Structurally constrained hybrid derivatives containing octahydrobenzo[g or f]quinoline moieties for dopamine D2 and D3 receptors: Binding characterization at D2/D3 receptors and elucidation of a pharmacophore model

Article abstract of DOI:10.1021/jm8008629

A series of structurally constrained analogues based on hybrid compounds containing octahydrobenzo[g or f]quinoline moieties were designed, synthesized, and characterized for their binding to dopamine D2 and D3 receptors expressed in HEK-293 cells. Among the newly developed constrained molecules, trans-octahydrobenzo[f]quinolin-7-ol (8) exhibited the highest affinity for D2 and D3 receptors, the (-)-isomer being the eutomer. Interestingly, this hybrid constrained version 8 showed significant affinity over the corresponding nonhybrid version 1 (representing a constrained version of the aminotetralin structure only) when assayed under same conditions (K_i of 49.1 and 14.9 nM for 8 vs 380 and 96.0 nM for 1 at D2 and D3, respectively). Similar results were found with other lead hybrid compounds, indicating a contribution of the piperazine moiety in the observed enhanced affinity. On the basis of the data of new lead constrained derivatives and other lead hybrid derivatives developed by us, a unique pharmacophore model was proposed consisting of three pharmacophoric centers, two with aromatic/hydrophobic and one with cationic features.

Full text of DOI:10.1021/jm8008629

7806
J. Med. Chem. 2008, 51, 7806–7819
Structurally Constrained Hybrid Derivatives Containing Octahydrobenzo[g or f]quinoline
Moieties for Dopamine D2 and D3 Receptors: Binding Characterization at D2/D3 Receptors
and Elucidation of a Pharmacophore Model
Dennis A. Brown,† Prashant S. Kharkar,† Ingrid Parrington,‡ Maarten E. A. Reith,‡ and Aloke K. Dutta*,†
Department of Pharmaceutical Sciences, Eugene Applebaum College of Pharmacy and Health Sciences, Wayne State UniVersity, Detroit,
Michigan 48202, and Department of Psychiatry, Millhauser Laboratories, New York UniVersity School of Medicine,
New York, New York 10016
ReceiVed July 13, 2008
A series of structurally constrained analogues based on hybrid compounds containing octahydrobenzo[g or
f]quinoline moieties were designed, synthesized, and characterized for their binding to dopamine D2 and
D3 receptors expressed in HEK-293 cells. Among the newly developed constrained molecules, trans-
octahydrobenzo[f]quinolin-7-ol (8) exhibited the highest affinity for D2 and D3 receptors, the (-)-isomer
being the eutomer. Interestingly, this hybrid constrained version 8 showed significant affinity over the
corresponding nonhybrid version 1 (representing a constrained version of the aminotetralin structure only)
when assayed under same conditions (Ki of 49.1 and 14.9 nM for 8 vs 380 and 96.0 nM for 1 at D2 and D3,
respectively). Similar results were found with other lead hybrid compounds, indicating a contribution of the
piperazine moiety in the observed enhanced affinity. On the basis of the data of new lead constrained
derivatives and other lead hybrid derivatives developed by us, a unique pharmacophore model was proposed
consisting of three pharmacophoric centers, two with aromatic/hydrophobic and one with cationic features.
Introduction
D3 subtype receptors at the molecular level, a number of
excellent studies with mutant D2 and D3 receptors have been
published that delineated key binding residues in trans-
membrane domains 3, 5, and 7 for receptor activation.12-14
Briefly, these studies indicated that two serine residues in TM-5
and one aspartate residue in TM-3 are critical for activation by
agonists of both D2 and D3 receptors. The Asp110 residue in
TM-3 has been shown to interact with the basic N-atom in DA
and aminotetralin molecules. In these studies, Ser192 has been
especially shown to be crucial for D3 interaction and activation.
All these results suggest the existence of multiple bioactive
conformations available for different agonists, resulting from
different degrees of interactions with these key residues produc-
ing one signaling pathway.
The dopamine (DAa) receptor system has been aggressively
targeted for drug development for the treatment of psychiatric
illnesses, neurodegeneration, drug abuse, and other therapeutic
areas.1,2 The DA receptors, belonging to a class of G-protein-
coupled receptors (GPCRs), are found in the central nervous
system (CNS) and in the periphery.3 In the CNS, DA receptors
can be classified as being either D1-like or D2-like. The D1-like
receptors include the D1 and D5 subtypes, and the D2-like
receptors include the D2, D3, and D4 subtypes. These clas-
sifications are made on the basis of receptor pharmacology and
function. Both D1-like and D2-like DA receptors share the same
effector molecule, adenylate cyclase. Upon receptor activation,
D1-like receptors activate adenylate cyclase whereas D2-like
receptors inhibit it.4
The influence of the freely rotating amino group in 2-ami-
notetralins has been explored by introducing further conforma-
tional constraint into derived structures by the addition of
another annulated ring, giving rise to monophenolic cis- and
trans-1,2,3,4,4a,5,10,10a-octahydrobenzo[g]quinolines15 and
monophenolic cis- and trans-1,2,3,4,4a,5,6,10b-octahydroben-
zo[f]quinolines.16 Extensive SAR studies based on in vitro
functional data have been conducted to determine the influence
of hydroxyl group position, alkyl group chain length, and the
geometry of ring fusion.15-19 These studies showed that for
phenolic 1,2,3,4,4a,5,10,10a-octahydrobenz[g]quinolines and
1,2,3,4,4a,5,6,10b-octahydrobenzo[f]quinolines, only compounds
having trans ring fusion were active (1, 2, 3, Figure 1)15,16 and
the corresponding cis analogues were almost completely inac-
tive. This is largely attributed to the protonated amino group
being unfavorably positioned to interact with the key aspartate
residue in the cis conformation.15 Compounds that showed
activity in the functional assays had hydroxyl group substitution
at either the 6 position [for benzo[g]quinolines (2, Figure 1)]
or the 7/9 position for benzo[f]quinolines (1 and 3, Figure 1),
corresponding to the R- and ꢀ-rotameric forms, respectively.15,17a
In this regard, it is important to mention that to the best of our
An enormous amount of work has been done toward the
development of DA agonists.5 The initial research was focused
on elucidating the bioactive conformation of the natural ligand,
DA.6-10 These efforts yielded the constrained class of com-
pounds known as 2-aminotetralins, including (S)-(-)-5-hydroxy-
2-(N,N-di-n-propylamino)tetralin [(S)-(-)-5-OH-DPAT] (Figure
1)6 and (R)-(+)-7-hydroxy-2-(N,N-di-n-propylamino)tetralin [(R)-
(+)-7-OH-DPAT] (Figure 1),11 corresponding to the R- and
ꢀ-rotamers of DA, respectively. These studies revealed not only
the bioactive conformation of the phenethyl side chain of DA
but also that among the two hydroxyl groups in DA, the
m-hydroxyl group is the most important in terms of receptor
activation.6 In regard to agonist interaction with the D2 and
* To whom correspondence should be addressed. Phone: 1-313-577-1064.
Fax: 1-313-577-2033. E-mail: adutta@wayne.edu.
†
Wayne State University.
New York University School of Medicine.
‡
a Abbreviations: DA, dopamine; GPCR, G-protein-coupled receptor;
CNS, central nervous system; MOE, Molecular Operating Environment;
HEK, human embryonic kidney; 5-OH-DPAT, 5-hydroxy-2-(dipropylami-
no)tetralin; 7-OH-DPAT, 7-hydroxy-2-(dipropylamino)tetralin; PMA, phos-
phomolybdic acid.
10.1021/jm8008629 CCC: $40.75
2008 American Chemical Society
Published on Web 11/18/2008

Structurally Constrained Hybrid DeriVatiVes
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 24 7807
Figure 1. Molecular structures of dopamine D2/D3 receptor ligands.
Figure 2. Hybrid template design.
knowledge no systematic receptor binding studies for DA
receptor subtypes (D2 and D3) have been reported with these
constrained derivatives, and hence, there is a lack of information
on precise binding affinity of these derivatives with the
respective receptor.
assumption that the aminotetralin moiety would interact with
the agonist binding site in the DA receptor and the arylpiperazine
fragment would interact with the accessory binding site residues
in the D3 receptor to impart selectivity. Based on this approach,
our group has been able to create highly potent DA D2/D3
agonists, with some compounds being very selective for the D3
receptor.21-23 Compounds (-)-7 and 11 (Figure 3), two of our
promising lead structures, exhibited high D3 selectivity in the
functional assays and high in ViVo potency in Parkinson’s disease
animal models.22,23 In our effort to develop a pharmacophore
model for our hybrid structures, we designed and synthesized
a series of compounds that contain varying degrees of confor-
mational constraint in the agonist moiety of the hybrid template.
These compounds were then evaluated for their binding affinities
at human D2L and D3 receptors in human embryonic kidney
(HEK)-293 cell lines. Using molecular modeling approaches,
we report in this paper the development of a three-point
pharmacophore model for the binding of hybrid compounds to
D2/D3 receptors.
The effect of N-alkyl chain length on functional activity has
also been investigated.15,16 Linearly fused trans compounds with
a hydroxyl group in the 6-position with alkyl groups n-propyl
or smaller showed good functional activity. However, any
compound containing an alkyl group larger than n-propyl was
completely inactive, indicating the presence of a small alkyl-
binding site in the receptor.15 Also, all 8-hydroxy compounds
(e.g., compound 4 in Figure 1) were inactive, regardless of alkyl
substitution. Angularly fused compounds, e.g., 1 and 3, like their
linear counterparts with 6-OH substitution, also showed func-
tional activity. Interestingly, 7-OH angular compounds tolerated
alkyl group size larger than n-propyl in functional assays. In
fact, n-butyl or larger substitution showed enhanced affinity,
showing that there is an additional “large” alkyl-binding site in
D2-like DA receptors.10,15 The relatively lower activity in
octahydrobenzo[g]quinolinol series, compared to the octahy-
drobenzo[f]quinolinol series of compounds, was attributed to
the unfavorable orientation of the nitrogen lone pair or proto-
nated nitrogen atom and a sterically unfavorable orientation of
the piperidine ring. Again, in all these studies, the precise nature
of interaction was not available due to the lack of quantitative
binding data for DA D2 and D3 receptor subtypes.
Chemistry
Scheme 1 outlines the synthesis of (()-cis-1,2,3,4,4a,5,10,10a-
octahydrobenzo[g]quinolin-8-ol derivatives. Amides 13a-c
were prepared via condensation with appropriately substituted
phenylpiperazines and chloroacyl chlorides in the presence of
triethylamine. Originally, our aim was to synthesize only trans
ring fused 1,2,3,4,4a,5,10,10a-octahydrobenzo[f and g]quinoline-
8-ol molecules. When following the reported preparation
procedure for trans-1,2,3,4,4a,5,10,10a-octahydrobenz[g]quino-
line,24 we were unable to obtain the intermediate 8-methoxy-
1,2,3,4,4a,5,10,10a-octahydrobenzo[g]quinoline in satisfactory
yield, so we set out on devising an alternative synthesis. In the
In an effort to develop novel, potent, and selective ligands
for the DA D3 receptor subtype, our laboratory employed a
hybrid structure approach, combining known DA agonist
moieties with the N-arylpiperazine moiety derived from known
D3 antagonists (Figure 2).20,21 This approach was based on the

7808 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 24
Brown et al.
Figure 3. Molecular structures of compounds used for development of pharmacophore hypothesis.
first step, 3-methoxybenzyl bromide 14 was treated with
activated zinc dust in THF to give zinc bromide 15, which
underwent a trans-metalation reaction with methyl 2-chloroni-
cotinate in the presence of (PPh3)2NiCl2 to give intermediate
16,25 which was then reduced catalytically with PtO2 to yield
substituted cis- and trans-piperidines 17. The mixture of cis-
and trans-isomers was not separated in the subsequent trans-
formations. The amine 17 was then protected by conversion to
its methyl carbamate. The ester was hydrolyzed by treatment
with LiOH in MeOH/H2O, and the resulting acid was converted
to its acid chloride by reaction with SOCl2. A Friedel-Crafts
acylation reaction with TiCl4 was performed to yield benzylic
ketone (()-19. The ketone was reduced catalytically with Pd/C
in the presence of perchloric acid, and then the carbamate was
cleaved by treatment with hydrazine and KOH in ethylene glycol
to give secondary amine (()-21.
The relative stereochemical assignments of octahydrobenzo-
quinoline intermediates in this paper were made by converting
the corresponding secondary amine intermediates, e.g., com-
pound 21 in Scheme 1, to their N-benzyl analogues and
observing the 1H NMR splitting pattern of the N-benzyl protons.
It has been reported that the N-benzyl protons in N-benzyl
substituted cis- and trans-fused octahydrobenzo[f and g]quino-
lines appear as an AB quartet and that the chemical shift
difference between the A and B portions differ between cis-
and trans-fused compounds.26 The observed chemical shift
difference for cis-fused compounds is around 0.6 δ ppm,
whereas the difference for the trans-fused counterparts is much
larger, around 0.9-1.0 δ ppm. The chemical shift differences
of our synthesized compounds are in accordance with these
reported values (see the Supporting Information for the syntheses
and NMR data for N-benzyl compounds).
pionitrile to give 27. This ester was decarboxylated in the
presence of LiCl and H2O in DMSO to yield ketone 28, which
was then protected by conversion to acetal. The cyano group
was reduced in the presence of Raney Ni to give amine 30,
which was then treated with HCl/MeOH to give iminium 31.
The second aim of this synthesis was accomplished by control-
ling the stereoselectivity of iminium reduction of 31 by using
NaCNBH3. The trans-fused secondary amine (()-32 was the
major product detected (stereochemistry assigned based on the
N-benzyl method).26 Amine (()-32 was alkylated with 13a-c
to yield amides (()-33a-c. These amides were reduced and
demethylated in the same manner as described above to produce
targets (()-35a-c.
Scheme 3 outlines the synthesis of (()-trans-1,2,3,4,4a,5,6,10b-
octahydrobenzo[f]quinolin-7-ol derivatives. Amine (()-36 was
synthesized as previously described.28 Racemic 37 was prepared
by N-alkylating (()-36 with chloride 13a to give amide (()-
37, which was then reduced and demethylated to give (()-8.
Compound (()-36 was next resolved into its enantiomers by
following the procedure described by Wikstro¨m et al.16 as shown
in Scheme 4. Experimental details for separation of enantiomers
are provided in the Supporting Information. Compounds (+)-8
and (-)-8 were obtained by following the same procedure as
for the racemic version. Compound (()-trans-4-propyl-
1,2,3,4,4a,5,6,10b-octahydrobenzo[f]quinolin-7-ol (1) was also
synthesized from amine (()-36 as a reference compound.
Scheme5outlinesthesynthesisof(()-trans-1,2,3,4,4a,5,10,10a-
octahydrobenzo[g]quinolin-6-ol derivatives. Secondary amine
40 was prepared as previously described.29 Completion of the
syntheses of (()-43a,b was carried out in a similar manner as
described for the previous compounds. The reference compound
2 was also synthesized from amine intermediate 40.
Amine (()-21 was condensed with chlorides 13a and 13b to
give amides (()-22a,b, respectively. The amide carbonyls were
then reduced with LiAlH4 to provide 23a,b. Demethylation of
methyl ethers by BBr3 in CH2Cl2 gave the final compounds (()-
24a,b.
In addition, the synthesis of (-)-5-OH-DPAT, (-)-6, and
racemic 9 were also carried out to use them as reference
compound and in our SAR studies. The detailed synthesis of 9
is given in the Supporting Information.
The second attempt to synthesize (()-trans-1,2,3,4,4a,5,10,10a-
octahydrobenzo[g]quinoline derivatives is described in Scheme
2. The first aim of this synthesis was to selectively alkylate
position 1 of 7-methoxy-2-tetralone 25. This was accomplished
by converting 25 to its enolate with NaH followed by trapping
with dimethyl carbonate to give compound 26.27 This was then
converted to its dianion with LDA and treated with chloropro-
Results and Discussion
Our previous articles on hybrid drug development approach
for D2/D3 receptors outlined the development of aminotetralin-
arylpiperazine-based compounds exhibiting D3 preferential
activity both in binding and in functional assays.22,23 One of
the very first compounds based on this hybrid template was

