23 (t, 9
H, J ) 7.3 Hz), 1.29 (tt, 2 H, J ) 7.6 Hz), 1.50 (tt, 2 H, J ) 7.3
Hz), 1.54 (tt, 2 H, J ) 7.5 Hz), 2.27 (t, 2 H, J ) 7.4 Hz), 3.21 (dt,
2 H, J ) 6.6 Hz), 3.38 (q, 6 H, J ) 7.2 Hz), 3.69 (t, 2 H, J ) 4.9
Hz), 4.03 (q, 2 H, J ) 7.2 Hz), 4.45 (t, 2 H, J ) 4.7 Hz), 7.03
(ddd, 2 H, J ) 2.4, 2.8, 9.8 Hz), 7.85 (ddd, 2 H, J ) 2.5, 2.9, 9.6
Hz), 8.30 (t, 1 H, J ) 5.7 Hz); 13C NMR (DMSO-d6) δ 7.45, 14.25,
24.35, 26.05, 28.98, 33.60, 39.06, 53.09, 55.29, 59.75, 61.34,
114.22, 127.85, 129.06, 159.61, 165.46, 172.96. Anal. (C23H39-
BrN2O4‚0.5H2O) C: calcd, 55.64; found, 55.17; H, N.
dine (1.0 mmol, 0.22 g) were dissolved in absolute ethanol (20
mL) and heated to 140 °C for 24 h in a glass reactor. After cooling
to room temperature and evaporation of the solvent in vacuo, the
remaining residue was taken up in absolute ethanol (10 mL) and
poured into ethyl acetate (300 mL) while stirring to precipitate the
final product. Compound 13 was obtained by suction filtration as
1
a yellow, hygroscopic powder (0.62 g, 90%). H NMR (DMSO-
d6) δ 1.22 (t, 9 H, J ) 7.1 Hz), 1.29 (tt, 2 H, J ) 7.6 Hz), 1.51 (tt,
2 H, J ) 7.3 Hz), 1.56 (tt, 2 H, J ) 7.6 Hz), 1.80 (bs, 4 H), 2.29
(t, 2 H, J ) 7.2 Hz), 2.69 (bs, 2 H), 3.09 (bs, 2 H), 3.20 (dt, 2 H,
J ) 6.2 Hz), 3.39 (q, 6 H, J ) 7.3 Hz), 3.70 (t, 2 H, J ) 4.9 Hz),
4.45 (t, 2 H, J ) 4.7 Hz), 7.03 (d, 2 H, J ) 8.9 Hz), 7.57 (ddd, 1
H, J ) 1.0, 7.3, 8.6 Hz), 7.85 (ddd, 1 H, J ) 1.0, 7.1, 8.6 Hz),
7.88 (d, 2 H, J ) 8.9 Hz), 8.09 (dd, 1 H, J ) 0.7, 8.7 Hz), 8.42 (t,
1 H, J ) 5.2 Hz), 8.69 (d, 1 H, J ) 8.8 Hz), 9.65 (s, 1 H), 11.15
(s, 1 H), 14.68 (s, 1 H); 13C NMR (DMSO-d6) δ 7.51, 20.26, 21.43,
24.18, 24.55, 26.32, 28.24, 29.08, 33.09, 39.13, 53.14, 55.35, 61.41,
111.13, 114.24, 115.10, 119.42, 124.27, 125.82, 127.83, 129.16,
132.96, 137.44, 152.61, 155.29, 159.64, 165.50, 171.90. Anal.
(C34H49BrClN5O3‚2H2O) C: calcd, 56.16; found, 55.73; H, N.
2-(4-(((6-(2-(6-Chloro-1,2,3,4-tetrahydroacridin-9-yl)hydrazino)-
6-oxohexyl)amino)carbonyl)phenoxy)-N,N,N-triethylethanam-
monium Bromide Hydrochloride (14). N,N,N-Triethyl-2-(4-(((6-
hydrazino-6-oxohexyl)amino)carbonyl)phenoxy)ethanammonium
bromide (11; 1.0 mmol, 0.47 g) and 6,9-dichloro-1,2,3,4-tetrahy-
droacridine (1.0 mmol, 0.25 g) were reacted as described above to
2-(3-(((6-Ethoxy-6-oxohexyl)amino)carbonyl)-5-((2-triethyl-
ammonio)ethoxy)phenoxy)-N,N,N-triethylethanammonium Di-
bromide (10). Ethyl 6-((3,5-bis(2-bromoethoxy)benzoyl)amino)-
hexanoate (8; 20.0 mmol, 10.18 g) were reacted as ethyl 6-((4-(2-
bromoethoxy)benzoyl)amino)hexanoate(7)toobtain10byprecipitation
of an alcoholic solution (50 mL) from ethyl acetate (900 mL) as a
1
white and hygroscopic powder (11.5 g, 81%). H NMR (DMSO-
d6) δ 1.16 (t, 3 H, J ) 7.1 Hz), 1.24 (t, 18 H, J ) 7.1 Hz), 1.29 (tt,
2 H, J ) 7.8 Hz), 1.52 (tt, 2 H, J ) 7.3 Hz), 1.54 (tt, 2 H, J ) 7.4
Hz), 2.27 (t, 2 H, J ) 7.4 Hz), 3.23 (dt, 2 H, J ) 6.6 Hz), 3.39 (q,
12 H, J ) 7.2 Hz), 3.69 (t, 4 H, J ) 4.7 Hz), 4.03 (q, 2 H, J ) 7.2
Hz), 4.46 (t, 4 H, J ) 4.7 Hz), 6.75 (t, 1 H, J ) 2.2 Hz), 7.15 (d,
2 H, J ) 2.2 Hz), 8.52 (t, 1 H, J ) 5.7 Hz); 13C NMR (DMSO-d6)
δ 7.50, 14.26, 24.33, 26.04, 28.94, 33.59, 39.15, 53.11, 55.32, 59.77,
61.72, 104.48, 106.79, 137.08, 158.50, 165.30, 172.96. Anal.
(C31H57Br2N3O5‚H2O) C, H, N.
