Synthesis of (nor)tropeine (di)esters and allosteric modulation of glycine receptor binding

Article abstract of DOI:10.1016/j.bmc.2007.10.097

(Hetero)aromatic mono- and diesters of tropine and nortropine were prepared. Modulation of [³H]strychnine binding to glycine receptors of rat spinal cord was examined with a ternary allosteric model. The esters displaced [³H]strychnine binding with nano- or micromolar potencies and strong negative cooperativity. Coplanarity and distance of the ester moieties of diesters affected the binding affinity being nanomolar for isophthaloyl-bistropane and nortropeines. Nortropisetron had the highest affinity (K_A ～ 10 nM). Two esters displayed negative cooperativity with glycine in displacement, while three esters of low-affinity and nortropisetron exerted positive cooperativity with glycine.

Full text of DOI:10.1016/j.bmc.2007.10.097

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry 16 (2008) 2086–2092
Synthesis of (nor)tropeine (di)esters and allosteric
modulation of glycine receptor binding
a,
b
b
a
*
G a´ bor Maksay, P e´ ter Nemes, Zolt a´ n Vincze and Timea B ´ı r o´
a
Department of Molecular Pharmacology, Institute of Biomolecular Chemistry, Chemical Research Center,
Hungarian Academy of Sciences, H-1525 Budapest, POB 17, Hungary
Department of Chemistry, Faculty of Veterinary Sciences, Szent Istv a´ n University, H-1400 Budapest, POB 2, Hungary
b
Received 12 June 2007; revised 19 October 2007; accepted 30 October 2007
Available online 4 November 2007
3
Abstract—(Hetero)aromatic mono- and diesters of tropine and nortropine were prepared. Modulation of [ H]strychnine binding to
3
glycine receptors of rat spinal cord was examined with a ternary allosteric model. The esters displaced [ H]strychnine binding with
nano- or micromolar potencies and strong negative cooperativity. Coplanarity and distance of the ester moieties of diesters affected
the binding affinity being nanomolar for isophthaloyl-bistropane and nortropeines. Nortropisetron had the highest affinity
(
K
A
ꢀ 10 nM). Two esters displayed negative cooperativity with glycine in displacement, while three esters of low-affinity and nor-
tropisetron exerted positive cooperativity with glycine.
2007 Elsevier Ltd. All rights reserved.
ꢀ
1
. Introduction
we have developed substituted benzoate esters of
nortropine as high-affinity, GlyR-selective allosteric
5
Glycine is the major inhibitory neurotransmitter in
spinal cord. Glycine receptors (GlyRs) function as pen-
tameric anion channels belonging to the ‘cysteine-loop’
superfamily of ionotropic neurotransmitter receptors.
Pharmacological fine-tuning of these receptors can be
performed by allosteric agents: for example, A-type c-
aminobutyric acid receptors by 1,4-benzodiazepines
and neurosteroids. GlyRs play predominant roles in
the processing of motor and sensory signals, neuronal
development, inflammatory pain sensitization, and in
inherited neurological disorders such as hyperekplexia.
Although they are potential targets of muscle relaxant,
sedative, and analgesic agents, they are still ‘therapeutic
orphans’. Due to the structural homology of cysteine-
loop receptors, especially of their binding cavities,
modulators. Since tropisetron has shown GlyR sub-
6
,7
unit-selectivity, tropeines might serve as a promising
lead to develop selective modulators of GlyRs. Here,
we extended the structure–activity analysis of tropeines
to modify their alcohol as well as acid parts. Further,
diesters were also examined because several advantages
of twin drugs might be exploited in medicinal chemis-
8
try. A ternary allosteric model was used which has been
successfully applied for the binding of the antagonist
3
9
[ H]strychnine to GlyRs. By this way we determined
not only the binding affinity of allosteric modulators
1
(K ) but also the cooperativity factor b of their alloste-
A
ric interactions with glycine which is a relevant property
in glycinergic neurotransmission.
2
5
-HT type serotonin receptor antagonists such as trop-
3
3
isetron potentiate GlyR function. Preliminary struc-
ture–activity analysis has concluded that tropeines
2. Results and discussion
2.1. Chemistry
(tropine esters) serve as a consensus structure of positive
allosteric modulation of GlyRs: most potent ‘Gly-posi-
tive’ agents are aromatic esters of 3a-hydroxy-tropane
Tropane alkaloids occur as esters of relatively simple
carboxylic acids and tropine, or W-tropine. A large num-
ber of different esters were also synthesized, owing to
their pharmacological significance. Several acylation
4
(
tropine). Starting from 5-HT3 receptor antagonists
Keywords: Glycine receptors; Tropeines; Ternary allosteric model;
Glycine displacement.
1
0,11
methods
are known for the preparation of tropeines
*
Corresponding author. Tel.: +36 1 325 7900/282; fax: +36 1 325
554; e-mail: maksay@chemres.hu
and nortropeines, which can be applied for the synthesis
of the compounds described in this paper.
7
0
968-0896/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2007.10.097

G. Maksay et al. / Bioorg. Med. Chem. 16 (2008) 2086–2092
2087
1
7
3
a-Succinyl-bis-N-ethyl-nortropane (2) and 3a-tereph-
roethyl chloroformate in refluxing dichloroethane
and methanol, respectively, to get norscopolamine (12,
Fig. 6).
thaloyl-bis-tropane (4) (Fig. 4) were prepared from N-
ethyl-nortropine and tropine, respectively, with the
corresponding dicarboxylic acid chloride according to
1
2
a known procedure.
