1,2,4-Triazol-3-yl-thiopropyl-tetrahydrobenzazepines: A series of potent and selective dopamine D3 receptor antagonists

Article abstract of DOI:10.1021/jm0705612

The discovery of new highly potent and selective dopamine D₃ receptor antagonists has recently permitted characterization of the role of the dopamine D₃ receptor in a wide range of preclinical animal models. A novel series of 1,2,4-triazol-3-yl-thiopropyl-tetrahydrobenzazepines demonstrating a high level of D₃ affinity and selectivity with an excellent pharmacokinetic profile is reported here. In particular, the pyrazolyl derivative 35 showed good oral bioavailability and brain penetration associated with high potency and selectivity in vitro. In vivo characterization of 35 confirmed that this compound blocks the expression of nicotine- and cocaine-conditioned place preference in the rat, prevents nicotine-triggered reinstatement of nicotine-seeking behavior in the rat, reduces oral operant alcohol self-administration in the mouse, increases extracellular levels of acetylcholine in the rat medial prefrontal cortex, and potentiates the amplitude of the relative cerebral blood volume response to d-amphetamine in a regionally specific manner in the rat brain.

Full text of DOI:10.1021/jm0705612

5
076
J. Med. Chem. 2007, 50, 5076-5089
1,2,4-Triazol-3-yl-thiopropyl-tetrahydrobenzazepines: A Series of Potent and Selective
Dopamine D Receptor Antagonists
3
,
§
§
‡
§
#
§
Fabrizio Micheli,* Giorgio Bonanomi, Frank E. Blaney, Simone Braggio, Anna Maria Capelli, Anna Checchia,
#
X
§
§
§
†
Ornella Curcuruto, Federica Damiani, Romano Di Fabio, Daniele Donati, Gabriella Gentile, Andy Gribble,
§
#
§
§
†
∇
∇
§
Dieter Hamprecht, Giovanna Tedesco, Silvia Terreni, Luca Tarsi, Andrew Lightfoot, Geoff Stemp, Gregor MacDonald,
∇
⊥
§
#
|
§
‡
Alex Smith, Michela Pecoraro, Marcella Petrone, Ornella Perini, Jacqui Piner, Tino Rossi, Angela Worby, Maria Pilla,
Enzo Valerio, Cristiana Griffante, Manolo Mugnaini, Martyn Wood, Claire Scott, Michela Andreoli, Laurent Lacroix,
Adam Schwarz, Alessandro Gozzi, Angelo Bifone, Charles R. Ashby, Jr., Jim J. Hagan, and Christian Heidbreder*
§
§
§
†
†
§
†
§
§
§
@
†
,§
Psychiatry Centre of Excellence, Molecular DiscoVery Research, and Safety Assessment, GlaxoSmithKline Medicine Research Centre,
Via Fleming 4, 37135 Verona, Italy, Psychiatry Centre of Excellence, Neurology Centre of Excellence, Molecular DiscoVery Research, and
Safety Assessment, GlaxoSmithKline Medicine Research Centre, NFSP, Harlow, UK, and Department of Pharmaceutical Sciences, St. John’s
UniVersity, Jamaica, New York 11439
ReceiVed May 14, 2007
The discovery of new highly potent and selective dopamine D
characterization of the role of the dopamine D receptor in a wide range of preclinical animal models. A
novel series of 1,2,4-triazol-3-yl-thiopropyl-tetrahydrobenzazepines demonstrating a high level of D affinity
and selectivity with an excellent pharmacokinetic profile is reported here. In particular, the pyrazolyl derivative
5 showed good oral bioavailability and brain penetration associated with high potency and selectivity in
3
receptor antagonists has recently permitted
3
3
3
vitro. In vivo characterization of 35 confirmed that this compound blocks the expression of nicotine- and
cocaine-conditioned place preference in the rat, prevents nicotine-triggered reinstatement of nicotine-seeking
behavior in the rat, reduces oral operant alcohol self-administration in the mouse, increases extracellular
levels of acetylcholine in the rat medial prefrontal cortex, and potentiates the amplitude of the relative
cerebral blood volume response to d-amphetamine in a regionally specific manner in the rat brain.
Introduction
cocaine-, nicotine-, alcohol-, and heroin-seeking behaviors.
Importantly, in contrast with dopamine D2 receptor antagonists,
selective antagonism at dopamine D3 receptors does not elicit
catalepsy, does not affect spontaneous or stimulant-induced
locomotion, does not increase serum prolactin levels, and does
Following the isolation and characterization of the cDNA for
the dopamine D3 receptor, subsequent studies indicated that
1
D3 receptors, as well as D3 receptor mRNA, are primarily
localized in limbic regions of the rat2 and human
,3
4-6
brain.
7
not increase dopamine levels in the neostriatum.
This finding led to the postulate that D3 receptors may be
involved in the pathophysiology of drug addiction and schizo-
phrenia. Thus, the high binding affinity of the dopamine D3
receptor to endogenous dopamine, its high expression in the
mesolimbic system, and its up-regulation in the ventral striatum
of schizophrenic patients that are off antipsychotics, as well as
its up-regulation in the ventral striatum of cocaine overdose
victims, and in rodents after cocaine self-administration or
behavioral sensitization to cocaine- or nicotine-associated cues
support the hypothesis that the dopamine D3 receptor is an
attractive new target for the pharmacotherapeutic management
of drug addiction and schizophrenia (for reviews see refs 7-11).
Recent studies have also shown that selective dopamine D3
receptor antagonists are efficacious in animal models of
The main medicinal chemistry work reported in this manu-
script is related to the modification of the cyclohexyl ethyl linker
of previously reported GSK amide molecules with a thio-
triazolyl moiety and with a major exploration of the benzazepine
scaffold. We herein describe the SAR analysis of a range of
9
1,2,4-triazol-3-yl-thiopropyl-tetrahydrobenzazepines, which has
provided highly potent and selective dopamine D3 receptor
antagonists.
Chemistry
During the past decade, part of GSK research in CNS drug
discovery was aimed at the discovery of novel chemical entities
able to selectively modulate the dopamine D3 receptor. The
successful result of this work led to the well characterized
trans-N-[4-[2-(6-cyano-1,2,3,4-tetrahydroisoquinolin-2yl)ethyl]-
*
Author to whom correspondence should be addressed. E-mail:
12
cyclohexyl]-4-quinolininecarboxamide (SB-277011, 1) and
Fabrizio.E.Micheli@gsk.com; Christian.A.Heidbreder@gsk.com.
§
trans-3-(2-(4-((3-(3-(5-methyl-1,2,4-oxadiazolyl))phenyl)car-
Psychiatry Centre of Excellence, GlaxoSmithKline Medicine Research
Centre, Verona, Italy.
boxamido)cyclohexyl)ethyl)-7-methylsulfonyl-2,3,4,5-tetrahy-
dro-1H-3-benzazepine (SB-414796, 2) (Figure 1).
While these molecules proved to be very useful for target
validation, part of their overall developability profile, including
#
Molecular Discovery Research, GlaxoSmithKline Medicine Research
13
Centre, Verona, Italy.
⊥
Safety Assessment, GlaxoSmithKline Medicine Research Centre,
Verona, Italy.
∇
Neurology Centre of Excellence, GlaxoSmithKline Medicine Research
a
Center, Harlow, UK.
Abbreviations: CPP, conditioned place preference; Ach, acetylcholine;
†
+
Psychiatry Centre of Excellence, GlaxoSmithKline Medicine Research
hERG, human ether-a go-go K channel; NCE, novel chemical entity; PK,
Centre, Harlow, UK.
pharmacokinetic; P450, cytochrome P450; hCli, human intrinsic clearance;
MW, molecular weight; clogD, calculated logD; PSA, polar surface area;
F%, bioavailability; B/B, brain/blood; Clb, blood clearance; Vd, distribution
volume; SDM, site-directed mutagenesis; rCBV, relative cerebral blood
volume; FLIPR, fluorescent imaging plate reader; GPCR, G-protein coupled
receptor; TM, transmembrane; SPA, scintillation proximity assay; ECG,
electrocardiogram.
‡
Molecular Discovery Research, GlaxoSmithKline Medicine Research
Centre, Harlow, UK.
|
Safety Assessment, GlaxoSmithKline Medicine Research Centre, Har-
low, UK.
@
St. John’s University.
Current address: European Patent Office, Munich, Germany.
X
1
0.1021/jm0705612 CCC: $37.00 © 2007 American Chemical Society
Published on Web 09/15/2007

Dopamine D
3
Receptor Antagonists
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 21 5077
consisted of the design of derivatives endowed with reduced
clogD and lower MW (4-7). While no major effect was
observed in terms of potency at the dopamine D3 receptor, much
higher hERG values were observed for compounds 4-6. In
contrast, derivative 7 showed a good in vitro profile, but its
hCli was relatively high (6.4 mL/min/g); therefore, no further
activity was planned on this compound.