1
obtain 14 as a yellow and hygroscopic powder (0.68 g, 94%). H
NMR (DMSO-d6) δ 1.22 (t, 9 H, J ) 7.3 Hz), 1.29 (tt, 2 H, J )
7.6 Hz), 1.52 (tt, 2 H, J ) 7.3 Hz), 1.56 (tt, 2 H, J ) 7.3 Hz), 1.80
(bs, 4 H), 2.30 (t, 2 H, J ) 7.3 Hz), 2.66 (bs, 2 H), 3.07 (bs, 2 H),
3.25 (dt, 2 H, J ) 6.3 Hz), 3.39 (q, 6 H, J ) 7.3 Hz), 3.70 (t, 2 H,
J ) 4.7 Hz), 4.45 (t, 2 H, J ) 4.6 Hz), 7.02 (d, 2 H, J ) 8.9 Hz),
7.64 (dd, 1 H, J ) 2.2, 9.5 Hz), 7.87 (d, 2 H, J ) 8.8 Hz), 8.19 (d,
1 H, J ) 2.2 Hz), 8.41 (t, 1 H, J ) 5.4 Hz), 8.73 (d, 1 H, J ) 9.5
Hz), 9.77 (s, 1 H), 11.18 (s, 1 H), 14.88 (s, 1 H); 13C NMR (DMSO-
d6) δ 7.49, 20.16, 21.28, 24.05, 24.49, 26.29, 28.25, 29.06, 33.10,
39.11, 53.14, 55.34, 61.40, 111.49, 113.64, 114.22, 118.27, 126.19,
126.68, 127.82, 129.13, 137.46, 138.23, 153.22, 155.11, 159.63,
165.50, 171.96. Anal. (C34H48BrCl2N5O3‚2H2O) C, H, N.
N,N,N-Triethyl-2-(4-(((6-hydrazino-6-oxohexyl)amino)carbo-
nyl)phenoxy)ethanammonium Bromide (11). 2-(4-(((6-Ethoxy-
6-oxohexyl)amino)carbonyl)phenoxy)-N,N,N-triethylethanammo-
nium bromide (9; 2.0 mmol, 0.97 g) was dissolved in a mixture of
hydrazine (51.4 mmol, 2.5 mL hydrazine hydrate 100%) and
absolute ethanol (20 mL), and the solution was heated to 80 °C for
72 h in a glass reactor. After cooling to room temperature, absolute
ethanol (80 mL) was added and the solution evaporated in vacuo.
The remaining residue was taken up in absolute ethanol (10 mL)
and poured into ethyl acetate (300 mL) at -20 °C while stirring.
Compound 11 was obtained by suction filtration as a white and
1
very hygroscopic powder (0.81 g, 86%). H NMR (DMSO-d6) δ
N,N,N-Triethyl-2-(3-(((6-(2-(1,2,3,4-tetrahydroacridin-9-yl)-
hydrazino)-6-oxohexyl)amino)carbonyl)-5-((2-triethylammonio)-
ethoxy)phenoxy)ethanammonium Dibromide Hydrochloride
(15). N,N,N-Triethyl-2-(3-(((6-hydrazino-6-oxohexyl)amino)carbo-
nyl)-5-((2-triethylammonio)ethoxy)phenoxy)ethanammonium di-
bromide (12; 0.5 mmol, 0.35 g) and 9-chloro-1,2,3,4-tetrahydroacri-
dine (0.5 mmol, 0.11 g) were reacted as described above to achieve
1.23 (t, 9 H, J ) 7.3 Hz), 1.26 (tt, 2 H, J ) 7.6 Hz), 1.50 (tt, 4 H,
J ) 7.2 Hz), 2.00 (t, 2 H, J ) 7.4 Hz), 3.21 (dt, 2 H, J ) 6.6 Hz),
3.38 (q, 6 H, J ) 7.3 Hz), 3.69 (t, 2 H, J ) 4.9 Hz), 4.12 (s, 2 H),
4.44 (t, 2 H, J ) 4.7 Hz), 7.03 (ddd, 2 H, J ) 2.4, 2.9, 9.8 Hz),
7.85 (ddd, 2 H, J ) 2.5, 2.9, 9.8 Hz), 8.30 (t, 1 H, J ) 5.7 Hz),
8.88 (s, 1 H); 13C NMR (DMSO-d6) δ 7.45, 25.13, 26.33, 29.12,
33.52, 39.19, 53.09, 55.29, 61.33, 114.22, 127.86, 129.07, 159.60,
165.44, 171.66. Anal. (C21H37BrN4O3‚0.5H2O) C, H, N: calcd,
11.61; found, 11.12.
1
15 as a yellow and hygroscopic powder (0.33 g, 72%). H NMR
(DMSO-d6) δ 1.23 (t, 18 H, J ) 7.1 Hz), 1.29 (tt, 2 H, J ) 7.8
Hz), 1.54 (tt, 2 H, J ) 7.6 Hz), 1.56 (tt, 2 H, J ) 7.6 Hz), 1.81 (bs,
4 H), 2.29 (t, 2 H, J ) 7.4 Hz), 2.69 (bs, 2 H), 3.09 (bs, 2 H), 3.22
(dt, 2 H, J ) 6.6 Hz), 3.40 (q, 12 H, J ) 7.2 Hz), 3.69 (t, 4 H, J
) 4.9 Hz), 4.49 (t, 4 H, J ) 4.6 Hz), 6.77 (t, 1 H, J ) 2.2 Hz),
7.24 (d, 2 H, J ) 2.3 Hz), 7.59 (ddd, 1 H, J ) 1.0, 7.1, 8.5 Hz),
7.87 (ddd, 1 H, J ) 1.0, 7.7, 8.4 Hz), 8.06 (dd, 1 H, J ) 1.0, 8.5
Hz), 8.69 (d, 1 H, J ) 8.5 Hz), 8.77 (t, 1 H, J ) 5.7 Hz), 9.65 (s,
1 H), 11.11 (s, 1 H), 14.51 (s, 1 H); 13C NMR (DMSO-d6) δ 7.53,
20.27, 21.40, 24.18, 24.53, 26.30, 28.27, 28.99, 33.07, 39.20, 53.14,
55.33, 61.81, 104.60, 106.91, 111.22, 115.10, 119.45, 124.22,
125.88, 133.04, 137.02, 137.42, 152.67, 155.31, 158.51, 165.28,
171.89. Anal. (C42H67Br2ClN6O4‚3H2O) C, H, N.
N,N,N-Triethyl-2-(3-(((6-hydrazino-6-oxohexyl)amino)carbo-
nyl)-5-((2-triethylammonio)ethoxy)phenoxy)ethanammonium Di-
bromide (12). 2-(3-(((6-Ethoxy-6-oxohexyl)amino)carbonyl)-5-((2-
triethylammonio)ethoxy)phenoxy)-N,N,N-triethylethanammonium
dibromide (10; 2.0 mmol, 1.42 g) was reacted as 2-(4-(((6-ethoxy-
6-oxohexyl)amino)carbonyl)phenoxy)-N,N,N-triethylethanammo-
nium bromide (9) to precipitate 12 from ethyl acetate (300 mL)