Compounds 1, 3, 6, 10, and 11 were donated by Profes-
sor L a´ szl o´ Gyermek (Harbor-UCLA Medical Center,
1
2,18
Acylation of the hydroxyl group of tropine can also be
performed by means of ethyl esters. For the synthesis
of 3a-(3 -pyridinecarbonyloxy)-tropane (5) the transe-
sterification method was used. Nicotinic acid ethyl ester
and tropine were heated at 120 ꢁC in the presence of cat-
alytic amount of sodium, under reduced pressure,
whereby the alcohol produced was distilled off. 3a-(3 -
Indole-3 -carbonyloxy)-tropane
obtained according to the procedure described. 3b-
Torrance, CA, USA).
0
13
2.2. Binding to glycine receptors
Most esters resulted in concentration-dependent, full
displacement of specific [ H]strychnine binding as shown
3
0
for compounds 3 and 8 in Figure 1. Since residual bind-
ing accompanying maximal displacement cannot be dif-
ferentiated accurately from non-specific binding, the
extent of strong negative cooperativity could not be ex-
actly determined via the ternary allosteric model and Eq.
2 resulting in a ꢁ 10 in Table 1.
0
(tropisetron)
was
14
0
(
9 -Xantenecarbonyloxy)-tropane was obtained in excel-
lent yield from W-tropine and from the acyl chloride in
CH Cl in the presence of TEA.
2
2
Several methods are known for the demethylation of
tropane derivatives.
Since tropeines proved to be most potent modulators of
GlyRs as benzoate esters, we examined their diesters as
twin agents. The 3a-isophthaloyl-bistropane (3) was
much more potent than 3a-terephthaloyl-bistropane 4
(Table 1) or most of the substituted benzoyloxy-mono-
tropanes to show the importance of the distance and
orientation of two coplanar ester moieties. The planar
trans orientation of the diester 3a-fumaroyl-bistropane
15,16
By use of ethyl chloroformate,
demethylation takes place readily under the formation
of methyl chloride and N-ethoxycarbonyl derivatives.
However, the removal of the ethoxycarbonyl group re-
quires vigorous reaction conditions, so that the ester
function is also cleaved. Therefore, the use of 2,2,2-tri-
chloroethyl chloroformate is much more appropriate
for the demethylation of tropeines, and it was used for
the preparation of 7, 8, and 9 (Scheme 1 and Fig. 5).
The removal of the trichloroethoxycarbonyl group from
the nitrogen atom can be achieved with zinc in acetic
acid solution (Scheme 1). This reductive procedure does
not affect the ester group, and nortropeines can be ob-
tained in good yield.
5
1 resulted in fair potency (K = 7.5 lM) but the decon-
A
jugation of ester moieties (e.g., succinyl) rendered com-
pound 2 completely inactive (Table 1). Extension of
the tropane skeleton of 3a-fumaroyl-bistropane (1) into
3a-fumaroyl-bisgranatane (11) also resulted in a reduced
binding affinity (from K = 7.5 to 20.8 lM in Table 1).
A
The acid parts of the esters were also modified. The het-
0
Demethylation of scopolamine and the subsequent
cleavage of the carbamate was performed using a-chlo-
eroaromatic nicotinic ester 3a-(3 -pyridinecarbonyloxy)-
0
tropane (5) resulted in K = 2.2 lM, while its 6 -chloro
A
N
H C
N
H C
N
3
3
5 R =
a
N
6
R =
Cl
Tropisetron
O
OH
R
R =
N
H
O
b
CCl CH OC
N
HN
3
2
c
O
R
O
N
R
7
8
R =
R =
O
O
N
H
Scheme 1. Synthesis of tropine and nortropine esters. Reagents and conditions: (a) for the preparation of compounds 5 (HBr) and 6, tropine,
2 3 2 3 3
RCOOEt, Na, 120 ꢁC; (b) for compounds 7 (HCl) and 8 (HCl), ClCOOCH CCl , K CO , toluene, reflux; (c) Zn, CH COOH, water, room
temperature.

2088
G. Maksay et al. / Bioorg. Med. Chem. 16 (2008) 2086–2092
3
8
8
demethylation. 3a-(Hydroxydiphenylacetoxy)-N-allyl-
1
.0
.8
.6
.4
.2
nortropane (10) with its non-conjugated phenyl rings
was quite inactive (Table 1). The separation of the phe-
nyl ring from the carbonyloxy moiety by a hydroxyalkyl
group in atropine and the annellation of an epoxide ring
to the tropane ring of atropine made scopolamine inac-
+Gly(10μM)
0
0
4
tive on GlyRs. However, N-demethylation restored the
affinity of norscopolamine (12) to K = 8 lM (Table 1).