Structurally Constrained Hybrid DeriVatiVes
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 24 7809
Scheme 1a
a Reagents and conditions: (a) Et3N, CH2Cl2, 0 °C; (b) Zn dust, THF, 0 °C; (c) methyl 2- chloronicotinate, (PPh3)2NiCl2, THF; (d) (i) HCl, Et2O; (ii) PtO2,
MeOH; (e) methyl chloroformate, K2CO3, CH2Cl2; (f) LiOH, MeOH, H2O; (g) SOCl2, CH2Cl2, cat. DMF; (h) TiCl4, CH2Cl2, 0°C; (i) Pd/C, AcOH, HClO4;
(j) NH2NH2, KOH, H2O, (CH2OH)2, reflux; (k) 13a/b, K2CO3, CH3CN; (l) LiAIH4, THF, reflux; (m) BBr3, CH2Cl2, -78 °C.
compound 6 and its corresponding enantiomers (-)-6 and (+)-
6. The initial D2/D3 binding data demonstrated very little
difference in affinity between these two enantiomers. However,
we recently realized that during the process of separation of
these two enantiomers, we inadvertently converted most of the
compounds back to its racemic version, and thus, the binding
data were erroneous. A newer approach to separate these
enantiomers has been adopted by us recently that gave us the
two enantiomers with high enantiomeric purity.22,30 The new
data show appreciable differences between the two enantiomers,
with the (+)-(R)-6 isomer expectedly showing the highest
affinity for the both D2 and D3 receptors. This is in line with
the binding data of the corresponding enantiomers of the parent
structure 7-OH-DPAT. Thus, the enantiomers (-)-6 and (+)-6
(Figure 3) possessed higher binding affinity for D3 receptors
and lower binding affinity for D2 receptors (Ki of 38.6 and 1.77
nM for D3 and Ki of 809 and 40.6 nM for D2). A similar trend
was observed for the hybrid structures derived from 5-OH-
DPAT, compounds (-)-7 and (+)-7 (Figure 3). The observed
differences in the inhibition constants for (-)-7 and its enan-
tiomer (+)-7 were ∼20-fold for the D3 receptor (18.4 nM versus
0.82 nM) and ∼9-fold for the D2L receptor (238 nM versus 26
nM).22 These data are also in line with the data found from
5-OH-DPAT affinity. However, compound (-)-7 exhibited more
than 4-fold higher affinity compared to (-)-5-OH-DPAT under
our binding assay conditions for both the D2 and D3 receptors
(Ki of 26 vs 220 nM for D2 and Ki of 0.82 vs 4.73 nM for D3,
respectively). This is an interesting finding, since it indicates
that the presence of the piperazine moiety further enhanced the
affinity of this hybrid compound while maintaining its strong
agonist activity.22 Thus, contribution of piperazine fragment
increased the interaction with the D2/D3 receptors.

7810 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 24
Brown et al.
Scheme 2a
a Reagents and conditions: (a) NaH, PhCH3, dimethyl carbonate, reflux; (b) 2 equiv of LDA, THF, -78 °C; (c) bromopropionitrile; (d) LiCl, H2O,
DMSO, reflux; (e) (CH2OH)2, TsOH, triethyl orthoformate, CH2Cl2; (f) Raney Ni, MeOH; (g) 6 M HCl, MeOH, reflux; (h) NaCNBH3, MeOH; (i) 13a/
13b/13c, K2CO3, CH3CN, cat. Kl, reflux; (j) LiAIH4, THF, reflux; (k) BBr3, CH2Cl2, -78 °C.
Scheme 3a
Scheme 4a
a Reagents and conditions: (a) NaOH, H2O/CH2Cl2; (b) crystallization
followed by HPLC; (c) (i) Na tert-butoxide, H2O, THF; (ii) HCl, MeOH.
earlier. With this background, it was of interest to us to apply
the hybrid structure approach to these octahydrobenzo[f]- and
[g]quinolines to investigate the following goals: (1) evaluate
binding affinity for the DA receptor subtypes with the aim of
delineating the factors responsible for the observed trend in
binding affinity and D2/D3 selectivity, if any, in these deriva-
tives and (2) develop a pharmacophore model for our hybrid
derivatives by using these conformationally constrained DA
analogues including aminotetralins.
The binding data of the target compounds for DA D2 and
D3 receptors are given in Table 1. In the case of hybrid (()-
trans-1,2,3,4,4a,5,10,10a-octahydrobenzo[g]quinolin-6-ol de-
rivatives, represented by compound (()-43a, the binding
affinities for the D3 and D2 receptors (Ki(D3) ) 736 and Ki(D2)
) 1712 nM) were found to be low. The corresponding N-n-
propyl analogue 2 (Figure 1), of the parent structure of 43a,
was reported to be active for the DA receptors in functional
assays.15 However, the cis-analogues of 2 were reported to be
a Reagents and conditions: (a) 13a, K2CO3, CH3CN, reflux; (b) LiAIH4
(c) BBr3, CH2Cl2, -78 °C.
Monophenolic cis- and trans-1,2,3,4,4a,5,10,10a-octahy-
drobenzo[g]quinolines had been evaluated for interactions with
D1-like and D2-like dopamine receptors and compared with the
effects of the corresponding unconstrained aminotetralins a
couple of decades ago.15 These molecules were mainly evaluated
for functional activity in the membrane preparations. Similarly,
cis- and trans-1,2,3,4,4a,5,6,10b-octahydrobenzo[f]quinolines
were tested for central DA and serotonin receptor-stimulating
activity using biochemical and behavioral tests in rats.15,16 With
the availability of cloned DA receptor subtypes, it is now
possible to investigate the affinity and selectivity of these
compounds for DA receptor subtypes, which was not possible

Structurally Constrained Hybrid DeriVatiVes
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 24 7811
Scheme 5a
times more potent at D2 compared to 43a and 4-fold more potent
at D3 compared to 43a (Ki of 136 nM vs 1712 nM, respectively).
In the case of (()-cis- and (()-trans-1,2,3,4,4a,5,10,10a-
octahydrobenzo[g]quinolin-8-ol hybrid derivatives, represented
by compounds (()-24a,b and (()-35a-c, respectively, a similar
trend was observed when compared to their corresponding 6-OH
analogues (43a,b). Previously, all 8-OH N-n-propyl derivatives
were reported to be inactive in DA receptor functional assays.15
Compound (()-24a, a cis derivative, exhibited very weak
binding affinity at D3 receptor (Ki ) 419 nM) and even weaker
affinity for the D2 receptor (Ki ) 3326 nM). As seen with 43b,
the corresponding dichloro analogue 24b improved the affinity
to a great extent, which can be attributed to the presence of
3,4-dichloro group possibly promoting hydrophobic interactions
with the DA receptors. This phenomenon was observed with
other known DA receptor ligands as well. A similar trend was
observed for the trans compounds (()-35a and (()-35b. The
trans compounds (()-35b and (()-35c were found to possess
moderate affinity for the D3 receptor (Table 1). A change of
spacer length between the octahydrobenzo[g]quinoline and the
piperazine moieties from 2 to 4, as depicted in 35a and35c, led
to ∼3-fold gain in affinity for the D3 and ∼9-fold for the D2
receptors (Ki of 473 vs 102 nM for D3 and Ki of 2522 vs 289
nM for D2, respectively) for 35c. These benzo[g]quinoline
derivatives exhibited moderate selectivity for the DA D3
receptor subtypes.
a Reagents and conditions: (a) 13a/b, K2CO3, CH3CN, reflux; (b) LiAIH4,
THF, reflux; (c) BBr3, CH2Cl2, -78°C.
Table 1. Affinity for Cloned D2L and D3 Receptors Expressed in HEK
Cells Measured by Inhibition of [3H]Spiperone Bindinga
Having investigated the benzo[g]quinolines, we started
analyzing the binding data for the hybrid 1,2,3,4,4a,5,6,10b-
octahydrobenzo[f]quinolin-7-ol and -9-ol derivatives. The most
active compound in the present study was (()-8 (Ki(D3) ) 14.9
nM, Ki(D2) ) 49.1 nM, respectively), which belongs to the
7-hydroxy series; the corresponding 9-OH compound 9 (Figure
3), which was synthesized by us, exhibited moderate affinity
for the D3 receptor (Ki(D3) ) 72 nM, Ki(D2) ) 219 nM). Even
though the compound (()-8 possessed higher potency, it lacked
high selectivity for the D3 receptor (Table 1). The trans-7-
hydroxy-N-n-propyl derivatives were reported to be more active
than their cis counterparts in this structural class.17a Also, the
change from n-propyl to n-butyl substitution was reported to
increase activity. We further resolved (()-8 and tested their
binding affinities. Thus, in accordance with what has been
previously reported,21 we observed higher binding affinity in
the (-)-8 isomer (corresponding to (4aS,10bS) configuration)
(Table 1). It is again important to point out here that complete
binding characterization of the parent compound 1 and its
analogues for D2/D3 receptors was never carried out. In
agreement with the higher binding affinity of (-)-7 (hybrid
structure based on (S)-(-)-5-OH-DPAT) compared to its
enantiomer, a good separation of affinity was found for the D3
and D2 receptors in (+)-8 and (-)-8 (Ki of 89.3 and 4.95 nM
for D3 and Ki of 835 and 23.6 nM for D2 receptor, Table 1). In
order to judge correctly the impact of the hybrid version of
1,2,3,4,4a,5,6,10b-octahydrobenzo[f]quinolin-7-ol derivatives in
binding, we synthesized and characterized the corresponding
nonhybrid derivative 1 to evaluate its binding affinity for the
D2/D3 receptors. Interestingly, as observed in the case of (-)-
5-OH-DPAT and (-)-7, the hybrid racemic version (()-8
exhibited more than 6-fold affinity compared to 1 (Ki of 14.9
vs 96 nM for D3 and Ki of 49 vs 380 nM for D2, respectively,
Table 1). This again demonstrated the unique contribution of
the piperazine fragment in enhancing affinity of hybrid com-
pounds compared to known ligands. Interestingly, hybrid
constrained derivatives did not contain any free N-alkyl group,
unlike nonhybrid constrained versions, which displayed an
Ki (nM)
compd
D2L [3H]spiperone
D3 [3H]spiperone
D2L/D3
(()-7-OH-DPAT
(-)-5-OH-DPAT
2
311 ( 47
220 ( 37.7
2780 ( 457
347 ( 51
6.19 ( 1.4
4.73 ( 0.64
395 ( 75
50.2
46.5
7.03
3.61
21
22.9
31.5
12.9
3
58.6
45.0
253
7.94
0.79
5.33
1.3
2.83
3.30
9.35
4.77
2.33
0.80
1
96.0 ( 4.7
38.6 ( 0.7
1.77 ( 0.42
0.825 ( 0.136
18.4 ( 1.0
72.2 ( 17.2
4.15 ( 0.76
44.0 ( 10.6
0.925 ( 0.234
419 ( 119
187 ( 24
473 ( 167
256 ( 69
102 ( 20
14.9 ( 4.3
89.3 ( 19.4
4.95 ( 1.08
736 ( 206
168 ( 24
(-)-6
809 ( 65
(+)-6
40.6 ( 3.6
26.0 ( 7.5
238 ( 14
219 ( 30
243 ( 65
1979 ( 567
234 ( 40
3326 ( 788
148 ( 30
2522 ( 554
338 ( 35
289 ( 56
49.1 ( 10.3
835 ( 182
23.6 ( 1.1
1712 ( 355
136 ( 23
(-)-7
(+)-7
9
(-)-10
(+)-10
11
(()-24a
(()-24b
(()-35a
(()-35b
(()-35c
(()-8
(+)-8
(-)-8
(()-43a
(()-43b
a Results are mean values ( SEM for three to eight experiments each
performed in triplicate.
inactive. Also, the transition from n-propyl to n-butyl led to
complete loss of functional activity for the racemate of 2.15 Since
no binding data were available for interaction of this compound
2 with DA receptors, this compound was resynthesized by us
to evaluate its binding interaction with the cloned DA receptor
subtypes. As shown in Table 1, compound 2 exhibited very poor
affinity for both D2 and D3 receptors. Thus, the data from the
hybrid structure (()-43a correlated well with the binding affinity
for the nonhybrid structure. No attempts were made to synthesize
cis-analogues of 43a. However, the dichloro derivative 43b was
significantly more potent and the effect was more significant
for D2 receptor compared to D3 receptor. Thus, 43b was 12