using an alcoholic solution (10 mL) at room temperature. Com-
pound 12 was obtained as a white and very hygroscopic powder
(1.07 g, 77%). 1H NMR (DMSO-d6) δ 1.24 (t, 18 H, J ) 7.1 Hz),
1.27 (m, 2 H), 1.51 (m, 4 H), 2.01 (t, 2 H, J ) 7.4 Hz), 3.22 (dt,
2 H, J ) 6.6 Hz), 3.40 (q, 12 H, J ) 7.3 Hz), 3.69 (t, 4 H, J ) 4.9
Hz), 4.42 (bs, 2 H), 4.47 (t, 4 H, J ) 4.6 Hz), 6.76 (t, 1 H, J ) 2.4
Hz), 7.16 (d, 2 H, J ) 2.2 Hz), 8.53 (t, 1 H, J ) 5.7 Hz), 8.95 (s,
1 H); 13C NMR (DMSO-d6) δ 7.50, 25.10, 26.30, 29.04, 33.49,
39.25, 53.11, 55.32, 61.74, 104.52, 106.80, 137.09, 158.50, 165.30,
171.65. Anal. (C29H55Br2N5O4‚H2O) C, H, N.
2-(3-(((6-(2-(6-Chloro-1,2,3,4-tetrahydroacridin-9-yl)hydrazino)-
6-oxohexyl)amino)carbonyl)-5-((2-triethylammonio)ethoxy)phe-
noxy)-N,N,N-triethylethanammonium Dibromide Hydrochloride
(16). N,N,N-Triethyl-2-(3-(((6-hydrazino-6-oxohexyl)amino)carbo-
nyl)-5-((2-triethylammonio)ethoxy)phenoxy)ethanammonium di-
bromide (12; 1.0 mmol, 0.70 g) and 6,9-dichloro-1,2,3,4-tetrahy-
droacridine (1.0 mmol, 0.25 g) were reacted as described above to
N,N,N-Triethyl-2-(4-(((6-(2-(1,2,3,4-tetrahydroacridin-9-yl)-
hydrazino)-6-oxohexyl)amino)carbonyl)phenoxy)ethanammo-
nium Bromide Hydrochloride (13). N,N,N-Triethyl-2-(4-(((6-
hydrazino-6-oxohexyl)amino)carbonyl)phenoxy)ethanammonium
bromide (11; 1.0 mmol, 0.47 g) and 9-chloro-1,2,3,4-tetrahydroacri-
1
yield 16 as a yellow and hygroscopic powder (0.88 g, 93%). H
NMR (DMSO-d6) δ 1.23 (t, 18 H, J ) 7.3 Hz), 1.30 (tt, 2 H, J )
7.9 Hz), 1.54 (tt, 2 H, J ) 7.6 Hz), 1.56 (tt, 2 H, J ) 7.6 Hz), 1.80
(bs, 4 H), 2.30 (t, 2 H, J ) 7.3 Hz), 2.66 (bs, 2 H), 3.07 (bs, 2 H),

5692 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 23
Elsinghorst et al.
3.22 (dt, 2 H, J ) 6.5 Hz), 3.40 (q, 12 H, J ) 7.2 Hz), 3.69 (t, 4
H, J ) 4.6 Hz), 4.49 (t, 4 H, J ) 4.3 Hz), 6.76 (t, 1 H, J ) 2.1
Hz), 7.23 (d, 2 H, J ) 2.2 Hz), 7.65 (dd, 1 H, J ) 2.2, 9.2 Hz),
8.18 (d, 1 H, J ) 2.2 Hz), 8.74 (d, 1 H, J ) 9.2 Hz), 8.76 (t, 1 H,
J ) 5.5 Hz), 9.77 (s, 1 H), 11.18 (s, 1 H), 14.79 (s, 1 H); 13C
NMR (DMSO-d6) δ 7.55, 20.17, 21.28, 24.08, 24.49, 26.29, 28.25,
28.96, 33.10, 39.20, 53.16, 55.36, 61.82, 104.63, 106.93, 111.53,
113.65, 118.26, 126.23, 126.70, 137.03, 137.48, 138.22, 153.25,
155.16, 158.51, 165.29, 171.96. Anal. (C42H66Br2Cl2N6O4‚4H2O)
C, H, N.
thoroughly evaporated in vacuo to allow the deprotection of the
intermediate. Hydrogenolysis was carried out using Pd/C (0.8 g)
and hydrogen at balloon pressure in anhydrous methanol for 6 h.
Subsequently, the catalyst was filtered off and the reaction mixture
was evaporated in vacuo. Potassium carbonate (150.0 mmol, 20.73
g) and 18-crown-6 (3.8 mmol, 1.00 g) were added to the remaining
oil, and the reaction apparatus was protected by argon, followed
by the addition of 1,2-dibromoethane (60 mL) and anhydrous
acetonitrile (60 mL). The suspension was heated to 80 °C for 36 h
and the inorganic salts were removed afterward. Evaporation of
the acetonitrile produced a suspension that was diluted by ethyl
acetate (50 mL) to obtain the precipitate 21 by suction filtration
(4.20 g, 50%) as a white and hygroscopic powder. 1H NMR
(DMSO-d6) δ 1.16 (t, 3 H, J ) 7.1 Hz), 1.24 (t, 18 H, J ) 7.1 Hz),
1.30 (tt, 2 H, J ) 7.6 Hz), 4.03 (q, 2 H, J ) 7.2 Hz), 1.54 (tt, 2 H,
J ) 7.3 Hz), 1.54 (tt, 2 H, J ) 7.6 Hz), 2.27 (t, 2 H, J ) 7.4 Hz),
3.25 (dt, 2 H, J ) 6.6 Hz), 3.42 (q, 12 H, J ) 7.2 Hz), 3.71 (t, 2
H, J ) 6.2 Hz), 3.74 (t, 4 H, J ) 4.9 Hz), 4.23 (t, 2 H, J ) 6.0
Hz), 4.50 (t, 4 H, J ) 4.7 Hz), 7.42 (s, 2 H), 8.62 (t, 1 H, J ) 5.8
Hz); 13C NMR (DMSO-d6) δ 7.55, 14.26, 24.35, 26.05, 29.03,
31.93, 33.60, 39.13, 53.18, 55.53, 59.77, 62.57, 72.70, 107.02,
130.49, 138.42, 150.98, 165.17, 172.97. Anal. (C33H60Br3N3O6‚
H2O) C, H, N.
2-(5-(((6-Ethoxy-6-oxohexyl)amino)carbonyl)-2,3-bis((2-tri-
ethylammonio)ethoxy)phenoxy)-N,N,N-triethylethanammoni-
um Tribromide (22). 2-(5-(((6-Ethoxy-6-oxohexyl)amino)carbonyl)-
2-(2-bromoethoxy)-3-((2-triethylammonio)ethoxy)phenoxy)-N,N,N-
triethylethanammonium dibromide (21; 2.0 mmol, 1.67 g) was
suspended in a mixture of triethylamine (36.0 mmol, 5.0 mL) and
anhydrous acetonitrile (20 mL). The reaction mixture was heated
to 100 °C for 48 h in a glass reactor. At first the educt dissolved
as the temperature raised, but soon the final product precipitated.