0
A
0
The cooperativity factor b of the allosters with glycine in
displacement has been found to correlate well with the
modulation of ionophore activity of GlyRs: allosteric
agents can be distinguished which have positive
1 -
0
10
10-9
tropeine] (M)
10-8
10-7
10-6
[
(
ities with glycine. Therefore, concentration-dependent
b < 1), negative (b > 1), and neutral (b ꢀ 1) cooperativ-
9
3
Figure 1. Concentration-dependent displacement of [ H]strychnine
binding by 3a-isophthaloyl-bistropane (3, j), nortropisetron (8, s)
and the effect of 10 lM glycine on nortropisetron displacement (•).
Points are means ± SEM of 3–4 experiments. Fitting to Eq. 2 resulted
3
displacement of [ H]strychnine binding by glycine was
examined simultaneously in the absence and presence
of the tropeines at low occupancy of their binding sites.
Figure 2 compares the distinctive effects of positive
versus negative allosteric agents. Representative com-
pounds were isophthaloyl-bistropane 3 and nortropise-
tron 8 in Figure 2. Negative modulators (b > 1)
resulted in displacement curves of glycine merging into
control (Fig. 2A), while positive modulators (b < 1)
shifted the curve downwards (Fig. 2B).
A
in K and a values of Table 1.
0
0
substitution decreased the potency of 3a-(6 -chloro-3 -
pyridinecarbonyloxy)-tropane (6) to K = 5.7 lM
(Table 1). N-Demethylation in 3a-(3 -pyridinecarbonyl-
oxy)-nortropane (7) strongly increased the affinity to
A
0
K = 82 nM (Table 1) as observed previously for tropine
A
5
benzoates.
The presence of 10 lM glycine has led to the appearance
of a high-affinity (nanomolar), but minor displacement
3
,4
3
4
Indolic tropeines are also active on GlyRs. Tropise-
tron is particularly promising since its femtomolar con-
phase of [ H]strychnine binding by tropisetron. There-
fore, nortropisetron displacement was also examined
in the presence of 10 lM glycine which resulted in dis-
placement to about 60% (Fig. 1). However, Figure 1
shows that no distinct high-affinity phase could be
observed for nortropisetron, although glycine
significantly decreased the slope value of displacement
to n = 0.75 ± 0.04 as well as the IC50 value to
9.3 ± 2.6 nM nortropisetron (mean ± SEM of three
experiments). Thus, nortropisetron displays one of the
highest affinities reported for GlyR ligands. The de-
creased slope and increased potency can be reconciled
with displacement heterogeneity due to the appearance
1
9
centrations potentiate a1 GlyRs and it has shown
subunit-selectivity for both potentiation and inhibition
6
,7
of GlyRs. Therefore, N-demethylated tropisetron (8)
was prepared. Nortropisetron 8 was the most potent dis-
placer with K = 17 nM, more than 100 times stronger
A
4
than tropisetron at GlyRs. A tricyclic xantene ester
was also prepared and examined. The binding site toler-
ated 3b-(9 -xantenecarbonyloxy)-nortropane 9 fairly
0
(
K = 3.1 lM in Table 1). The disadvantages of 3b ori-
A
5
entation of the ester group and non-planarity of its two
phenyl rings were compensated by the advantageous N-
3
Table 1. Binding affinity and cooperativity factors of tropane derivatives with [ H]strychnine (a) and glycine (b) for GlyRs of rat spinal cord
a
a
Compounds
K
A
(lM)
b
b
1
2
3
4
5
6
7
8
9
1
1
1
. 3a-Fumaroyl-bistropane
7.5 ± 2.4
ꢁ100
2.6 ± 0.4
ND
. 3a-Succinyl-bis-N-ethyl-nortropane
. 3a-Isophthaloyl-bistropane (50 nM)
. 3a-Terephthaloyl-bistropane (2.5 lM)
0.197 ± 0.013
24.6 ± 3.0
2.2 ± 0.1
5.7 ± 0.2
0.082 ± 0.019
0.017 ± 0.003
3.1 ± 0.5
ꢀ52
ꢁ10
3.4 ± 0.7
0.14 ± 0.01
ꢁ10
0
b
. 3a-(3 -Pyridinecarbonyloxy)-tropane
6.9 ± 1.5
ꢁ10
ND
0.45 ± 0.08
2.2 ± 1.1
0
0
0
0
0
. 3a-(6 -Chloro-3 -pyridinecarbonyloxy)-tropane (1 lM)
. 3a-(3 -Pyridinecarbonyloxy)-nortropane (10 nM)
ꢁ10
c
. 3a-(3 -Indolecarbonyloxy)-nortropane (5 nM)
ꢁ10
0.73 ± 0.09
0.26 ± 0.05
. 3b-(9 -Xantenecarbonyloxy)-nortropane (0.3 lM)
ꢁ10
0. 3a-(Hydroxydiphenylacetoxy)-N-allyl-nortropane
1. 3a-Fumaroyl-bisgranatane (6 lM)
2. Norscopolamine
20.8 ± 2.5
8.0 ± 0.6
ꢁ10
ꢁ10
1.2 ± 0.7
b
ND
d
Data are means ± SEM fitted to binding data of three to six experiments. Fitting to Eq. 2 resulted in K
via Eq. 1. K values were 18.9 ± 8.6 lM (mean ± SEM of 10 experiments).
Most of the esters resulted in full displacement of specific binding not distinguishable from non-specific binding (see Fig. 1) therefore their a values
A
and a values, while b values were determined
L
a
could not be exactly determined (a ꢁ 10).
b
c
ND, not determined.