7812 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 24
Brown et al.
influence of nature of N-alkyl substitution on activity.15,17a The
fact that an alkyl group could be replaced by a substituted
piperazine fragment with enhancement of binding activity may
indicate development of additional cooperative binding interac-
tions of the hybrid derivatives with D2/D3 receptors.
aminotetralins structures, along with (-)-8, a hybrid tricyclic
aminotetralin, and (-)-11, representing a 2-aminothiazole-based
fused bicyclic system. Of the several pharmacophore hypotheses
generated, two representative hypotheses were selected on the
basis of the overall alignment score, nature of features present,
and the molecules covered. Only the top ranked of the two
hypotheses is shown in Figure 4b along with the hybrid
analogues aligned onto these features. These two hypotheses
differ in interfeature distances as shown in Figure 4c, which
depicts the features and the corresponding interfeature distance
ranges. The range of distances results from changes in the
conformation of the ethylene bridge connecting the cationic N
with the arylpiperazine portion. In Figure 4b, this ethylene linker
adopted the gauche conformation whereas in the second
hypothesis (not shown), it adopted the anti conformation. These
interfeature distances are shown in Table 2. These distances
were measured from the gauche and anti conformations of the
individual molecules, whereas interfeature distance ranges
shown in Figure 4c were measured from the individual
pharmacophore hypotheses (representing gauche and anti con-
formations of the ethylene bridge). The distances between
hydrophobic features with one centered on the aminotetralin
aryl portion (Ar1) and the other on the phenyl ring of
arylpiperazine portion (Ar2), as well as the distances between
cationic N and the arylpiperazine hydrophobic features, were
greater in the case of the anti conformation (Table 2). The
distances between the cationic N and Ar1 were reduced in the
case of the anti conformation compared to the gauche confor-
mation (shown in Figure 4b). We strongly believe that these
conformations and the associated interfeature distances coupled
with other factors govern the selectivity for D2/D3, provided
the above-mentioned features (shown in Figure 4b) are present
in the molecules. Figure 4b represents the DA receptor (D2 and
D3) ligand pharmacophore depicting three features consisting
of two aromatic/hydrophobic and one cationic feature. Along
with these three features, it shows the Don2 feature, which
represents the direction in which the H-bond donor attached to
the aromatic portion should be oriented for optimal interaction
with the receptor(s). The aromatic/hydrophobic features repre-
sent (1) Ar1, the phenyl ring in hybrid aminotetralins (+)-6,
(-)-7, and (-)-8 and 2-aminothiazole ring in (-)-11, and (2)
Ar2, the phenyl ring attached to N-4 of piperazine. One set of
aromatic/hydrophobic and cationic features is common to parts
a and b of Figure 4. This set of two features is similar to that
described for the D3 ligand pharmacophore by Varady, et al.31
However, compared to the pharmacophore features shown in
Figure 4a, the directional character of the H attached to a
cationic N is absent in Figure 4b, even though in principle the
H attached to a quaternary N is oriented in one direction only.
The lack of directional feature in this case was probably due to
the pharmacophore generation process itself (the methodological
part). In addition, our current proposed model bears certain
resemblance to earlier models.15,17a
The data taken together represent our SAR study with hybrid
derivatives developed for D2/D3 receptors to understand the
molecular mode of interactions and to derive a pharmacophore
model. Thus, we designed and synthesized conformationally
constrained versions in which the aminotetralin moiety was
converted into trans-1,2,3,4,4a,5,10,10a-octahydrobenzo[g]quin-
oline and trans-1,2,3,4,4a,5,6,10b-octahydrobenzo[f]quinoline
structures. Hybrid compounds with a trans-1,2,3,4,4a,5,6,10b-
octahydrobenzo[f]quinoline structure exhibited more affinity for
D2/D3 receptors than those with a trans-1,2,3,4,4a,5,10,10a-
octahydrobenzo[g]quinoline, indicating preference for one con-
strained conformation over the other. Overall, hybrid derivatives
exhibited a significant increase of affinity compared to the
corresponding nonhybrid counterparts, reflecting an important
role of the piperazine moiety in affinity for dopamine receptors.
Having acquired all the necessary data, our next task was to
carry out a computational analysis to derive a reliable pharma-
cophore model of hybrid compounds for their binding to D2/
D3 receptors.
Next, we extended our studies toward the generation of
pharmacophore hypotheses based on these conformationally
constrained bi- and tricyclic hybrid structures. There are few
reports of DA receptor pharmacophoric requirements based on
various ligands including conformationally constrained tricyclic
aminotetralins.15-17a The central theme of these receptor models
was the importance of various positions taken by the N-
substituents. One orientation disfavored any groups bigger than
n-propyl (sterically defined), whereas other orientations tolerated
bigger groups (e.g., McDermed’s DA-receptor model,15 Wik-
stro¨m’s extended model16) (sterically favorable). The hybrid
compounds (linear and angular) described in the present
investigation follow the general requirements depicted by above-
mentioned DA-receptor models. For additional description and
illustration with the help of representative hybrid compounds
of this DA-receptor model, readers are encouraged to see the
Supporting Information.
In the initial trial runs of pharmacophore generation with the
known D2/D3 ligands (S)-5-OH-DPAT, R-(+)-7-OH-DPAT,
and 5 (R-(+)-PD128907)17b (Figure 1), the best three-point
pharmacophore (based on overall alignment score), as depicted
in Figure 4a, shows the presence of one aromatic/hydrophobic
feature occupied by the aromatic ring (green), another hydro-
phobic feature present near the N-n-propyl groups (green)
common to all three ligands, and the quaternary N as the cationic
feature (blue). Shown as purple are the two features depicting
the directions of the phenolic hydroxyl H-bond donor and the
H attached to cationic N (required for reinforced H-bonding).
Figure 4a clearly shows the orientation of the H atom attached
to cationic N in the direction of the purple sphere, representing
a possible interaction with the Asp residue in TM-3. Similarly,
another purple sphere located near phenolic hydroxyls represents
the Ser residue(s) of the DA receptor subtypes. The distance
between the aromatic/hydrophic feature near the aromatic ring
and the cationic N was found in the narrow range of 5.06-5.16
Å. This distance falls within the range of 4.1-6.1 Å reported
for the same interfeature distance in a pharmacophore derived
from known D3 ligands.31
The factors responsible for the D3 preference for most of
these compounds ((+)-6, (-)-7, and 1) cannot be addressed by
the present pharmacophore completely. In our opinion, the
highest selectivity for D3 exhibited by (-)-11 may be due to
the presence of an additional aryl ring extending beyond those
present in rest of the molecules. Also, as seen from the
representative top scoring hypotheses (gauche and anti confor-
mations of the ethylene linker), we strongly believe that the
conformation of this linker is critical for interaction with either
or both D2/D3 receptors. In addition, at this point in time, no
information is available regarding the role of either or both
A pharmacophore model was generated on the basis of the
hybrid compounds (+)-6 and (-)-7, representing bicyclic hybrid

Structurally Constrained Hybrid DeriVatiVes
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 24 7813
Figure 4. (a) Three-point pharmacophore generated using known D2/D3 ligands, (b) three-point pharmacophore derived from D2/D3 ligand hybrid
analogues, and (c) three-point pharmacophore derived from hybrid analogues with corresponding interfeature distances: Aro/Hyd, Aromatic/
Hydrophobic; cat., cationic; Don2, the directional feature toward which the H-bond donor should be oriented. Ar1 denotes the aromatic ring of
either aminotetralin or 2-aminothiazole moieties, and Ar2 is the phenyl ring attached to one of the piperazine nitrogens.
Table 2. Pharmacophoric Distances for the Gauche and
Anti-Conformations of the Hybrid Analogues
interfeature distance (Å)a
gaucheb
compd N+-Ar1 N+-Ar2 Ar1-Ar2 N+-Ar1 N+-Ar2 Ar1-Ar2
pertaining to selectivity, it may shed light on requirements for
potency and aid in the design of potent and selective D2/D3
ligands/agonists. Certainly, the arylpiperazine portion of the
hybrid structures has a vital, yet complex, role to play in
imparting the higher potency compared to nonhybrid DA ligands
and also potentially plays a role in endowing selectivity for one
receptor over the other. Efforts are underway to identify
molecular determinants for potency and selectivity of these
hybrid structures, and our forthcoming publications will attempt
to address this with the help of carefully designed structures.
This will further help us to refine the preliminary pharmacophore
proposed in this article.
antib
(+)-6
(-)-6
11
5.06
5.14
4.92
5.13
8.50
8.57
8.51
8.52
12.26
12.62
12.00
12.66
5.00
5.13
4.89
5.15
8.73
9.26
8.59
9.26
12.95
13.66
12.24
13.72
(-)-8
a N+ represents the cationic feature. Ar1 denotes the aromatic ring of
either aminotetralin or 2-aminothiazole moieties, and Ar2 is the phenyl ring
attached to one of the piperazine nitrogens. The distance N+-Ar1 is the
distance between the cationic N and the ring centroid of Ar1. b The terms
gauche and anti represent the conformation of the ethylene bridge connecting
cationic N with the arylpiperazine moiety.
Conclusion
In this manuscript we have carried out further modifications of
our hybrid derivatives by converting them into structurally con-
strained versions to understand their mode of interaction in general
with DA D2 and D3 receptors and also to obtain an insight into
bioactive conformational structures of these molecules. For this
purpose, several 1,2,3,4,4a,5,10,10a-octahydrobenz[g]quinoline-ol
and 1,2,3,4,4a-5,6,10b-octahydrobenzo[f]quinolin-ol based hybrid
derivatives were prepared and biologically characterized. In this
piperazine N-atom(s) on affinity and selectivity. Our future
efforts will be focused on addressing these critical questions
and further refining this pharmacophore model. The hybrid
analogue concept representing D2/D3 receptor interaction
requirements could possibly address otherwise difficult to
explore selectivity requirements. In addition to further enhancing
our understanding of D2/D3 ligand structural requirements