After cooling to room temperature 22 was collected by suction
filtration (1.30 g, 69%) as a white and hygroscopic powder and
washed once with cooled acetonitrile. 1H NMR (DMSO-d6) δ 1.16
(t, 3 H, J ) 7.1 Hz), 1.23 (t, 9 H, J ) 7.3 Hz), 1.25 (t, 18 H, J )
7.3 Hz), 1.30 (tt, 2 H, J ) 7.9 Hz), 1.51-1.59 (m, 4 H), 2.27 (t, 2
H, J ) 7.4 Hz), 3.25 (dt, 2 H, J ) 6.7 Hz), 3.45 (q, 12 H, J ) 7.3
Hz), 3.47 (q, 6 H, J ) 7.3 Hz), 3.61 (t, 2 H, J ) 4.9 Hz), 3.73 (t,
4 H, J ) 5.5 Hz), 4.03 (q, 2 H, J ) 7.2 Hz), 4.26 (t, 2 H, J ) 4.7
Hz), 4.55 (t, 4 H, J ) 5.4 Hz), 7.48 (s, 2 H), 8.71 (t, 1 H, J ) 5.7
Hz); 13C NMR (DMSO-d6) δ 7.57, 7.59, 14.26, 24.34, 26.05, 28.99,
33.59, 39.13, 53.20, 55.06, 55.83, 59.76, 62.19, 65.94, 106.72,
131.10, 137.45, 150.97, 165.05, 172.96. Anal. (C39H75Br3N4O6‚
H2O) C, H, N.
3,4,5-Tris(acetyloxy)benzoic Acid (18). 3,4,5-Trihydroxyben-
zoic acid (17; 0.15 mol, 25.5 g) was reacted as 4-hydroxybenzoic
acid (1) to yield 18 (38.8 g, 87%) as a white powder, mp 172 °C,
1
lit.32 163 °C. H NMR (DMSO-d6) δ 2.29 (s, 6 H), 2.32 (s, 3 H),
7.74 (s, 2 H), 13.39 (s, 1 H); 13C NMR (DMSO-d6) δ 20.50, 122.06,
129.07, 138.42, 143.36, 165.51, 167.04, 168.12. Anal. (C13H12O8)
C, H, N.
Ethyl 6-((4-Benzyloxy-3,5-dihydroxybenzoyl)amino)hexanoate
(19). 3,4,5-Tris(acetyloxy)benzoic acid (18; 30.0 mmol, 8.89 g) and
ethyl 6-aminohexanoate hydrochloride (30.0 mmol, 5.88 g) were
reacted following the standard procedure described before. Evapo-
ration in vacuo produced a crude oil that was dissolved in anhydrous
acetone (100 mL). Benzyl chloride (60.0 mmol, 6.9 mL), potassium
carbonate (90.0 mmol, 12.4 g), and potassium iodide (9.0 mmol,
1.5 g) were added, and the reaction mixture was heated to reflux
for 18 h. Subsequent removal of the inorganic salts by suction
filtration and evaporation in vacuo led to an oily residue that was
washed with n-hexane to remove excess benzyl chloride. A solution
of potassium carbonate (30.0 mmol, 4.1 g) in water (40 mL) was
added to a solution of the residue in a mixture of ethyl acetate
(200 mL) and methanol (50 mL). After refluxing for 1 h and
washing with water (150 mL) containing concentrated hydrochloric
acid (5 mL), the reaction mixture was subjected to flash column
chromatography using ethyl acetate. Recrystallization from chlo-
roform and n-pentane produced 19 (7.47 g, 62%) as a white solid,
mp 119 °C. 1H NMR (DMSO-d6) δ 1.15 (t, 3 H, J ) 7.1 Hz), 1.27
(tt, 2 H, J ) 7.7 Hz), 1.46 (tt, 2 H, J ) 7.3 Hz), 1.53 (tt, 2 H, J )
7.5 Hz), 2.26 (t, 2 H, J ) 7.4 Hz), 3.15 (dt, 2 H, J ) 6.5 Hz), 4.03
(q, 2 H, J ) 7.0 Hz), 5.00 (s, 2 H), 6.78 (s, 2 H), 7.28 (tt, 1 H, J
) 1.4, 7.4 Hz), 7.33 (tt, 2 H, J ) 1.4, 6.6 Hz), 7.50 (d, 2 H, J )
6.9 Hz), 8.09 (t, 1 H, J ) 5.5 Hz), 9.24 (s, 2 H); 13C NMR (DMSO-
d6) δ 14.25, 24.36, 26.03, 28.91, 33.63, 39.01, 59.77, 73.21, 106.95,
127.79, 128.11, 128.31, 130.38, 136.57, 138.04, 150.54, 166.26,
172.97. Anal. (C22H27NO6) C: calcd, 65.82; found, 65.02; H, N.
Ethyl 6-((4-Benzyloxy-3,5-bis-(2-bromoethoxy)benzoyl)ami-
no)hexanoate (20). Ethyl 6-((4-benzyloxy-3,5-dihydroxybenzoyl)-
amino)hexanoate (19; 15.0 mmol, 6.02 g), potassium carbonate
(150.0 mmol, 20.73 g), and 18-crown-6 (3.8 mmol, 1.00 g) were
suspended in 1,2-dibromoethane (120 mL) and heated to 80 °C for
36 h while stirring. Subsequent removal of the inorganic salts by
suction filtration, washing with dichloromethane, and evaporation
in vacuo resulted in a crude product that was subjected to column
chromatography using petrol ether and ethyl acetate in a ratio of
1:2. Compound 20 was achieved from the first fractions as a white
N,N,N-Triethyl-2-(5-(((6-hydrazino-6-oxohexyl)amino)carbo-
nyl)-2,3-bis((2-triethylammonio)ethoxy)phenoxy)-
ethanammonium Tribromide (23). 2-(5-(((6-Ethoxy-6-oxohexyl)-
amino)carbonyl)-2,3-bis((2-triethylammonio)ethoxy)phenoxy)-N,N,N-
triethylethanammonium tribromide (22; 1.0 mmol, 0.94 g) was
dissolved in a mixture of hydrazine (51.4 mmol, 2.5 mL hydrazine
hydrate 100%) and absolute ethanol (20 mL), and the solution was
heated to 80 °C for 72 h in a glass reactor. After cooling to room
temperature, absolute ethanol (80 mL) was added and the solution
was evaporated in vacuo. The remaining residue was taken up in
absolute ethanol (10 mL) and poured into ethyl acetate (300 mL)