In order to address the hERG liability associated with
derivatives 4-6, the isoxazolyl moiety was chosen to replace
the oxazolyl moiety to assess the role of the relative position
of the heteroatoms, and derivative 8 was prepared. Also in this
case (Table 2) a good in vitro profile was achieved in terms of
dopamine D3 vs D2 potency. While B/B remained good (3.1),
compound 8 showed the great advantage of a reduced distribu-
tion volume (3.4 L/kg); this profile was also associated with a
lower clogD (5.2). Rat F% was similar to 3 (24% vs 27%), and
a reduced Clb was also observed (38 vs 59 mL/min/kg).
Accordingly, compound 8 was tested in vivo and produced a
significant decrease in a rat model of nicotine-triggered relapse
to nicotine-seeking behavior. However, the hERG pIC50 (pIC50
Figure 1. Structures of the previously reported GSK selective
dopamine D receptor antagonists.
3
human ether-a go-go K+ channel (hERG ) affinity and the
potentially related modification of the QTc51 profile, was
suboptimal for further development. Considerable efforts were,
therefore, devoted to the identification of NCEs endowed with
drug-like properties. Our main focus was directed toward the
modification of our lead molecule 2, removing the cyclohexyl
ethyl linker, the sulfone and the amide moiety, which were
believed to be the major cause of the poor developability profile
of compound 2. To that purpose, among the various possibilities
offered by the general knowledge in the field (see ref 9 and
cited references), we selected the thiotriazole group as a
candidate to be introduced in the molecule, while different
linkers were chosen with a focus on linear ones. Compounds
of the series (Tables 1 to 3) were prepared according to the
general Scheme 1 reported below. Additional information can
also be found in refs 14-18, while detailed experimental parts
for all of the intermediates is available as Supporting Informa-
tion.
a
)
6.3) still exceeded the desired selectivity profile. Functional
activity at the hERG channel was also confirmed electrophysi-
2
2
ologically with an in vitro EC50 of 0.12 µM (pIC50 ) 6.9), a
value which produced a clear QTc prolongation in vivo when
assessed in the anesthetized guinea pig.
2
3
Despite the good potency and selectivity achieved, all the
compounds reported in Table 2 that exceeded hERG dofetilide
binding values of 6.5 were not progressed further along the
screening cascade. The compounds with values lower than 6.5
were progressed and either failed in the developability screen
or showed a suboptimal PK profile in rats. The results from
the SAR exploration showed quite clearly (e.g., 20-22) a
general trend whereby increases in dopamine D3 affinity were
often paralleled by increases in potency at the hERG ion
channel, thus suggesting common interacting features for the
thiotriazole benzazepine template both at the dopamine D3
receptor and at the hERG ion channel. Receptor modeling
studies were then performed to identify a rational way forward
in the exploration of this scaffold.
Compounds were assayed for their agonistic and antagonistic
properties using a functional GTPγS assay expressing the human
dopamine D3 receptor. Key features of NCEs were as follows:
at least 100-fold selectivity vs dopamine D2 and histamine H1
receptors (functional assays), as well as a minimal requirement
of 100-fold selectivity vs the hERG ion channel (dofetilide
binding assay). Generic developability screens such as CYPEX
bactosome P450 inhibition and rat and human in vitro clearance
in liver microsomes were included early in the screening
cascade.
Molecular Modeling Studies
Over the last 15 years, receptor modeling and docking has
become well established in the field of GPCR drug design, and
more recently several computational techniques have also been
Results and Discussion
2
4
From the data reported in Table 1, it became evident that the
newly designed thiotriazole scaffolds showed promising proper-
ties from an in vitro perspective. Compound 3 showed an almost
nanomolar potency at the desired target and was endowed with
a greater than 100-fold selectivity over D2, H1, and hERG. The
in vitro pharmacokinetic developability screens of compound
developed to address hERG liability. In the work described
here a combination of docking experiments and ligand-based
in silico models were used to optimize the series’ properties
for both the D3 receptor and the hERG channel. The docking
studies in particular were able to establish a molecular level
understanding of the ligand-receptor/channel interactions.
Recent site-directed mutagenesis work described by Mitcheson
3
revealed a cytochrome P450 profile with no inhibition on
2
5
single isoforms below 10 µM and a moderate intrinsic clearance
in human liver microsomes (4.7 mL/min/g). From the in silico
point of view, compound 3 had a molecular weight of 525 Da,
a calculated logD19 of 5.9, and a polar surface area of 73 Aˆ 2.
Accordingly, 3 was tested in vivo21 in rats to assess its in vivo
PK profile. The results proved to be very promising, suggesting
that 3 might represent an ideal template to replace 2: the
molecule was actually endowed with a moderate bioavailability
in rat (27%) and had a high brain/ blood ratio (3.7). Nonetheless,
being the first in a series, compound 3 had a relatively high
blood clearance (Clb 59 mL/min/kg), and its distribution volume
was relatively high (7.9 L/kg). Considering the results of the in
vitro profiling, the first optimization step for derivative 3
et al. showed that the residues Tyr652 and Phe656 on the S6
helices are the most important ones for binding a variety of
ligands. They form two concentric rings of aromatic residues
which line the pore cavity immediately below the ion filter.
Thr623 and Val625 in the pore turn region are also important,
especially for sulfonamide ligands such as MK-499.25 There is
still some uncertainty as to whether hERG blockers bind in the
open or closed form of the channel, although most evidence
20
26
would favor the open state. Recent publications have described
2
5
27
docking studies in both closed and open state models
although the latter required the extensive inclusion of explicit
water. In the current study, models were constructed of both
the open and closed forms. Following some preliminary studies,

5
078 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 21
Micheli et al.
Table 1. Functional Activity at the Human Dopamine D3 Receptor and Selectivity for 5-Oxazolyl Derivatives
hH1-FLIPRc
entry
R
hD3-GTPγS fpKib
hD2-GTPγS fpKib
pKb
hERG pIC50
PSA, Å
cLogD
1
2
3
4
5
6
7
not applicable
not applicable
2-methyl -5-quinolinyl
2-thienyl
2-indolyl
5-indolyl
8.4
9.2
8.7
8.5
8.9
8.8
8.8
6.4
6.1
< 6.1
6.6
6.0
6.2
5.9
6.2
NT
6.3
6.4
5.7
5.7
5.6
6.1
6.5
7.3
7.4
6.3
69
105
73
60
76
3.8
2.9
5.9
5.2
5.1
5.5
4.2
6.2
6.0
76
65
N-methyl-2-pyrrolyl
a
b
SEM for D3 GTPγS, H1 FLIPR and hERG data sets is (0.1 and for the D2 GTPγS data is (0.2. fpKi ) functional pKi obtained from the GTPγS
c
functional assay. FLIPR ) fluorescent imaging plate reader.
Table 2. Functional Activity at the Human Dopamine D3 Receptor and Selectivity for 5-Isoxazolyl Derivatives
hD3-GTPγS
hD2-GTPγS
fpKib
hH1-FLIPRc
hERG
pIC50
b
entry
R
fpKi
pKb
PSA, Å
cLogD
8
9
2-methyl-5-quinolinyl
2-methyl-6-quinolinyl
2,3-dimethyl-6-quinoxalinyl
Ph
8.7
9.0
8.2
8.8
8.4
8.5
8.8
8.6
7.6
8.9
9.1
8.6
8.8
9.2
9.0
8.8
7.4
8.0
8.0
8.3
8.7
8.9
8.9
8.5
9.1
9.4
8.6
6.0
5.7
5.2
6.4
6.0
6.3
6.4
<5.8
6.0
6.6
<6.4
6.5
6.7
6.4
6.3
5.9
<5.9
6.8
6.2
6.8
6.6
6.9
5.9
7.1
8.1
6.7
5.9
6.2
NT
NT
7.0
NT
5.7
6.5
6.7
6.2
7.1
7.4
6.9
7.4
7.2
6.9
6.6
6.1
6.9
7.7
6.5
7.1
6.7
5.8
6.7
6.9
NT
6.60
6.3
NT
6.5
6.8
7.7
7.2
7.6
6.6
6.1
6.9
6.6
7.2
7.1
7.6
8.0
6.6
6.1
6.5
5.9
6.4
6.1
6.9
5.4
7.4
NT
7.4
6.8
73
73
86
60
84
60
60
73
73
60
65
60
60
60
60
60
60
60
60
60
60
60
69
69
69
69
82
5.2
5.2
4.9
4.9
4.5
5.8
5.3
3.8
3.6
4.5
3.5
5.0
5.7
5.6
6.3
5.1
2.7
3.7
3.5
3.9
4.2
5.2
2.8
4.6
5.0
5.0
4.6
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
4-cyano Ph
4-trifluoromethylphenyl
4-methylphenyl
2-pyridinyl
4-pyridinyl
2-thienyl
1-methyl-pyrrol-2-yl
2-chlorophenyl
3-chorophenyl
4-chlorophenyl
3,4-dichlorophenyl
2-fluorophenyl
Me
CF3
i-Pr
t-Bu
cyclopentyl
4-methyl-cyclohexyl
tetrahydropyran-4-yl
2-methoxyphenyl
3-methoxyphenyl
4-methoxyphenyl
6-methoxy-3-pyridinyl
a
b
SEM for D3 GTPγS, H1 FLIPR and hERG data sets is (0.1 and for the D2 GTPγS data is (0.2. fpKi ) functional pKi obtained from the GTPγS
c
functional assay. FLIPR ) fluorescent imaging plate reader.
however, most of the docking of the receptor antagonists was
performed in the closed form of the channel, as only in this
state could they interact with several of the key aromatic residues
simultaneously. In addition to these docking studies, other
parameters were monitored in the design phase, in particular
lipophilicity and basicity.