7814 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 24
Brown et al.
at 0 °C under a N2 atmosphere for 2 h. Stirring was stopped, and
the excess zinc was allowed to settle. The benzylzinc bromide was
then added via cannulation to a suspension of bis(triphenylphoshine)n-
ickel(II) chloride (1.78 g, 27.3 mmol) and methyl 2-chloronicotinate
(1.17 g, 6.82 mmol) in 100 mL of dry THF. This mixture was stirred
at ambient temperature for 48 h. The reaction was quenched by
the addition of 10% NH4Cl (100 mL), and the product was extracted
with ethyl acetate, dried (Na2SO4), filtered, and concentrated. The
crude mixture was dissolved in ether, and ethereal HCl was added.
The salt was recovered by filtration and used without further
regard, as pointed out earlier, octahydrobenzo[f and g]quinolinol
related derivatives as constrained molecules of aminotetralin
structures (7-OH-DPAT and 5-OH-DPAT) were synthesized before
for interaction with DA receptors. However, none of those
derivatives were evaluated in detailed binding interaction with D2
and D3 receptors, as they were mainly characterized in functional
assays and in some cases in binding assay for the D2 receptor.
Therefore, precise information on binding interactions with D2/
D3 receptors was not available. The present results from constrained
hybrids indicate that a compound with structure 1,2,3,4,4a,5,6,10b-
octahydrobenzo[f]quinolinol, e.g., (-)-8, exhibits high affinity for
D2/D3 receptors and in this regard is significantly more potent
than the corresponding nonhybrid version. Similar results were
found with other lead compounds when compared with the
corresponding nonhybrid version. Our study with the hybrid version
of these constrained derivatives as well as other lead hybrid
compounds provides some unique information about the interaction
of these compounds with DA receptors. These results clearly
indicate a significant contribution of the piperazine moiety in
enhancing interaction of hybrid derivatives with DA receptor
subtypes. Our results prompt the proposal of a unique pharma-
cophore structure for hybrid derivatives consisting of three phar-
macophoric centers. This unique pharmacophore structure will be
further refined by incorporating the results of future studies.
1
purification to yield 1.57 g of 16 (free base, 90%). H NMR (400
MHz, CDCl3) δ 3.75 (s, 3H), 3.87 (s, 3H), 4.56 (s, 2H), 6.70-6.73
(m, 1H), 6.81-6.85 (m, 1H), 7.15-7.19 (t, 1H, J ) 8 Hz),
7.22-7.26 (m, 2H), 8.16-8.18 (dd, 1H, J ) 1.6 Hz, J ) 8 Hz),
8.68-8.7 (dd, 1H, J ) 1.6 Hz, J ) 8 Hz).
cis- and trans-2-(3-Methoxybenzyl)piperidine-3-carboxylic
Acid Methyl Ester (17). The hydrochloride salt of pyridine 16
(5.75 g, 19.2 mmol) was dissolved in 200 mL of MeOH, and 200
mg of PtO2 was added. The mixture was then hydrogenated at 45
psi for 24 h. The mixture was filtered though a pad of Celite and
the solvent removed in vacuo to yield a mixure of cis- and trans-
amine 17 isomers as an oil (4.77 g, 95%), which were used without
further purification and were not separated in the subsequent
transformations. 1H NMR (CDCl3) δ 7.18-7.25 (t, 1H, J ) 8 Hz),
6.72-6.79 (m, 3H), 3.80 (s, 3H) 3.72 (s, 3H), 3.0-3.12 (m, 2H),
2.84- 2.9 (m, 1H), 2.57-2.77 (m, 3H), 1.8-2.2 (m, 1H), 1.76-1.82
(m, 2H), 1.66-1.73 (m, 2H), 1.42-1.46 (m, 1H).
cis- and trans-2-(3-Methoxybenzyl)piperidine-1,3-dicarboxylic
Acid Dimethyl Ester (18). Amine 17 (1.58 g, 6.03 mmol) was added
to a suspension of 1.9 g (13 mmol) of K2CO3 in 20 mL of
dichloromethane. Methyl chloroformate (0.93 mL, 12 mmol) was
added, and the mixture was stirred for 3 h at ambient temperature.
Water (20 mL) was added and the product was extracted with
CH2Cl2, dried (Na2SO4), filtered, and concentrated to yield the crude
carbamate 18 as an oil (1.9 g, 100%), which was used without
Experimental Section
Analytical silica gel-coated TLC plates (silica gel 60 F254) were
purchased from EMD Chemical, Inc., and were visualized with UV
light by treatment with phosphomolybdic acid (PMA), Dragendor-
ff’s reagent, or ninhydrin. Flash column chromatography was carried
1
out on Whatman Purasil 60A silica gel 230-400 mesh. H NMR
spectra were routinely obtained on Varian 400 MHz FT NMR
equipment. The NMR solvent used was CDCl3, CD3OD, or DMSO-
d6 as indicated. TMS was used as an internal standard. Elemental
analysis was performed by Atlantic Microlab, Inc., and were within
(0.4% of the theoretical value. Optical rotations were recorded
on Perkin-Elmer 241 polarimeter.
1
further purification. H NMR (400 MHz, CDCl3) δ 1.66-2.0 (m,
3H), 2.50-2.54 (m, 1H), 2.64-2.77 (m, 2H), 2.87-2.99 (m, 2H),
3.33 (m, 2H), 3.58-3.60 (m, 2H), 3.69 (s, 3H), 3.80 (s, 3H),
4.8-4.98 (m, 1H), 4.1-4.16 (m, 1H), 6.62-6.80 (m, 3H),
7.15-1.18 (t, 1H, J ) 8 Hz).
(()-cis-8-Methoxy-5-oxo-3,4,4a,5,10,10a-hexahydro-2H-benzo-
[g]quinoline-1-carboxylic Acid Methyl Ester (19). Methyl ester
18 (910 mg, 2.83 mmol) was dissolved in 30 mL of MeOH. LiOH
(102 mg, 4.25 mmol) and water (7 mL) were added, and the mixture
was allowed to stir at room temperature for 24 h. Methanol was
then evaporated, and the mixture was partioned between water and
ethyl acetate. The aqueous phase was then acidified and extracted
with several portions of ethyl acetate. The combined organic
fractions were dried (Na2SO4), filtered, and concentrated to yield
the crude acid, which was used without further purification. Then
800 mg (2.6 mmol) of the acid was dissolved in 30 mL of
dichloromethane. SOCl2 (0.23 mL, 13.4 mmol) was added, and the
mixture was refluxed until no more HCl gas was produced. The
mixture was cooled and concentrated to yield the crude acid
chloride, which was dissolved in 40 mL of dry CH2Cl2 and cooled
to 0 °C. TiCl4 (1 M in CH2Cl2, 5.2 mL, 5.2 mmol) was added
dropwise, and this mixture was stirred in an ice bath for 2 h. The
reaction was quenched by the addition of saturated NaCl (20 mL),
and the product was extracted with CH2Cl2. The organic layer was
dried (Na2SO4), filtered, and concentrated. The product was purified
by column chromatography (94:5:1 EtOAc/MeOH/Et3N) to yield
only the cis-isomer 19 as a yellow solid (480 mg, 63%). 1H NMR
(400 MHz, CDCl3) δ 1.59-1.62 (m, 1H), 1.78-1.82 (m, 1H),
1.91-1.97 (m, 1H), 2.69-2.72 (m, 1H), 2.72-2.82 (m, 1H),
2.92-3.1 (m, 1H), 3.3-3.4 (m, 1H), 3.72 (s, 3H), 3.89 (s, 3H),
4.1-4.18 (m, 1H), 4.79-4.81 (m, 1H), 6.65-6.67 (m, 1H),
6.83-6.87 (m, 1H), 8.01-8.04 (d, 1H, J ) 8 Hz).
Procedure A. 2-Chloro-1-(4-phenylpiperazin-1-yl)ethanone
(13a). 1-Phenylpiperazine 12a (3 g, 18.5 mmol) and triethylamine
(3.34 mL, 22.2 mmol) were dissolved in 100 mL of CH2Cl2 and
cooled to 0 °C. Chloroacetyl chloride (1.9 mL, 22.2 mmol) was
added slowly over the course of 5 min. The mixture was allowed
to stir at 0 °C for 30 min, at which time saturated NaHCO3 (100
mL) was added. The organic layer was separated, dried (Na2SO4),
filtered, and concentrated. The crude solid was then recrystallized
from EtOAc to give 4.3 g (97%) of 2-chloro-1-(4-phenylpiperazin-
1
1-yl)ethanone 13a. H NMR (400 MHz, CDCl3) δ 3.47 (bs, 4H),
3.59 (bs, 4H), 4.25 (s, 2H), 6.53-6.55 (d, 1H, J ) 7.6 Hz),
6.64-6.66 (d, 1H, J ) 7.6 Hz), 6.83-6.87 (t, 1H, J ) 8 Hz).
2-Chloro-1-[4-(2,3-dichlorophenyl)piperazin-1-yl]ethanone
(13b). This compound was prepared following procedure A using
2 g (7.47 mmol) of 1-(2,3-dichlorophenyl)piperazine, 0.92 mL (8.22
mmol) of chloroacetyl chloride, and 2 mL (15 mmol) of triethy-
lamine to give 2.6 g (87%) of 13b as a solid. 1H NMR (400 MHz,
CDCl3) δ 3.43 (bs, 4H), 4.27 (s, 2H), 3.61 (bs, 4H), 6.41-4.49
(m, 2H), 6.90-6.93 (t, 1H, J ) 8 Hz).
4-Chloro-1-(4-phenylpiperazin-1-yl)butan-1-one (13c). This
compound was prepared following procedure A using 2 g (12.3
mmol) of 1-phenylpiperazine, 1.91 g (13.6 mmol) of 4-chlorobu-
tanoyl chloride, and 2 mL (15 mmol) of triethylamine to give 3.19 g
(97%) of solid 13c. 1H NMR (400 MHz, CDCl3) δ 1.94-1.98 (m,
2H), 2.32-2.34 (t, 2H, J ) 8 Hz), 3.33-3.35 (m, 4H), 3.48-3.50
(m, 4H), 3.67-3.67 (t, 2H, J ) 8 Hz), 6.53-6.55 (d, 1H, J ) 7.6
Hz), 6.64-6.66 (d, 1H, J ) 7.6 Hz), 6.83-6.87 (t, 1H, J ) 8 Hz).
2-(3-Methoxybenzyl)nicotinic Acid Methyl Ester (16). Acti-
vated zinc dust (1.75 g, 27.3 mmol) was added to 20 mL of dry
THF and the suspension cooled to 0 °C. 3-Methoxybenzyl bromide
14 (1.90 mL, 13.6 mmol) was added, and this mixture was stirred
(()-cis-8-Methoxy-3,4,4a,5,10,10a-hexahydro-2H-benzo[g]-
quinoline-1-carboxylic Acid Methyl Ester (20). Ketone (()-
19 (610 mg, 2.1 mmol) and 1.5 mL of 70% perchloric acid were
dissolved in 25 mL of acetic acid. Then 60 mg of Pd/C was then

Structurally Constrained Hybrid DeriVatiVes
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 24 7815
added and the mixture was hydrogenated at 45 psi for 34 h. The
catalyst was filtered through a pad of Celite, and acetic acid was
evaporated. The residue was washed with 20% NaOH, dried
(Na2SO4), filtered, and concentrated to yield 510 mg (88%) of 20
(s, 3H), 6.60-6.62 (m, 1H), 6.66-6.69 (m, 1H), 6.83-6.87 (m,
1H), 6.95-6.97 (dd, 1H, J ) 1.8 Hz, J ) 8 Hz), 7.14-7.20 (m,
2H).
Procedure D. cis- 1-[2-(4-Phenylpiperazin-1-yl)ethyl]-1,2,3,
4,4a,5,10,10a-octahydrobenzo[g]quinolin-8-ol (24a). Compound
(()-23a (310 mg, 0.76 mmol) was dissolved in 40 mL of
dichloromethane and cooled to -78 °C. Then 1.5 mL of BBr3 (1
M in CH2Cl2) was added dropwise and the mixture was allowed to
warm to ambient temperature and stirred overnight. Saturated
NaHCO3 (50 mL) was added, and the product was extracted with
CH2Cl2. The organic extracts were dried (Na2SO4), filtered, and
concentrated. The titled compound was then purified by column
chromatography using 94:5:1 EtOAc/MeOH/Et3N to yield (()-24a
(233 mg, 78%) as a tan solid. 1H NMR (400 MHz, CDCl3) δ
1.35-1.43 (m, 2H), 1.6-1.77 (m, 2H), 2.1-2.19 (m, 1H),
2.55-2.84 (m, 14H), 2.92-3.0 (m, 1H), 3.09-3.2 (m, 1H),
3.18-3.22 (m, 4H), 6.36-6.38 (m, 1H), 6.52-6.55 (m, 1H),
6.82-6.91 (m, 4H), 7.22-7.28 (m, 2H). The free base was
1
as an oil, which was used without further purification. H NMR
(400 MHz, CDCl3) δ 1.40-1.6 (m, 3H), 1.6-1.71 (m, 1H),
2.03-2.07 (m, 1H), 2.55-2.59 (d, 1H, J ) 8 Hz), 2.85-3.1 (m,
3H), 2.71-2.82), 3.71 (m, 3H), 3.76 (m, 3H), 4.16-4.21 (m, 1H),
4.45-4.51 (m, 1H), 6.59-6.6 (m, 1H), 6.68-6.71 (m, 1H),
6.96-6.98 (d, 2H, J ) 8 Hz).
(()-cis-8-Methoxy-1,2,3,4,4a,5,10,10a-octahydrobenzo[g]-
quinoline (21). Carbamate 20 (350 mg, 1.27 mmol) was added to
a solution of KOH (240 mg, 7.64 mmol), water (0.13 mL, 7.64
mmol), and hydrazine hydrate (0.24 mL, 7.64 mmol) in ethylene
glycol (20 mL). This mixture was refluxed for 3 h, at which time
the solution was cooled and poured into water and the product was
extracted with several portions of ethyl acetate. After the mixture
was dried, filtered, and concentrated, 182 mg (90%) of the titled
product 21 was recovered as a sticky oil. 1H NMR (400 MHz,
CDCl3) δ 1.38-1.43 (m, 1H), 1.55 (s, 1H), 1.66-1.74 (m, 3H),
1.96-2.2 (m, 1H), 2.5-2.75 (m, 3H), 2.92-3.10 (m, 4H), 3.72 (s,
3H), 6.55-6.2 (m, 1H), 6.68-6.67 (m, 1H), 6.96-7.00 (d, 2H, J
) 8 Hz).
converted to its HCl salt, mp
(C25H33N3O·2HCl) C, H, N.
) 174-177 °C. Anal.
(()-cis-1-{2-[4-(2,3-Dichlorophenyl)piperazin-1-yl]ethyl}-1,
2,3,4,4a,5,10,10a-octahydrobenzo[g]quinolin-8-ol (24b). Com-
pound 22b (338 mg, 0.714 mmol) was demethylated following
1
procedure D, giving (()-24b, 267 mg (80%), as a tan solid. H
Procedure B. (()-cis-2-(8-Methoxy-3,4,4a,5,10,10a-hexahydro-2-
H-benzo[g]quinolin-1-yl)-1-(4-phenylpiperazin-1-yl)ethanone
(22a). A mixture of amine (()-21 (390 mg, 1.8 mmol), chloride
13a (514 mg, 2.16 mmol), and K2CO3 (745 mg, 5.4 mmol) in
acetonitrile (40 mL) was heated at reflux for 1 h. The mixture was
then cooled, filtered, and concentrated. The product was purified
by column chromatography using ethyl acetate as an eluent to yield
22a (460 mg, 61%) as a solid. 1H NMR (400 MHz, CDCl3) δ
1.5-1.6 (m, 4H), 1.72-1.8 (m, 1H), 2.9-3.1 (m, 4H), 2.7-2.82
(m, 7H), 3.2-3.29 (m, 1H), 3.38-3.5 (m, 2H), 3.54-3.58 (m, 1H),
3.78 (s, 3H), 7.25-7.30 (m, 2H), 6.62-6.64 (m, 1H), 6.68-6.7
(m, 1H), 6.84-6.9 (m, 3H), 6.96-7.0 (d, 1H, J ) 8 Hz).
(()-cis-1-[4-(2,3-Dichlorophenyl)piperazin-1-yl]-2-(8-methoxy-
3,4,4a,5,10,10a-hexa-hydro-2H-benzo[g]quinolin-1-yl)ethanone
(22b). Amine (()-21 (200 mg, 0.8 mmol), chloride 13b (290 mg,
0.95 mmol), and K2CO3 (328 mg, 2.4 mmol) were reacted according
to procedure B. The product was purified by column chromatog-
raphy using ethyl acetate as an eluent to yield 354 mg (92%) of
22b as a solid. 1H NMR (400 MHz, CDCl3) δ 1.50-1.61 (m, 3H),
1.76-1.83 (m, 1H), 2.14 (bs, 1H), 2.3-2.4 (m, 1H), 2.52-2.8 (m,
10H), 3.01-3.07 (m, 2H), 3.32-3.40 (m, 2H), 3.51-3.56 (m, 2H),
3.75 (s, 3H), 6.61-6.62 (m, 1H), 6.65-6.68 (m, 1H), 6.82-6.86
(m, 1H), 6.94-6.96 (dd, 1H, J ) 1.8 Hz, J ) 8 Hz), 7.13-7.19
(m, 2H).
NMR (400 MHz, CDCl3) δ 1.35-1.43 (m, 2H), 1.6-1.77 (m, 2H),
2.1-2.19 (m, 1H), 2.55-2.84 (m, 14H), 2.92-3.0 (m, 1H),
3.09-3.2 (m, 1H), 3.18-3.22 (m, 4H), 6.36-6.38 (m, 1H),
6.52-6.55 (m, 1H), 6.82-6.91 (m, 4H), 7.22-7.28 (m, 2H). The
free base was then converted to its HCl salt, mp ) 161-165 °C.
Anal. (C25H31Cl2N3O·2HCl) C, H, N.
2-Hydroxy-7-methoxy-3,4-dihydronaphthalene-1-carboxylic
Acid Methyl Ester (26). 7-Methoxy-2-tetralone 25 (12.5 g, 76
mmol) and NaH (5.6 g, 0.145 mol) were refluxed in benzene (200
mL) for 30 min, at which time dimethyl carbonate (12 mL, 0.145
mol) was added and refluxing was continued for 3 h. The mixture
was cooled and then carefully quenched by the addition of water
(50 mL), and the product was extracted with ethyl acetate, dried
(Na2SO4), filtered, and concentrated to yield the product, which
was purified by column chromatography (1:1 hexane/EtOAc) to
give 16 g of 26 (96%) as a liquid. 1H NMR (400 MHz, CDCl3) δ
2.50-2.54 (t, 2H, J ) 8 Hz), 2.73-2.77 (t, 2H, J ) 8 Hz), 3.8 (s,
3H), 3.92 (s, 3H), 6.60-6.63 (dd, 1H, J ) 2 Hz, 8 Hz), 7.02-7.05
(d, 1H, J ) 8 Hz), 7.30-7.31 (d, 2H, J ) 2 Hz), 13.38 (s, 1H).
3-(2-Cyanoethyl)-2-hydroxy-7-methoxy-3,4-dihydronaphth-
alene-1-carboxylic Acid Methyl Ester (27). n-Butyllithium (55.6
mL, 0.138mol) was added to a solution of diisopropylamine (19.3
mL, 0.138 mol) in dry THF (100 mL) at -78 °C and stirred for 15
min, then warmed to 0 °C and stirred for an additional 15 min.
Next, a solution of 26 (12.9 g, 0.055 mol) dissolved in THF (20
mL) was added slowly and the mixture was stirred at 0 °C for 1 h.
Chloropropionitrile (5.6 mL, 0.071 mol) was then added, and this
solution was stirred at room temperature for 1 h and carefully
quenched with 50 mL of 1 M HCl. The product was extracted with
ethyl acetate (3 × 50 mL), dried (Na2SO4), filtered, and concen-
trated. The crude oil was then purified by column chromatography
(1:1 hexane/EtOAc) to yield 27, 7.28 g (46%), as an oil. 1H NMR
(400 MHz, CDCl3) δ 1.70-1.76 (m, 1H), 1.88-1.94 (m, 1H),
2.41-2.50 (m, 2H), 2.57-2.67 (m, 1H), 2.82-2.86 (t, 1H, J )
6.8 Hz), 2.97-3.01 (m, 1H), 3.80 (s, 3H), 3.93 (s, 3H), 13.4 (s,
1H), 6.63-6.65 (dd, 1H, J ) 2 Hz, J ) 8 Hz), 7.03-7.05 (d, 1H,
J ) 8 Hz), 7.30-7.31 (d, 1H, J ) 8 Hz).
Procedure C. (()-cis-1-[2-(4-Phenylpiperazin-1-yl)ethyl]-1,
2,3,4,4a,5,10,10a-octahydrobenzo[g]quinolin-8-ol (23a). Lith-
ium aluminum hydride (244 mg, 6.44 mmol) was added to dry THF
(50 mL), and the suspension was cooled to 0 °C. A solution of
amide 22a (450 mg, 1.0 mmol) in THF (15 mL) was added
dropwise, and the mixture was refluxed for 3 h. The mixture was
cooled in an ice bath and quenched with 10% NaOH, and the
resulting solids were filtered. The filtrate was dried (Na2SO4),
filtered, and concentrated to yield amine 23a (235 mg, 98%) as an
oil, which was used without further purification. 1H NMR (400
MHz, CDCl3) δ 1.51-1.62 (m, 4H), 1.70-1.81 (m, 1H), 2.92-3.11
(m, 4H), 2.72-2.86 (m, 9H), 3.22-3.30 (m, 1H), 3.37-3.50 (m,
2H), 3.55-3.58 (m, 1H), 3.77 (s, 3H), 7.24-7.29 (m, 2H),
6.63-6.65 (m, 1H), 6.67-6.7 (m, 1H), 6.85-6.9 (m, 3H),
6.97-7.01 (d, 1H, J ) 8 Hz).
(()-cis-1-[4-(2,3-Dichloro-phenyl)piperazin-1-yl]-2-(8-meth-
oxy-3,4,4a,5,10,10a-hexahydro-2H-benzo[g]quinolin-1-yl)etha-
none (23b). Compound 22b (354 mg, 0.725 mmol) was reduced
following procedure C to yield 338 mg (98%) of 23b as an oil. 1H
NMR (400 MHz, CDCl3) δ 1.51-1.60 (m, 3H), 1.75-1.82 (m,
1H), 2.15 (bs, 1H), 2.31-2.4 (m, 1H), 2.50-2.87 (m, 12H),
3.010-3.05 (m, 2H), 3.30-3.39 (m, 2H), 3.52-3.55 (m, 2H), 3.78
3-(6-Methoxy-3-oxo-1,2,3,4-tetrahydronaphthalen-2-yl)pro-
pionitrile (28). Methyl ester 27 (430 mg, 1.5 mmol), LiCl (63
mg, 1.5 mmol), water (2 mL), and DMSO (1 mL) were refluxed
for 4 h. The solution was then cooled and poured into water (15
mL) and the product extracted with several portions of ethyl acetate
(3 × 15 mL). The organic extracts were pooled, dried (Na2SO4),
filtered, and concentrated to yield 243 mg (71%) of 28 as an oil,
which was used directly in the next step. 1H NMR (400 MHz,
CDCl3) δ 1.63-1.79 (m, 1H), 2.16-2.21 (m, 1H), 2.54-2.56 (m,
2H), 2.62-2.67 (m, 1H), 2.76-2.88 (m, 1H), 3.05-3.10 (dd, 1H,