Nortropisetron.
d
6b,7b-Epoxy-1aH,5aH-nortropan-3a-ol (ꢂ)tropate.

G. Maksay et al. / Bioorg. Med. Chem. 16 (2008) 2086–2092
01
2089
1
A
β>1
control
isophthaloyl
bistropane(50nM)
(
11)
1.0
0.8
0.6
0.4
0.2
3
+
7
100
ß
(
8)
6
9
4
1
0-1
1
0-8
10-7
10-6
KA (M)
10-5
Figure 3. Correlation between the logarithms of binding affinity (K
A
)
and cooperativity factors with glycine (b) of tropine esters for
3
[
H]strychnine binding to rat spinal GlyRs. Data are logarithms of
2
1 -
0
5
10-4
the means ± SEM in Table 1. Correlation coefficient: r = 0.89.
Nortropisetron and 3a-fumaroyl-bisgranatane in parentheses (8 and
[Gly] (M)
1
1) did not fit in the correlation and were excluded.
β<1
control
B
R
N
N
R
1
R = Me X=
+
nortropisetron(5 nM)
1
0
0
0
0
.0
.8
.6
.4
.2
2
3
R = Et, X= (CH2)2
R = Me X=
O
O
X
4
R = Me, X=
O
O
Figure 4. The structures of dicarboxylic acid bis-tropine esters 1, 2, 3, 4
HCl).
(
ilar correlation was found for a structurally homoge-
5
neous group of tropine benzoates. In contrast, the
bisgranatane ester 11 with about neutral cooperativity
(
Table 1) did not fit in the correlation of Figure 3. This
1 -
0
6
10-5
Gly] (M)
confirms the preference for tropeine structures in the
allosteric modulation of GlyRs. Nortropisetron did
not fit in the correlation either, possibly because of its
indole-specific, high-affinity, positive allosteric interac-
tions with glycine.
[
3
Figure 2. Displacement of [ H]strychnine binding by glycine in the
absence (control, s) and presence of 50 nM 3a-isophthaloyl-bistro-
pane (A) and 5 nM nortropisetron (B) representing negative (A, b > 1)
and positive (B, b < 1) cooperativities, respectively, with glycine.
3
It should be noted that nortropeines including nortrop-
isetron have recently shown nanomolar displacing
potencies on [ H]strychnine binding to homomeric
Ratios of [ H]strychnine binding in the presence of glycine and the
S
tropeines (BSAL) over control (B ). Data are means ± SEM of three
3
experiments. Fitting to Eq. 1 resulted in the b values of Table 1.
GlyRs containing a , the major human subunit, tran-
1
2
0
siently expressed in human embryonic kidney 293 cells
and these potencies have been amplified by the hyper-
of a minor high-affinity phase. We can conclude that
glycine ‘creates’ a minor component of ‘superhigh’-affin-
ity displacement by nortropisetron but it cannot be sep-
arated from the major nanomolar phase. Consequently,
low nanomolar concentrations of nortropisetron show
mutual positive binding cooperativity with glycine
which is promising for the development of high-affinity,
subunit-selective positive modulators of GlyRs.
ekplexia mutation R271L of a subunits. Consequently,
1
3
displacement of [ H]strychnine binding offers an in vitro
method to evaluate agents against neurological disor-
2
ders associated with inherited mutations of GlyRs.
0
3. Conclusions
Tropeines 3 and 8 with high, nanomolar affinity dis-
played negative cooperativity with glycine, while the
ones with low, micromolar affinity (4, 6, and 9) showed
positive cooperativity (Table 1). Figure 3 demonstrates a
This study extended the structure–activity analysis of
tropine esters and allosteric modulation of GlyR bind-
ing with the modification of acid and alcohol parts.
The tropeine binding site tolerated tricyclic carboxylate
esters and twin tropeines as well. Not only nortropeines
but also properly oriented bistropine esters with conju-
2
good correlation (r = 0.89) between the logarithms of
K and b values of mono and bis (nor)tropeines. A sim-
A

2090
G. Maksay et al. / Bioorg. Med. Chem. 16 (2008) 2086–2092
N
HN
O
O
Ph
O
OH
Ph
O
9
10
O
0
Figure 5. The structures of 3b-(9 -xantenecarbonyloxy)-nortropane (9) and 3a-(hydroxydiphenylacetoxy)-N-allyl-nortropane (10).
H C
with powdered K CO , and then it was extracted with
2 3
3
N
HN
H C
CH Cl (2· 5 mL). The combined organic phases were
2
2
3
N
washed with H O (4 mL), dried over MgSO , and evap-
orated under reduced pressure to yield the free base,
which was crystallized from diisopropyl ether (1.51 g,
2
4
OH
O
O
1
3
8%), mp: 104 ꢁC. H NMR (CDCl ) d (ppm): 5.02 (t,
3
O
J = 5.2 Hz, 2H, H-3), 3.21 (br s, 4H, H-1 and H-5),
.61 (s, 4H), 2.39 (q, J = 7.2 Hz, 4H, CH –CH ), 2.14
O
O
12
2
11
2
3
O
(t, J = 4.2 Hz, 2H, CH –CH ), 2.07 (t, J = 4.2 Hz, 2H,
2 2
O
CH –CH ), 1.93 (br s, 8H), 1.05–1.71 (m, 4H), 1.08 (t,
2 2
1
3
J = 7.2 Hz, 6H, CH –CH ); C NMR d (ppm): 171.85
2
3
Figure 6. The structures of 3a-fumaroyl-bisgranatane (11) and nor-
scopolamine (12).