The ligands were docked manually in the hERG cavity
described above, with full conformational flexibility of both the
ligand and protein taken into consideration. Only low energy
states of the former were allowed while the protein side chain
Although the ligand could be docked in a linear conformation,
this was only possible immediately below the ion gate, where
very few interactions with the aromatic residues above were
found. An alternative pose with the ligand oriented along the
channel axis led to an unlikely result where one end of the ligand
extended below the bottom of the helical bundle. By contrast
most of the resulting poses had compound 10 in a “U-shaped”
orientation as seen in Figure 2A. This is undoubtedly due to an
induced fit of the ligand to the aromatic cavity. Thus, not only
does the compound exhibit good interactions with all four of
the Phe656 residues as well as two of the Tyr652s but the
intramolecular π-stacking of the ligand is also entropically
favored. From Figure 2A it can be seen that the protonated
rotamer angles were set within the limits defined by the Karplus
rotamer library.28 Multiple starting poses were set up which were
then fully minimized. The results for compound 10 are discussed
below.

Dopamine D
3
Receptor Antagonists
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 21 5079
Table 3. Functional Activity at the Human Dopamine D3 Receptor and Selectivity for 5-Pyrazolyl Derivatives
hD3-GTPγS
fpKib
hD2-GTPγS
fpKib
hH1-FLIPRc
pKb
hERG
pIC50
entry
R
R1
n
PSA, Å
cLogD
3
3
3
3
3
4
4
4
5
6
7
8
9
0
1
2
2-methyl-5-quinolinyl
3,4-difluorophenyl
H
H
H
H
H
H
H
Me
1
1
1
1
0
0
0
1
8.8
8.8
8.1
7.9
7.8
8.3
8.7
8.8
6.5
<6.2
6.3
<6.1
<6.2
<6.2
<6.2
<6.3
6.1
5.7
6.3
5.7
6.6
7.5
5.6
5.7
5.7
6.3
5.9
65
52
52
78
65
78
52
65
5.3
6.0
5.4
3.4
5.8
4.2
5.8
5.6
4-trifluoromethylphenyl
4-methyl-1,3-oxazol-5-yl
2-methyl-5-quinolinyl
5-methyl-2-pyrazinyl
3,4-difluorophenyl
5.9
<5.6
<5.6
<5.6
5.9
2-methyl-5-quinolinyl
a
b
SEM for D3 GTPγS, H1 FLIPR and hERG data sets is (0.1 and for the D2 GTPγS data is (0.2. fpKi ) functional pKi obtained from the GTPγS
c
functional assay. FLIPR ) fluorescent imaging plate reader.
Scheme 1. General Synthetic Procedures for the Preparation of Compounds 3-42a
a
i: (1) C3H3NH2, N-(3-dimethylaminopropyl)-N-ethylcarbodiimide (EDC), HOBt, CH2Cl2, RT, overnight; (2) refluxing AcOH., cat. Hg(OAc)2, 2 h; ii:
(1) TFA; CH2Cl2, RT, 2 h; (2): Br(CH2)3Cl or Br(CH2)2Cl, or Br(CH2)4Cl ,TEA, refluxing THF, 2 h; iii: DMF, cat. NaI, K2CO3, 60 °C, 24 h; iv (1)
NH2OH·HCl/Py, RT, 3 h; (2) N-chlorosuccinimide (NCS), DMF, 40 °C, 1.5 h; (3) C3H5Cl, TEA, RT, overnight; v: (1) K2CO3, MeOH/H2O, 50 °C; 2 h; :
Br(CH2)3Cl or Br(CH2)2Cl, TEA, refluxing THF, 2 h; vi (1) C9H18N2O2; LDA; -78 °C to RT; (2) TFA; CH2Cl2, RT, 2 h; (3) : Br(CH2)3Cl or Br(CH2)2Cl,
TEA, refluxing THF, 2 h.
nitrogen and the triazole and quinoxaline rings all form hydrogen
bonds with serines in the pore.
D3 receptor are also the key features which interact “negatively”
with the hERG channel. To further test this hypothesis,
derivatives 24-28 were prepared and a clear association
between potency and lipophilicity was observed for both targets.
To maintain the potency at D3 it was therefore important to
keep these key interactions (Figure 2B) while, at the same time,
introducing new features which might reduce activity at the
hERG channel.
Compound 10 was found to adopt a similar “U-shaped”
conformation” when docked into a model of the dopamine D3
receptor. As clearly depicted in Figure 2B, the best putative
binding mode showed a strong salt bridge between the basic
benzazepine nitrogen and Asp110 on transmembrane helix 3
(TM3). This aspartate is believed to be the primary binding site
for most basic compounds binding to the monoamine neu-
rotransmitter receptors. Hydrogen bond interactions were ob-
served between the phenylisoxazole moiety and Ser192 (TM5)
and between the quinoxaline nitrogen and Ser182 (second
extracellular loop) while π-stacking interactions were seen for
the quinoxalinyl-triazole moiety with Phe345 and His349 (TM6).
The D3 selectivity of these compounds can be explained by the
hydrogen bond between the triazole ring and Thr368. The
corresponding residue in the D2 receptor is a phenylalanine (Phe)
which would be expected to cause a severe steric clash with
these ligands.
Medicinal Chemistry Strategies To Address hERG Li-
ability. A number of approaches were used to tackle this issue.
Among those approaches, two will be described based on the
results of the docking studies and on general knowledge about
the hERG channel. In the first one, new molecules were
designed to increase the hydrophilicity (Figure 3A), by replacing
the heterocyclic substituents of the template on the thiotriazole
end. Favorable results were achieved in terms of reduction of
hERG activity while keeping relatively high potency at the
dopamine D receptor. For example, compound 30 had a very
3
good in vitro profile with a high potency at the dopamine D3
The models, in agreement with the SAR, suggest that in
compound 10 the same fragments (basic nitrogen; quinoline and
oxazole) necessary for potency and selectivity at the dopamine
receptor (pKi ) 8.9) and a 1000-fold selectivity over dopamine
D2 and H1 receptors. This high selectivity was also observed
with respect to the hERG channel. Based on this encouraging

5
080 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 21
Micheli et al.
profile and achieved the appropriate selectivity over the hERG
channel. Further exploration of this subseries demonstrated a
SAR consistent with previously obtained results: a clear trend
was once again observed between lipophilicity and hERG
affinity. A slight increase in clogD with respect to 35 (e.g., 36
clogD ) 5.5) led to the same dopamine D3 affinity and a slightly
higher hERG value, while a marked increase in clogD (e.g., 37
clogD ) 6.1) was associated with a slight decrease in the
dopamine D3 affinity, but a significant increase in the hERG
value (pIC50 ) 7.5 vs 5.2). Once again, an appropriately chosen
hydrophilic substituent (38) led to a slight decrease in the
dopamine D3 affinity but to a marked reduction in the hERG
pIC50. According to the receptor and channel models, a reduction
in the linker length should have provided active compounds,
with a slightly reduced hERG activity. Compounds 39-41 were
in agreement with the predicted potencies. Finally, the introduc-
tion of an R-methyl group (42) should have been beneficial in
terms of hERG reduction and actually an in vitro profile similar
to 35 was obtained; unfortunately a very high hCli was also
observed, probably due to the possible metabolic liability of
the methyl group.
Compound 35 was also endowed with a P450 profile with
no inhibition on single isoforms below 10 µM and ideal hCli
(1.8 mL/min/g liver) and was therefore selected for further
characterization. The PK profile in rat showed a moderate Clb
(
(
38 mL/min/kg), acceptable F% (15%), a low-to-moderate Vd
3.7 L/kg), and a good half-life (T1/2 ) 1.7 h); moreover, as
predicted by our calculations, the B/B ratio of 35 was high (2.5).