7816 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 24
Brown et al.
J ) 5.6 Hz, J ) 8 Hz), 3.55-3.70 (q, 2H, J ) 6 Hz), 6.65-6.66
(m, 1H), 3.80 (s, 3H), 6.76-6.79 (dd, 1H, J ) 2.4 Hz, J ) 8 Hz),
7.11-7.13 (d, 1H, J ) 8 Hz).
CDCl3) δ 1.20-1.34 (m, 1H), 1.79-1.81 (m, 1H), 1.91-2.23 (m,
5H), 2.41-2.50 (m, 5H), 2.79-2.82 (m, 1H), 3.0-3.41 (m, 9H),
3.62-3.80 (m, 7H), 6.60-6.63 (m, 1H), 6.68-6.72 (m, 1H),
6.89-6.95 (m, 4H), 7.26-7.31 (m, 2H).
(()-trans-8-Methoxy-1-[2-(4-phenylpiperazin-1-yl)ethyl]-1,
2,3,4,4a,5,10,10a-octahydrobenzo[g]quinoline (34a). Amide
33a (110 mg, 0.262 mg) was reduced with lithium aluminum
hydride (70 mg, 1.31 mmol) following procedure C to yield 105
mg of solid 34a in 99% yield. 1H NMR (400 MHz, CDCl3) δ
1.13-1.23 (m, 1H), 1.61-1.76 (m, 2H), 1.82-1.90 (m, 1H),
2.22-2.41 (m, 3H), 2.52-2.81 (m, 10H), 3.0-3.20 (m, H),
3.21-3.30 (m, 5H), 3.78 (s, 3H), 6.62-672 (m, 2H), 6.82-7.10
(m, 4H), 7.22-7.30 (m, 2H).
(()-trans-1-{2-[4-(2,3-Dichlorophenyl)piperazin-1-yl]ethyl}-
8-methoxy-1,2,3,4,4a,-5,10,10a-octahydrobenzo[g]quinoline (34b).
Amide 33b (130 mg, 0.266 mmol) was reduced using 51 mg (1.33
mmol) of lithium aluminum hydride following procedure C to give
solid 34b, 120 mg (97%). 1H NMR (400 MHz, CDCl3) δ 1.10-1.22
(m, 1H), 1.62-1.77 (m, 2H), 1.83-1.91 (m, 1H), 2.21-2.40 (m,
3H), 2.53-2.80 (m, 10H), 3.1-3.20 (m, H), 3.23-3.32 (m, 5H),
3.78 (s, 3H), 6.63-673 (m, 2H), 7.68-7.10 (m, 4H),7.22-7.30
(m, 2H).
(()-trans-8-Methoxy-1-[4-(4-phenylpiperazin-1-yl)butyl]-1,
2,3,4,4a,5,10,10a-octahydrobenzo[g]quinoline (34c). Amide 33c
(170 mg, 0.38 mmol) was reduced using 74 mg (1.9 mmol) of
lithium aluminum hydride according to procedure C. Column
chromatography using EtOAc/MeOH (95:5) gave 34c, 159 mg
(96%). 1H NMR (400 MHz, CDCl3) δ 1.12-1.25 (m, 2H),
1.51-1.73 (m, 6H), 1.84-1.90 (m, 1H), 2.22-2.30 (m, 2H),
2.39-2.41 (m, 3H), 2.58-2.62 (m, 5H), 2.63-2.90 (m, 3H),
2.91-3.11 (m, 1H), 3.14-3.22 (m, 5H), 3.78 (s, 3H), 6.63-6.70
(m, 2H), 6.83-6.86 (t, 1H, J ) 8 Hz), 6.92-6.97 (t, 3H, J ) 8
Hz), 7.24-7.28 (t, 2H, J ) 8 Hz).
3-(7′-Methoxy-3′,4′-dihydro-1′H-spiro[[1,3]dioxolane-2,2′-
naphthalen]-3′-yl)propionitrile (29). A solution of ketone 28 (280
mg, 1.22 mmol), triethyl orthoformate (0.71 mL, 4.3 mmol),
ethylene glycol (1.2 mL), and TsOH (5 mg, 0.012 mmol) in
dichloromethane (20 mL) was stirred overnight at room temperature.
The reaction mixture was then poured into water (30 mL) and the
organic layer was separated, dried (Na2SO4), filtered, and concen-
trated to yield an oil which was purified by column chromatography
1
(EtOAc) to yield 29 332 mg (98%). H NMR (400 MHz, CDCl3)
δ 1.50-1.62 (m, 1H), 2.01-2.19 (m, 2H), 2.40-2.59 (m, 2H),
2.65-2.71 (m, 1H), 2.82-3.32 (m, 2H), 3.76 (m, 3H), 3.80-4.09
(m, 4H), 6.56-6.57 (m, 1H), 6.70-6.72 (dd, 1H, J ) 2.8 Hz, J )
8 Hz), 7.00-7.03 (d, 1H, J ) 8 Hz).
3-(7′-Methoxy-3′,4′-dihydro-1′H-spiro[[1,3]dioxolane-2,2′-
naphthalen]-3′-yl)-propylamine (30). Nitrile 29 (350 mg, 1.28
mmol) was dissolved in 30 mL of MeOH. Raney nickel (200 mg)
was added, and the mixture was hydrogenated overnight. The
reaction solution was filtered through a pad of Celite and concen-
trated to give 300 mg of 30 (62%) as a solid which was used without
further purification. 1H NMR (400 MHz, CDCl3) δ 1.10-1.79 (m,
6H), 1.92-2.0 (m, 1H), 2.60-3.16 (m, 6H), 3.67-3.79 (m, 3H),
3.82-4.10 (m, 4H), 6.56-6.57 (m, 1H), 6.70-6.72 (dd, 1H, J )
2.8 Hz, J ) 8 Hz), 7.01-7.04 (d, 1H, J ) 8 Hz).
(()-trans-8-Methoxy-1,2,3,4,4a,5,10,10a-octahydrobenzo[g]-
quinoline (32). Amine 30 (300 mg, 1.1 mmol) was dissolved in
20 mL of MeOH, and 5 mL of 6 M HCl was added. The solution
was refluxed for 2 h, cooled, and concentrated to yield the iminium
salt intermediate 31. The crude solid was then dissolved in 40 mL
of MeOH. NaCNBH3 (65 mg, 1.02 mmol) was added and the
solution stirred for 1 h at ambient temperature. Methanol was then
evaporated, and the residue was partioned between saturated
NaHCO3 (50 mL) and ethyl acetate (50 mL). The organic layer
was separated, dried (Na2SO4), filtered, and concentrated. The crude
amine was then purified by column chromatography (95:4:1 EtOAc/
(()-trans-1-[2-(4-Phenylpiperazin-1-yl)ethyl]-1,2,3,4,4a,5,10,
10a-octahydrobenzo[g]quinolin-8-ol (35a). Ether 34a (100 mg,
0.47 mmol) was demethylated following procedure D. The crude
solid was purified by crystallization (EtOAc/MeOH) to give 35a,
75 mg (78%). 1H NMR (400 MHz, CDCl3) δ 1.12-1.24 (m, 1H),
1.62-1.75 (m, 2H), 1.83-1.92 (m, 1H), 2.24-2.42 (m, 3H),
2.50-2.82 (m, 10H), 3.13-3.20 (m, H), 3.20-3.31 (m, 5H),
6.63-6.74 (m, 2H), 6.83-7.11 (m, 4H), 7.21-7.30 (m, 2H). The
free base was then converted to its HCl salt, mp ) 167-170 °C.
Anal. (C25H33N3O·2HCl·H2O) C, H, N.
1
MeOH/Et3N) to give 32, 187 mg (80%), in solid form. H NMR
(400 MHz, CDCl3) δ 1.11-1.26 (m, 1H), 1.46-1.73 (m, 4H),
1.91-1.95 (m, 1H), 2.37-2.43 (m,1H), 2.57-2.76 (m, 4H),
2.83-2.88 (m, 1H), 3.10-3.14 (m, 1H), 3.77 (s, 3H), 6.61-6.62
(m, 1H), 6.67-6.97 (dd, 1H, J ) 2.8 Hz, J ) 8 Hz), 6.96-6.98
(d, 1H, J ) 8 Hz).
(()-trans-2-(8-Methoxy-3,4,4a,5,10,10a-hexahydro-2H-ben-
zo[g]quinolin-1-yl)-1-(4-phenylpiperazin-1-yl)ethanone (33a).
Amine 32 (110 mg, 0.5 mmol), chloride 13a (202 mg, 0.66 mg),
and K2CO3 (175 mg, 1.27 mmol) were refluxed following procedure
B. The crude solid 33a (190 mg) was used without further
(()-trans-1-{2-[4-(2,3-Dichlorophenyl)piperazin-1-yl]ethyl}-
1,2,3,4,4a,5,10,10a-octahydrobenzo[g]quinolin-8-ol (35b). Ether
34b (140 mg, 0.30 mmol) was demethylated following procedure
D. The crude solid was purified by crystallization (EtOAc/MeOH)
to give 35b, 100 mg (73%). 1H NMR (400 MHz, CDCl3) δ
1.14-1.24 (m, 1H), 1.63-1.79 (m, 2H), 1.84-1.92 (m, 1H),
2.20-2.41 (m, 3H), 2.55-2.83 (m, 10H), 3.2-3.23 (m, H),
3.22-3.34 (m, 5H), 6.60-6.72 (m, 2H), 6.92-7.14 (m, 3H),
7.33-7.40 (m, 1H). The free base was then converted to its HCl
salt, mp ) 182-185 °C. Anal. (C25H31Cl2N3O·2HCl) C, H, N.
(()-trans-1-[4-(4-Phenylpiperazin-1-yl)butyl]-1,2,3,4,4a,5,10,
10a-octahydrobenzo[g]quinolin-8-ol (35c). Methyl ether 34c (90
mg, 0.207 mmol) was demethylated according to procedure D to
give 35c, 61 mg (70%), after crystallization from EtOAc/MeOH.
1H NMR (400 MHz, CDCl3) δ 1.112-1.24 (m, 2H), 1.50-1.72
(m, 6H), 1.85-1.90 (m, 1H), 2.21-2.30 (m, 2H), 2.37-2.40 (m,
3H), 2.58-2.63 (m, 5H), 2.62-2.91 (m, 3H), 2.91-3.11 (m, 1H),
3.15-3.23 (m, 5H), 6.62-6.71 (m, 2H), 6.84-6.87 (t, 1H, J ) 8
Hz), 6.93-6.98 (t, 3H, J ) 8 Hz), 7.24-7.28 (t, 2H, J ) 8 Hz).
The free base was converted to its HCl salt, mp ) 178-183 °C.
Anal. (C27H37N3O) C, H, N.
1
purification. H NMR (400 MHz, CDCl3) δ 1.50-1.20 (m, 1H),
1.6-1.7 (m, 1H), 1.84-1.90 (m, 1H), 2.30-2.48 (m, 3H), 2.17 (s,
H), 2.62-2.79 (m, 1H), 2.50-3.1 (m, 5H), 3.5-3.36 (m, 2H),
3.53-3.62 (m, 1H), 3.68-3.74 (m, 1H), 3.79 (s, 3H), 3.95-4.17
(m, 1H), 3.99-4.13 (m, 1H), 6.62-6.70 (m, 2H), 6.89-6.97 (m,
4H), 7.26-7.31 (m, 2H).
(()-trans-1-[4-(2,3-Dichlorophenyl)piperazin-1-yl]-2-(8-meth-
oxy-3,4,4a,5,10,10a-hexahydro-2H-benzo[g]quinolin-1-yl)ethanone
(33b). Amine 32 (70 mg, 0.277 mmol) and chloride 13b (93 mg,
0.30 mmol) were treated with K2CO3 (76 mg, 0.55 mmol) following
procedure B to yield 126 mg of crude solid 33b, which was used
without further purification. 1H NMR δ (400 MHz, CDCl3)
1.50-1.20 (m, 1H), 1.6-1.7 (m, 1H), 1.84-1.90 (m, 1H), 2.17 (s,
H), 2.30-2.48 (m, 3H), 2.62-2.79 (m, 1H), 2.80-3.1 (m, 5H),
3.5-3.36 (m, 2H), 3.53-3.62 (m, 1H), 3.68-3.74 (m, 1H), 3.79
(s, 3H), 3.95-4.10 (m, 1H), 4.12-4.14 (m, 1H), 6.62-6.70 (m,
2H), 6.89-6.97 (m, 4H), 7.26-7.31 (m, 2H).
(()-trans-(8-Methoxy-3,4,4a,5,10,10a-hexahydro-2H-benzo
[g]quinolin-1-yl)-1-(4-phenylpiperazin-1-yl)butan-1-one (33c). This
compound was prepared using 200 mg of amine 32 (0.79 mmol),
288 mg of chloride 13c (0.945 mmol), and 327 mg of K2CO3 (2.37
mmol) according to procedure B. Column chromatography using
EtOAc as eluent afforded 33c 330 mg (93%). 1H NMR (400 MHz,
(()-trans-2-(7-Methoxy-2,3,4a,5,6,10b-hexahydro-1H-benzo
[f]quinolin-4-yl)-1-(4-phenylpiperazin-1-yl)ethanone (37). This
compound was prepared using 200 mg of amine 3630 (0.79 mmol),
288 mg of chloride 13a (0.945 mmol), and 327 mg of K2CO3 (2.37
mmol) according to procedure B. Column chromatography using
1
EtOAc as eluent afforded 330 mg of (()-37 (93%) as an oil. H
NMR (CDCl3) δ 1.13-1.23 (m, 1H), 1.61-1.76 (m, 2H), 1.82-1.90

