Synthesis and preliminary pharmacological evaluation of 5-Hydroxy- and 5,6-dihydroxy-1,2,3,7,12,12a-hexahydrobenzo[5,6]cyclohepta[1,2,3-ij]isoquinoline derivatives as dopamine receptor ligands

Article abstract of DOI:10.1016/S0968-0896(01)00016-5

A series of 5-hydroxy- and 5,6-dihydroxy-1,2,3,7,12,12a-hexahydrobenzo[5,6]cyclohepta[1,2,3-ij]isoquinoline derivatives 5a-e and 6a-e) were synthesized as conformationally rigid analogues of 1-benzyltetrahydroisoquinoline and evaluated for their affinity and D₁ and D₂ dopamine receptors. All compounds showed lower D₁ and D₂ affinities than dopamine. The 5-hydroxy-1-methyl-2,3,12,12a-hexahydrobenzo[5,6]cyclohepta[1,2,3-ij]isoquinoline 5a and the 5,6-dihydroxy analogue 6a showed D₂ agonist activity. This was proved by their effects on prolactin release from primary cultures of rat anterior pituitary cells. Molecular modeling studies showed that the geometric parameters namely the distances from meta and para hydroxyl oxygens to the nitrogen and the height of nitrogen from the hydroxylated phenyl ring plane) of the dopaminerigc pharmacophore embedded in our components have lower values in comparison with those observed in D₁ and D₂ selective ligands. Copyright

Full text of DOI:10.1016/S0968-0896(01)00016-5

Bioorganic & Medicinal Chemistry 9 (2001) 1447±1458
Synthesis and Preliminary Pharmacological Evaluation of
5-Hydroxy- and 5,6-Dihydroxy-1,2,3,7,12,12a-hexahydrobenzo
[5,6]cyclohepta[1,2,3-ij]isoquinoline Derivatives as Dopamine
Receptor Ligands
Gian Mario Cingolani,a Antonio Di Stefano,b Fabrizio Napolitani,a
Barbara Mosciatti,a Gianfabio Giorgioni,a Nunzia Cinone,b Luigi Brunetti,b
Grazia Luisi,b Barbara Michelotto,b Giustino Orlando,b Barbara Costa,c
Antonio Lucacchini,c Claudia Martinic and Francesco Claudia,*
a
Á
Dipartimento di Scienze Chimiche, Universita di Camerino, Via S. Agostino 1, 62032 Camerino (MC), Italy
b
Á
Dipartimento di Scienze del Farmaco, Universita ``G. D'Annunzio'', Via dei Vestini 31, 66100 Chieti, Italy
c
Á
Dipartimento di Psichiatria, Neurobiologia, Farmacologia e Biotecnologie, Universita di Pisa, 56126 Pisa, Italy
Received 9 October 2000; accepted 15 January 2001
AbstractÐA series of 5-hydroxy- and 5,6-dihydroxy-1,2,3,7,12,12a-hexahydrobenzo[5,6]cyclohepta[1,2,3-ij]isoquinoline derivatives
(5a±e and 6a±e) were synthesized as conformationally rigid analogues of 1-benzyltetrahydroisoquinoline and evaluated for their
anity at D1 and D2 dopamine receptors. All compounds showed lower D1 and D2 anities than dopamine. The 5-hydroxy-1-
methyl-2,3,12,12a-hexahydrobenzo[5,6]cyclohepta[1,2,3-ij]isoquinoline 5a and the 5,6-dihydroxy analogue 6a showed D2 agonist
activity. This was proved by their e€ects on prolactin release from primary cultures of rat anterior pituitary cells. Molecular mod-
eling studies showed that the geometric parameters (namely the distances from meta and para hydroxyl oxygens to the nitrogen and
the height of nitrogen from the hydroxylated phenyl ring plane) of the dopaminergic pharmacophore embedded in our compounds
have lower values in comparison with those observed in D1 and D2 selective ligands. # 2001 Elsevier Science Ltd. All rights
reserved.
Introduction
R-(À)-Apomorphine (1; Fig. 1) is a D1 and D2 DA
receptor agonist. This compound has been the subject of
extensive SARstudies with regard to its interactions
with DA receptors.2 The apomorphine tetracyclic sys-
tem incorporates the 1-benzyltetrahydroisoquinoline
moiety that is embedded also in the D1 selective
antagonists 1-(2,5-dimethoxy-4-propylbenzyl)-6-chloro-
7-hydroxy-2-methyl-1,2,3,4-tetrahydroisoquinoline (2),3
Dopamine-mediated neurotransmission plays an impor-
tant role in several psychiatric and neurological dis-
orders. For this reason great interest has been focused
on the search for novel dopamine (DA) receptor ago-
nists and antagonists. DA receptors are classi®ed into
two families: D1-like (D1 and D5) and D2-like (D2, D3,
and D4), on the basis of amino acid sequences and of
their ability to activate or inhibit the enzyme adenylyl
cyclase, respectively.1 DA antagonists are used for
treating schizophrenia, mania, delirium and Hunting-
ton's disease. Clinical applications of central DA ago-
nists include the treatment of Parkinson's disease and
neuroendocrine disorders.
and
1-(2-bromo-4,5-dimethoxybenzyl)-7-hydroxy-6-
methoxy-2-methyl-1,2,3,4-tetra-hydroisoquinoline (3).4
In a previous work, we described the synthesis of some 5
-hydroxy-2,3,12,12a-tetrahydro-1H-[1]benzoxepino[2,
3,4ij]isoquinolines as monophenolic ligands for DA
receptors.5 The N-methyl derivative 4 bound to the D2-
like receptors with the same anity as DA and showed
D2 agonist activity. As an extension of our research in
this ®eld, here we report the synthesis of 5-hydroxy- and
5,6-dihydroxy-1,2,3,7,12,12a-hexahydrobenzo[5,6]cyclo-
*Corresponding author. Tel.:+39-(0)737-402238; fax: +39-(0)737-
637345; e-mail: claudi@camserv.unicam.it
0968-0896/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(01)00016-5

1448
G. M. Cingolani et al. / Bioorg. Med. Chem. 9 (2001) 1447±1458
Figure 1. Structures of R-(À)-apomorphine (1), tetrahydrobenzylisoquinolines (2, 3), tetrahydrobenzoxepinoisoquinoline (4), and hexahydro-
benzocycloheptaisoquinolines (5a±e, 6a±e).
Scheme 1. Reagents: (a) (CH3)2SO4, tetrabutylammonium bromide, CH2Cl2, NaOH 10%; (b) POCl3, CH3CN, re¯ux; (c) RX, CH3CN; (d) NaBH4,
CHCl3, CH3OH; (e) H2SO4 98%, À40 ꢀC; (f) CH3SO3H, methionine, rt.

G. M. Cingolani et al. / Bioorg. Med. Chem. 9 (2001) 1447±1458
1449
Scheme 2. Reagents: (a) H2, Pd/C 10%, CH3COOH; (b) CH3SO3H, methionine, rt.
hepta[1,2,3-ij]isoquinoline derivatives (5a±e and 6a±e), a
new class of rigid congeners of 1-benzyltetrahy-
droisoquinoline. The new compounds are isosteres of
previously reported tetrahydrobenzoxepino [2,3,4-ij]iso-
quinolines,5 and were synthesized with the aim to verify
whether the isosteric replacement of oxygen with a
methylene could a€ect the anity and selectivity for
DA receptors. The monohydroxy derivatives were pre-
pared in order to develop monophenolic ligands for DA
receptors and to compare their anities with catechol
compounds which, as a rule, show higher anities and
D1 selectivity. Isoquinoline nitrogen was substituted
with methyl, ethyl, n-propyl, and benzyl groups in view
of the fact that they can modulate the D1 and D2
anities.
hepta[1,2,3-ij]isoquinolines was established on the basis
1
of H NMRspectra.
Tetrahydroisoquinoline 15a (Scheme 1) showed three
signals in the aromatic area ascribed to the hydrogens at
C-5, C-7, and C-8. The signal of the hydrogen at C-5
(d=6.85) appeared as a doublet with a coupling con-
stant of 2.64 Hz. This hydrogen is coupled to the
hydrogen on C-7 and the constant value is consistent
with a meta coupling. The hydrogen at C-7 gave a
double doublet (d=6.50) with two coupling constants of
2.64 and 8.45 Hz. This hydrogen is coupled to the
hydrogens on C-5 (J=2.64 Hz) and on C-8 (ortho cou-
pling, J=8.45 Hz). At d 6.18 a doublet with a coupling
constant of 8.45 Hz was present and was ascribed to the
hydrogen on C-8. The spectrum of hexahydrocyclo-
heptaisoquinoline 17a showed two doublets at d 6.85
and 6.74. The coupling constants value was 2.61, thus
indicating a meta coupling between the hydrogen at C-4
and C-6.
Results and Discussion
The 5-hydroxy- and 5,6-dihydroxyhexahydrobenzo[5,6]
cyclohepta[1,2,3-ij]isoquinolines 5a±e and 6a±e were
synthesized successfully as illustrated in Schemes 1 and
2. The hydroxyamides 7 and 8 were prepared in good
yields from 2-[(3-methoxy- or 3,4-dimethoxy)phenyl]
ethylamine and 3-isochromanone in re¯uxing xylene, by
a modi®cation of the method previously described.6
Then, the benzylic hydroxy group was protected by
methylation with dimethylsulphate in a two-phase sys-
tem according to the Merz procedure.7
NMRspectra of the oxalate salts 5a±e and 6a±e showed
a very broad signal (basically a deformation of the
baseline). These signals disappeared after D2O addition,
and they can be assigned to hydroxy and ammonium
groups.
Table 1 gives the binding anities for the novel
hydroxybenzocycloheptaisoquinolines (5a±e, 6a±e) and
methoxylated precursors (17a±e, 18a±e) determined in
DA D1 and D2 receptor assays. Anities are referred to
D1-like and D2-like families of DA receptors. Porcine
striatal membranes were used as the tissue source.
[3H]SCH 23390 and [3H]Spiperone were used as radio-
ligands for the D1 and D2 receptors, respectively.
The amides 9 and 10 were subjected to Bischler±Napier-
alski cyclization with phosphorus oxychloride in dry
acetonitrile. The resulting 3,4-dihydroisoquinolines 11 and
12 were alkylated with the appropriate alkyl iodide
(methyl, ethyl, propyl iodide) or benzyl bromide to give the
corresponding dihydroisoquinolinium salts 13a±d and
14a±d. Reduction with sodium borohydride in CHCl3/
CH3OH a€orded the tetrahydroisoquinolines 15a±d and
16a±d. The hexahydrobenzo[5,6]cyclohepta[1,2,3-ij]isoqui-
nolines 17a±d and 18a±d were obtained by intramolecular
cyclization in concentrated sulfuric acid at low tempera-
ture under nitrogen. The hydroxy derivatives 5a±e and 6a±
e were obtained by demethylation with methanesulfonic
acid and methionine.8 Catalytic hydrogenation of benzyl
derivatives 17d and 18d in glacial acetic acid in presence of
10% Pd/C, followed by O-demethylation, produced the
unsubstituted derivatives 5e and 6e (Scheme 2).
Furthermore, the compound 5a and the dihydroxy ana-
logue 6a were pharmacologically studied in vitro by
evaluating their e€ects on prolactin (PRL) release in
primary cultures of rat anterior pituitary cells to assess
their agonist activity. Previous studies indicated that D2
receptors on lactotroph pituitary cells inhibit PRL
release via inhibition of adenylate cyclase. On the other
hand, there is evidence for a dual role of DA in both
stimulating and inhibiting PRL secretion through acti-
vation of the same DA receptor.9,10
As far as the anities for D1 receptor are concerned, all
compounds show lower binding anity than DA or R-
(À)-apomorphine. Substituents on the nitrogen atom
The cyclization of 1-(2-methoxymethylbenzyl)-5-meth-
oxy-1,2,3,4-tetrahydroisoquinolines to benzo[5,6]cyclo-