(C@O), 68.32 (C-3), 57.54 (C-1 and C-5), 46.07 (CH2–
CH ), 36.56 (C-2 and C-4), 29.96 (CH –CH ), 26.33
(C-6 and C-7), 14.13 (CH –CH ). Anal. Calcd for
2
2
3
2 3
gated aromatic dicarboxylic acids in meta position
showed nanomolar displacing potencies. The highest
C H N O : C, 67.32; H, 9.24; N, 7.14. Found: C,
22 36 2 4
67.01; H, 9.29; N, 6.99.
affinity of nortropisetron with K ꢀ 10 nM is a promis-
A
ing lead to develop GlyR subunit-selective agents.
4.1.3. 3a-Terephthaloyl-bistropane (4). To a solution of
tropine (1.41 g, 10 mmol) in CH Cl (5 mL) is added
dropwise terephthaloyl chloride (1.02 g, 5 mmol) in
CH Cl (10 mL). The mixture was stirred at room tem-
2
2
4. Experimental
2
2
perature for 1 h. The solvent was then removed in vac-
uum, and to the oily residue acetone (5 mL) was
added. The precipitate was filtered to yield white crystal-
4
.1. Chemistry
1
The structures of the final products as free bases were
confirmed by H and C NMR. Spectra were recorded
on a Bruker Avance 250 (250 MHz) spectrometer, in
line substance (1.15 g, 56%) mp >240 ꢁC (HCl salt). H
1
13
NMR (CDCl ) d (ppm): 8.03 (s, 4H, H-Ar), 5.20 (t,
3
J = 4.9 Hz, 2H, H-3), 3.12 (br s, 4H, H-1 and H-5),
1
3
CDCl solutions. Chemical shifts (d) are expressed in
2.90 (s, 6H, N–CH ), 1.61–2.25 (m, 16H); C NMR d
3
3
ppm relative to the internal standard TMS. Infrared
spectra were obtained with a Perkin-Elmer 1605 FT-
IR spectrometer. The purity of the compounds was con-
trolled with microanalysis, which was carried out on a
Heraeus Micro Rapid CHN. Melting points were deter-
mined on a B u¨ chi SMP 20 apparatus.
(ppm): 165.23 (C@O), 134.75 (C-Ar), 129.71 (C-Ar)
68.95 (C-3), 59.99 (C-1 and C-5), 40.68 (N–CH ),
3
36.86 (C-2 and C-4), 26.08 (C-6 and C-7). Anal. Calcd
for C H N O : C, 69.88; H, 7.82; N, 6.79. Found: C,
69.55; H, 7.65; N, 6.97.
2
4
32
2
4
0
4
.1.4. 3a-(3 -Pyridinecarbonyloxy)-tropane (5). Nicotinic
4
.1.1. General procedure for the preparation of the free
bases. Powdered K CO (138 mg, 1 mmol) was added to
acid ethyl ester (1.52 g, 10 mmol), tropine (1.55 g,
11 mmol), and sodium (0.05 g 1.15 mmol) were heated
at 120 ꢁC for 24 h at 15 mm in vacuum, whereby ethanol
produced was distilled off. Methanol was added to the
mixture, and it was stirred for 10 min. The solvent was
then removed, water was added to the residue, and it
was extracted with diethyl ether. The ethereal solution
was dried and evaporated to give thick yellow oil
2
3
a solution of the salt (0.5 mmol) in water (2 mL) and the
mixture was extracted with CH Cl (2 · 3 mL). The
2
2
combined organic phases were washed with water
(
4 mL), dried (MgSO ), and the solvent was evaporated
4
under reduced pressure to yield the free base, which was
used for NMR and microanalysis.
(1.7 g, 69%). It was dissolved in acetone and acidified
with cc. HBr solution to afford white crystals (1.85 g),
4
tion of N-ethyl-nortropine (1.55 g, 10 mmol) in CH Cl
.1.2. 3a-Succinyl-bis-N-ethyl-nortropane (2). To a solu-
1
recrystallized from ethanol, mp >240 ꢁC. H NMR
2
2
0
(
5
5 mL) is added dropwise succinyl chloride (0.775 g,
mmol) in CH Cl (10 mL). The mixture was stirred
(CDCl ) d (ppm): 9.23 (d, J = 1.3 Hz, 1H, H-2 ), 8.81
3
0
(dd, J = 4.7 Hz, J = 1.3 Hz, 1H, H-4 ), 8.32 (d,
0
2
2
at room temperature for 1 h. The solvent was then re-
moved in vacuum, and to the oily residue water
J = 7.9 Hz, 1H, H-6 ), 7.44 (dd, J = 4.7 Hz, 7.9 Hz,
0
1H, H-5 ), 5.31 (t, J = 5.1 Hz, 1H, H-3), 3.20 (br s,
2H, H-1 and H-5), 2.33 (s, 3H, N–CH ), 1.84–2.17 (m,
(
10 mL) was added. The solution was made alkaline
3

G. Maksay et al. / Bioorg. Med. Chem. 16 (2008) 2086–2092
2091
1
3
0
0
8
1
5
3
H); C NMR d (ppm): 164.79 (C@O), 153.65 (C-6 ),
4.1.7. 3b-(9 -Xantenecarbonyloxy)-nortropane (9). W-
Tropine (1.41 g, 0.01 mol) and TEA (1.1 g, 11 mmol)
were dissolved in CH Cl (15 mL), and 9-xantenecarb-
0
0
0
51.00 (C-2 ), 137.27 (C-4 ), 126.88 (C-3 ), 123.71 (C-
0
), 69.06 (C-3), 60.06 (N–CH ), 40.80 (C-1 and C-5),
3
2
2
7.00 (C-2 and C-4), 26.07 (C-6 and C-7). Anal. Calcd
oxylic acid chloride (2.69 g, 11 mmol) was added drop-
wise, at room temperature in CH Cl (10 mL). The
solution was then stirred for 2 h and extracted with sat-
for C H N O : C, 68.27; H, 7.37; N, 11.37. Found:
2
1
4
18
2
2
2
C, 68.39; H, 7.15; N, 11.60.