Before submission of the compound to in vivo disease models,
compound 35 was first submitted to hERG electrophysiology
determination. The reported affinity was 0.43 µM (pIC50 ) 6.5),
leading to an almost 4-fold improvement with respect to
compound 8 in accordance with the computational models. More
importantly, when tested in vivo in the anesthetized guinea pig
model, compound 35 showed no QTc prolongation or any
cardiac liability despite a high exposure in blood (Cmax of 34.4
µg/mL @ 25 mg/kg).
Figure 2. (A) Interactions of compound 10 in a “U-shaped” conforma-
tion which is energetically preferred over an extended conformation
because of increased interactions with aromatic residues in the cavity.
The backbone ribbon of the pore domain is included but that of the S5
and S6 helices has been removed for clarity. (B) A similar “U-shaped”
conformation is found in docking to the dopamine D
Again the backbone has been removed for clarity.
3
receptor model.
Following these positive results, compound 35 was thoroughly
characterized.
Further in Vitro Characterization of Compound 35.
Compound 35 displayed high selectivity over D2 receptors
Table 4) and a 100-fold selectivity over a wide range of
profile, a PK evaluation of 30 in the rat was performed.
Bioavailability was good (F ) 50%), with low Vd (2.4 L/kg)
and low Clb (33 mL/min/kg). Unfortunately, brain penetration
was poor (B/B ) 0.1) and so was the actual brain concentration.
Considering the fact that the MW of compound 30 was 468 Da
with a similar PSA to compound 8 which had a B/B ) 3.1
(
receptors and enzymes (see detailed Cerep Report in the
35
Supporting Information). Moreover, 35 did not stimulate [ S]-
GTPγS binding above basal levels in human recombinant
dopamine D3 receptors when tested alone up to a concentration
of 10 µM (Table 4). Derivative 35 behaved as a potent
competitive antagonist with a pKb derived from Schild analysis
of 7.93 (slope 1.0), comparable to the value of pKi obtained
also from filtration binding data (D3 ) 7.99 and D2 ) 5.02)
(PSA ) 69 vs 73 Aˆ 2), the low brain concentration was probably
related to the much reduced clogD (2.8 vs 5.2), unless some
specific efflux mechanism was involved for this particular
compound.
In the second approach, (Figure 3B), the effort was related
to the reduction of π interaction within the hERG channel of
the isoxazolyl moiety. Accordingly, specific compounds were
designed to achieve a greater dihedral angle with the benza-
zepine moiety, also reducing PSA while maintaining MW and
clogD. In house developed computational tools were used to
select the appropriate bioisosteric replacements for the isoxazolyl
moiety in the benzazepine class and synthetic feasibility was
used to filter among the different templates. The “top scorer”
of this exercise was represented by an N-methylated pyrazolyl
template, and this result is clearly related to the benzazepine
template here described in agreement with the specific SAR
here reported.
1
25
(
Table 4). Finally, 35 was tested in a [ I]-7-OH-PIPAT
competition binding assay on brain homogenates from rat
nucleus accumbens and olfactory tubercles. In these saturation
125
binding experiments, [ I]-7-OH-PIPAT bound to a single class
of receptor sites with a pKD value of 9.34 ( 0.11 and a Bmax of
-1
1
44 ( 16 fmol mg protein (n ) 8). Compound 35 completely
125
displaced, in a monophasic manner, the specific [ I]-7-OH-
PIPAT binding to rat membrane preparation. A pKi value of
.48 ( 0.08 (n ) 4) was found, thus confirming a high affinity
to the rat dopamine D3 receptor.
Further in Vivo Assays of Compound 35. An in vivo study
was conducted to assess the efficacy of compound 35 on the
expression of nicotine-induced conditioned place preference in
the rat. In the CPP paradigm, animals are given an injection of
8
When compound 35 (MW ) 530 Da, clogD ) 5.4, PSA )
5 Aˆ 2) was prepared, it showed (Table 3) the desired in vitro
6

Dopamine D
3
Receptor Antagonists
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 21 5081
Figure 3. Alternative medicinal chemistry strategies to address hERG liability.
35
Table 4. Profile of Compound 35 in Human Recombinant DA D3 Radioligand Binding, [ S]GTPγS Binding in Cell Membranes Expressing hDA D3
1
25
Receptors, and [ I]-7-OH-PIPAT Competition Binding Assay on Brain Homogenates from Rat Brain
pKi GTPγS pKi filtration binding
pKb,
hDA D3
rat native tissue,
r DA D3
hDA D3
9.0
hDA D2
6.5
hDA D3
8.0
hDA D2
5.0
35
7.93
8.48
Table 5. Effect of a Single i.p. Administration of Vehicle and 35 (1, 3,
0 mg/kg) on the Expression of the Conditioned Place Preference
Response to 0.4 mg/kg s.c. of (-)-Nicotine in Adult Male
Sprague-Dawley Rats
period, the drug is no longer available, but the rats are free to
try to obtain the drug (a period of “extinction training”). After
extinction of responding, the ability of various events to reinitiate
drug-seeking can be investigated. In both human addicts and
animal reinstatement models, a return to drug use can be
precipitated by three distinct types of stimuli: (1) re-exposure
to the drug itself, (2) exposure to environmental cues that had
been previously associated with drug intake, and (3) exposure
to stress. In this experiment, we showed that compound 35 (1,
3, 10 mg/kg i.p.) could prevent nicotine-triggered reinstatement
of nicotine seeking behavior in a dose-dependent manner. An
overall ANOVA showed that there was a significant difference
between treatment groups (F[4,44] ) 5.36, P < 0.001]. The
Fisher’s LSD post-hoc test revealed that priming with nicotine-
induced a significant increase in active lever pressing compared
to priming with saline (P < 0.001). Furthermore, compound
1
time spent in chambers (min)
treatment pairings drug given on test day
paired
unpaired
vehicle/vehicle
vehicle/nicotine
vehicle/nicotine
vehicle/nicotine
vehicle/nicotine
vehicleb
vehicle
35, 1 mg/kg
35, 3 mg/kg
35, 10 mg/kg
7.55 ( 0.22
10.07 ( 0.26
7.19 ( 0.22
6.71 ( 0.25
7.53 ( 0.22
a
c
7.44 ( 0.22
4.92 ( 0.26
7.80 ( 0.22
8.29 ( 0.25
7.47 ( 0.22
d
d
d
a
Each value represents the mean number of minutes spent in each
b
chamber ( SEM. The vehicle was 1 mL/kg s.c. of deionized, distilled
water for the pairings. The vehicle used on the test day was 1 mL/kg i.p.
c
of 20% hydroxypropyl-â-cyclodextrin. Significantly greater than vehicle/
vehicle pairings and vehicle on the test day, P < 0.0001 ANOVA and
d
Student-Newman-Keuls test. Significantly less than vehicle/nicotine
pairings + vehicle on the test day, P < 0.0001, ANOVA and Student-
Newman-Keuls test.
3
5 (3 and 10 mg/kg) produced a significant decrease in the
number of active lever presses compared to the vehicle/nicotine
group (P < 0.05).
a drug or vehicle and confined or “paired” to a specific
environment with distinct cues. This pairing of the animal with
specific cues after being given vehicle or drug is repeated.
Subsequently, on the test day, animals are not given any
treatment (drug-free) and are allowed to freely explore the CPP
apparatus to determine if they prefer an environment in which
they previously received drug compared to an environment in
which they previously received vehicle. If the animals approach
and spend a significantly greater amount of time in the drug-
paired environment, one can reasonably infer that the drug was
appetitive and this appetitive value is coded in the brain and is
accessible in the drug-free state.
Compound 35 (1, 3, 10 mg/kg, i.p.) significantly reduced
nicotine CPP in a dose-dependent manner (one-way ANOVA:
F[4,76] ) 31.7; P < 0.0001), thus confirming that it can
completely reverse the incentive motivational properties of
nicotine (Table 5).
To exclude the possibility that these positive results were
influenced by any impairment in locomotor activity or motor
coordination (typical effect of mixed D3/ D2 receptor antago-
nists), 35 was also tested in the rat Rotarod. Compound 35 at
a high dose of 10 mg/kg i.p. failed to affect endurance
performance at any of the posttreatment times tested in the
Rotarod. An ANOVA with a main factor of treatment and a
repeated measurements factor of test time showed that there
was no significant effect of treatment (F[2,19] ) 0.1], no
significant effect of test time (F[2,38] ) 2.02], and no significant
treatment by test time interaction (F[4,38] ) 0.35].
To confirm the ability of the dopamine D3 mechanism to act
vs different drugs of abuse, compound 35 was also tested on
the expression of cocaine-induced CPP in the rat. As previously
seen with nicotine, compound 35 (1, 3, 10 mg/kg, i.p.) also
significantly blocked cocaine CPP in a dose-dependent manner
(one-way ANOVA: F[4,36] ) 18.09; P < 0.0001), thus confirm-
ing that it can completely reverse the incentive motivational
properties of cocaine (Table 6).