Structurally Constrained Hybrid DeriVatiVes
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 24 7817
(m, 1H), 2.22-2.41 (m, 3H), 2.52-2.81 (m, 10H), 3.0-3.20 (m,
H), 3.21-3.30 (m, 5H), 3.78 (s, 3H), 6.62-672 (m, 2H), 6.82-7.10
(m, 4H), 7.22-7.30 (m, 2H).
(EtOAc/MeOH) to give (+)-8, 200 mg (82%). [R]D +43.6° (c 0.86
in MeOH). 1H NMR (400 MHz, CDCl3) δ 1.71-1.70 (m, 2H),
2.13-2.21 (m, 1H), 2.35-2.53 (m, 3H), 2.60-2.77 (m, 8H),
2.97-3.11 (m, 3H), 3.20-3.22 (t, 4H, J ) 5.2 Hz), 6.71-6.73 (d,
1H, J ) 8 Hz), 6.84-6.88 (t, 1H, J ) 8 Hz), 6.92-6.94 (m, 3H),
7.12-7.20 (t, 1H, J ) 8 Hz), 7.25-7.31 (t, 2H, J ) 8 Hz). The
free base was converted to its HCl salt, mp ) 163-167 °C. Anal.
(C25H33N3O·2HCl) C, H, N.
(-)-(4aS,10bS)-trans-4-(2-(4-Phenylpiperazin-1-yl)ethyl)-1,2,
3,4,4a,5,6,10b-octa-hydrobenzo[f]quinolin-7-ol ((-)-8). Compound
(-)-38 (265 mg, 0.653 mmol) was demethylated following
procedure D. The crude solid was purified by recrystallization
(EtOAc/MeOH) to give (-)-8, 215 mg (84%). [R]D -45.6° (c 0.55
in MeOH). 1H NMR (400 MHz, CDCl3) δ 1.71-1.68 (m, 2H),
2.12-2.21 (m, 1H), 2.32-2.49 (m, 3H), 2.61-2.77 (m, 8H),
2.95-3.05 (m, 3H), 3.21-3.23 (t, 4H, J ) 5.2 Hz), 6.71-6.73 (d,
1H, J ) 8 Hz), 6.80-6.84 (t, 1H, J ) 8 Hz), 6.95-6.97 (m, 3H),
7.16-7.20 (t, 1H, J ) 8 Hz), 7.22-7.26 (t, 2H, J ) 8 Hz). The
free base was converted to its HCl salt, mp ) 166-169 °C. Anal.
(C25H33N3O·2HCl) C, H, N.
(+)-(4aR,10aR)-trans-2-(7-Methoxy-2,3,4a,5,6,10b-hexahydro-
1H-benzo[f]quinolin-4-yl)-1-(4-phenylpiperazin-1-yl)ethanone
((+)-37). This compound was prepared following procedure B,
using 160 mg (0.632 mmol) of the (+)-3616 isomer (HCl salt) and
166 mg (0.70 mmol) of chloride 13a to give 260 mg (98%) of
1
(+)-37 as an oil. H NMR (400 MHz, CDCl3) δ 1.14-1.24 (m,
1H), 1.63-1.75 (m, 2H), 1.80-1.91 (m, 1H), 2.23-2.41 (m, 3H),
2.51-2.81 (m, 10H), 2.1-3.22 (m, H), 3.20-3.31 (m, 5H), 3.75
(s, 3H), 6.64-6.73 (m, 2H), 6.83-7.12 (m, 4H), 7.21-7.31 (m,
2H).
(-)-(4aS,10bS)-trans-2-(7-Methoxy-2,3,4a,5,6,10b-hexahydro-
1H-benzo[f]quinolin-4-yl)-1-(4-phenylpiperazin-1-yl)ethanone
((-)-37). This compound was prepared following procedure B,
using 150 mg (0.632 mmol) of the (-)-3616 isomer (HCl salt) and
183 mg (0.70 mmol) of chloride 13a to give 293 mg (96%) of
1
(-)-37 as an oil. H NMR (400 MHz, CDCl3) δ 1.15-1.24 (m,
1H), 1.63-1.74 (m, 2H), 1.81-1.92 (m, 1H), 2.23-2.43 (m, 3H),
2.53-2.85 (m, 10H), 3.2-3.26 (m, H), 3.23-3.33 (m, 5H), 3.80
(s, 3H), 6.63-6.70 (m, 2H), 6.81-7.11 (m, 4H), 7.23-7.31 (m,
2H).
trans-4-Propyl-1,2,3,4,4a,5,6,10b-octahydrobenzo[f]quinolin-
7-ol (1). To a solution of compound (()-36 (0.217 g, 1 mmol) and
TEA (0.5 mL) in dichloromethane (15 mL) at 0 °C was added
propionyl chloride (0.111 g, 1.2 mmol) dropwise. The reaction
mixture was allowed to stir for an additional 1 h. The reaction
mixture was washed with saturated NaHCO3 solution, followed by
water. The organic layer was dried (Na2SO4) and the solvent
removed in vacuo to yield a viscous liquid (0.2 g) which was
subjected to reduction with excess LiAlH4 following procedure C.
Demethylation of the tertiary amine thus obtained was carried out
in refluxing aqueous HBr (48%) for 3 h. The acid was removed in
vacuo. The residue obtained was converted to free base using
aqueous Na2CO3. The free base was purified using column
chromatography (dichloromethane/MeOH 8:2) to yield 187 mg of
(()-trans-7-Methoxy-4-(2-(4-phenylpiperazin-1-yl)ethyl)-1,
2,3,4,4a,5,6,10b-octahydrobenzo[f]quinoline (38). Compound 37
(361 mg, 0.833 mmol) was reduced with lithium aluminum hydride
(250 mg, 6.66 mmol) according to procedure C to give (()-38,
335 mg (98%), as an oil, which was sufficiently pure to use in the
next step. 1H NMR (400 MHz, CDCl3) δ 1.77-1.82 (m, 2H),
2.17-2.22 (m, 1H), 2.35-2.50 (m, 3H), 2.57-2.75 (m, 8H),
2.94-3.08 (m, 3H), 3.20-3.22 (t, 4H, J ) 5.2 Hz), 3.81 (s, 3H),
6.69-6.71 (d, 1H, J ) 8 Hz), 6.83-6.87 (t, 1H, J ) 8 Hz),
6.92-6.94 (m, 3H), 7.13-7.17 (t, 1H, J ) 8 Hz), 7.24-7.28 (t,
2H, J ) 8 Hz).
(+)-(4aR,10aR)-trans-7-Methoxy-4-(2-(4-phenylpiperazin-1-
yl)ethyl)-1,2,3,4,4a,5,6,-10b-octahydrobenzo[f]quinoline ((+)-38).
Compound (+)-37 (260 mg, 0.620 mmol) was reduced with lithium
aluminum hydride (188 mg, 5.0 mmol) according to procedure C
to give (+)-38, 251 mg (99%), as an oil, which was sufficiently
pure to use in the next step. 1H NMR (400 MHz, CDCl3) δ
1.75-1.80 (m, 2H), 2.16-2.23 (m, 1H), 2.34-2.51 (m, 3H),
2.58-2.76 (m, 8H), 2.94-3.10 (m, 3H), 3.20-3.22 (t, 4H, J )
5.2 Hz), 3.81 (s, 3H), 6.68-6.74 (d, 1H, J ) 8 Hz), 6.84-6.85 (t,
1H, J ) 8 Hz), 6.90-6.94 (m, 3H), 7.12-7.18 (t, 1H, J ) 8 Hz),
7.22-7.29 (t, 2H, J ) 8 Hz).
(-)-(4aS,10bS)-trans-7-Methoxy-4-(2-(4-phenylpiperazin-1-
yl)ethyl)-1,2,3,4,4a,5,6,-10b-octahydrobenzo[f]quinoline ((-)-38).
Compound (-)-37 (293 mg, 0.70 mmol) was reduced with lithium
aluminum hydride (212 mg, 5.59 mmol) according to procedure C
to give (-)-38, 265 mg (93%), as an oil, which was sufficiently
pure to use in the next step. 1H NMR (400 MHz, CDCl3) δ
1.75-1.86 (m, 2H), 2.18-2.26 (m, 1H), 2.34-2.51 (m, 3H),
2.54-2.72 (m, 8H), 2.92-3.06 (m, 3H), 3.22-3.24 (t, 4H, J )
5.2 Hz), 3.81 (s, 3H), 6.69-6.70 (d, 1H, J ) 8 Hz), 6.85-6.89 (t,
1H, J ) 8 Hz), 6.91-6.95 (m, 3H), 7.12-7.18 (t, 1H, J ) 8 Hz),
7.23-7.28 (t, 2H, J ) 8 Hz).
1
(()-1 (92%). H NMR (400 MHz, CDCl3) δ 1.04-1.07 (t, 3H, J
) 8 Hz), 1.46-1.49 (m, 1H), 1.79 (m, 3H), 1.98-2.14 (m, 3H),
2.59-2.77 (m, 3H), 3.03-3.16 (m, 5H), 3.6-3.63 (m, 1H),
6.63-6.65 (d, 1H, J ) 7.2 Hz), 6.18-6.83 (d, 1H, J ) 8 Hz),
7.00-7.04 (t, 1H, J ) 7.2 Hz). The free base was then converted
to its oxalte salt, mp ) 156-160 °C. Anal. (C18H25NO5) C, H, N
(()-trans-2-(6-Methoxy-3,4,4a,5,10,10a-hexahydro-2H-ben-
zo[g]quinolin-1-yl)-1-(4-phenylpiperazin-1-yl)ethanone (41a).
The hydrochloride salt of amine 4029 (350 mg, 1.38 mmol), chloride
13a (383 mg, 1.25 mmol), and K2CO3 (573 mg, 4.15 mmol) in
CH3CN were reacted following procedure B to yield 41a, 501 mg
(95%), as a yellow oil. 1H NMR (400 MHz, CDCl3) δ 1.62-1.69
(m, 3H), 1.90-1.93 (m, 1H), 2.14-2.36 (m, 4H), 2.65-2.72 (m,
1H), 2.90-2.96 (m, 2H), 3.00-3.12 (m, 3H), 3.17-3.22 (m, 1H),
3.29 (bs, 1H), 3.57-3.60 (m, 1H), 3.69-3.73 (m, 2H), 3.81 (s,
3H), 3.88-3.92 (d, 1H, J ) 13 Hz), 3.99-4.01 (m, 1H), 6.65-6.67
(d, 1H, J ) 8 Hz), 6.70-6.72 (d, 1H, J ) 8 Hz), 6.88-6.95 (m,
3H), 7.07-7.11 (t, 1H, J ) 8 Hz), 7.23-7.31 (m, 2H).
(()-trans-1-[4-(2,3-Dichlorophenyl)piperazin-1-yl]-2-(6-meth-
oxy-3,4,4a,5,10,10a-hexahydro-2H-benzo[g]quinolin-1-yl)etha-
none (41b). The hydrochloride salt of amine 40 (100 mg, 0.386
mmol), chloride 13b (118 mg, 0.386 mmol), and K2CO3 (106 mg,
0.772 mmol) in CH3CN were reacted following procedure B to
yield 41b, 170 mg (90%), as an oil, which was used without further
(()-trans-4-(2-(4-Phenylpiperazin-1-yl)ethyl)-1,2,3,4,4a,5,6,
10b-octahydrobenzo[f]quinolin-7-ol (8). Compound 38 (335 mg,
8.27 mmol) was demethylated following procedure D. The crude
solid was purified by recrystallization (EtOAc/MeOH) to give (()-
1
purification. H NMR (400 MHz, CDCl3) δ 1.61-1.68 (m, 3H),
1.91-1.94 (m, 1H), 2.13-2.35 (m, 4H), 2.66-2.73 (m, 1H),
2.91-2.97 (m, 2H), 3.01-3.13 (m, 3H), 3.17-3.22 (m, 1H), 3.29
(bs, 1H), 3.57-3.60 (m, 1H), 3.69-3.73 (m, 2H), 3.81 (s, 3H),
3.88-3.92 (d, 1H, J ) 13 Hz), 3.99-4.01 (m, 1H), 6.58-6.59 (d,
1H, J ) 8 Hz), 6.63-6.65 (d, 1H, J ) 8 Hz), 6.85-6.84 (d, 1H,
J ) 8 Hz), 7.00-7.04 (t, 1H, J ) 8 Hz), 7.07-7.13 (m, 2H).
(()-trans-6-Methoxy-1-[2-(4-phenylpiperazin-1-yl)ethyl]-1,2,
3,4,4a,5,10,10a-octahydrobenzo[g]quinoline (42a). Amide 41a
(500 mg, 1.38 mmol) was reduced with lithium aluminum hydride
(280 mg, 7.0 mmol) following procedure C to give 52, 480 mg
(99%) of 42a, which was used without further purification. 1H NMR
(400 MHz, CDCl3) δ 1.60-1.72 (m, 4H), 2.21-2.27 (m, 2H),
1
8, 264 mg (81%). H NMR (400 MHz, CDCl3) δ 1.72-1.70 (m,
2H), 2.11-2.20 (m, 1H), 2.34-2.51 (m, 3H), 2.62-2.79 (m, 8H),
2.96-3.10 (m, 3H), 3.22-3.24 (t, 4H, J ) 5.2 Hz), 6.70-6.72 (d,
1H, J ) 8 Hz), 6.81-6.85 (t, 1H, J ) 8 Hz), 6.94-6.96 (m, 3H),
7.15-7.19 (t, 1H, J ) 8 Hz), 7.26-7.30 (t, 2H, J ) 8 Hz). The
free base was converted to its HCl salt, mp ) 165-169 °C. Anal.
(C25H33N3O·2HCl·0.5H2O) C, H, N.
(+)-(4aR,10aR)-trans-4-(2-(4-Phenylpiperazin-1-yl)ethyl)-1,
2,3,4,4a,5,6,10b-octa-hydrobenzo[f]quinolin-7-ol ((+)-8). Com-
pound (+)-38 (250 mg, 0.616 mmol) was demethylated following
procedure D. The crude solid was purified by recrystallization