urated NaHCO solution. The organic phase was sepa-
3
4
.1.5. General procedure for the demethylation of trope-
ines. Tropeines (5 mmol) were dissolved in toluene
10 mL) and a catalytic amount of K CO was added.
rated, dried with Na SO , and evaporated at reduced
4
2
pressure. The W-tropine ester was demethylated and
the carbamate was cleaved according to the general pro-
cedure to yield white crystalline substance (2.85 g, 85%),
(
2
3
The solution was boiled and trichloroethyl chlorofor-
,16
5
1
mate
(7.5 mmol) in toluene (5 mL) was added drop-
mp 141 ꢁC; H NMR (CDCl ) d (ppm): 6.92–7.21 (m,
3
0
wise in 20 min while stirring. The mixture was refluxed
for additional 2 h, cooled, and extracted twice with
8H, H-Ar), 5.30 (s, 1H, H-9 ), 4.66–4.95 (m, 1H, H-3),
3.57 (br s, 2H, H-1 and H-5), 1.79–1,41 (m, 8H);
1
3
C
1
rated, dried over Na SO , and evaporated. The oily
material was crystallized from diisopropyl ether and
used without further purification. Yield: 85–90%.
0% acetic acid solution. The organic phase was sepa-
NMR d (ppm): 171.83 (C@O), 151.63, 129.48, 129.18,
123.65, 118.55, 117.37 (aromatics), 68.15 (C-3), 54.40
(C-1 and C-5), 37.02 (C-2 and C-4), 28.31 (C-6 and C-
7). Anal. Calcd for C H NO : C, 75.20; H, 6.31; N,
2
4
2
1
21
3
4.18. Found: C, 75.07; H, 6.18; N, 4.19.
4.1.6. General procedure for the removal of the trichlo-
rocarboethoxy group. N-Trichlorocarboethoxy-nor-
tropeine (4.5 mmol) dissolved in acetic acid (47 mL)
was added dropwise to a suspension of zinc dust
4.1.8. Norscopolamine (12). To a solution of scopolamine
(6b,7b-epoxy-1aH,5aH-tropan-3a-ol
5
(ꢂ)tropate)
(0.31 g, 1.0 mmol) in 1,2-dichloroethane (5 mL), a-chlo-
roethyl chloroformate (0.17 g, 1.2 mmol) and TEA
(0.12 g, 1.2 mmol) were added. The solution was refluxed
for 10 h and then it was evaporated under reduced pres-
sure. To the residue methanol (5 mL) was added and re-
fluxed for 2 h. The solution was then evaporated,
saturated K CO solution was added, and it was extracted
(
1.46 g, 22 mmol) in water (5 mL) at room temperature.
After stirring for 2 h, most of the acetic acid was re-
moved in vacuum, and the residue was cooled to
1
0 ꢁC, and made alkaline by adding an aqueous solution
of 15% of NaOH and 5% of potassium–sodium tartrate.
The solution was then saturated with K CO and ex-
tracted three times with CH Cl . The organic phases
2
3
2
3
with CH Cl . The solution was dried over Na SO and
2 2 2 4
2
2
were combined, the solution was dried over Na SO
2
evaporated to give a yellow oil, which was crystallized
from diisopropyl ether to give white crystalline substance
4
and evaporated to give oily materials that could be
either crystallized, or were converted into their hydro-
chloride salts. Yield: 50–61%.
1
(0.20 g, 69%), mp 95 ꢁC; H NMR (CDCl ) d (ppm): 7.22–
3
7.36 (m, 5H, H-Ar), 5.00 (t, J = 4.9 Hz, 1H, H-3), 4.14–
0
4.15 (m, 1H, H-2 ), 3.71–3.82 (m, 2H, CH –OH), 3.26
2
0
4
.1.6.1. 3a-(3 -Pyridinecarbonyloxy)-nortropane (7).