1450
G. M. Cingolani et al. / Bioorg. Med. Chem. 9 (2001) 1447±1458
seem to a€ect the D1 anities. Bulky groups decrease
the anity. The N-ethyl, the N-methyl and the unsub-
stituted derivatives are the most active. Methoxy groups
demethylation does not signi®cantly increase the a-
nity. A slight increase is observed in the catechol 6b and
in the monophenolic derivative 5e that is the most active
in the series.
agonist, which has been shown to stimulate PRL release
only at low concentrations (1±100 pM).11
From a comparison of binding anities of 5-hydroxy-1-
methyl-2,3,12,12a-tetrahydro-1H-[1]benzoxepino[2,3,4-
ij]isoquinoline 4 (Fig. 1, D1, Ki=2.88 mM; D2, Ki=
0.27 mM) and 5-hydroxy-1-methyl-2,3,12,12a-hexahy-
drobenzo[5,6]cyclohepta[1,2,3-ij]isoquinoline 5a (D1,
Ki=6.4 mM; D2, Ki=8.9 mM) it is evident that the iso-
steric replacement of oxygen with a methylene decreases
both anity and selectivity.
Data obtained for D2 receptor did not di€er sub-
stantially from those for D1 receptor. The dimethoxy N-
benzyl derivative 18d showed D2 anity similar to that
of DA, while all the others were less potent than DA or
R-(À)-apomorphine. As regards the in vitro e€ects on
PRL release from primary cultures of rat anterior
pituitary cells, the compounds 5a and 6a at low con-
centrations stimulated, and at higher concentrations
inhibited PRL release (Fig. 2). Both compounds showed
agonist activity at D2 receptors.
Binding data obtained with our compounds indicated
that the presence of a phenol or a catechol group did
not modify the anity for D1 and D2 receptor. More-
over, the anities of methoxy derivatives did not di€er
signi®cantly from those of hydroxy derivatives. These
results suggest that all compounds have weak interac-
tions with the receptors. A similar trend was observed
with the previously synthesized 5-hydroxy- and 10-
hydroxy-2,3,12,12a-tetrahydro-1H-[1]benzoxepino[2,3,-
4-ij]isoquinolines which showed low anity and selec-
tivity at both the D1-like and D2-like receptors.5 In
these compounds the lack of the catechol nucleus could
not be invoked as the only factor responsible for anity
to D1 and D2 receptors. Indeed, the ®ndings presented
in the present paper con®rmed that the catechol nucleus
The dihydroxy derivative 6a stimulates PRL release at a
concentration of 1 nM and is more e€ective than 5a. It
can be hypothesized that the PRL-stimulating activity
demonstrated by 5a and 6a is due to the higher anity
to D2 stimulatory receptors, an e€ect which is over-
whelmed at higher concentrations, where D2 inhibitory
e€ects counteract stimulatory ones. This ®nding is simi-
lar to the e€ect of quinpirole, a speci®c D2 receptor
Table 1. Inhibition of [3H]SCH 23390 (D1-like) and [3H]Spiperone (D2-like) binding to porcine striatal membranes of the hexahydrobenzo-
[5,6]cyclohepta[1,2,3-ij]isoquinoline derivatives
Ki (mM)a
Ratio
Compound
RR
R2
D1-like
D2-like
D1-like/D2-like
1
5a
5b
5c
5d
5e
6a
6b
6c
6d
6e
17a
17b
17c
17d
17e
18a
CH3
C2H5
n-C3H7
CH2C6H5
H
CH3
C2H5
n-C3H7
CH2C6H5
H
CH3
C2H5
n-C3H7
CH2C6H5
H
CH3
C2H5
n-C3H7
CH2C6H5
H
H
H
H
H
H
OH
OH
OH
OH
OH
H
H
H
H
H
OCH3
OCH3
OCH3
OCH3
OCH3
OH
OH
OH
OH
OH
OH
OH
OH
6.4Æ0.4
21.8Æ1.8
29.8Æ1.2
41.9Æ2.2
3.7Æ0.1
8.6Æ0.3
6.6Æ0.2
11.6Æ0.8
21.6Æ1.2
6.3Æ0.3
6.1Æ0.3
8.3Æ0.4
22.3Æ1.7
55.4Æ2.1
10.8Æ1.3
7.2Æ0.5
27.4Æ1.8
72.2Æ4.1
61.8Æ3.2
4.3Æ0.2
1.22Æ0.1
0.36Æ0.02
8.9Æ0.6
14.8Æ1.1
40.3Æ3.1
32.3Æ2.0
22.1Æ1.7
8.2Æ0.4
0.72
1.47
0.74
1.29
0.16
1.06
0.68
0.85
3.38
0.33
0.41
0.50
3.33
3.18
1.35
0.31
0.97
5.19
12.6
0.33
0.28
2.0
9.7Æ0.5
13.7Æ0.7
6.4Æ0.4
OH
OH
18.9Æ0.9
14.7Æ1.1
16.5Æ1.2
6.7Æ0.2
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
17.4Æ1.3
8.0Æ0.5
23.3Æ1.5
28.2Æ2.0
13.9Æ0.8
4.9Æ0.2
18b
18c
18d
18e
dopamine
apomorphine
13.1Æ0.9
4.38Æ0.2
0.18Æ0.01
aThe Ki values are meansÆS.E.M. of at least three experiments.

G. M. Cingolani et al. / Bioorg. Med. Chem. 9 (2001) 1447±1458
1451
Figure 2. E€ect of 5a and 6a on PRL release from anterior pituitary cell cultures in vitro after a 1 h incubation. Group meansÆSEM, n=6±8,
*p<0.02 versus control.
was an important but not a sucient feature to display
high anity at D1 receptor (see compound 6a, Table 1).
and D2 receptors, namely one or both the hydroxyl
groups, and the cationic nitrogen.
In order to identify the molecular features responsible
for the low anity and selectivity toward D1 and D2
receptors, molecular modeling studies were performed
on the representative compounds 4, 5a and 6a (Fig. 1).
Although highly congeneric, these compounds alter-
natively displayed all the pharmacophore features
required for the binding of an agonist to both the D1
MACROMODEL 6.512 was used for the compounds
modeling reported in the present study. Initial struc-
tures, built in their protonated forms, were energy
minimized within the MM2* force ®eld,13 using the
Steepest Descendent minimizer, allowing for 1000 itera-
tions per structure, until a gradient of 0.05 kJ/A was
reached.