while stirring to precipitate 23 as a white and very hygroscopic
1
solid (8.33 g, 90%), mp 89 °C. H NMR (DMSO-d6) δ 1.15 (t, 3
H, J ) 7.3 Hz), 1.30 (tt, 2 H, J ) 7.7 Hz), 1.51 (tt, 2 H, J ) 7.6
Hz), 1.55 (tt, 2 H, J ) 7.6 Hz), 2.27 (t, 2 H, J ) 7.4 Hz), 3.23 (dt,
2 H, J ) 6.6 Hz), 3.84 (t, 4 H, J ) 5.4 Hz), 4.03 (q, 2 H, J ) 7.2
Hz), 4.37 (t, 4 H, J ) 5.4 Hz), 5.07 (s, 2 H), 7.18 (s, 2 H), 7.30 (tt,
1 H, J ) 1.6, 7.3 Hz), 7.36 (tt, 2 H, J ) 1.3, 7.0 Hz), 7.53 (m, 2
H), 8.35 (t, 1 H, J ) 5.7 Hz); 13C NMR (DMSO-d6) δ 14.24, 24.36,
26.07, 28.99, 31.64, 33.58, 39.21, 59.76, 68.97, 74.35, 106.59,
128.00, 128.23, 128.51, 130.19, 137.67, 139.27, 151.68, 165.32,
172.94. Anal. (C26H33Br2NO6) C, H, N.
1
powder (0.84 g, 91%) that was collected by suction filtration. H
NMR (DMSO-d6) δ 1.23 (t, 9 H, J ) 7.2 Hz), 1.25 (t, 20 H, J )
6.9 Hz), 1.51 (tt, 2 H, J ) 7.6 Hz), 1.54 (tt, 2 H, J ) 7.6 Hz), 2.01
(t, 2 H, J ) 7.4 Hz), 3.25 (dt, 2 H, J ) 6.7 Hz), 3.45 (q, 12 H, J
) 7.0 Hz), 3.47 (q, 6 H, J ) 7.3 Hz), 3.61 (t, 2 H, J ) 4.9 Hz),
3.72 (t, 4 H, J ) 5.5 Hz), 4.11 (bs, 2 H), 4.26 (t, 2 H, J ) 4.9 Hz),
4.55 (t, 4 H, J ) 5.4 Hz), 7.47 (s, 2 H), 8.68 (t, 1 H, J ) 5.8 Hz),
8.90 (s, 1 H); 13C NMR (DMSO-d6) δ 7.57, 7.59, 25.14, 26.32,
29.10, 33.52, 39.26, 53.20, 55.07, 55.83, 62.18, 65.94, 106.70,
131.14, 137.43, 150.98, 165.07, 171.66.
N,N,N-Triethyl-2-(5-(((6-(2-(1,2,3,4-tetrahydroacridin-9-yl)-
hydrazino)-6-oxohexyl)amino)carbonyl)-2,3-bis((2-triethylam-
monio)ethoxy)phenoxy)ethanammonium Tribromide Hydro-
chloride (24). N,N,N-Triethyl-2-(5-(((6-hydrazino-6-oxohexyl)-
2-(5-(((6-Ethoxy-6-oxohexyl)amino)carbonyl)-2-(2-bromo-
ethoxy)-3-((2-triethylammonio)ethoxy)phenoxy)-N,N,N-trieth-
ylethanammonium Dibromide (21). Ethyl 6-((4-benzyloxy-3,5-
bis-(2-bromoethoxy)benzoyl)amino)hexanoate (20; 10.0 mmol, 6.15
g) was heated to 60 °C for 24 h in a solution of triethylamine (100.0
mmol, 14.0 mL) and anhydrous nitromethane (100 mL). After
cooling to room temperature, the solvent and excess amine were

Gallamine-Tacrine Hybrid
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 23 5693
amino)carbonyl)-2,3-bis((2-triethylammonio)ethoxy)phenoxy)-
ethanammonium tribromide (23; 0.4 mmol, 0.37 g) and 9-chloro-
1,2,3,4-tetrahydroacridine (0.4 mmol, 0.09 g) were dissolved in
absolute ethanol (10 mL) and heated to 140 °C for 24 h in a glass
reactor. After cooling to room temperature the reaction mixture
was poured into ethyl acetate (300 mL) while stirring to precipitate
24 as a yellow, hygroscopic powder (0.45 g, 98%) that was collected
by suction filtration. 1H NMR (DMSO-d6) δ 1.23 (t, 9 H, J ) 7.0
Hz), 3.75 (t, 4 H, J ) 5.5 Hz), 1.24 (t, 18 H, J ) 7.3 Hz), 1.30 (tt,
2 H, J ) 7.4 Hz), 1.52-1.61 (m, 4 H), 1.81 (bs, 4 H), 2.29 (t, 2 H,
J ) 7.3 Hz), 2.70 (bs, 2 H), 3.09 (bs, 2 H), 3.24 (dt, 2 H, J ) 6.6
Hz), 3.46 (q, 12 H, J ) 7.2 Hz), 3.48 (q, 6 H, J ) 7.1 Hz), 3.63
(t, 2 H, J ) 4.7 Hz), 4.27 (t, 2 H, J ) 4.7 Hz), 4.58 (t, 4 H, J )
5.4 Hz), 7.60 (s, 2 H), 7.60 (ddd, 1 H, J ) 1.0, 5.4, 7.6 Hz), 7.87
(ddd, 1 H, J ) 0.7, 7.1, 8.4 Hz), 8.05 (dd, 1 H, J ) 0.7, 8.5 Hz),
8.70 (d, 1 H, J ) 8.8 Hz), 9.06 (t, 1 H, J ) 5.7 Hz), 9.65 (s, 1 H),
11.09 (s, 1 H), 14.46 (s, 1 H); 13C NMR (DMSO-d6) δ 7.60, 7.62,
20.26, 21.40, 24.18, 24.55, 26.31, 28.23, 28.98, 33.08, 39.20, 53.22,
55.10, 55.89, 62.31, 66.01, 106.92, 111.20, 115.09, 119.39, 124.26,
125.89, 131.02, 133.04, 137.39, 137.47, 150.99, 152.64, 155.35,
164.98, 171.89. Anal. (C50H85Br3ClN7O5‚3H2O) C, H, N.
Molecular Docking Studies. Crystal structures of several
acetylcholinesterase complexes (1ACJ,20 1ZGB,19 and 1N5M22)
were retrieved from the Protein Data Bank. Unnecessary information
from the PDB files was dropped by keeping only the ATOM entries,
missing side chain atoms were generated using the repair feature
of the ADT suite,34 and polar hydrogens were added. Using the
CORINA35 web service, three-dimensional coordinates were ob-
tained for the ligands. OpenBabel36 was used to convert these PDB
files to MOL2 format, which allows the addition of partial charges
that were calculated using MOPAC.37 It can be observed from the
crystal structures above and one obtained for the precursor 6,9-
dichloro-1,2,3,4-tetrahydroacridine38 that only one tacrine conformer
is energetically favored. Therefore, the right conformer and the
protonation at the endocyclic nitrogen has to be considered when
generating these coordinates. As a result of the energy minimization,
the structures from the CORINA algorithm tend to be stretched,
which gives a good bias toward the anticipated binding mode.