7818 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 24
Brown et al.
1.90-1.94 (m, 1H), 2.59-2.74 (m, 8H), 2.33-2.39 (m, 1H),
2.91-3.09 (m, 3H), 3.22-3.27 (m, 5H), 3.81 (s, 3H), 6.65-6.67
(d, 1H, J ) 8 Hz), 6.70-6.72 (d, 1H, J ) 8 Hz), 6.88-6.95 (m,
3H), 7.07-7.11 (t, 1H, J ) 8 Hz), 7.23-7.31 (m, 2H).
function until a root-mean-square (rms) deviation of 0.001 kcal/
(mol Å) was achieved. Ring conformations of bi- and tricyclic
aminotetralin derivatives were generated using simulated annealing
protocol with default settings (no. of cycles, 10; heating, 700 K
for 1000 fs; annealing, 50 K for 1000 fs; annealing function,
exponential) in Sybyl 7.1. The minimum energy conformations for
each of the bi- and tricyclic ring systems were further minimized
in Sybyl using the settings mentioned above. Further, rotatable
bonds in all molecules were searched from 0 to 359° in 10°
increments using systematic search protocol in Sybyl. The minimum
energy conformations were minimized further as described above.
The molecules were imported in TriposMol2 (.mol2) format in
MOE 2007.09 and stored in a molecular database. This database
was used as an input in the Pharmacophore Elucidation functionality
in MOE. The structures of molecules used for generating pharma-
cophore queries are shown Figures 1 and 3.
The objective of Pharmacophore Elucidation is to generate all
popular pharmacophore queries (with coverage n, typically 90% of
all active molecules) considering all possible discrete geometries with
all possible combinations of input query expressions. The Pharma-
cophore Elucidator operates on 3D conformations of the molecules
present in the input database. In the present study the Conformations
option in Pharmacophore Elucidation panel was set to bond rotation
wherein the rotatable bonds of each molecule were set systematically
to specific torsion angles from a collection of rules. The ring
conformations were not searched. The default annotation scheme CHD
was used for determining the pharmacophore features of each molecule.
Pharmacophore Elucidator generates pharmacophores compatible with
the specified scheme. Other parameter values in the Pharmacophore
Elucidation panel were kept at their default values. The alignment
options (emphasize aromatic atoms and emphasize donor and acceptor
atoms) that influence the overlap scoring were enabled. The results of
the pharmacophore generation were written to an output database,
which was analyzed further to select appropriate pharmacophore
hypothesis.
The set of compounds used for pharmacophore generation consisted
of conformationally constrained (bi- and tricyclic aminotetralin deriva-
tives) structures shown in Figure 3 with their dopamine D2 and D3
receptor binding affinities listed in Table 1. To determine the
appropriate annotation scheme, several runs with other annotation
schemes given in Pharmacophore Elucidation functionality such as
PCH, CHD, PCHD, PPCH, PCHD, Unified, etc. were tried, keeping
all other parameters constant (data not shown). For these initial trial
runs, compounds R-(+)-7-OH-DPAT, S-(-)-5-OH-DPAT, and R-(+)-
5, known D2/D3 agonists, were used. It is well-known from earlier
dopamine pharmacophores that an aromatic H-bond donor and cationic
nitrogen (reinforced H-bonding) are necessary to interact with dopam-
ine receptor serine and aspartate residues, respectively.15 To include
these highly directional pharmacophore requirements such as reinforced
H-bonding, the annotation scheme CHD was selected. This scheme
avoids use of atomic H-bond/acceptor features but includes the
directional character of the atomic H-bond donor/acceptor features.
This is particularly useful for DA ligands, since these possess varying
nature of the H-bond donor, which interacts with serine. Also, enabling
the alignment options emphasize aromatic atoms and emphasize donor
and acceptor atoms led to meaningful pharmacophore hypotheses. The
final pharmacophores were generated using the CHD annotation
scheme with above-mentioned alignment options. For the generation
of pharmacophore hypotheses based on hybrid analogues, compounds
(+)-6, (-)-7, (-)-8 (4aS, 10bS), and (-)-11 (S-enantiomer) were used.
Selection of the appropriate hypotheses was based on overall alignment
score and the associated pharmacophore features.
(()-trans-1-a{2-[4-(2,3-Dichlorophenyl)piperazin-1-yl]ethyl}-
6-methoxy-1,2,3,4,4a-5,10,10a-octahydrobenzo[g]quinoline (42b).
Amide 41b (170 mg, 0.349 mmol) was reduced with lithium aluminum
hydride (55 mg, 1.43 mmol) following procedure C to give 42b, 152
mg (92%) of the titled product, which was used without further
purification. 1H NMR (400 MHz, CDCl3) δ 1.59-1.66 (m, 3H),
1.90-1.93 (m, 1H), 2.11-2.33 (m, 4H), 2.64-2.71 (m, 1H), 2.90-2.96
(m, 2H), 3.00-3.12 (m, 3H), 3.12-3.29 (m, 8H), 3.81 (s, 3H),
6.58-6.59 (d, 1H, J ) 8 Hz), 6.63-6.65 (d, 1H, J ) 8 Hz), 6.85-6.84
(d, 1H, J ) 8 Hz), 7.00-7.04 (t, 1H, J ) 8 Hz), 7.07-7.13 (m, 2H).
(()-trans-1-[2-(4-Phenylpiperazin-1-yl)ethyl]-1,2,3,4,4a,5,10,
10a-octahydrobenzo[g]quinolin-6-ol (43a). Compound 42a (470
mg, 1.16 mmol) was demethylated following procedure D. The
crude solid was purified by column chromatography (1:10 MeOH/
EtOAc) to give 43a, 400 mg (88%). 1H NMR (400 MHz, CDCl3)
δ 1.13-1.15 (m, 2H), 1.90-1.92 (m, 1H), 2.15-2.27 (m, 3H),
2.33-2.39 (m, 1H), 2.60-2.75 (m, 9H), 2.85-2.91 (m, 1H),
3.03-3.11 (m, 2H), 3.17-3.23 (m, 5H), 6.53-6.55 (d, 1H, J )
7.6 Hz), 6.64-6.66 (d, 1H, J ) 8 Hz), 7.24-7.28 (t, 2H, J ) 8
Hz), 6.83-6.87 (t, 1H, J ) 8 Hz), 6.91-6.99 (m, 3H). The free
base was then converted to its HCl salt, mp ) 174-179 °C. Anal.
(C25H33N3O·3HCl) C, H, N.
(()-trans-1-{2-[4-(2,3-Dichlorophenyl)piperazin-1-yl]ethyl}-
1,2,3,4,4a,5,10,10a-octahydrobenzo[g]quinolin-6-ol (43b). Com-
pound 42b (152 mg, 0.321 mmol) was demethylated following
procedure D. The crude solid was purified by column chromatog-
raphy (1:10 MeOH/EtOAc) to give 43b, 102 mg (69%). 1H NMR
(400 MHz, CDCl3) δ 1.12-1.25 (m, 2H), 1.82-1.91 (m, 1H),
2.21-2.42 (m, 3H), 2.51-2.63 (m, 9H), 2.78-2.81 (m, 1H),
2.98-3.12 (m, 8H), 6.53-6.54 (d, 1H, J ) 8 Hz), 6.58-6.60 (d,
1H, J ) 8 Hz), 6.80-6.79 (d, 1H, J ) 8 Hz), 6.95-7.00 (t, 1H, J
) 8 Hz), 7.02-7.09 (m, 2H). The free base was then converted to
its HCl salt, mp ) 161-165 °C. Anal. (C25H31N3OCl2 ·2HCl) C,
H, N.
(()-trans-1-Propyl-1,2,3,4,4a,5,10,10a-octahydrobenzo[g]quin-
olin-6-ol (2). To a solution of compound (()-40 (0.217 g, 1 mmol)
and TEA (0.5 mL) in dichloromethane (15 mL) at 0 °C was added
propionyl chloride (0.111 g, 1.2 mmol) dropwise. The reaction
mixture was allowed to stir for an additional 1 h. The reaction
mixture was washed with saturated NaHCO3 solution, followed by
water. The organic layer was dried (Na2SO4) and the solvent
removed in vacuo to yield a thick oil (0.2 g), which was subjected
to reduction with excess LiAlH4 following procedure C. Demethy-
lation of the tertiary amine thus obtained was carried out in refluxing
aqueous HBr (48%) for 3 h. The acid was removed in vacuo. The
residue obtained was converted to free base using aqueous Na2CO3.
The free base was purified using column chromatography (dichlo-
romethane/MeOH 8:2) to give 0.1 g of 2. The purified free base
1
was converted to the HCl salt. H NMR (400 MHz, CD3OD): δ
1.0-1.04 (t, 3H), 1.45-1.51 (m, 2H), 1.67-1.86 (m, 5H),
1.91-1.94 (m, 1H), 2.10-2.16 (m, 1H), 2.74-2.78 (m, 2H),
2.89-3.04 (m, 5H), 6.52-6.55 (m, 2H), 6.85-6.89 (m, 1H). The
free base was converted to HCl salt, mp ) 165-167 °C. Anal.
(C16H25ClNO·HCl·0.3C2H5OC2H5) C, H, N.
Molecular Modeling. All molecular modeling studies reported
herein were performed on a Hewlett-Packard xw4300 computer
workstation with main memory of 2 GB and Intel Pentium 4 CPU
of 3.4 GHz running under the operating system Linux Red Hat 3.
The molecular modeling software packages Sybyl 7.132 and
Molecular Operating Environment (MOE) 2007.0933 were em-
ployed for the present work.
The structures used in the present study were constructed either
from X-ray crystal structure or using fragments in Sybyl’s fragment
library. All the structures were generated in their N-protonated
forms. Partial atomic charges were calculated using Gasteiger-Hu¨ckel
method in Sybyl 7.1. Each structure was fully geometry-optimized
using the Tripos force field34 with a distance-dependent dielectric
Measurement of Affinity in Inhibiting [3H]Spiperone
Binding to Dopamine D2 and D3 Receptors. Binding affinities
were assessed according to the general procedure described in our
previous study.22 Briefly, membranes from HEK 293 cells express-
ing rat D2L and D3 receptors were incubated with each test
compound and [3H]spiperone (0.4 nM, 15 Ci/mmole, Perkin-Elmer)
for 1 h at 30 °C in 50 mM Tris-HCl (pH 7.4), with 0.9% NaCl,
and 0.025% ascorbic acid in the absence of GTP. (+)-Butaclamol
(2 M) was used to define nonspecific binding. Assays were