1452
G. M. Cingolani et al. / Bioorg. Med. Chem. 9 (2001) 1447±1458
Table 2. ÁE and geometry calculations values of the representative
structures of the most populated and energetically accessible con-
formationally related families of 4, 5a and 6a
formationally related families were taken into account
to evaluate and compare geometric parameters of the
pharmacophore patterns. The parameters referring to
the pharmacophore patterns were in accordance with
those used by other authors: (a) the distances between
the cationic nitrogen and the hydroxyl oxygens meta
and para to the ethylamine chain; (b) the height of the
nitrogen above the plane of the ring to which the oxy-
gens are bonded.15,16
Compound Conformation ÁE Timesa
Distance
(A)
Heightd
(A)
1b
2c
4
1
2
3
4
0.00
0.42
0.65
4.85
63
42
44
41
6.48
6.46
6.45
6.41
0.07
0.17
0.37
0.88
The pharmacophore map which had been hypothesized
by Wilcox et al.16 for both the D1 and D2 receptors had
highlighted a distinction between these two receptors in
terms of the pharmacophore geometric parameters
values. For high anity and selective D1 ligands,
such as 3-allyl-6-bromo-7,8-dihydroxy-1-phenyl-2,3,4,5-
tetrahydro-1H-3-benzazepine (Br-APB), these authors
reported: (a) a distance between the nitrogen and meta
and para hydroxyl oxygens of the catechol ring of
around 7.0 and 6.7 A, respectively; (b) a height of the
nitrogen above the plane of the catechol ring around
1.28 A. In the pharmacophore map for D2 agonists, the
distances reported for N-n-propylnorapomorphine
(NPA), a high anity and selective D2 ligand, were 7.6
A for N-m-OH and 6.4 A for N-p-OH; the height of
the nitrogen above the plane of the catechol ring was
1.83 A.
5a
6a
1
2
3
4
0.00
0.92
1.26
4.36
72
47
73
39
6.48
6.46
6.46
6.43
0.02
0.30
0.30
0.85
1
2
3
4
0.00
0.85
1.75
4.45
23
23
20
26
6.48 6.21
6.46 6.18
6.45 6.21
6.44 6.21
0.01
0.31
0.29
0.31
aTimes number that each conformation was found.
bDistance between nitrogen and oxygen of 5-OH.
cDistance between nitrogen and oxygen of 6-OH.
dHeight of the nitrogen from the phenyl ring plane.
In general, the compounds acting on the same receptors
are likely to have a common topology for the pharma-
cophore moieties. In our case, the accuracy in ®nding
their common pharmacophore conformations was
facilitated by the use of highly congeneric rigid com-
pounds. Since the ligands probably do not bind to the
receptor in their global minimum conformations, we
were interested in ®nding families of candidate con-
formations of each compound which were within the
region of conformational space energetically accessible
(ÁE=5 kcal/mol).
Interestingly, we found that the low energy conforma-
tions of the least selective derivatives 4, 5a and 6a had
lower values for the postulated pharmacophore geo-
metry parameters, compared with those observed in D1
and D2 selective ligands. In fact, for any low energy
conformation, the distances between the meta and para
hydroxyl group (5-OH and 6-OH) and nitrogen atom
(Fig. 3) were around 6.5 A (lower compared to the
values of 7 and 7.6 A found in D1 and D2 selective
ligands, respectively), and 6.2 A (lower compared to the
values 6.7 and 6.4 A found in D1 and D2 selective
ligands, respectively).
To search conformational space, a molecular simulation
following a stochastic approach, called Monte Carlo,14
was performed. This approach is based on exploring the
energy surface by randomly probing the geometry of a
molecular system. This led us to reach a more realistic
picture of the molecules, which is likely far from the
static, idealized image provided by molecular mechanics
but where all the atoms are in constant motion. The
Monte Carlo searching algorithm was used in combi-
nation with an energy minimization procedure to
explore the energetically accessible conformations.
The nitrogen atoms were nearly coplanar with the
phenyl ring, approximately 0.01±0.9 A away from the
phenyl ring plane; these values are lower than the
1.28 and 1.83 A found for D1 and D2 ligands, such
as Br-APB and NPA. Consequently, the di€erences
between D1 and D2 selective ligands and D1/D2
unselective ligands lie in the pharmacophore patterns
geometry.
Thus the Monte Carlo/energy minimization protocol of
the MACROMODEL12 program was used until a
number of 500 ®nal structures (ÁE=50 kcal/mol) was
collected. The ring closure bonds were used by the
protocol in six- and seven-membered non-aromatic
rings to allow torsion angles within these rings to be
varied as well. The energy minimization convergence
criterion was set to an RMS gradient of 0.05 kJ/A. All
the heavy atoms were superimposed in order to elimi-
nate duplicate conformations, and thus the output of
this protocol application is an ensemble of representa-
tive structures of conformationally related families
(Table 2).
The combination of these di€erences, in terms of
geometry pharmacophore features, could play a crucial
role in de®ning D1/D2 receptor selectivity. Since the
unselective ligands present lower distance values com-
pared to those found in the pharmacophore patterns of
D1 and D2 selective ligands, it is reasonable to assume
that these geometry parameters values allow ligands to
be recognized by both receptors. Thus, these pharma-
cophore parameters values could help in understanding
the low anity and selectivity at D1 and D2 receptors
displayed by compounds 4, 5a, 6a. Finally, these para-
meters could be important for designing new selective
ligands.
The search results were ®ltered on energy (ÁEꢁ5.0 kcal/
mol). The representative structures of the con-

G. M. Cingolani et al. / Bioorg. Med. Chem. 9 (2001) 1447±1458
1453
6.76 (m, 1H, ArH), 6.62 (m, 2H, ArH), 6.20 (br s, 1H,
NH), 4.61 (s, 2H, CH2O), 3.68 (s, 3H, OCH3), 3.59 (s,
2H, CH2CO), 3.49 (q, J=6.85 Hz, 2H, CH2N), 2.72 (t,
J=6.85 Hz, 2H, ArCH2CN), 2.65 (br s, 1H, OH). Anal.
calcd for (C18H21NO3): C, 72.22; H, 7.07; N, 4.68;
found: C, 71.95; H, 6.91; N, 4.84.
General procedure for the preparation of compounds 9
and 10. The two-phase system consisting of (a) a solu-
tion of the amides 7 or 8 (2.20 mmol) and tetra-
butylammonium bromide (0.005 g, 0.016 mmol) in CH2Cl2
(2.7 mL) and (b) 50% NaOH (0.5 mL) was equilibrated by
vigorous stirring for 15±30 min. Dimethylsulfate (0.25 mL,
2.60 mmol) was then added dropwise with cooling. The
mixture was stirred for 3 h at room temperature and,
after addition of 30% NH4OH (0.1 mL), for a further 30
min. The mixture was poured into water and the
organic phase was separated, washed with water, dried,
and evaporated under reduced pressure.
2-[2-(Methoxymethyl)phenyl]-N-(3-methoxyphenethyl)-
acetamide 9. The crude product was puri®ed by chro-
matography on silica gel (eluent AcOEt/cyclohexane
Figure 3. Superimposed candidate pharmacophore conformations of
low selective compounds 4 (shown in green), 5a (shown in magenta)
and 6a (shown in yellow). The oxygen atoms are shown in red and the
nitrogen in blue. The arrows highlight the distances (A) between
cationic nitrogen and hydroxyl oxygen at position 5 and 6.
1
6:4) to obtain an oily residue in 58% yield. H NMR
(CDCl3) d 7.29 (m, 4H, ArH), 7.13 (m, 1H, ArH), 6.72
(m, 1H, ArH), 6.60 (m, 2H, ArH), 6.12 (br s, 1H, NH),
4.38 (s, 2H, CH2O), 3.76 (s, 3H, OCH3), 3.58 (s, 2H,
CH2CO), 3.40 (q, J=6.6 Hz, 2H, CH2N), 3.26 (s, 3H,
OCH3), 2.67 (t, J=6.6 Hz, 2H, ArCH2CN). Anal. calcd
for (C19H23NO3): C, 72.82; H, 7.39; N, 4.47; found: C,
72.64; H, 7.58; N, 4.23.
Experimental
Chemistry
Melting points were measured on a Buchi 510 apparatus
and are uncorrected. Microanalyses were performed on
a 1106 Carlo Erba CHN Analyzer. Analyses indicated
by the symbols of the elements were withinÆ0.4% of
the calculated values.
2-[2-(Methoxymethyl)phenyl]-N-(3,4-dimethoxyphenethyl)
acetamide 10. The residue was puri®ed by silica-gel col-
umn chromatography (eluent AcOEt) and then recrys-
ꢀ
1
tallized from cyclohexane. Yield 63%, mp 93±95 C. H
NMR(CDCl 3) d 7.27 (m, 4H, ArH), 6.68 (m, 1H, ArH),
6.55 (m, 2H, ArH), 6.10 (br s, 1H, NH), 4.38 (s, 2H,
CH2O), 3.82 (s, 3H, OCH3), 3.80 (s, 3H, OCH3), 3.56 (s,
2H, CH2CO), 3.40 (q, J=6.65 Hz, 2H, CH2N), 3.29 (s,
3H, OCH3), 2.62 (t, J=6.65 Hz, 2H, ArCH2CN). Anal.
calcd for (C20H25NO4): C, 69.95; H, 7.34; N, 4.08;
found: C, 70.19; H, 7.51; N, 4.26.
1H NMRspectra were recorded on a Varian VXR200-
MHz spectrometer. Chemical shift values are reported
in parts per million (d) down®eld from the internal
standard tetramethylsilane (Me4Si). The IRspectra
were run on a Perkin-Elmer Model 297 spectrometer as
Nujol mulls or liquid ®lms. The identity of all new
compounds was con®rmed by both elemental analyses
and NMRdata. TLC was performed with Merck 60
F254 precoated silica on glass. Solutions were routinely
dried over anhydrous sodium sulfate prior to evapora-
tion. Chromatographic puri®cations were performed by
Merck-60 70±230 mesh ASTM silica gel columns with
the reported solvent. Oxalate salts were formed by
dissolving the base in ether and adding 1 equivalent of
oxalic acid dissolved in a small amount of ether.
General procedure for the preparation of dihydroisoquino-
lines 11 and 12. The amide 9 or 10 (1.28 mmol) in dry
acetonitrile (6.4 mL) was treated with phosphoryl chlo-
ride (0.39 g, 2.57 mmol) and re¯uxed under N2 for 15
min. Solvent and excess of reagent were removed in
vacuo. A solution of the residue in CHCl3 (10 mL) was
rapidly washed with 1 N NaOH, dried, and evaporated
under reduced pressure. The residue was converted to the
oxalate salt which was recrystallized from EtOH/Et2O.
2-[2-(Hydroxymethyl)phenyl]-N-(3-methoxyphenethyl)-
acetamide 7. A mixture of 2-(3-methoxyphenyl) ethyla-
mine (0.54 g, 3.60 mmol) and 3-isochromanone (0.56 g,
3.80 mmol) in xylene (2 mL) was re¯uxed under N2 for
2 h with azeotropic removal of water. Solvent was
removed under vacuum. The residue was puri®ed by
silica gel column chromatography (eluent AcOEt) and
recrystallized from CHCl3/cyclohexane. Yield 56%, mp
6-Methoxy-1-[2-(methoxymethyl)benzyl]-3,4-dihydroiso-
quinoline oxalate 11. Yield 90%, mp 175±177 ꢀC. 1H
NMR(DMSO- d6) d 7.80 (d, J=8.79 Hz, 1H, ArH), 7.26
(m, 4H, ArH), 7.06 (s, 1H, ArH), 6.96 (d, J=8.79 Hz,
1H, ArH), 4.51 (s, 2H, CH2O), 3.87 (s, 3H, OCH3), 3.75
(t, J=7.70 Hz, 2H, CH2N+), 3.29 (s, 3H, OCH3), 3.00
(t, J=7.70 Hz, 2H, ArCH2CN). Anal. calcd for
103±105 ꢀC. IR(®lm) n (cmÀ1) 3275 (NH), 3185 (OH),
(C19H21NO2 H2C2O4): C, 65.44; H, 6.02; N, 3.63;
found: C, 65.21; H, 6.19; N, 3.81.
.
1
1630 (CˆO). H NMR(CDCl ) d 7.25 (m, 5H, ArH),
3