AutoDock18 was used to discover possible binding modes, as it
implements a more meticulous scoring procedure than the GOLD39
software used in previous investigations.4 In addition to the
evaluation of hydrophobic and hydrogen-bonding contacts, there
are additional terms regarding electrostatic interactions, the ligands’
torsional entropy, and desolvation upon binding. This was thought
to drive the docking process into the right direction with respect to
the quaternary side chains. The docking grid (180 000 points, 50
× 60 × 60, 0.375 Å) was centered in between Tyr121 and Phe330
(PDB 1ZGB, Torpedo californica) to stretch the complete distance
from the active to the peripheral binding site. A population of 100
chromosomes and a set of 100 000 generations with the rigid root
placed at the center of mass was chosen for 100 runs. This resulted
in negative binding energies ∆Gbind for all compounds, with a
stepwise decrease of freed energy as the substitution pattern
increased. This is probably due to the increasing amount of torsional
energy, which exceeds the energy estimation for the binding process.
To validate the applied docking procedures, a classical redocking
approach was undertaken. Tacrine and gallamine were removed
from their corresponding crystal structures (1ACJ, 1N5M) and
rebuilt by the same procedure as applied for ligand preparation.
They were then docked into the gorge without any bias toward the
active or peripheral binding site, and the top-scored poses were
compared with their original position in the crystal structure. The
predicted positioning of tacrine comes very close to the one
observed in 1ACJ. The rmsd calculated by PyMOL40 adds up to
0.561 Å, which is accounted by the literature41 to be a very good
fit. Looking at gallamine, the matching is not as good as with tacrine
but still acceptable. The crystal structure of acetylcholinesterase in
complex with gallamine (1N5M) does not state the conformation
of the quaternary side chains. The rmsd comparison refers therefore
only to the tris-substituted benzene moiety and gives a value of
2.498 Å. From a total of 100 runs, 87% located the gallamine ligand
at the peripheral binding site, which demonstrates its clear bias to
bind peripherally.
2-(5-(((6-(2-(6-Chloro-1,2,3,4-tetrahydroacridin-9-yl)hydra-
zino)-6-oxohexyl)amino)carbonyl)-2,3-bis((2-triethylammonio)-
ethoxy)phenoxy)-N,N,N-triethylethanammonium Tribromide
Hydrochloride (25). N,N,N-Triethyl-2-(5-(((6-hydrazino-6-oxo-
hexyl)amino)carbonyl)-2,3-bis((2-triethylammonio)ethoxy)phen-
oxy)ethanammonium tribromide (23; 0.4 mmol, 0.37 g) and 6,9-
dichloro-1,2,3,4-tetrahydroacridine (0.4 mmol, 0.10 g) were reacted
as described above to obtain 25 as a yellow, hygroscopic powder
1
(0.46 g, 98%). H NMR (DMSO-d6) δ 1.23 (t, 9 H, J ) 7.3 Hz),
1.24 (t, 18 H, J ) 7.1 Hz), 1.30 (tt, 2 H, J ) 7.9 Hz), 1.52-1.62
(m, 4 H), 1.81 (bs, 4 H), 2.30 (t, 2 H, J ) 7.4 Hz), 2.66 (bs, 2 H),
3.07 (bs, 2 H), 3.24 (dt, 2 H, J ) 6.5 Hz), 3.46 (q, 12 H, J ) 7.0
Hz), 3.48 (q, 6 H, J ) 7.1 Hz), 3.63 (t, 2 H, J ) 4.8 Hz), 3.74 (t,
4 H, J ) 5.5 Hz), 4.27 (t, 2 H, J ) 4.6 Hz), 4.57 (t, 4 H, J ) 5.2
Hz), 7.59 (s, 2 H), 7.67 (dd, 1 H, J ) 2.2, 9.1 Hz), 8.14 (d, 1 H,
J ) 2.2 Hz), 8.74 (d, 1 H, J ) 9.5 Hz), 9.03 (t, 1 H, J ) 5.7 Hz),
9.77 (s, 1 H), 11.13 (s, 1 H), 14.64 (s, 1 H); 13C NMR (DMSO-d6)
δ 7.59, 7.61, 20.16, 21.26, 24.06, 24.47, 26.29, 28.26, 28.97, 33.08,
39.20, 53.21, 55.09, 55.88, 62.29, 65.99, 106.90, 111.57, 113.65,
118.27, 126.27, 126.67, 131.02, 137.46, 137.51, 138.19, 150.98,
153.27, 155.20, 164.98, 171.95. Anal. (C50H84Br3Cl2N7O5‚3H2O)
C, H: calcd, 7.39; found, 7.86; N.
Cholinesterase Inhibition Assays. Cholinesterase inhibition was
assayed spectrophotometrically in duplicate on a minimum of at
least four different inhibitor concentrations at 25 °C according to
a literature method.33 Assay buffer was 100 mM sodium phosphate,
100 mM NaCl, pH 7.3. Stock solutions of acetylcholinesterase
(Electrophorus electricus, ∼100 U/mL; Torpedo californica, ∼3
U/mL; Homo sapiens, ∼3 U/mL) and butyrylcholinesterase (Homo
sapiens, ∼10 U/mL) in assay buffer were kept at 0 °C. Appropriate
dilutions were prepared immediately before starting the measure-
ment. Acetyl- or butyrylthiocholine (10 mM) and 5,5′-dithio-bis-
(2-nitrobenzoic acid) (DTNB; 7 mM) were dissolved in assay buffer
and kept at 0 °C. Stock solutions of the inhibitors were prepared
in assay buffer. IC50 values were calculated from the linear steady-
state turnover of the substrate using eq 1
Radioligand Binding Assay. Membranes were prepared from
porcine hearts at 4 °C as described before.14,29 Ventricular tissue
(40 g) was homogenized in a 0.32 M sucrose solution and
centrifuged for 11 min at 300 g (2000 rpm in a Beckman rotor
35). Thereafter, the supernatant was centrifuged for 40 min at 80 000
g (32 000 rpm in a Beckman rotor 35), and the resulting pellet was
resuspended in 4 mM Na2HPO4, 1 mM KH2PO4, pH 7.4 (4 mL/g
original tissue wet weight). Aliquots (0.5 mL) were shock frozen
in liquid nitrogen and stored at -80 °C; the protein content ranged
between 0.3 and 0.9 mg/mL.