Structurally Constrained Hybrid DeriVatiVes
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 24 7819
octahydrobenzo[f]quinolines: structural and stereochemical consider-
ations for centrally acting pre- and postsynaptic dopamine-receptor
agonists. J. Med. Chem. 1985, 28, 215–225.
terminated by addition of ice-cold buffer and filtration in the MACH
3-96 Tomtec harvester (Wallac, Gaithersburg, MD). IC50 values
were estimated by nonlinear regression analysis with the logistic
model in the least-squares fitting program ORIGIN and converted
to inhibition constants (Ki) by the Cheng-Prusoff equation.35 In
this conversion, the Kd values for [3H]spiperone binding were 0.057
nM for D2 receptors and 0.125 nM for D3 receptors.
(17) (a) Wikstro¨m, H.; Andersson, B.; Elebring, T.; Svensson, K.; Carlsson,
A.; Largent, B. N-Substituted 1,2,3,4,4a,5,6,10b-octahydrobenzo[f]quin-
olines and 3-phenylpiperidines: effects on central dopamine and σ
receptors. J. Med. Chem. 1987, 30, 2169–2174. (b) Witkin, J. M.;
Levant, B.; Zapata, A.; Kaminski, R.; Gasior, M. The dopamine D3/
D2 agonist (+)-PD-128,907 [(R-(+)-trans-3,4a,10b-tetrahydro-4-pro-
pyl-2H,5H-1benzopyrano-[4,3-b]-1,4-oxazin-9-ol)] protects against
acute and cocaine-kindled seizures in mice: further evidence for the
involvement of D3 receptors. J. Pharmacol. Exper. Ther. 2008, 326,
930–938.
(18) Cannon, J. G.; Hamer, R. L.; Ilhan, M.; Bhatnagar, R. K.; Long, J. P.
Conformationally restricted congeners of dopamine derived from
octahydrobenzo[g]quinoline and octahydrobenzo[f]quinoline. J. Med.
Chem. 1984, 27, 190–195.
Acknowledgment. This work is supported by National
Institute of Neurological Disorders and Stroke/National Institutes
of Health (Grant NS047198, A.K.D.). We thank Drs. Swati Biswas
and Fernando Fernandez for synthesizing compounds (+)-6 and
9. We are grateful to Dr. K. Neve, Oregon Health and Science
University, Portland, OR, for D2L and D3 expressing HEK cells.
(19) Wikstro¨m, H.; Sanchez, D.; Lindberg, P.; Arvidsson, L.-E.; Hacksell,
U.; Johansson, A.; Nilsson, J. L. G.; Hjorth, S.; Carlsson, A.
Monophenolic octahydro-benzo[f]quinolines: central dopamine- and
serotonin-receptor stimulating activity. J. Med. Chem. 1982, 25, 925–
931.
(20) Dutta, A. K.; Fei, X.-S.; Reith, M. E. A. A novel series of hybrid
compounds derived by combining 2-aminotetralin and piperazine
fragments: binding activity at D2 and D3 receptors. Bioorg. Med. Chem.
Lett. 2002, 12, 619–622.
(21) Dutta, A. K.; Venkataraman, S. K.; Fei, X.-S.; Kolhatkar, R.; Zhang,
S.; Reith, M. E. A. Synthesis and biological characterization of novel
hybrid 7-{[2-(4-phenylpiperazin-1-yl)ethyl]-propylamino}-5,6,7,8-tet-
rahydronaphthalen-2-ol and their heterocyclic bioisosteric analogues
for dopamine D2 and D3 receptors. Bioorg. Med. Chem. 2004, 12,
4361–4373.
(22) Biswas, S.; Zhang, S.; Fernandez, F.; Ghosh, B.; Zhen, J.; Kuzhikan-
dathil, E.; Reith, M. E. A.; Dutta, A. K. Further structure-activity
relationships study of hybrid 7-{[2-(4-phenylpiperazin-1-yl)ethyl]pro-
pylamino}-5,6,7,8-tetrahydronaphthalen-2-ol analogues: identification
of a high-affinity D3-preferring agonist with potent in vivo activity
with long duration of action. J. Med. Chem. 2008, 51, 101–117.
(23) Biswas, S.; Hazeldine, S.; Ghosh, B.; Parrington, I.; Kuzhikandathil,
E.; Reith, M. E.; Dutta, A. K. Bioisosteric heterocyclic versions of
7-{[2-(4-phenylpiperazin-1-yl)ethyl]propylamino}-5,6,7,8-tetrahy-
dronaphthalen-2-ol: identification of highly potent and selective
agonists for dopamine D3 receptor with potent in vivo activity. J. Med.
Chem. 2008, 51, 3005–3019.
(24) Beres, J. A.; Cannon, J. G. Regiospecific carboxylation of certain
methoxylated 2-tetralones. Synth. Commun. 1979, 9, 819–824.
(25) Khatib, S.; Tal, S.; Godsi, O.; Peskin, U.; Eichen, Y. Site selective
processes: a combined theoretical and experimental investigation of
thermally activated tautomerization processes in 2-(2,4-dinitroben-
zyl)pyridine derivatives. Tetrahedron 2000, 56, 6753–6761.
(26) Walsh, D. A.; Smissman, E. E. Observable magnetic nonequivalence
of diastereotopic protons as a stereochemical probe. J. Org. Chem.
1974, 39, 3705–3708.
(27) Xia, Y.; Reddy, E. R.; Kozikowski, A. P. Synthesis of the benzenoid
analogue of the Chinese nootropic agent huperzine a. Tetrahedron
Lett. 1989, 30, 3291–3294.
(28) Cannon, J.; Chang, Y.-a.; Amoo, V.; Walker, K. Stereospecific route
to trans-1,2,3,4,4a-5,6,10b-octahydrobenzo[f]quinolines. Synthesis
1986, 494–496.
Supporting Information Available: Elemental analysis data for
all final targets and additional synthesis information and analytical
data. This material is available free of charge via the Internet at
http://pubs.acs.org.
References
(1) Emilien, G.; Maloteaux, J.-M.; Geurts, M.; Hoogenberg, K.; Cragg,
S. Dopamine receptors. Physiological understanding to therapeutic
intervention potential. Pharmacol. Ther. 1999, 84, 133–156.
(2) Missale, C.; Nash, S. R.; Robinson, S. W.; Jaber, M.; Caron, M. G.
Dopamine receptors: from structure to function. Physiol. ReV. 1998,
78, 189–225.
(3) D’Ischia, M.; Prota, G. Biosynthesis, structure, and function of
neuromelanin and its relation to Parkinson’s disease: a critical update.
Pigm. Cell Res. 1997, 10, 370–376.
(4) Gurevich, E. V.; Joyce, J. N. Distribution of dopamine D3 receptor
expressing neurons in the human forebrain: comparison with D2 receptor
expressing neurons. Neuropsychopharmacology 1999, 20, 60–80.
(5) Boeckler, F.; Gmeiner, P. The structural evolution of dopamine D3
receptor ligands: structure-activity relationships and selected neu-
ropharmacological aspects. Pharmacol. Ther. 2006, 112, 281–333.
(6) McDermed, J. D.; McKenzie, G. M.; Freeman, H. S. Synthesis and
dopaminergic activity of (()-, (+)-, and (-)-2-dipropylamino-5-hydroxy-
1,2,3,4-tetrahydronaphthalene. J. Med. Chem. 1976, 19, 547–549.
(7) Tedesco, J. L.; Seeman, P.; McDermed, J. D. The conformation of
dopamine at its receptor: binding of monohydroxy-2-aminotetralin
enantiomers and positional isomers. Mol. Pharmacol. 1979, 16, 369–
381.
(8) Cannon, J. G.; Lee, T.; Goldman, H. D.; Costall, B.; Naylor, R. J.
Cerebral dopamine agonist properties of some 2-aminotetralin deriva-
tives after peripheral and intracerebral administration. J. Med. Chem.
1977, 20, 1111–1116.
(9) McDermed, J. D.; McKenzie, G. M.; Phillips, A. P. Synthesis and
pharmacology of some 2-aminotetralins. Dopamine receptor agonists.
J. Med. Chem. 1975, 18, 362–367.
(10) Hacksell, U.; Svensson, U.; Nilsson, J. L. G.; Hjorth, S.; Carlsson,
A.; Wikstro¨m, H.; Lindberg, P.; Sanchez, D. N-Alkylated 2-aminote-
tralins: central dopamine-receptor stimulating activity. J. Med. Chem.
1979, 22, 1469–1475.
(11) Damsma, G.; Bottema, T.; Westerink, B. H.; Tepper, P. G.; Dijkstra,
D.; Pugsley, T. A.; MacKenzie, R. G.; Heffner, T. G.; Wikstro¨m, H.
Pharmacological aspects of R-(+)-7-OH-DPAT, a putative dopamine
D3 receptor ligand. Eur. J. Pharmacol. 1993, 249, R9–R10.
(12) Cox, B. A.; Henningsen, R. A.; Spanoyannis, A.; Neve, R. L.; Neve,
K. A. Contributions of conserved serine residues to the interactions
of ligands with dopamine D2 receptors. J. Neurochem. 1992, 59, 627–
635.
(29) Mellin, C.; Hacksell, U. Stereoselective synthesis of methoxy substi-
tuted 1,2,3,4,4a-5,10,10a-octahydrobenzo[g]quinolines. Tetrahedron
1987, 43, 5443–5450.
(30) ten Hoeve, W.; Wynberg, H. The design of resolving agents. Chiral
cyclic phosphoric acids. J. Org. Chem. 1985, 50, 4508–4514.
(31) Varady, J.; Wu, X.; Fang, X.; Min, J.; Hu, Z.; Levant, B.; Wang, S.
Molecular modeling of the three-dimensional structure of dopamine
3 (D3) subtype receptor: discovery of novel and potent D3 ligands
through a hybrid pharmacophore- and structure-based database search-
ing approach. J. Med. Chem. 2003, 46, 4377–4392.
(32) SYBYL Molecular Modeling System, version 7.1; Tripos Inc., St. Louis,
MO 63144-2913.
(33) Molecular Operating Environment (MOE) 2007.09 is available from
Chemical Computing Group, Inc., 1010 Sherbrooke Street W, Suite
910, Montreal, Quebec H3A 2R7, Canada.
(13) Sartania, N.; Strange, P. G. Role of conserved serine residues in the
interaction of agonists with D3 dopamine receptors. J. Neurochem.
1999, 72, 2621–2624.
(14) Lundstrom, K.; Turpin, M. P.; Large, C.; Robertson, G.; Thomas, P.;
Lewell, X. Q. Mapping of dopamine D3 receptor binding site by
pharmacological characterization of mutants expressed in CHO cells
with the Semliki Forest virus system. J. Recept. Signal Transduction
Res. 1998, 18, 133–150.
(15) Seiler, M. P.; Markstein, R.; Walkinshaw, M. D.; Boelsterli, J. J.
Characterization of dopamine receptor subtypes by comparative
structure-activity relationships: dopaminomimetic activities and solid
state conformation of monohydroxy-1,2,3,4,4a,5,10,10a-octahydro-
benz[g]quinolines and its implications for a rotamer-based dopamine
receptor model. Mol. Pharmacol. 1989, 35, 643–651.
(16) Wikstro¨m, H.; Andersson, B.; Sanchez, D.; Lindberg, P.; Arvidsson,
L. E.; Johansson, A. M.; Nilsson, J. L.; Svensson, K.; Hjorth, S.;
Carlsson,A.Resolvedmonophenolic2-aminotetralinsand1,2,3,4,4a,5,6,10b-
(34) Clark, M.; Cramer, R. D., III; Van Opdenbosh, N. Validation of the
general purpose Tripos 5.2 force field. J. Comput. Chem. 1989, 10,
982–1012.
(35) Cheng, Y.; Prusoff, W. H. Relationship between the inhibition constant
(Ki) and the concentration of inhibitor which causes 50% inhibition
(I50) of an enzymatic reaction. Biochem. Pharmacol. 1973, 22,
3099–3108.
JM8008629