1454
G. M. Cingolani et al. / Bioorg. Med. Chem. 9 (2001) 1447±1458
6,7-Dimethoxy-1-[2-(methoxymethyl)benzyl)-3,4-dihy-
droisoquinoline oxalate 12. Yield 83%, mp 197±198 ꢀC.
1H NMR(DMSO- d6) d 7.30 (m, 5H, ArH), 7.10 (s, 1H,
ArH), 4.51 (s, 2H, CH2O), 3.86 (s, 3H, OCH3), 3.80 (t,
J=7.49 Hz, 2H, CH2N+), 3.68 (s, 3H, OCH3), 3.30 (s,
3H, OCH3), 3.04 (t, J=7.49 Hz, 2H, ArCH2CN). Anal.
calcd for (C20H23NO.3H2C2O4): C, 63.60; H, 6.07; N,
3.37; found: C, 63.78; H, 5.87; N, 3.13.
J=7.21 Hz, 2H, CH2N+), 3.90 (s, 3H, OCH3), 3.70 (s,
3H, OCH3), 3.60 (s, 3H, OCH3), 3.38 (s, 3H, N+CH3),
3.20 (t, J=7.21 Hz, 2H, ArCH2CN+). Anal. calcd for
(C21H26INO3): C, 53.97; H, 5.61; N, 3.00; found: C,
53.83; H, 5.42; N, 3.13.
6,7-Dimethoxy-2-ethyl-1-[2(methoxymethyl)benzyl]-3,4-
dihydroisoquinolinium iodide 14b. Recrystallized from
i-PrOH. Yield 60%, mp 181±183 ꢀC. 1H NMR(DMSO-
d6) d 7.48 (m, 1H, ArH), 7.30 (m, 3H, ArH), 7.20 (s, 1H,
ArH), 6,90 (m, 1H, ArH), 4.72 (s, 2H, ArCH2CˆN),
4.60 (s, 2H, ArCH2O), 4.18 (t, J=7.08 Hz, 2H,
CH2N+), 4.02 (q, J=7.12 Hz, 2H, N+CH2C), 3.92 (s,
3H, OCH3), 3.52 (s, 3H, OCH3), 3.41 (s, 3H, OCH3),
3.20 (t, J=7.08 Hz, 2H, ArCH2CN+), 1.34 (t, J=7.12 Hz,
3H, N+CCH3). Anal. calcd for (C22H28INO3): C, 54.89;
H, 5.86; N, 2.91; found: C, 54.66; H, 5.91; N, 2.97.
General procedure for the preparation of alkyliodides
13a±c and 14a±c. The appropriate alkyl iodide (methyl
iodide, ethyl iodide, propyl iodide) (4.37 mmol) was
added to a solution of the base 11 or 12 (2.20 mmol) in
dry acetonitrile. The mixture was re¯uxed for 3 h. The
solvent was removed to give a yellow solid which was
recrystallized.
6-Methoxy-1-[2(methoxymethyl)benzyl]-2-methyl-3,4-di-
hydroisoquinolinium iodide 13a. Recrystallized from
i-PrOH/Et2O. Yield 60%, mp 172±174 ꢀC. 1H NMR
(DMSO-d6) d 7.83 (d, J=8.79 Hz, 1H, ArH), 7.48 (d,
J=6.96 Hz, 1H, ArH), 7.20 (m, 2H, ArH), 7.15 (d,
J=2.28 Hz, 1H, ArH), 7.02 (dd, J=2.28, 8.79 Hz, 1H,
ArH), 6.88 (d, J=6.96 Hz, 1H, ArH), 4.64 (s, 4H,
ArCH2O, ArCH2CˆN+), 4.20 (t, J=7 Hz, 2H,
CH2N+), 3.90 (s, 3H, OCH3), 3.65 (s, 3H, OCH3), 3.40
(s, 3H, N+CH3), 3.25 (t, J=7 Hz, 2H, ArCH2CN+).
Anal. calcd for (C20H24INO2): C, 54.93; H, 5.53; N,
3.20; found: C, 55.11; H, 5.64; N, 3.44.
6,7-Dimethoxy-1-[2(methoxymethyl)benzyl]-2-propyl-6,7-
dimethoxy-3,4-dihydroisoquinolinium iodide 14c. Recrys-
tallized from i-PrOH. Yield 51%, mp 190±192 ꢀC. H
1
NMR(DMSO- d6) d 7.45 (m, 1H, ArH), 7.28 (m, 3H,
ArH), 7.20 (s, 1H, ArH), 6.88 (m, 1H, ArH), 4.75 (s,
2H, ArCH2CˆN), 4.60 (s, 2H, ArCH2O), 4.15 (t,
J=7.21 Hz, 2H, CH2N+), 3.90 (m, 5H, N+CH2CC,
OCH3), 3.60 (s, 3H, OCH3), 3.38 (s, 3H, OCH3), 3.20 (t,
J=7.21 Hz, 2H, ArCH2CN+), 1.79 (m, 2H,
N+CCH2C), 0.92 (t, J=7.4 Hz, 3H, N+CCCH3). Anal.
calcd for (C23H30INO3): C, 55.76; H, 6.10; N, 2.83;
found: C, 56.01; H, 5.89; N, 2.92.
2-Ethyl-6-methoxy-1-[2(methoxymethyl)benzyl]-3,4-dihy-
droisoquinolinium iodide 13b. Recrystallized from
EtOH/Et2O. Yield 80%, mp 135±137 ꢀC. 1H NMR
(DMSO-d6) d 7.82 (d, J=8.76 Hz, 1H, ArH), 7.45 (m,
1H, ArH), 7.30 (m, 2H, ArH), 7.14 (d, J=2.27 Hz, 1H,
ArH), 7.02 (dd, J=2.27, 8.76 Hz, 1H, ArH), 6.85 (d,
J=6.92 Hz, 1H, ArH), 4.69 (s, 2H, ArCH2O), 4.60 (s,
2H, ArCH2CˆN), 4.19 (t, J=7 Hz, 2H, CH2N+), 3.94
(m, 5H, N+CH2C, OCH3), 3.39 (s, 3H, OCH3), 3.21 (t,
J=7 Hz, 2H, ArCH2CN+), 1.50 (t, J=7.04 Hz, 3H,
N+-C-CH3). Anal. calcd for (C21H26INO2): C, 55.88;
H, 5.81; N, 3.10; found: C, 55.82; H, 5.93; N, 2.92.
General procedure for the preparation of N-alkyltetrahy-
droisoquinoline 15a±c and 16a±c. Sodium borohydride
(0.048 g, 1.27 mmol) was added in portion to a stirred
solution of the appropriate alkyliodide 13a±c or 14a±c
(1.27 mmol) in CHCl3 (1.5 mL) and CH3OH (7 mL).
The mixture was stirred for 30 min at room tempera-
ture. After evaporation of solvent, water and CHCl3 were
added. The organic phase was separated, dried, and eva-
porated. The residue was converted to the oxalate salt.
6-Methoxy-1-[2-(methoxymethyl)benzyl]-2-methyl-1,2,3,-
4-tetrahydroisoquinoline oxalate 15a. Recrystallized
from EtOH/Et2O. Yield 84%, mp 130±132 ꢀC. 1H
NMR(DMSO- d6) d 10.93 (br s, 1H, NH+), 7.28 (m,
3H, ArH), 7.10 (m, 1H, ArH), 6.85 (d, J=2.64 Hz, 1H,
ArH), 6.50 (dd, J=2.64, 8.45 Hz, 1H, ArH), 6.18 (d,
J=8.45 Hz, 1H, ArH), 4.45 (m, 1H, CHN+), 4.20 (d,
J=11.8 Hz, 1H, ArCHO), 4.00 (d, J=11.8 Hz, 1H,
ArCHO), 3.72 (m, 5H, CH2N+, OCH3), 3.40 (m, 2H,
ArCH2), 3.22 (s, 3H, OCH3), 3.06 (m, 2H, ArCH2CN+)
2.80 (s, 3H, N+CH3). Anal. calcd for (C20H25NO2H2
C2O4): C, 65.82; H, 6.78; N, 3.49; found: C, 63.66; H,
6.63; N, 3.27.
6-Methoxy-1-[2(methoxymethyl)benzyl]-2-propyl-3,4-di-
hydroisoquinolinium iodide 13c. Recrystallized from
EtOH/Et2O. Yield 78%, mp 162±164 ꢀC. 1H NMR
(DMSO-d6) d 7.88 (d, J=8.75 Hz, 1H, ArH), 7.48 (m,
1H, ArH), 7.29 (m, 2H, ArH), 7.15 (d, J=2.3 Hz, 1H,
ArH), 7.01 (dd, J=2.3, 8.75 Hz, 1H, ArH), 6.85 (d,
J=6.94 Hz, 1H, ArH), 4.69 (s, 2H, ArCH2O), 4.62 (s,
2H, ArCH2CˆN), 4.20 (t, J=7.12 Hz, 2H, CH2N+),
3.88 (m, 5H, N+CH2CC, OCH3), 3.40 (s, 3H, OCH3),
3.23 (t, J=7.12 Hz, 2H, ArCH2CN+), 1.78 (m, 2H,
N+CCH2C), 0.90 (t, J=7.3 Hz, 3H, N+CCCH3). Anal.
calcd for (C22H28INO2): C, 56.78; H, 6.06; N, 3.01;
found: C, 57.05; H, 5.81; N, 2.82.
2-Ethyl-6-methoxy-1-[2-(methoxymethyl)benzyl]-1,2,3,4-
tetrahydroisoquinoline oxalate 15b. Recrystallized from
i-PrOH. Yield 80%, mp 140±141 ꢀC. 1H NMR(DMSO-
d6) d 10.05 (br s, 1H, NH+), 7.28 (m, 3H, ArH), 7.06
(m, 1H, ArH), 6.86 (d, J=2.51 Hz, 1H, ArH), 6,51 (dd,
J=2.51, 8.40 Hz, 1H, ArH), 6.10 (d, J=8.40 Hz, 1H,
ArH), 4.46 (m, 1H, CHN+), 4.30 (d, J=12 Hz, 1H,
ArCHO), 4.02 (d, J=12 Hz, 1H, ArCHO), 3.71 (m, 5H,
6,7-Dimethoxy-1-[2(methoxymethyl)benzyl]-2-methyl-
3,4-dihydroisoquinolinium iodide 14a. Recrystallized
from i-PrOH. Yield 83%, mp 192±194 ꢀC. 1H NMR
(DMSO-d6) d 7.45 (m, 1H, ArH), 7.30 (m, 3H, ArH),
7.18 (s, 1H, ArH), 6.92 (d, J=6.9 Hz, 1H, ArH), 4.70 (s,
2H, ArCH2C7N), 4,60 (s, 2H, ArCH2O), 4.15 (t,