A second study also assessed the efficacy of compound 35
in preventing reinstatement of nicotine-seeking behavior. In
reinstatement experiments, rats are trained to self-administer
drugs intravenously by pressing a lever. During a subsequent
We have also recently shown that 1 (10, 30 mg/kg i.p.)
significantly reduced alcohol preference in alcohol Preferring

5
082 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 21
Micheli et al.
Figure 4. Region-specific potentiation of phMRI amphetamine (1 mg/kg i.v.) response by acute pretreatment (30 min previously) with 35 (20
mg/kg i.p.). (a) Statistical map of vehicle/amphetamine [n ) 13] vs 35/amphetamine [n ) 11] (P < 0.01). (b) Mean ((SEM) rCBV changes
following amphetamine challenge for vehicle/amphetamine [n ) 13] and 35/amphetamine [n ) 11] groups in selected ROIs, including the mPFC
as a nonpotentiated control region (t-test veh/amp vs 35/amp; *** P < 0.001). Abbreviations: NAc ) nucleus accumbens, dCPU ) dorsal caudate
putamen, Amyg ) amygdala, vSub ) ventral subiculum, VPThal ) ventroposterolateral thalamic nuclei, mPFC ) medial prefrontal cortex.
Table 6. Effect of a Single i.p. Administration of Vehicle and 35 (1, 3, 10 mg/kg) on the Expression of the Conditioned Place Preference Response to
1
5 mg/kg i.p. of (-)-Cocaine HCl in Adult Male Sprague-Dawley Rats
time spent in chambers (min)
treatment pairings
drug given on test day
paired
unpaired
vehicle/vehicle
vehicle/cocaine
vehicle/cocaine
vehicle/cocaine
vehicle/cocaine
vehicleb
vehicle
35, 1 mg/kg
35, 3 mg/kg
35, 10 mg/kg
7.42 ( 0.40a
11.26 ( 0.48c
7.58 ( 0.33d
7.58 ( 0.40
3.74 ( 0.48
7.42 ( 0.33
7.94 ( 0.38
6.86 ( 0.43
d
7.06 ( 0.38
8.14 ( 0.43d
a
Each value represents the mean number of minutes spent in each chamber ( SEM. b The vehicle was 1 mL/kg s.c. of deionized, distilled water for the
c
pairings. The vehicle used on the test day was 1 mL/kg i.p. of 20% hydroxypropyl-â-cyclodextrin. Significantly greater than vehicle/vehicle pairings and
vehicle on the test day, P < 0.0001 ANOVA and Student-Newman-Keuls test. d Significantly less than vehicle/nicotine pairings + vehicle on the test day,
P < 0.0001, ANOVA and Student-Newman-Keuls test.
(
P) and Non-Preferring (NP) rats in a dose-dependent manner.29
amplitude of the relative cerebral blood volume (rCBV) response
32
Similarly, 35 (1, 3, 10 mg/kg i.p.) also significantly attenuated
oral operant alcohol self-administration in the C57BL/6N mouse.
Our results show that compound 35 significantly decreased the
number of active lever presses associated with alcohol intake
to d-amphetamine in a regionally specific manner), compound
35 (20 mg/kg i.p.) potentiated the amphetamine response in the
D3-rich NAc, but also in a number of structures outside the focal
distribution of the D3 receptor, and associated with reinstatement
of drug seeking behavior, such as the ventral subiculum region
of the hippocampus, the amygdale, and striatum (Figure 4).
(
one-way ANOVA: F[3,68] ) 3.86; P < 0.05) without altering
the number of inactive lever presses (one-way ANOVA: F[3,68]
)
1.19; P ) 0.32).
Previous studies have shown that 1 significantly improves
Conclusions
the learning deficit produced by the nonselective muscarinic
receptor antagonist scopolamine and the anxiogenic benzodi-
azepine inverse agonist FG-7142 without altering the normal
learning process in nonimpaired rats.30 Accordingly, the effect
of compound 35 (3, 10 mg/kg i.p.) on dialysate levels of ACh
in the rat medial prefrontal cortex was examined. Microdialysis
samples were analyzed using liquid chromatography coupled
with tandem mass spectrometry (LC-MS/MS) for the detection
Starting from the potent and selective dopamine D3 receptor
antagonist previously reported (2), a new series of 1,2,4-triazol-
3-yl-thiopropyl-tetrahydrobenzazepines was identified. Rational
design and computational tools helped us to overcome hERG-
related issues and to design a number of products endowed with
high affinity and selectivity for the dopamine D3 receptor. In
addition, selected compounds of this family series proved to
have appropriate developability characteristics (P450, Cli, F%)
and good CNS penetration. In particular, compound 35 showed
reduced hERG liability in vitro and no QTc prolongation in
vivo. Compound 35 was shown to be a potent and selective
dopamine D3 receptor antagonist in vivo with no sign of
locomotor liability, a typical effect produced by nonselective
D3/D2 receptor antagonists.
3
1
of ACh without the use of acetylcholinesterase inhibitors.
Compound 35 (10 mg/kg i.p.) significantly increased the levels
of ACh up to about 250%. These findings demonstrate that
compound 35, like other selective DA D3 receptor antagonists
from different chemical series,3 increases extracellular con-
centrations of ACh in the rat medial prefrontal cortex and further
strengthen the likelihood that selective DA D3 receptor antago-
nists may have cognitive enhancing properties.
0,31
Provided that in vivo preclinical evidence can be extrapolated
to human, selective DA D3 receptor antagonists belonging to
this new chemical series show the highest promise for the
treatment of drug addiction, psychosis, and schizophrenia.
Finally, in analogy to what was shown in phMRI experiments
in rats after acute pretreatment with 1 (i.e., potentiation of the

Dopamine D
3
Receptor Antagonists
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 21 5083
Experimental Section
P450 CYPEX Assay. Inhibition (IC50) of human CYP1A2, 2C9,
C19, 2D6, and 3A4 was determined using Cypex Bactosomes
expressing the major human P450s. A range of concentrations (0.1,
.2, 0.4, 1, 2, 4, and 10 µM) of test compound were prepared in
2
Biological Test Methods. In Vitro Studies. Human Recom-
3
binant D Radioligand Binding Assays. Radioligand binding
0
3
33
studies using [ H]-FLB-457 were performed according to the
following protocol. Test compounds (10 point, half log serially
diluted in 5 mM HCl) were incubated with assay buffer (50 mM
methanol and preincubated at 37 °C for 10 min in 50 mM potassium
phosphate buffer (pH 7.4) containing recombinant human CYP450
microsomal protein (0.1 mg/mL; Cypex Limited, Dundee, UK) and
probe-fluorescent substrate. The final concentration of solvent was
between 3 and 4.5% of the final volume. Following preincubation,
NADPH regenerating system (7.8 mg of glucose 6-phosphate, 1.7
mg of NADP and 6 units of glucose 6-phosphate dehydrogenase/
mL of 2% (w/v) NaHCO3; 25 µL) was added to each well to start
the reaction. Production of fluorescent metabolite was then
measured over a 10-min time-course using a Spectrafluor plus plate
reader. The rate of metabolite production (AFU/min) was deter-
mined at each concentration of compound and converted to a
percentage of the mean control rate using Magellan (Tecan
software). The inhibition (IC50) of each compound was determined
from the slope of the plot using Grafit v5 (Erithacus software, UK).
Miconazole was added as a positive control to each plate. CYP450
isoform substrates used were ethoxyresorufin (ER; 1A2; 0.5 µM),
TRIS, 10 mM NaCl, 5 mM KCl, 1 mM MgCl
2
, 2 mM CaCl
.4), 0.2 nM [ H]-FLB-457, and CHO cell membranes containing
either hD or rD cloned receptors for 45 min at 37 °C. Total binding
2
, pH
3
7
3
3
was defined using 5 mM HCl, and nonspecific binding was defined
using 10 µM haloperidol. The reaction was terminated by rapid
filtration through GF/B filterplates presoaked in double distilled
H
2
O, followed by 3 × 1 mL washes with ice cold 50 mM TRIS
preset buffer (pH 7.7). Bound radioactivity was determined by
scintillation spectrometry.
[
35S] GTPγS Functional Binding Assay in Cell Membranes
3
Expressing hD Receptors. In vitro functional studies were
35
performed according to the following [ S]-GTPγS protocol. Test
compounds (10 point, half log serially diluted in assay buffer) were
incubated with 10 µM GDP and CHO cell membranes containing
either hD
[
3
2
or hD
S]-GTPγS was added before a second incubation for 30 min at
0 °C. Basal binding was defined using assay buffer, and
3
cloned receptors for 30 min at 30 °C. 100 pM
7
-methoxy-4-triflouromethylcoumarin-3-acetic acid (FCA; 2C9; 50
35
µM), 3-butyryl-7-methoxycoumarin (BMC; 2C19; 10 µM), 4-
methylaminomethyl-7-methoxycoumarin (MMC; 2D6; 10 µM),
diethoxyflourescein (DEF; 3A4; 1 µM) and 7-benzyloxyquinoline
nonspecific binding was defined using 20 µM unlabeled GTPγS.