[I]
IC50
)
(1)
V
V
0 - 1
For measurement of [3H]NMS binding cardiac membranes (60-
100 µg protein/mL) were incubated with 0.2 nM [3H]NMS in 5
mM Na,K,Pi-buffer, pH 7.4 at 23 °C. Atropine (1 µM) served to
define nonspecific binding. Under control conditions specific
binding of [3H]N-methylscopolamine was characterized by a pKD
) 9.8 ( 0.1 (0.15 nM) and Bmax ) 98 ( 8 fmol/mg protein (means
( SEM, n ) 12). The assays aimed to measure radioligand
dissociation (t0.5,diss,control ) 5.1 ( 0.2, mean ( SEM, n ) 42) were
prepared in a larger volume; after a preincubation period with
where [I] is the inhibitor concentration and V0 and V are the rates
in the absence and presence of the inhibitor, respectively. Into a
cuvette containing 825 µL of assay buffer, 50 µL of the DTNB
solution, 60 µL of acetonitrile, 10 µL of an inhibitor solution, and
10 µL of a cholinesterase solution (∼3 U/mL) were added and
thoroughly mixed. After incubation for 15 min at 25 °C, the reaction
was initiated by adding 50 µL of the acetyl- or butyrylthiocholine
solution.

5694 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 23
Elsinghorst et al.
radioligand of 30 min, the time course of [3H]NMS dissociation
was visualized by adding 1 µM atropine. Test compounds were
added together with atropine. At time intervals suited to describe
the time course of dissociation, 1 mL aliquots were separated and
filtered rapidly (glass fiber filters No. 6, Schleicher and Schu¨ll,
Dassel, FRG). After washing the filters (twice with 5 mL of ice
cold incubation buffer), they were placed into scintillation vials,
and the radioactivity was determined in the presence of 5 mL of
Ready Protein (Beckman) by liquid scintillation counting in a
Beckman LS 6500. Data of individual experiments were analyzed
by nonlinear regression analysis using the Prism software (ver. 4.0,
Graph Pad, San Diego, U.S.A.). Dissociation data was fitted to a
monoexponential decay function; biexponential curve fitting did
not yield better results (F-test, p g 0.05, data not shown). The rate
constant of radioligand dissociation was calculated from the half-
lives using eq 2
(13) Tra¨nkle, C.; Weyand, O.; Voigtla¨nder, U.; Mynett, A.; Lazareno,
S.; Birdsall, N. J. M.; Mohr, K. Interactions of orthosteric and
allosteric ligands with [3H]dimethyl-W84 at the common allosteric
site of muscarinic M2 receptors. Mol. Pharmacol. 2003, 64, 180-
190.
(14) Tra¨nkle, C.; Dittmann, A.; Schulz, U.; Weyand, O.; Buller, S.; Jo¨hren,
K.; Heller, E.; Birdsall, N. J. M.; Holzgrabe, U.; Ellis, J.; Ho¨ltje, H.
D.; Mohr, K. Atypical muscarinic allosteric modulation: cooperativity
between modulators and their atypical binding topology in muscarinic
M2 and M2/M5 chimeric receptors. Mol. Pharmacol. 2005, 68, 1597-
1610.
(15) Darvesh, S.; Hopkins, D. A.; Geula, C. Neurobiology of butyryl-
cholinesterase. Nat. ReV. Neurosci. 2003, 4, 131-138.
(16) Sarkar, A.; Ilankumaran, P.; Kisanga, P.; Verkade, J. G. First synthesis
of a highly basic dendrimer and its catalytic application in organic
methodology. AdV. Synth. Catal. 2004, 346, 1093-1096.
(17) Pearson, A. J.; Bruhn, P. R. Studies on the synthesis of aryl ethers
using arene-manganese chemistry. J. Org. Chem. 1991, 56, 7092-
7097.
(18) Morris, G. M.; Goodsell, D. S.; Halliday, R.; Huey, R.; Hart, W. E.;
Belew, R. K.; Olson, A. J. Automated docking using a Lamarckian
genetic algorithm and a empirical binding free energy function. J.
Comb. Chem. 1998, 19, 1639-1662.
ln 2
t0.5
k-1
)
(2)
(19) PDB ID: 1ZGB. Haviv, H.; Wong, D. M.; Greenblatt, H. M.; Carlier,
P. R.; Pang, Y.-P.; Silman, I.; Sussman, J. L. Crystal packing mediates
enantioselective ligand recognition at the peripheral site of acetyl-
cholinesterase. J. Am. Chem. Soc. 2005, 127, 11029-11036.
(20) PDB ID: 1ACJ. Harel, M.; Schalk, I.; Ehret-Sabatier, L.; Bouet, F.;
Goeldner, M.; Hirth, C.; Axelsen, P. H.; Silman, I.; Sussman, J. L.
Quaternary ligand binding to aromatic residues in the active-site gorge
of acetylcholinesterase. Proc. Natl. Acad. Sci. U.S.A. 1993, 90, 9031-
9035.
Concentration effect curves for the effect of the modulators on
the rate constant k-1 of [3H]NMS dissociation were analyzed by a
four-parameter logistic equation. It was checked whether the curve
slope deviated from unity (F-test). Multiple comparisons were
carried out by one-way ANOVA with Tukey’s multiple comparison
test. A p value < 0.05 was taken as the criterion for statistical
significance.
(21) Campiani, G.; Fattorusso, C.; Butini, S.; Gaeta, A.; Agnusdei, M.;
Gemma, S.; Persico, M.; Catalanotti, B.; Savini, L.; Nacci, V.;
Novellino, E.; Holloway, H. W.; Greig, N. H.; Belinskaya, T.;
Fedorko, J. M.; Saxena, A. Development of molecular probes for
the identification of extra interaction sites in the mid-gorge and
peripheral sites of butyrylcholinesterase (BuChE). Rational design
of novel, selective, and highly potent BuChE inhibitors. J. Med.
Chem. 2005, 48, 1919-1929.
(22) PDB ID: 1N5M, 1N5R. Bourne, Y.; Taylor, P.; Radi, Z.; Marchot,
P. Structural insights into ligand interactions at the acetylcholin-
esterase peripheral anionic site. EMBO J. 2003, 22, 1-12.
(23) Data not shown; for compounds, see ref 4.
(24) Alonso, D.; Dorronsoro, I.; Rubio, L.; Mun˜oz, P.; Garc´ıa-Palomero,
E.; Del Monte, M.; Bidon-Chanal, A.; Orozco, M.; Luque, F. J.;
Castro, A.; Medina, M.; Mart´ınez, A. Donepezil-tacrine hybrid related
derivatives as new dual binding site inhibitors of AChE. Bioorg. Med.
Chem. 2005, 13, 6588-6597.
(25) Marco-Contelles, J.; Leo´n, R.; de Los R´ıos, C.; Guglietta, A.;
Terencio, J.; Lo´pez, M. G.; Garc´ıa, A. G.; Villarroya, M. Novel
multipotent tacrine-dihydropyridine hybrids with improved acetyl-
cholinesterase inhibitory and neuroprotective activities as potential
drugs for the treatment of Alzheimer’s disease. J. Med. Chem. 2006,
49, 7607-7610.