G. M. Cingolani et al. / Bioorg. Med. Chem. 9 (2001) 1447±1458
1455
CH2N+, OCH3), 3.40 (m, 2H, CH2N+), 3.22 (s, 3H,
OCH3), 3.02 (m, 4H, two ArCH2CN+), 1.23 (t, J=7.3 Hz,
3H, N+CCH3). Anal. calcd for (C21H27NO.2 H2C2O4): C,
66.49; H, 7.03; N, 3.37; found: C, 66.21; H, 7.24; N, 3.16.
(0.3 mL, 3.20 mmol) in dry acetonitrile (7 mL) was
re¯uxed, under N2, for 15 min. The solvent and the
excess of reagent were removed under reduced pressure.
A solution of the residue in CHCl3 (10 mL) was rapidly
washed with 1 N NaOH, dried and evaporated under
reduced pressure. The residue was puri®ed by silica gel
column chromatography (eluent AcOEt). After removal
of the solvent a mixture of the product, KI (0.87 g,
5.26 mmol), benzyl bromide (0.28 g, 1.66 mmol) and dry
acetone (36 mL) was heated at 70 ꢀC, under N2, for 2
days. The solvent was removed under vacuum. The
residue was dissolved in CH3OH (22 mL) containing a
trace of H2O, and NaBH4 (0.045 g, 1.19 mmol) was
added, under stirring, in small portions over 30 min at
room temperature. Stirring was maintained for an
additional 60 min. The solvent was removed, the residue
was diluted with H2O and extracted with CH2Cl2. The
organic extracts were dried and evaporated. The residue
was puri®ed by column chromatography on silica gel
(eluent AcOEt). Evaporation of pure fractions gave a
oily residue. This was converted to the oxalate salt
which was recrystallized from EtOH/Et2O.
6-Methoxy-1-[2-(methoxymethyl)benzyl]-2-propyl-1,2,3,-
4-tetrahydroisoquinoline oxalate 15c. Recrystallized
from EtOH/Et2O. Yield 71%, mp 161±163 ꢀC. 1H
NMR(DMSO- d6) d 10.40 (br s, 1H, NH+), 7.26 (m,
3H, ArH), 7.02 (m, 1H, ArH), 6.81 (d, J=2.71 Hz, 1H,
ArH), 6.51 (dd, J=2.71, 8.56 Hz, 1H, ArH), 6.10 (d,
J=8.56 Hz, 1H, ArH), 4.44 (m, 1H, CHN+), 4.29 (d,
J=12.3 Hz, 1H, ArCHO), 4.15 (d, J=12.3 Hz, 1H,
ArCHO), 3.70 (m, 5H, CH2N+, OCH3), 3.39 (m, 2H,
ArCH2), 3.25 (s, 3H, OCH3), 2.92 (m, 4H, two
ArCH2CN+), 1.69 (m, 2H, N+CCH2C), 0.82 (t,
J=7.51 Hz, 3H, N+CCCH3). Anal. calcd for
(C22H29NO.2H2C2O4): C, 67.11; H, 7.28; N, 3.26; found:
C, 67.24; H, 7.37; N, 3.09.
6,7-Dimethoxy-1-[2-(methoxymethyl)benzyl)-2-methyl-
1,2,3,4-tetrahydroisoquinoline oxalate 16a. Recrys-
tallized from EtOH/Et2O. Yield 73%, mp 119±121 ꢀC.
1H NMR(DMSO- d6) d 9.90 (br s, 1H, NH+), 7.28 (m,
3H, ArH), 7.10 (m, 1H, ArH), 6.80 (s, 1H, ArH), 5.52
(s, 1H, ArH), 4.40 (m, 1H, CHN+), 4.25 (d,
J=12.41 Hz, 1H, ArCHO), 3.98 (d, J=12.41 Hz, 1H,
ArCHO), 3.68 (m, 5H, OCH3, CH2N+), 3.41 (m, 2H,
ArCH2), 3.22 (s, 3H, OCH3), 3.20 (s, 3H, OCH3), 3.03
(m, 2H, ArCH2CN+), 2.80 (s, 3H, N+CH3). Anal.
calcd for (C21H27NO.3H2C2O4): C, 64.02; H, 6.77; N,
3.25; found: C, 63.87; H, 6.53; N, 3.06.
2-Benzyl-6-methoxy-1-[2-(methoxymethyl)benzyl]-1,2,3,-
4-tetrahydroisoquinoline oxalate 15d. Yield 30%, mp
ꢀ
1
192±194 C. H NMR(DMSO- d6) d 10.90 (br s, 1H,
NH+), 7.24 (m, 8H, ArH), 7.04 (m, 1H, ArH), 6.80 (d,
J=2.1 Hz, 1H, ArH), 6.60 (dd, J=2.1, 8.5 Hz, 1H,
ArH), 6.42 (d, J=8.5 Hz, 1H, ArH), 4.33±3.40 (m, 6H,
ArCH2O, CHN+, OCH3), 3.23±2.60 (m, 11H, two
ArCH2, two CH2N+, OCH3). Anal. calcd for
.
(C26H29NO2 H2C2O4): C, 70.42; H, 6.54; N, 2.93;
found: C, 70.16; H, 6.71; N, 2.78.
6,7- Dimethoxy - 2- ethyl -1-[2 -(methoxymethyl)benzyl]-
1,2,3,4-tetrahydroisoquinoline oxalate 16b. Recrys-
2-Benzyl-5,6-dimethoxy-1-[2-(methoxymethyl)benzyl]-
1,2,3,4-tetrahydroisoquinoline oxalate 16d. Yield 25%,
ꢀ
1
tallized from i-PrOH. Yield 87%, mp 146±147 C. H
NMR(DMSO- d6) d 9.80 (br s, 1H, NH+), 7.29 (m, 3H,
ArH), 7.06 (m, 1H, ArH), 6.83 (s, 1H, ArH), 5.52 (s,
1H, ArH), 4.49 (m, 1H, CHN+), 4.30 (d, 1H,
J=12.25 Hz, ArCHO), 4.00 (d, J=12.25 Hz, 1H,
ArCHO), 3.73 (s, 3H, OCH3), 3.48 (m, 2H, CH2N+),
3.30±2.91 (m, 12H, two OCH3, two ArCH2CN+,
N+CH2C), 1.30 (t, J=7.4 Hz, 3H, N+CCH3). Anal.
calcd for (C22H29NO.3H2C2O4): C, 64.71; H, 7.01; N,
3.14; found: C, 64.58; H, 6.95; N, 3.29.
mp 175±177 ꢀC. H NMR(DMSO- d6) d 7.28 (m, 8H,
1
ArH), 7.02 (m, 1H, ArH), 6.80 (s, 1H, ArH), 5.85 (s,
1H, ArH), 4.02 (m, 5H, ArCH2O, ArCH2N+, CHN+),
3.75 (s, 3H, OCH3), 3.68±2.62 (m, 9H, two ArCH2,
+
.
CH2N , OCH3). Anal. calcd for (C27H31NO3 H2C2O4):
C, 68.62; H, 6.55; N, 2.76; found: C, 68.51; H, 6.76; N,
2.92.
General procedure for the preparation of methoxyben-
zo[5,6]cyclohepta[1,2,3-ij]isoquinolines 17a±d and 18a±d.
The appropriate 2-alkyl-1-[2-(methoxymethyl)benzyl]-6-
methoxy- (or 6,7-dimethoxy)-1,2,3,4-tetrahydroisoquino-
line (2.12 mmol) was added in portions to cold 98%
sulfuric acid (50 mL) at À40 ꢀC, under N2, with vigorous
mechanical stirring. After the addition, the reaction
mixture was stirred at À10 ꢀC for 2 h. Then, the mixture
was poured into crushed ice and left stirring for 2 h. The
solution was made alkaline with 2 N NaOH and extrac-
ted with CHCl3. The extracts were dried and evaporated
under vacuum. The oil was treated with ethereal oxalic
acid and the oxalate salt was recrystallized from EtOH/
Et2O.
6,7-Dimethoxy-1-[2-(methoxymethyl)benzyl]-2-propyl-
1,2,3,4-tetrahydroisoquinoline oxalate 16c. Recrystallized
ꢀ
1
from EtOH/Et2O. Yield 71%, mp 75±78 C (dec). H
NMR(DMSO- d6) d 9.90 (br s, 1H, NH+), 7.28 (m, 3H,
ArH), 7.07 (m, 1H, ArH), 6.72 (s, 1H, ArH), 5.51 (s,
1H, ArH), 4.46 (m, 1H, CHN+), 4.30 (d, J=12.3 Hz,
1H, ArCHO), 4.00 (d, J=12.3 Hz, 1H, ArCHO), 3.70
(s, 3H, OCH3), 3.48 (m, 2H, CH2N+), 3.22 (s, 3H,
OCH3), 3.20 (s, 3H, OCH3), 2.98 (m, 6H, two ArCH2,
N+CH2CC), 1.75 (m, 2H, N+CCH2C), 0.90 (t,
J=7.42 Hz, 3H, N+CCCH3). Anal. calcd for
.
(C23H31NO3 H2C2O4): C, 65.34; H, 7.24; N, 3.05; found:
C, 65.46; H, 7.03; N, 2.91.
5 - Methoxy - 1 - methyl - 1,2,3,7,12,12a - hexahydrobenzo-
[5,6]cyclohepta[1,2,3-ij]isoquinoline oxalate 17a. Yield
35%, mp 128±130 ꢀC. 1H NMR(DMSO- d6) d 8.20 (br s,
1H, NH+), 7.18 (m, 4H, ArH), 6.85 (d, J=2.61, 1H,
General procedure for the preparation of the benzyltetra-
hydroisoquinolines oxalate 15d and 16d. A solution of
amide 9 or 10 (1.60 mmol) and phosphoryl chloride