The reaction was terminated by rapid filtration through GF/B
(
7-BQ; 3A4; 25 µM). The test was performed in three replicates.
filterplates presoaked in ddH
with ice cold 20 mM HEPES, 10 mM MgCl
radioactivity was determined by scintillation spectrometry.
2
0, followed by 3 × 1 mL washes
Intrinsic Clearance (Cli) Assay. Intrinsic clearance (Cl ) values
i
2
buffer (pH 7.4). Bound
were determined in rat and human liver microsomes. Test com-
pounds (0.5 µM) were incubated at 37 °C for 30 min in 50 mM
potassium phosphate buffer (pH 7.4) containing 0.5 mg microsomal
protein/ mL. The reaction was started by addition of cofactor
125
[
I]-7-OH-PIPAT Competition Binding Assay on Brain
Homogenates from Rat Brain. Homogenates from frozen nucleus
accumbens and olfactory tubercles were prepared as described by
(
NADPH; 8 mg/mL). The final concentration of solvent was 1%
34
Burris et al. (1994). In saturation experiments, increasing con-
centrations of [1 I]-7-OH-PIPAT (15 pM to 2 nM) were incubated
with 12 µg/well of homogenates for 45 min at 37 °C in a final
volume of 200 µL of 50 mM Tris-HCl (pH7.0), 50 mM NaCl, 100
of the final volume. At 0, 3, 6, 9, 15, and 30 min an aliquot (50
µL) was taken, quenched with acetonitrile containing an appropriate
internal standard and analyzed by HPLC-MS/MS. The intrinsic
clearance (Cl ) was determined from the first-order elimination
i
constant by nonlinear regression using Grafit v5 (Erithacus software,
UK), corrected for the volume of the incubation and assuming 52.5
mg microsomal protein/g liver for all species. Values for Cl were
i
expressed as mL/min/g liver. The lower limit of quantification of
clearance was determined to be when <15% of the compound had
been metabolized by 30 min, and this corresponded to a Cl value
i
25
125
µM Gpp(NH)p and 0.02% BSA, i.e., conditions which inhibit [ I]-
7-OH-PIPAT binding to D
2
and 5HT1A receptors.34 Nonspecific
binding was determined by the presence of 1 µM 1. In competition
binding experiments, increasing concentrations of 35 (17 pM to 1
µM) were incubated as above in the presence of 0.1 nM [1 I]-7-
OH-PIPAT.
25
Reactions were stopped by filtration through GF/C 96-well filter
plates presoaked in 0.3% polyethylenimine using a cell harvester.
Filters were washed three times with 1 mL of ice-cold 50 mM Tris-
HCl (pH 7.7), and radioactivity was counted in a microplate
scintillation counter (Top Count, Perkin-Elmer).
of 0.5 mL/min/g liver. The upper limit was 50 mL/min/g liver.
Patch Clamp hERG Electrophysiology Assay. On the days of
experimentation, a weighed amount of the test substance was
formulated in dimethyl sulfoxide (DMSO; lot no. U10781; Sigma-
Aldrich, UK) by shaking, to give a stock concentration of 10 mM
of the test compound. The 10 mM stock solution was serially diluted
in DMSO to give further stock solutions of 1 and 0.1 mM. Aliquots
of these stock solutions were added to bath solution to achieve final
perfusion concentrations of 0.1, 1, and 10 µM test compound. The
corresponding vehicle concentration in all test substance perfusion
solutions was 0.1% DMSO. Test substance stock formulations were
freshly prepared on each day of experimentation, stored at room
temperature, and protected from light.
Reference Substance: E-4031 (batch no. MLE9446; Wako Pure
Chemical Industries Ltd.). Stock solutions of E-4031 (100 µM) were
prepared in reverse osmosis water, aliquoted, and stored at
approximately -20 °C until use. On the day of use, the 100 µM
stock solution was added to bath solution to give a final perfusion
concentration of 100 nM.
Radioligand binding data were analyzed by nonlinear regression
analysis using GraphPad Prism 4.0 (GraphPad Software, CA).
125
D
Determination of K and Bmax of [ I]-7-OH-PIPAT from saturation
experiments was assessed by using one-site binding (hyperbola)
equation. Curve fitting from competition-binding experiments was
determined by using the one-site competition equation after
checking with the F test (P < 0.05) that the Hill slope in the four-
parameter logistic equation was not statistically different from 1.0.
i
Under this condition, IC50 values were converted to K using the
3
3
Cheng-Prusoff equation. Results are expressed as mean pK (
i
SEM.
[
125I]-7-OH-PIPAT ([125I](R)-trans-7-hydroxy-2-[N-propyl-N-(3′-
iodo-2′-propenyl)amino]tetralin, 81.4 TBq/mmol) was purchased
from Perkin-Elmer Life and Analytical Sciences. Haloperidol and
Gpp(NH)p (guanylyl-5′-imidodiphosphate) were obtained from
Sigma Chemicals.
Bath and Pipet Solutions: The composition of bath solution
(mM): NaCl 137; KCl 4; CaCl2 1.8; MgCl2 1.0; D-glucose 10;
N-2-hydroxyethylpiperazine-N′-2-ethanesulfonic acid (HEPES) 10;
pH 7.4 with 1 M NaOH. Pipet solution was prepared in batches,
aliquoted, and stored frozen until the day of use. The composition
3
hERG- H-Dofetilide Binding Assay. hERG activity was
3
measured using H-dofetilide binding in a scintillation proximity
assay (SPA) format. The activity was measured with a Perkin-Elmer
Viewlux imager.
of pipet solution was (mM): KCl 130; MgCl 1.0; ethylene glycol
2
H1-FLIPR Assay. A functional response in CHO-hH1 cells was
measured using cytoplasmic calcium indicator Fluo-4. The change
in cell fluorescence being measured in a FLIPR (λex ) 488 nm,
bis(â-aminoethyl ether)-N,N,N′,N′-tetraacetic acid (EGTA) 5; MgATP
5; HEPES 10; pH 7.2 with 1 M KOH.
Study Design: Cells (passage number: 48) were transferred to
the recording chamber and continuously perfused (at approximately
λem ) 540 nm, Molecular Devices, UK).

5
084 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 21
-2 mL/min) with bath solution at room temperature. High
Micheli et al.
1
The animals were not exposed to the chambers prior to the start of
the pairings. During days 1-3, animals were acclimated to the
animal facility. During handling (days 4-6), animals were trans-
ported to the laboratory and handled for 5 min each. During
conditioning (days 7-14), animals were exposed to once-daily
conditioning sessions. For each conditioning session, animals (10
rats per group) were injected with nicotine (0.6 mg/kg s.c. in a
volume of 1 mL/kg) or vehicle (1 mL/kg s.c.) and then immediately
confined for 30 min in an appropriate cue-specific chamber. During
conditioning, nicotine was always paired with one cue-specific
environment, and vehicle was paired with the other; nicotine or
vehicle exposure (and appropriate environmental pairing) alternated
from day to day. This was done over an 8-day period, i.e., animals
were given four pairings with nicotine, and there was a 24-h
separation between exposure to vehicle and nicotine. The animals
in each group were randomly assigned to a 2 × 2 factorial design
with one factor being the pairing chamber and the other factor being
the order of conditioning. In the counterbalanced procedure, the
animals were randomly assigned to one of the two pairing chambers,
so that half of the subjects received the drug in one compartment
(white walls with bedding on the floor) and the other half in the
other compartment (alternating black and white boxes with a smooth
chamber floor). This procedure resulted in the animals receiving
equal exposure to the two compartments and, because of random
assignment, controlled for their side preference. Another group of
10 animals were paired with vehicle in both chambers of the
apparatus. On the test day (day 15), animals were randomly divided
into 5 groups of 10 and received compound 35 (1, 3, 10 mg/kg,
i.p.) or vehicle (20% 2-hydroxypropyl-â-cyclodextrin) in the home
cage 30 min before they were placed in the apparatus and allowed
free access to both chambers for 15 min. The amount of time spent
in each chamber was determined using an automated timing system.
resistance seals (seal resistances >1.5 GΩ) were formed between
the patch electrodes (resistance range: 1.4-5.5 MΩ) and individual
cells. The membrane across the electrode tip was then ruptured,
and the whole-cell patch-clamp configuration was established. Once
a stable patch had been achieved, recording commenced in voltage-
clamp mode, with the cell initially clamped at -80 mV. Currents
were evoked by stepping the membrane potential to +20 mV and
then to -50 mV (tail current). Test compound at 10, 1, and 0.1
µM was used to produce (if any) inhibition of hERG tail current
and therefore to investigate the concentration-response relationship
(n ) 3 cells/concentration). The effect of the vehicle (0.1% DMSO)
was investigated in 3 cells. The effect of 100 nM E-4031 was
investigated in 2 of the vehicle-treated cells to confirm the
sensitivity of the test system to an agent known to block hERG
current. All perfusion solutions were applied for approximately 10
min.