(d, J = 2.9 Hz, 1H, H-6), 3.17 (br s, 1H, H-5), 3.02 (br s,
1H, H-1), 2.12 (d, J = 2.9 Hz, 1H, H-7), 1.95–2.13 (m,
2H), 1.67 (br d, J = 15.2 Hz, 1H), 1.39 (br d,
White crystal (0. 52 g, 50%); mp > 240 ꢁC (HCl salt);
1
H NMR (CDCl ) d (ppm): 9.16 (d, J = 1.4 Hz, 1H,
3
0
0
13
H-2 ), 8.72 (dd, J = 4.8 Hz, J = 1.4 Hz, 1H, H-4 ), 8.22
0
J = 15.2 Hz, 1H). C NMR d (ppm): 171.99 (C@O),
136.23, 129.32, 128.42, 128.29 (aromatics), 67.14 (C-3),
(
d, J = 7.9 Hz, 1H, H-6 ), 7.35 (dd, J = 4.8 Hz, 7.9 Hz,
0
0
1
2
H, H-5 ), 5.30 (t, J = 4.8 Hz, 1H, H-3), 3.53 (br s,
63.96 (CH –OH), 54.83 (C-6), 54.26 (C-7), 53.79 (C-2 ),
2
1
3
H, H-1 and H-5), 1.82–2.41 (m, 8H); C NMR d
52.12 (C-1), 52.00 (C-5), 31.55 (C-2), 31.13 (C-4). Anal.
Calcd for C H NO : C, 66.42; H, 6.62; N, 4.84. Found:
C, 66.28; H, 6.71; N, 4.72.
0
0
(
1
ppm): 163.56 (C@O), 152.38 (C-6 ), 149.75 (C-2 ),
1
6
19
4
0
0
0
35.98 (C-4 ), 125.60 (C-3 ), 122.41 (C-5 ), 68.48 (C-3),
2.27 (C-1 and C-5), 36.38 (C-2 and C-4), 28.42 (C-6
5
and C-7). Anal. Calcd for C H N O : C, 67.22; H,
4.2. Biology: membrane preparation and binding studies
Crude synaptic membranes were prepared from male
1
3
16
2
2
6
.94; N, 12.06. Found: C, 66.98; H, 6.85; N, 12.02.
0
9
4
.1.6.2. 3a-(3 -Indolecarbonyloxy)-nortropane (8).
Wistar rats as described. Briefly, spinal cords homoge-
Tropisetron was demethylated and transformed to the
nor derivative according to the general procedure de-
scribed above. Its hydrochloride salt was recrystallized
nized in 5 mM Tris–HCl buffer (pH 7.4) were incubated
on ice for 20 min and centrifuged at 30,000g for 20 min.
The pellets were resuspended in 50 mM Tris–HCl buffer
(pH 7.4), centrifuged similarly four times, and frozen.
The thawed membranes were centrifuged in 50 mM
Tris–HCl buffer containing 200 mM KSCN (pH 7.4)
at 10,000g for 10 min before binding assays.
1
from ethanol; (0.74 g, 61%); mp >240 ꢁC; H NMR
(
CDCl ) d (ppm): 11.17 (br s, 1H, NH), 7.66–7.81 (m,
3
1
6
3
H, H-Ar), 7.50 (s, 1H, H-Ar), 7.11 (m, 1H, H-Ar),
.78–6.86 (m, 2H, H-Ar), 4.90 (t, J = 4.7 Hz, 1H, H-3),
1
3
.19 (2H, br s, H-1, H-5), 1.51–1.86 (m, 8H);
0
C
NMR d (ppm): 164.55 (C@O), 136.79 (C-8 ), 131.69
The binding assay was performed in 1 mL membrane
suspensions in 50 mM Tris–HCl plus 200 mM KSCN
0 0 0 0
(
C-9 ), 128.1 (C-2 ), 126.20 (C-5 ), 122.62 (C-4 ),
0
0
0
3
1
21.49 (C-6 ), 120.81 (C-7 ), 112.35 (C-3 ), 66.95 (C-3),
3.28 (C-1 and C-5), 37.49 (C-2 and C-4), 29.36 (C-6
with 3 nM [ H]strychnine (10 lCi/mmol, Dupont-
5
and C-7). Anal. Calcd for C H N O : C, 71.09; H,
NEN). Samples were incubated at 0 ꢁC for 35 min.
Duplicate aliquots were filtered on Whatman GF/B fil-
ters under vacuum and washed by 3· 3-mL ice-cold buf-
1
6
18
2
2
6
.71; N, 10.36. Found: C, 70.83; H, 6.68; N, 10.40.

2092
G. Maksay et al. / Bioorg. Med. Chem. 16 (2008) 2086–2092
fer. Radioactivity of the filters was measured by scintil-
lation spectrometry. Non-specific binding was deter-
mined in the presence of 2 mM glycine.
References and notes
1. Betz, H.; Laube, B. J. Neurochem. 2006, 97, 1600.
2
. (a) Laube, B.; Maksay, G.; Schemm, R.; Betz, H. Trends
Pharmacol. Sci. 2002, 23, 519; (b) Gelmi, M. L.; Pocar, D.;
Pontremoli, G.; Pellegrino, S.; Bombardelli, E.; Fontana,
G.; Riva, A.; Balduini, W.; Carloni, S.; Cimino, M.;
Johnson, F. J. Med. Chem. 2006, 49, 5571; (c) Gelmi, M.