1456
G. M. Cingolani et al. / Bioorg. Med. Chem. 9 (2001) 1447±1458
ArH), 6.74 (d, J=2.61 Hz, 1H, ArH), 5.08 (dd, J=4.03,
12.09 Hz, 1H, CHN+), 4.32 (d, J=15.02 Hz, 1H,
ArCHAr), 3.79 (d, J=15.02 Hz, 1H, ArCHAr), 3.72 (s,
3H, OCH3), 3.47 (m, 4H, two ArCH2), 3.01 (m, 2H,
CH2N+), 2.90 (s, 3H, N+CH3). Anal. calcd for
5,6-Dimethoxy-1-propyl-1,2,3,7,12,12a-hexahydroben-
zo[5,6]cyclohepta[1,2,3-ij]isoquino-line oxalate 18c. Yield
41%, mp 104±106 ꢀC. 1H NMR(DMSO- d6) d 8.75 (br s,
1H, NH+), 7.12 (m, 4H, ArH), 6.81 (s, 1H, ArH), 5.08
(dd, J=4.35, 12.5 Hz, 1H, CHN+), 4.20 (d, J=14.6 Hz,
1H, ArCHAr), 4.02 (d, J=14.6 Hz, 1H, ArCHAr), 3.80
(s, 3H, OCH3), 3.70 (s, 3H, OCH3), 3.66±2.75 (m, 8H,
two ArCH2, two N+CH2), 1.78 (m, 2H, N+CCH2C),
0.92 (t, J=7.32 Hz, 3H, N+CCH3). Anal. calcd for
.
(C19H21NO H2C2O4): C, 68.28; H, 6.28; N, 3.79; found:
C, 68.55; H, 6.47; N, 3.73.
1-Ethyl-5-methoxy-1,2,3,7,12,12a-hexahydrobenzo[5,6]-
cyclohepta[1,2,3-ij]isoquinoline oxalate 17b. Yield 28%,
.
(C22H27NO2 H2C2O4): C, 67.43; H, 6.84; N, 3.28;
found: C, 67.22; H, 6.71; N, 3.09.
mp 102±105 ꢀC. H NMR(DMSO- d6) d 8.40 (br s, 1H,
1
NH+), 7.18 (m, 4H, ArH), 6.85 (d, J=2.49 Hz, 1H,
ArH), 6.73 (d, J=2.49 Hz, 1H, ArH), 5.12 (dd, J=4.2,
11.7 Hz, 1H, CHN+), 4.48 (d, 1H, J=14.7 Hz, ArCHAr),
3.74±2.80 (m, 12H, ArCHAr, OCH3, two ArCH2,
CH2N+, N+CH2C), 1.32 (t, J=7.5 Hz, 3H, N+CCH3).
Anal. calcd for (C20H23NO.H2C2O4): C, 68.92; H, 6.57;
N, 3.65; found: C, 68.67; H, 6.76; N, 3.58.
1-Benzyl-5,6-dimethoxy-1,2,3,7,12,12a-hexahydroben-
zo[5,6]cyclohepta[1,2,3-ij]isoquino-line oxalate 18d. Yield
26%, mp 149±151 ꢀC. H NMR(DMSO- d6) d 10.60 (br
1
s, 1H, NH+), 7.42 (m, 5H, ArH), 7.15 (m, 4H, ArH),
6.80 (s, 1H, ArH), 4.82 (dd, J=4.3, 12.65 Hz, 1H,
CHN+), 4.15 (s, 2H, ArCH2N+), 4.05 (m, 2H,
ArCH2Ar), 3.80 (s, 3H, OCH3), 3.65 (s, 3H, OCH3),
3.60±2.60 (m, 6H, two ArCH2, CH2N+). Anal. calcd for
5-Methoxy-1-propyl-1,2,3,7,12,12a-hexahydrobenzo[5,6]-
cyclohepta[1,2,3-ij]isoquinoline oxalate 17c. Yield 14%,
mp 115±118 ꢀC. 1H NMR(DMSO- d6) d 10.21 (br s, 1H,
NH+), 7.18 (m, 4H, ArH), 6.82 (d, J=2.2 Hz, 1H,
ArH), 6.73 (d, J=2.2 Hz, 1H, ArH), 5.10 (m, 1H,
CHN+), 4.46 (d, J=15.2 Hz, 1H, ArCHAr), 3.87±2.85
(m, 12H, ArCHAr, OCH3, two ArCH2, CH2N+,
N+CH2CC), 1.80 (m, 2H, N+CCH2C), 0.95 (t,
J=7.3 Hz, 3H, N+CCCH3). Anal. calcd for
(C21H25NO.H2C2O4): C, 69.50; H, 6.85; N, 3.52; found:
C, 69.64; H, 7.01; N, 3.73.
.
(C26H27NO2 H2C2O4): C, 70.72; H, 6.15; N, 2.95;
found: C, 70.66; H, 6.04; N, 2.81.
General procedure for the demethylation
A mixture of the appropriate 1-alkyl-1,2,3,7,12,12a-hexa-
hydro-5-methoxy- (or 5,6-dimethoxy-)benzo[5,6]cyclo-
hepta[1,2,3-ij]isoquinoline (0.40 mmol), methanesulfonic
acid (7.1 g, 74 mmol), H2O (0.3 mL), and methionine
(0.69 g, 4.6 mmol) was stirred at 40 ꢀC for 4 days. The
solution was poured into ice±H2O and adjusted to pH
8.0 with 15% NH4OH. The mixture was extracted with
AcOEt. The organic extracts, washed with aqueous
NaHSO3 and H2O, were dried and evaporated. The oil
was treated with ethereal oxalic acid or EtOH/HCl and
the salt was recrystallized from EtOH/Et2O.
1-Benzyl-5-methoxy-1,2,3,7,12,12a-hexahydrobenzo[5,6]-
cyclohepta[1,2,3-ij]isoquinoline oxalate 17d. Yield 22%,
mp 174±176 ꢀC.1H NMR(DMSO- d6) d 10.40 (br s, 1H,
NH+), 7.24 (m, 9H, ArH), 6.85 (d, J=2.15 Hz, 1H,
ArH), 6.70 (d, J=2.15 Hz, 1H, ArH), 4.80 (m, 1H,
CHN+), 4.30 (d, J=14.8 Hz, 1H, ArCHAr), 3.76±2.62
(m, 12H, ArCHAr, OCH3, two ArCH2, two CH2N+).
Anal. calcd for (C25H25NO.H2C2O4): C, 72.79; H, 6.11;
N, 3.14; found: C, 72.64; H, 6.05; N, 2.92.
5 - Hydroxy - 1 - methyl - 1,2,3,7,12,12a - hexahydrobenzo-
[5,6]cyclohepta[1,2,3-ij]isoquinoline oxalate 5a. Yield
29%, mp 150±152 ꢀC. H NMR(DMSO- d6) d 7.18 (m,
1
4H, ArH), 6.62 (d, J=2.28 Hz, 1H, ArH), 6.52 (d,
J=2.28 Hz, 1H, ArH), 5.02 (dd, J=3.7, 11.6 Hz, 1H,
CHN+), 4.30 (d, J=15.1 Hz, 1H, ArCHAr), 3.72 (d,
J=15.1 Hz, 1H, ArCHAr), 3.40 (m, 4H, two ArCH2),
2.92 (m, 2H, N+CH2), 2.88 (s, 3H, N+CH3). Anal.
calcd for (C18H19NO.H2C2O4): C, 67.59; H, 5.96; N,
3.94; found: C, 67.32; H, 6.17; N, 3.71.
5,6-Dimethoxy-1-methyl-1,2,3,7,12,12a-hexahydroben-
zo[5,6]cyclohepta[1,2,3-ij]isoqui-noline oxalate 18a. Yield
54%, mp 181±183 ꢀC. 1H NMR(DMSO- d6) d 9.25 (br s,
1H, NH+), 7.18 (m, 4H, ArH), 6.85 (s, 1H, ArH), 5.10
(dd, J=4.4, 12.45 Hz, 1H, CHN+), 4.15 (d, J=14.1 Hz,
1H, ArCHAr), 4.05 (d, J=14.1 Hz, 1H, ArCHAr), 3.75
(s, 3H, OCH3), 3.65 (s, 3H, OCH3), 3.60±2.92 (m, 6H,
two ArCH2, CH2N+), 2.85 (s, 3H, N+CH3). Anal.
calcd for (C20H23NO.2H2C2O4): C, 66.15; H, 6.31; N,
3.51; found: C, 66.32; H, 6.16; N, 3.34.
1-Ethyl-5-hydroxy-1,2,3,7,12,12a-hexahydrobenzo[5,6]-
cyclohepta[1,2,3-ij]isoquinoline oxalate 5b. Yield 66%,
mp 144±146 ꢀC. H NMR(DMSO- d6) d 7.18 (m, 4H,
1
ArH), 6.62 (d, J=2.25 Hz, 1H, ArH), 6.52 (d,
J=2.25 Hz, 1H, ArH), 5.20 (dd, J=3.4, 11.8 Hz, 1H,
CHN+), 4.40 (d, J=14.8 Hz, 1H, ArCHAr), 3.60 (d,
J=14.8 Hz, 1H, ArCHAr), 3.59±2.74 (m, 8H, two
ArCH2, two N+CH2), 1.30 (t, J=7.9 Hz, 3H,
N+CCH3). Anal. calcd for (C19H21NO.H2C2O4): C,
68.28; H, 6.28; N, 3.79; found: C, 68.47; H, 6.13; N,
3.95.
5,6-Dimethoxy-1-ethyl-1,2,3,7,12,12a-hexahydrobenzo-
[5,6]cyclohepta[1,2,3-ij]isoqui-noline oxalate 18b. Yield
42%, mp 103±105 ꢀC. H NMR(DMSO- d6) d 7.15 (m,
1
4H, ArH), 6.86 (s, 1H, ArH), 6.25 (br s, 1H, NH+), 5.12
(dd, J=4.5, 12.28 Hz, 1H, CHN+), 4.21 (d, J=14.3 Hz,
1H, ArCHAr), 4.06 (d, J=14.3 Hz, 1H, ArCHAr), 3.78
(s, 3H, OCH3), 3.65 (s, 3H, OCH3), 3.60±2.80 (m, 8H,
two ArCH2, two N+CH2), 1.32 (t, J=7.24 Hz, 3H,
+
.
N CCH3). Anal. calcd for (C21H25NO2 H2C2O4): C,
5-Hydroxy-1-propyl-1,2,3,7,12,12a-hexahydrobenzo[5,6]-
cyclohepta[1,2,3-ij]isoquinoline oxalate 5c. Yield 44%,
mp 133±135 ꢀC (dec). 1H NMR(DMSO- d6) d 8.80 (br s,
66.81; H, 6.58; N, 3.39; found: C, 66.59; H, 6.42; N,
3.35.