Anesthetized Guinea Pig Model for the Assessment of QT
Prolongation. Prior to the experiment, six male Hartley guinea pigs
(Charles River, France) weighing between 564 and 605 g on the
day of the test were anesthetized with urethane (1-1.5 g/kg IP),
tracheotomized, artificially ventilated with a tidal volume of
approximately 1 mL/100 g at a rate of 54 cycles/min, and surgically
prepared for the experiment (i.e., insertion of catheters in appropriate
blood vessels, see below, and placement of surface electrodes for
lead II electrocardiogram recording). The animal’s temperature was
kept constant between 36.8 and 38.0 °C. Once physiological
parameters were stabilized (approximately 30 min following
surgery), each animal was given either the vehicle (5% w/v dextrose
in aqueous 0.9 w/v sodium chloride) or test compound by the
intravenous route (via the jugular vein) over 15 min/treatment.
The following parameters were recorded: arterial pressure (via
the carotid artery), heart rate, electrocardiography (i.e., RR, PQ,
QRS, and QT interval duration; the QT interval was corrected for
heart rate changes according to Fridericia, Bazett, and Van de
Water’s formula). Parameters were recorded prior to dosing to
establish baseline measurements and continuously during the
infusion period, but data were reported every 5 min during each
infusion period. In addition, blood samples (0.3 mL) were collected
via the femoral artery at the end of each infusion period for
toxicokinetic evaluations.
(
-)-Nicotine (+)-bitartrate salt was used in all experiments
(
Sigma-Aldrich Corporation, St. Louis, MO) and was dissolved in
sterile physiological saline and the pH adjusted to 7.4 with NaOH.
The doses of nicotine were expressed as free base. 2-Hydroxypro-
pyl-â-cyclodextrin was purchased from Tocris-Cookson Chemical
(St. Louis, MO). Data were analyzed with a one-way ANOVA with
a main factor of dose. Statistical significance was set at a probability
level of P < 0.05.
Reinstatement of Nicotine-Seeking Behavior. Male Lister
Hooded rats (Charles River, Germany) were individually housed
in a temperature-controlled environment with lights on from 06.00
to 18.00 h. During the experiments, water was continuously
available and animals were maintained at a constant body weight
of 240-260 g (85% of their ad libitum body weight).
In Vivo Studies. All experiments were pre-reviewed and
approved by a local animal care committee in accordance with the
guidelines of the “Principles of Laboratory Animal Care” (NIH
publication No 86-23, revised 1985) and with a Project License
that was obtained according to Italian law (Art. 7, Legislative Decree
no. 116, Jan 27, 1992), which acknowledges European Directive
Behavioral testing (self-administration sessions and reinstatement
phase) was conducted according to the same methodology previ-
ously described and published.39 Compound 35 was tested at 1, 3,
and 10 mg/kg i.p. (30-min pretreatment time). The vehicle was a
solution of or vehicle 20% 2-hydroxypropyl-â-cyclodextrin. Data
were analyzed by a two-way ANOVA with main factors of dose
and time and repeated measurements over time. Statistical signifi-
cance was set at P < 0.05 for all tests.
Rat Rotarod. Male Lister Hooded rats (Charles River, Germany)
were individually housed in a temperature-controlled environment
with lights on from 06.00 to 18.00 h. During the experiments, water
and food were continuously available. Rats were trained on the
accelerating Rotarod (4-40 rpm over 270 s; 7750, Ugo Basile,
Italy) twice daily for two consecutive days. On the test day (day
8
6/609/EEC on the care and welfare of laboratory animals.
Nicotine CPP. Male Sprague-Dawley rats (200 g at the start
of the pairings, Taconic Farms, Germantown, NY) were used in
all experiments. Animals were housed two per cage, and there was
no more than a 5% difference in body weight between the
cagemates. Animals were kept on a 12 h lights on/12 h lights off
schedule (lights on at 09.00 h). Food and water were freely
available. The conditioning and testing of all animals was carried
out between 11.00 and 18.00 h. All rats were naive and used only
once.
An automated, two-chambered, Plexiglas CPP apparatus was
used as previously described,3
two pairing chambers of the apparatus were identical in dimensions
6-38
with modifications. Briefly, the
(
25 × 14 × 36 cm) and were separated by removable Plexiglas
3), rats were treated with compound 35 (10 mg/kg, i.p.) (n ) 10
guillotine doors. The pairing chambers were composed of distinct
visual and tactile cues. The walls of one of the pairing chambers
were white with cage bedding on the floor, and the walls of the
second chamber consisted of alternating white and black boxes (1.2
rats/group). Rats were then repeatedly tested for their endurance
performance on the Rotarod 30, 60, and 120 min after treatment.
Rotarod latencies were measured with a 300 s cutoff time.
×
1.8 cm) in a chessboard pattern and a Plexiglas floor. The two
Cocaine CPP. The cocaine CPP procedure was identical to the
one described for nicotine except that for each conditioning session
animals were injected with (-)-cocaine HCl (15 mg/kg i.p. in a
volume of 1 mL/kg) or vehicle (1 mL/kg i.p. of deionized, distilled
water). (-)-Cocaine HCl was purchased from Sigma Chemicals
(St. Louis, MO).
pairing chambers were separated by a third, neutral connecting
tunnel, with one-half of the wall white and the other half with black
and white boxes.
Expression studies with compound 35 or vehicle were divided
into four phases: acclimation, handling, conditioning, and testing.

Dopamine D
3
Receptor Antagonists
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 21 5085
Table 7
singlet (br s), broad multiplet (br m). All mass spectrometric
measurements were performed using a Micromass Platform LCZ
analytical column
mobile phase
Zorbax SB-C18,
(Waters, Manchester, UK) mass spectrometer operated in positive
4
.6 × 50 mm (1.8 µm)
5 mM ammonium acetate
electrospray ionization mode. When LC/MS detection was per-
formed, analytical conditions were used as reported in Table 7.
General Synthetic Procedures. For a description of the known
intermediates, please refer to refs 14-18 or to the detailed
experimental parts in the Supporting Information.
To a stirred solution of the appropriately decorated (oxazole,
isoxazole, methylpyrazole, respectively, intermediates 44, 47, 50
reported in Scheme 1) benzazepine scaffold in THF, containing
+
+
0.1% formic acid/acetonitrile
0.1% formic acid
97/3 f 36/64 v/v in 3.5 min
f 10/90 in 3.5 min
2.0 mL/min
DAD, 210-350 nm
ES+
gradient
flow rate
detection
MS
1.1 equiv of triethylamine, was added 1-bromo-3-chloropropane
Alcohol Self-Administration. Male C57BL/6N mice (Charles
River, Italy), weighing approximately 27 g at the start of their
training, were used in the present studies. All mice were experi-
mentally naive, housed individually in a temperature controlled
room (20-22 °C) with a 12-h light-dark cycle (06.00-18.00 light
on), and had free access to tap water in their home cage.
The experiments were conducted according to the same meth-
odology previously described and published.40 The respective effects
of acute treatment with compound 35 were analyzed using a two-
way ANOVA with main factors of dose and session (pre vs post).
The differences between individual means were assessed with the
post-hoc Fisher’s PLSD test. Statistical significance was set at a
probability level of p < 0.05 for all tests.
Extracellular Levels of Acetylcholine in the Medial Prefrontal
Cortex (mPFC). Male Sprague-Dawley rats (Charles River, Italy)
weighing 250-300 g were group housed at 21 ( 1 °C with 50%
humidity on a 12/12 light-dark cycle (light-on at 6 a.m.);
experiments were carried out during the light phase.
The experiments were conducted according to the same meth-
odology previously described and published.31 The effect of
compound 35 on extracellular levels of ACh was analyzed by
ANOVA consisting of a between-subjects factor of treatment and
a repeated measurements factor of time. In addition, one-way
ANOVA on the area under the curve (AUC) was performed to
assess the main effect of drug treatment on ACh levels. The post-
hoc Fisher’s protected least significant difference pairwise com-
parison test was used where appropriate. Statistical significance was
set at a probability level of p < 0.05 for all tests.
(or appropriate alkylating agent) (1.1 equiv), and the solution was
refluxed for 2 h. After workup and chromatographic purification,
the resulting alkylating agent (respectively, 45, 48, 51 in Scheme
1) was dissolved in DMF. A catalytic amount of NaI was added
and the appropriately substituted thiotriazole added together with
2 3
1.2 equiv of K CO . The temperature was raised to 60 °C and the
suspension stirred for 24 h. The compound was purified through
column chromatography (silica gel, Merck) using the appropriate
mixture of cyclohexane/AcOEt. Hydrochlorides for biological
testing were prepared, when necessary, dissolving the desired
compounds in dichloromethane; an equimolar amount of HCl (1
M solution in Et O) was added at room temperature, and the solvent
2
was evaporated under reduced pressure to give the desired
compounds.