L.; Fontana, G.; Pocar, D.; Pontremoli, G.; Pellegrino, S.;
Bombardelli, E.; Riva, A.; Balduini, W.; Carloni, S.;
Cimino, M. J. Med. Chem. 2007, 50, 2245.
3. Chesnoy-Marchais, D. Br. J. Pharmacol. 1996, 118, 2115.
4. Maksay, G. Neuropharmacology 1998, 37, 1633.
5. Maksay, G.; Nemes, P.; B ´ı r o´ , T. J. Med. Chem. 2004, 47,
384.
6. Maksay, G.; Laube, B.; Betz, H. J. Neurochem. 1999, 73,
802.
. Supplisson, S.; Chesnoy-Marchais, D. Mol. Pharmacol.
000, 58, 763.
. (a) Bourguignon, J. J. In The Practice of Medicinal
Chemistry; Wermuth, C. G., Ed.; Academic Press: Lon-
don, 1996; pp 261–293; (b) Gyermek, L.; Lee, C.; Cho, Y.
M.; Nguyen, N.; Tsai, S. K. Med. Chem. Res. 2002, 11,
1
. Maksay, G.; B ´ı r o´ , T. Neuropharmacology 2002, 43, 1087.
0. Langlois, M.; Soulier, J. L.; Yang, D.; Bremont, B.;
Florac, C.; Rampillon, V.; Giudice, A. Eur. J. Med. Chem.
1993, 28, 869.
4
.3. Data analysis
Non-linear regression programs NLREG (PH Sherrod,
Nashville, TN) and GraphPad Prism Version 4.0 (San
Diego, CA) were used for fitting. The ternary allosteric
9
model of GlyR binding was described previously.
Briefly, it contains three dissociation constants (KS,
K , and K ) for the binding of the antagonist
L
A
3
[ H]strychnine (S), the agonist glycine (L), and the allo-
steric agents (A) as well as the cooperativity factors of A
with [ H]strychnine (a) and with glycine (b). Eq. 1 ex-
presses the ratio of specific [ H]strychnine binding in the
6
3
21
3
7
8
presence of three ligands (B ) over control (B , in the
SAL
S
2
3
presence of [ H]strychnine):
B
B
SAL
½Sꢃ þ K
S
¼
ð1Þ
ð2Þ
Sꢃ þ K ꢄ ½
LꢃðKAþ½Aꢃ=bÞþKLðKAþ½AꢃÞ
KLðKAþ½Aꢃ=aÞ
S
½
S
16.
9
In the absence of agonists Eq. 1 is simplified to:
½Sꢃ þ K
1
B
B
SA
S
¼
KSðKAþ½AꢃÞ
KAþ
S
½
Sꢃ þ
11. Wallace, J. L.; Kidd, M. R.; Cauthen, S. E.; Woodyard, J.
D. J. Heterocycl. Chem. 1978, 15, 315.
½
Aꢃ
a
1
1
1
2. Gyermek, L.; Lee, C.; Cho, Y.-M. U.S. Patent 6,376,510;
Chem. Abstr. 2002, 136, 340861.
3. Zeile, K.; Adickes, F.; Wick, H. U.S. Patent 2,824,106;
Chem. Abstr. 1958, 52, 7365b.
4. Donatsch, P.; Engel, G.; H u¨ gi, B.; Richardson, B. P.;
Stadler, H. R.; Stadler, B. M.; Stadler, S. A.; Breuleux, G.
U.S. Patent 4,789,673, Chem. Abstr. 1988, 114, 81598.
K and a values were determined via Eq. 2. Then concen-
A
tration-dependent displacement by glycine (control, in
the absence of A) resulted in the determination of KL
via Eq. 1. Parallel displacement studies in the absence
and presence of A at low occupancy (between 0.1 and
0
.4 K ) enabled us to determine b values via Eq. 1.
A
15. Kraiss, G.; N a´ dor, K. Tetrahedron Lett. 1971, 1, 57.
1
6. Wallace, J. L.; Kidd, M. R.; Cauthen, S. E.; Woodyard, J.
D. J. Pharm. Sci. 1980, 69, 1357.
Acknowledgments
17. Olofson, F. A.; Martz, J. T.; Senet, J.-P.; Piteau, M.;
Malfroot, T. J. Org. Chem. 1984, 49, 2081.
1
1
2
2
8. Friedman, A. H.; Smith, C. M. Arch. Int. Pharmacodyn.
Ther. 1959, 120, 160.
This work was supported by a Grant K 62203 of the
Hungarian Science Research Fund OTKA. Donation
of samples of compounds 1, 3, 6, 10, and 11 is gratefully
thanked to Professor L a´ szl o´ Gyermek (Harbor-UCLA
Medical Center, Torrance, CA, USA). Technical assis-
tance is acknowledged to Mrs. Ilona Kawka and Mrs.
M a´ ria Kuti.
9. Yang, Z.; Ney, A.; Cromer, B. A.; Ng, H. L.; Parker, M.
W.; Lynch, J. W. J. Neurochem. 2007, 100, 758.
0. Maksay, G.; B ´ı r o´ , T.; Laube, B.; Nemes, P. Neurochem.
Int., doi:10.1016/j.neuint.2007.06.009, in press.
1. Jakub ´ı k, J.; Bacakova, L.; El-Fakahany, E. E.; Tu cˇ ek, S.
Mol. Pharmacol. 1997, 52, 172.