G. M. Cingolani et al. / Bioorg. Med. Chem. 9 (2001) 1447±1458
1457
1H, NH+), 7.18 (m, 4H, ArH), 6.64 (d, J=2.6 Hz, 1H,
ArH), 6.53 (d, J=2.6 Hz, 1H, ArH), 5.00 (dd, J=3.5,
11.3 Hz, 1H, CHN+), 4.39 (d, J=14.3 Hz, 1H,
ArCHAr), 3.69 (d, J=14.3 Hz, 1H, ArCHAr), 3.45 (m,
4H, two ArCH2), 2.98 (m, 4H, two N+CH2), 1.77 (m,
2H, N+CCH2C), 0.92 (t, J=7.45 Hz, 3H, N+CCCH3).
Anal. calcd for (C20H23NO.H2C2O4): C, 68.91; H, 6.57;
N, 3.65; found: C, 69.12; H, 6.36; N, 3.78.
1- Benzyl - 5,6 - dihydroxy- 1,2,3,7,12,12a- hexahydroben-
zo[5,6]cyclohepta[1,2,3-ij]isoquino-line oxalate 6d. Yield
70%, mp 165±167 ꢀC. 1H NMR(DMSO- d6) d 7.90 (br s,
3H, two OH, NH+), 7.45 (m, 5H, ArH), 7.12 (m, 4H,
ArH), 6.52 (s, 1H, ArH), 4.78 (dd, J=3.2, 11.5 Hz, 1H,
CHN+), 4.16 (s, 2H, ArCH2N+), 4.12 (d, J=13.64 Hz,
1H, ArCHAr), 3.86 (d, J=13.64 Hz, 1H, ArCHAr),
3.58±2.48 (m, 6H, two ArCH2, N+CH2). Anal. calcd for
.
(C24H23NO2 H2C2O4): C, 69.79; H, 5.63; N, 3.13;
found: C, 69.88; H, 5.41; N, 2.92.
1-Benzyl-5-hydroxy-1,2,3,7,12,12a-hexahydrobenzo[5,6]-
cyclohepta[1,2,3-ij]isoquinoline oxalate 5d. Yield 22%,
mp 125±127 ꢀC. H NMR(DMSO- d6) d 7.48 (m, 5H,
5,6-Dihydroxy-1,2,3,7,12,12a-hexahydrobenzo[5,6]cyclo-
hepta[1,2,3-ij]isoquinoline oxalate 6e. Yield 60%, mp
1
ArH), 7.14 (m, 4H, ArH), 6.60 (d, J=2.4 Hz, 1H, ArH),
6.50 (d, J=2.4 Hz, 1H, ArH), 4.76 (dd, J=3.2, 11.3 Hz,
1H, CHN+), 4.28 (d, J=13.9 Hz, 1H, ArCHAr), 4.15
(s, 2H, N+CH2Ar), 3.60 (d, J=13.9 Hz, 1H, ArCHAr),
3.50±2.62 (m, 6H, two ArCH2, N+CH2). Anal. calcd for
(C24H23NO.H2C2O4): C, 72.37; H, 5.84; N, 3.25; found:
C, 72.14; H, 5.62; N, 3.32.
130±132 ꢀC. H NMR(DMSO- d6) d 8.10 (br s, 4H, two
1
OH, NH+2 ), 7.12 (m, 4H, ArH), 6.55 (m, 1H, ArH), 4.82
(dd, J=3.28, 11.66 Hz, 1H, CHN+), 4.10 (s, 2H,
ArCH2Ar), 3.50±2.69 (m, 6H, two ArCH2, N+CH2).
Anal. calcd for (C17H17NO.2H2C2O4): C, 63.86; H, 5.36;
N, 3.92; found: C, 63.58; H, 5.13; N, 4.11.
5-Hydroxy-1,2,3,7,12,12a-hexahydrobenzo[5,6]cyclohep-
ta[1,2,3-ij]isoquinoline oxalate 5e. Yield 52%, mp 126±
128 ꢀC (hygroscopic solid). 1H NMR(DMSO- d6) d 7.22
(m, 4H, ArH), 6.68 (d, J=2.6 Hz, 1H, ArH), 6,58 (d,
J=2.6 Hz, 1H, ArH), 4.67 (dd, J=3.4, 11.2 Hz, 1H,
CHN+), 4.09 (m, 2H, ArCH2Ar), 3.61±2.72 (m, 6H, two
ArCH2, N+CH2). Anal. calcd for (C17H17NO.H2C2O4):
C, 66.85; H, 5.62; N, 4.10; found: C, 67.11; H, 5.43; N,
4.27.
General procedure for the preparation of the compounds
17e and 18e. A mixture of 17d or 18d (0.20 mmol), ace-
tic acid (10 mL), and 10% Pd/C (100 mg) was shaken
under 25 psi of H2, in a Parr ¯ask at room temperature
for 3 h. After ®ltration, the solvent was evaporated
under reduced pressure. The residue was dissolved in
water. The solution was basi®ed with 15% NH4OH to
pH 9 and extracted with CHCl3. The extracts were dried
and evaporated. The residue was converted to the oxa-
late, and the salt was recrystallized from EtOH/Et2O.
1-Methyl-5,6-dihydroxy-1,2,3,7,12,12a-hexahydroben-
zo[5,6]cyclohepta[1,2,3-ij]isoquino-line hydrochloride 6a.
Yield 25%, mp 159±161 ꢀC (hygroscopic solid). 1H NMR
(DMSO-d6) d 11.30 (m, 1H, NH+), 9.58 (s, 1H, OH), 8.50
(s, 1H, OH), 7.12 (m, 4H, ArH), 6.57 (s, 1H, ArH), 5.18
(dd, J=3.2, 11.6Hz, 1H, CHN+), 4.14 (d, J=13.7 Hz, 1H,
ArCHAr), 4.02 (d, J=13.7Hz, 1H, ArCHAr), 3.70±3.12
(m, 4H, two ArCH2), 2.85 (m, 5H, N+CH2, N+CH3).
Anal. calcd for (C18H19NO.2HCl): C, 68.03; H, 6.34; N,
4.41; found: C, 67.81; H, 6.22; N, 4.33.
5-Methoxy-1,2,3,7,12,12a-hexahydrobenzo[5,6]cyclohep-
ta[1,2,3-ij]isoquinoline oxalate 17e. Yield 79%, mp 123±
+
125 ꢀC. H NMR(DMSO- d6) d 8.55 (br s, 2H, NH2 ),
7.20 (m, 4H, ArH), 6.85 (d, J=2.61 Hz, 1H, ArH), 6.75
(d, J=2.61 Hz, 1H, ArH), 4.70 (dd, J=6.06, 10.1 Hz,
1H, CHN+), 4.17 (d, J=14.8 Hz, 1H, ArCHAr), 4.04
(d, J=14.8 Hz, 1H, ArCHAr), 3.75 (s, 3H, OCH3),
3.51±2.80 (m, 6H, two ArCH2, N+CH2). Anal. calcd for
(C18H19NO.H2C2O4): C, 67.59; H, 5.96; N, 3.94; found:
C, 67.41; H, 6.12; N, 3.69.
1
1 - Ethyl - 5,6 - dihydroxy - 1,2,3,7,12,12a - hexahydroben-
zo[5,6]cyclohepta[1,2,3-ij]isoquino-line oxalate 6b. Yield
25%, mp 176±178 ꢀC. 1H NMR(DMSO- d6) d 8.75 (br s,
3H, two OH, NH+), 7.10 (m, 4H, ArH), 6.48 (s, 1H,
ArH), 4.90 (dd, J=3.1, 11.4 Hz, 1H, CHN+), 4.12 (d,
J=14.1 Hz, 1H, ArCHAr), 4.00 (d, J=14.1 Hz, 1H,
ArCHAr), 3.50±2.51 (m, 8H, two ArCH2, two N+CH2),
1.26 (t, J=7.9 Hz, 3H, N+CCH3). Anal. calcd for
5,6-Dimethoxy-1,2,3,7,12,12a-hexahydrobenzo[5,6]cyclo-
hepta[1,2,3-ij]isoquinoline oxalate 18e. Yield 75%, mp
123±125 ꢀC. 1H NMR(DMSO- d6) d 7.70 (br s, 2H,
NH+2 ), 7.21 (m, 4H, ArH), 6.88 (s, 1H, ArH), 4.93 (dd,
J=5.28, 11.2 Hz, 1H, CHN+), 4.14 (s, 2H, ArCH2Ar),
3.78 (s, 3H, OCH3), 3.66 (s, 3H, OCH3), 3.54±2.83 (m,
6H, two ArCH2, N+CH2). Anal. calcd for
.
.
(C19H21NO2 H2C2O4): C, 65.44; H, 6.02; N, 3.63;
found: C, 65.65; H, 5.86; N, 3.89.
(C19H21NO2 H2C2O4): C, 65.44; H, 6.02; N, 3.63;
found: C, 65.67; H, 6.24; N, 3.45.
5,6-Dihydroxy-1-propyl-1,2,3,7,12,12a-hexahydroben-
zo[5,6]cyclohepta[1,2,3-ij]isoquino-line oxalate 6c. Yield
41%, mp 149±151 ꢀC. 1H NMR(DMSO- d6) d 8.90 (br s,
3H, two OH, NH+), 7.12 (m, 4H, ArH), 6.52 (s, 1H,
ArH), 5.01 (dd, J=3.35, 11.6 Hz, 1H, CHN+), 4.12 (d,
J=13.9 Hz, 1H, ArCHAr), 4.02 (d, J=13.9 Hz, 1H,
ArCHAr), 3.65±2.60 (m, 8H, two ArCH2, two N+CH2),
Pharmacological methods
Binding assays: dopamine D1 and D2 receptors. [3H]SCH
23390 (86Ci/mmol) and [3H]Spiperone (96 Ci/mmol) were
purchased from Amersham International (England),
dopamine HCl from Sigma, R-(À)-apomorphine HCl and
ketanserin tartrate from Research Biochemicals Inc. All
other reagents were obtained from commercial suppliers.
1.78 (m, 2H, N+CCH2C), 0.98 (t, J=7.25 Hz, 3H,
+
.
N CCCH3). Anal. calcd for (C20H23NO2 H2C2O4): C,
66.15; H, 6.31; N, 3.51; found: C, 66.05; H, 6.53; N,
3.77.
Striatal tissue was isolated from porcine brains. The stria-
tum membranes were prepared as previously described.17