(26) Gallamine reference values for the investigated enzyme sources were
not reported; for other mammalian enzymes, see for example: Kamal,
M. A.; Al-Jafari, A. A. Mode of inhibition of bovine retinal
acetylcholinesterase by gallamine triethiodide in vitro. Pak. J. Biol.
Sci. 2000, 3, 767-771.
(27) Savini, L.; Gaeta, A.; Fattorusso, C.; Catalanotti, B.; Campiani, G.;
Chiasserini, L.; Pellerano, C.; Novellino, E.; McKissic, D.; Saxena,
A. Specific targeting of acetylcholinesterase and butyrylcholinesterase
recognition sites. Rational design of novel, selective, and highly potent
cholinesterase inhibitors. J. Med. Chem. 2003, 46, 1-4.
(28) Gregor, V. E.; Emmerling, M. R.; Lee, C.; Moore, C. J. The synthesis
and in vitro acetylcholinesterase and butyrylcholinesterase inhibitory
activity of tacrine (Cognex) derivatives. Bioorg. Med. Chem. Lett.
1992, 2, 861-864.
(29) Tra¨nkle, C.; Kostenis, E.; Burgmer, U.; Mohr, K. Search for lead
structures to develop new allosteric modulators of muscarinic
receptors. J. Pharmacol. Exp. Ther. 1996, 279, 926-933.
(30) Yerlikaya, Z.; Aksoy, S.; Bayramli, E. Structure and properties of
fully aromatic thermotropic liquid-crystalline copolyesters containing
m-hydroxybenzoic acid units. J. Appl. Polym. Sci. 2003, 90, 3260-
3269.
(31) Andrus, M. B.; Liu, J.; Meredith, E. L.; Nartey, E. Synthesis of
resveratrol using a direct decarbonylative Heck approach from
resorcylic acid. Tetrahedron Lett. 2003, 44, 4819-4822.
Supporting Information Available: Elemental analysis results
for 13-16, 24, and 25 as well as for intermediates. This material
is available free of charge via the Internet at http://pubs.acs.org.
References
(1) Holzgrabe, U.; Kapkova´, P.; Alptu¨zu¨n, V.; Scheiber, J.; Kugelmann,
E. Targeting acetylcholinesterase to treat neurodegeneration. Expert
Opin. Ther. Targets 2007, 11, 161-179.
(2) Martinez, A.; Castro, A. Novel cholinesterase inhibitors as future
effective drugs for the treatment of Alzheimer’s disease. Expert Opin.
InVest. Drugs 2006, 15, 1-12.
(3) Schliebs, R.; Arendt, T. The significance of the cholinergic system
in the brain during aging and in Alzheimer’s disease. J. Neural.
Transm. 2006, 113, 1625-1644.
(4) Elsinghorst, P. W.; Gonza´lez Tanarro, C. M.; Gu¨tschow, M. Novel
heterobivalent tacrine derivatives as cholinesterase inhibitors with
notable selectivity toward butyrylcholinesterase. J. Med. Chem. 2006,
49, 7540-7544.
(5) Mun˜oz-Torrero, D.; Camps, P. Dimeric and hybrid anti-Alzheimer
drug candidates. Curr. Med. Chem. 2006, 13, 399-422.
(6) Castro, A.; Martinez, A. Peripheral and dual binding site acetylcho-
linesterase inhibitors: implications in treatment of Alzheimer’s
disease. Mini-ReV. Med. Chem. 2001, 1, 267-272.
(7) Bovet, D.; Depierre, F.; de Lestrange, Y. Curarizing properties of
phenolic ethers with quarternary ammonium groups. Compt. Rend.
1947, 225, 74-76.
(8) Clark, A. L.; Mitchelson, F. The inhibitory effect of gallamine on
muscarinic receptors. Br. J. Pharmacol. 1976, 58, 323-331.
(9) Ellis, J.; Huyler, J.; Brann, M. R. Allosteric regulation of cloned
M1-M5 muscarinic receptor subtypes. Biochem. Pharmacol. 1991,
42, 1927-1932.
(10) Huang, X. P.; Prilla, S.; Mohr, K.; Ellis, J. Critical amino acid residues
of the common allosteric site on the M2 muscarinic acetylcholine
receptor: More similarities than differences between the structurally
divergent agents gallamine and bis(ammonio)alkane-type hexame-
thylene-bis-[dimethyl-(3-phthalimidopropyl)ammonium]dibromide.
Mol. Pharmacol. 2005, 68, 769-778.
(11) Prilla, S.; Schrobang, J.; Ellis, J.; Ho¨ltje, H. D.; Mohr, K. Allosteric
interactions with muscarinic acetylcholine receptors: Complex role
of the conserved tryptophan M2 422Trp in a critical cluster of amino
acids for baseline affinity, subtype selectivity, and cooperativity. Mol.
Pharmacol. 2006, 70, 181-193.
(12) Potter, L. T.; Ferrendelli, C. A.; Hanchett, H. E.; Hollifield, M. A.;
Lorenzi, M. V. Tetrahydroaminoacridine and other allosteric antago-
nists of hippocampal M1 muscarine receptors. Mol. Pharmacol. 1989,
35, 652-660.

Gallamine-Tacrine Hybrid
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 23 5695
(32) Gazit, A.; Yaish, P.; Gilon, C.; Levitzki, A. Tyrphostins I: Synthesis
and biological activity of protein tyrosine kinase inhibitors. J. Med.
Chem. 1989, 32, 2344-2352.
(38) Elsinghorst, P. W.; Nieger, M.; Gu¨tschow, M. 6,9-Dichloro-1,2,3,4-
tetrahydroacridine. Acta Crystallogr., Sect. E: Struct. Rep. Online
2007, 63, 1818-1820.
(39) Jones, G.; Willett, P.; Glen, R. C. Molecular recognition of receptor
sites using a genetic algorithm with a description of desolvation. J.
Mol. Biol. 1995, 245, 43-53.
(40) PyMOL Molecular Graphics System 0.99, http://www.pymol.org
(October 8, 2007).
(33) Ellman, G. L.; Courtney, K. D.; Andres, V.; Feather-Stone, R. M. A
new and rapid colorimetric determination of acetylcholinesterase
activity. Biochem. Pharmacol. 1961, 7, 88-95.
(34) AutoDockTools 1.3, http://www.scripps.edu/∼sanner/python/adt/au-
totoolsoverview.html (October 8, 2007).
(35) CORINA 3.2, http://www.molecular-networks.com/online_demos/
corina_demo.html.
(36) OpenBabel 2.0.0, http://openbabel.sourceforge.net (October 8, 2007).
(37) MOPAC 2002 1.01, http://www.fqs.pl/chemistry/mopac_2002
(October 8, 2007).
(41) Nissink, J. W. M.; Murray, C.; Hartshorn, M.; Verdonk, M. L.; Cole,
J. C.; Taylor, R. A new test set for validating predictions of protein-
ligand interaction. Proteins 2002, 49, 457-471.
JM070859S