1458
G. M. Cingolani et al. / Bioorg. Med. Chem. 9 (2001) 1447±1458
In brief, tissue was homogenised in 20 volumes of ice-cold
50 mM Tris±HCl bu€er at pH 7.4 (bu€er T), containing
protease inhibitors (20 mg/mL soybean trypsin inhibitor,
200 mg/mL, and 160 mg/mL benzamidine), using an
Ultra-Turrax TP-1810 (3Â20 s). The homogenate was
centrifuged for 10 min at 50,000 g at 4 ꢀC. The
resulting pellet was then washed once by resuspension
in fresh bu€er T and by centrifugation as before. The
®nal pellet was frozen at À20 ꢀC until the time of
assay.
Acknowledgements
This work was supported by the Ministero della Uni-
versita e della Ricerca Scienti®ca e Tecnologica (Pro-
getto Nazionale Tecnologie Farmaceutiche). Dr. A.F.
Parlow, NIDDK's National Hormone and Pituitary
Program, kindly supplied the rat PRL RIA kit. The
authors thank Professor M. Cardellini for his colla-
boration.
For D1 and D2 dopamine binding assay, the porcine
striatal pellets were suspended in the incubation bu€er
50 mM Tris±HCl bu€er at pH 7.4. [3H]SCH 23390
binding to D1 receptors was assayed in a ®nal incuba-
tion volume of 500 mL, which contained crude mem-
branes (ꢂ0.2 mg of protein), radioligand (ꢂ0.5 nM;
Kd=1.3 nM) and the tested compound in the range
10À8±10À3 M.
References
1. Sibley, D. R.; Monsma, J. F. Trends Pharmacol. Sci. 1992,
13, 61.
2. Neumeyer, J. L. In The Chemistry and Biology of Iso-
quinoline Alkaloids; Phillipson J. D.; Roberts M. F.; Zenk,
M. H., Eds., Springer-Verlag, Berlin/Heidelberg, 1985; pp
146±170.
3. Riddall, D. R. Eur. J. Pharmacol. 1992, 210, 279.
4. Kerkman, D. J.; Ackerman, M.; Artman, L. D.; Mac-
Kenzie, R. G.; Johnson, M. C.; Bednarz, L.; Montana, W.;
Asin, K. E.; Stamp¯i, H.; Kebabian, J. W. Eur. J. Pharmacol.
1989, 166, 481.
5. Claudi, F.; Di Stefano, A.; Napolitani, F.; Cingolani,
G. M.; Giorgioni, G.; Fontenla, J. A.; Montenegro, G. Y.;
Rivas, M. E.; Rosa, E.; Michelotto, B.; Orlando, G.; Brunetti,
L. J. Med. Chem. 2000, 43, 599.
[3H]Spiperone binding for D2 receptors was assayed in a
®nal incubation volume of 500 mL. Porcine striatal
membranes (ꢂ0.2 mg of protein) were incubated with
radioligand (ꢂ0.3 nM, Kd=0.71 nM), 50 nM Ketan-
serin, to prevent binding of radioligand to 5-HT2A
receptors, and various concentrations (10À8±10À3 M) of
the tested compound. All assays were performed in
duplicate. Dopamine was essayed in presence of the
monoamine oxidase inhibitor pargyline (10 mM).
6. Chatterjiee, A.; Ghosh, S. Synthesis 1981, 10, 818.
7. Merz, A. Angew. Chem., Int. Ed. Engl. 1973, 12, 846.
8. Irie, H.; Fujii, N.; Ogawa, H.; Yajima, H. J. C. S Chem.
Comm 1976, 922.
For both radioligand binding assays, the incubation was
at 37 ꢀC for 60 min and was terminated by dilution to
5 mL with ice-cold bu€er T, followed immediately by
rapid ®ltration through glass ®bre Whatman GF/C
®lters. The ®lters were then washed (3Â5 mL) with
bu€er T, and the amount of radioactivity retained on
the ®lters was determined by Packard 1600 TRliquid
scintillation counter at 66% eciency. Non speci®c
binding was de®ned in the presence of 2.5 mM
dopamine.
9. Enjalbert, A.; Bockaert, J. Mol. Pharmacol. 1983, 23, 576.
10. Chang, A.; Shin, S. H.; Pang, S. Endocrine 1997, 7,
177.
11. Tagawa, R.; Takahara, J.; Sato, M.; Niimi, M.; Murao,
K.; Ishida, T. Life Sci. 1992, 51, 727.
12. Mohamadi, F.; Richards, N. G. J.; Guida, W. C.; Lis-
kamp, R.; Lipton, M.; Chang, G.; Hendirckson, T.; Still, W. C.
J. Comput. Chem. 1990, 11, 440.
13. Allinger, N. L. J. Am. Chem. Soc. 1977, 99, 8127.
14. Chang, G.; Guida, W. C.; Still, W. C. J. Am. Chem. Soc.
1989, 111, 4379.
15. Brusniak, M. K.; Pearlman, R. S.; Neve, K.; Wilcox, R. E.
J. Med. Chem. 1996, 39, 850.
16. Wilcox, R. E.; Tseng, T.; Brusniak, M. K.; Ginsburg, B.;
Pearlman, R. S.; Teeter, M.; DuRand, C.; Starr, S.; Neve,
K. A. J. Med. Chem. 1998, 41, 4385.
17. Campiani, G.; Nacci, V.; Bechelli, S.; Ciani, S. M.; Gar-
olfo, A.; Fiorini, I.; Wikstrom, H.; de Boer, P.; Liao, Y.;
Tepper, P. G.; Cagnotto, A.; Mennini, T. J. Med. Chem. 1998,
41, 3763.
18. Cheng, Y.; Pruso€, W. H. Biochem. Pharmacol. 1973, 22,
3099.
The compounds were routinely dissolved in ethanol. The
level of ethanol did not exceed 1% and was maintained
constant in all tubes. For the active compounds, the IC50
values were determined and Ki values were derived
according to the equation of Cheng and Pruso€.18 Protein
concentration was assayed by the method of Lowry et al.19
PRL release from anterior pituitary cell cultures. The
assay was performed as previously described.20 PRL
concentrations were measured by a radioimmunoassay
kit supplied by NIDDK's National Hormone and
Pituitary Program, USA. The sensitivity of the assay
was 50 pg/tube.
19. Lowry, O. H.; Rosenbrough, N. J.; Farr, A. L.; Randall,
R . J. J. Biol. Chem. 1951, 193, 265.
20. Brunetti, L.; Orlando, G.; Michelotto, B.; Recinella, L.;
Ragazzoni, E.; Vacca, M. A. Toxicol. Lett. 2000, 116, 231.