3-(3-{[4-Methyl-5-(2-methyl-5-quinolinyl)-4H-1,2,4-triazol-3-
yl]thio}propyl)-7-(5-methyl-1,3-oxazol-2-yl)-2,3,4,5-tetrahydro-
1H-3-benzazepine (3). The compound was obtained as a white
1
foam (55% yield). H NMR (DMSO-d
6
) δ: 2.26 (m, 2H), 2.35 (d,
3H), 2.72 (s, 3H), 3.00-3.5 (br m, 10H), 3.43 (s, 3H), 3.69 (br m,
2H), 6.96 (m, 1H), 7.35 (d, 1H), 7.55 (d, 1H), 7.72 (dd, 1H), 7.80
(m, 2H), 7.91 (t, 1H), 8.17 (d, 1H), 8.26 (br s, 1H), 10.69 (br s,
+
1H). MS. m/z )525 [M + H] . Anal. (C30
32 6
H N OS·HCl) C, H, N.
7-(5-Methyl-1,3-oxazol-2-yl)-3-(3-{[4-methyl-5-(2-thienyl)-4H-
1,2,4-triazol-3-yl]thio}propyl)-2,3,4,5-tetrahydro-1H-3-benza-
zepine (4). The compound was obtained as a light brown foam
1
(35% yield); H NMR (DMSO-d
6
) δ: 2.2 (m, 2H), 2.39 (s, 3H),
3.08 (m, 4H), 3.24 (m, 4H), 3.43 (m, 2H), 3.47 (m, 2H), 3.76 (s,
3H), 6.99 (m, 1H), 7.28 (dd, 1H), 7.37 (d, 1H), 7.67 (dd, 1H), 7.76
Pharmacological MRI. Male Sprague-Dawley rats (250 g to
50 g; Charles-River, Como, Italy) were scanned in a Bruker
(dd, 1H), 7.8-7.83 (m, 2H), 10.83 (br s, 1H). MS. m/z 466 [M +
+
3
H] . Anal. (C24
H
27
N OS
5 2
‚HCl) C, H, N.
Biospec 4.7T MRI scanner under 0.8% maintenance halothane
anaesthesia and neuromuscular blockade; animal preparation and
monitoring, and MRI setup and acquisition, were the same as
published previously,32 except that the time series scan comprised
3-(3-{[5-(1H-Indol-2-yl)-4-methyl-4H-1,2,4-triazol-3-yl]thio}-
propyl)-7-(5-methyl-1,3-oxazol-2-yl)-2,3,4,5-tetrahydro-1H-3-
benzazepine (5). The compound was obtained as a light yellow
1
foam (66% yield); H NMR (DMSO-d
6
) δ: 2.2 (m, 2H), 2.39 (d,
16 contiguous slices of 1 mm thickness and a lower temporal
3H), 3.05-3.2 (m, 4H), 3.28 (m, 4H), 3.3-3.4 (m, 2H), 3.71 (m,
2H), 3.84 (s, 3H), 6.99 (m, 1H), 7.06 (s, 1H), 7.08 (mt, 1H), 7.22
(mt, 1H), 7.38 (d, 1H), 7.49 (d, 1H), 7.66 (d, 1H), 7.76 (dd, 1H),
resolution (80 s). The time series images were sensitized to changes
in relative cerebral blood volume (rCBV) by the injection of a blood
pool contrast agent (Endorem, Guerbet, France; 2.67 mL/kg)
following 5 reference image frames.3 Following 10 min equili-
bration, animals were administered either 20 mg/kg compound 35
7.82 (d, 1H), 10.16 (br s, 1H), 11.88 (s, 1H). MS. m/z 499 [M +
2,41
+
H] . Anal. (C28
H
30
6
N OS‚HCl) C, H, N.
3-(3-{[5-(1H-Indol-5-yl)-4-Methyl-4H-1,2,4-triazol-3-yl]thio}-
propyl)-7-(5-methyl-1,3-oxazol-2-yl)-2,3,4,5-tetrahydro-1H-3-
(
N ) 11) or vehicle saline (N ) 13) i.p., followed 30 min later by
a 1 mg/kg i.v. d-amphetamine challenge.
benzazepine (6). The compound was obtained as a light brown
1
The time series signal changes upon d-amphetamine injection
were converted into rCBV changes on a pixelwise basis.4
Following spatial coregistration of the image data, the data were
smoothed with a Gaussian kernel of full-width-half-maximum
equivalent to twice the in-plane pixel dimension, and the amphet-
amine response amplitude in each pixel was quantified by means
of a general linear model analysis using the AFNI package (v2.23).
Bilateral region of interest time courses corresponding to specific
anatomical structures were also extracted from each subject and
the rCBV changes quantified in the same way. Differences in the
amphetamine response magnitude between the two groups were
assessed using t-tests.
foam (71% yield) H NMR (DMSO-d
6
) δ: 2.24 (m,2H), 2.39 (d,
1,42
3H), 3.0-3.8 (m, 12H), 3.67 (s, 3H), 6.58 (m, 1H), 6.99 (m, 1H),
7.38 (d, 1H), 7.43 (dd, 1H), 7.5 (m, 1H), 7.58 (d, 1H), 7.77 (dd,
1H), 7.82 (m, 1H), 7.91 (d, 1H), 10.41 (br s, 1H), 11.44 (s, 1H).
+
MS. m/z 499 [M + H] . Anal. (C28
30 6
H N OS‚HCl) C, H, N.
3-(3-{[4-Methyl-5-(1-methyl-1H-pyrrol-2-yl)-4H-1,2,4-triazol-
3-yl]thio}propyl)-7-(5-methyl-1,3-oxazol-2-yl)-2,3,4,5-tetrahydro-
1H-3-benzazepine (7). The compound was obtained as a brown
1
foam (35% yield). H NMR (DMSO-d
6
) δ: 2.22 (m, 2H), 2.39 (d,
3H), 3.0-3.5 (m, 12H), 3.6 (s, 3H), 3.78 (s, 3H), 6.21 (dd, 1H),
6.56 (dd, 1H), 6.99 (m, 1H), 7.05 (dd, 1H), 7.38 (d, 1H), 7.76 (dd,
1H), 7.82 (d, 1H), 10.43 (br s, 1H). MS. m/z 463 [M + H] . Anal.
30 6
(C25H N OS‚HCl) C, H, N.
+
Chemical Procedures. General. Experimental. NMR spectra
were obtained on Varian INOVA spectrometers (300 MHz, 400
MHz, and 500 MHz). Chemical shifts are expressed in δ (ppm)
units and peak multiplicity is expressed as follows: singlet (s),
doublet (d), doublet of doublets (dd), triplet (t), multiplet (m), broad
7-(5-Methyl-3-isoxazolyl)-3-(3-{[4-methyl-5-(2-methyl-5-quin-
olinyl)-4H-1,2,4-triazol-3-yl]thio}propyl)-2,3,4,5-tetrahydro-1H-
3-benzazepine (8). The compound was obtained as a yellow foam
1
(30% yield). H NMR (DMSO-d
6
) δ: 2.27 (m, 2H), 2.45 (s, 3H),

5
086 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 21
Micheli et al.
2
7
1
5
.79 (br s, 3H), 3.09 (m, 4H), 3.3-3.8 (m, 11H), 6.74 (s, 1H),
3.25 (t, 2H), 3.3 (m, 2H), 3.39 (m, 2H), 3.7 (br m, 2H), 3.76 (s,
.36 (d, 1H), 7.66 (d, 2H), 7.73 (s, 1H), 7.89 (br m, 1H), 8 (br m,
3H), 6.75 (m, 1H), 7.28 (dd, 1H), 7.36 (d, 1H), 7.65-7.67 (m,
H), 8.25 (br m, 1H), 8.42 (br s, 1H), 10.61 (br s, 1H). MS. m/z
2H), 7.73 (d, 1H), 7.81 (dd, 1H), 10.55 (br s, 1H). MS. m/z 466
+
+
25 [M + H] . Anal. (C30
H
32
N
6
OS‚HCl) C, H, N.
[M + H] . Anal. (C24
H N OS
27 5 2
‚HCl) C, H, N.
7
-(5-Methyl-3-isoxazolyl)-3-(3-{[4-methyl-5-(2-methyl-6-quin-
olinyl)-4H-1,2,4-triazol-3-yl]thio}propyl)-2,3,4,5-tetrahydro-1H-
-benzazepine (9). The compound was obtained as a colorless oil:
7-(5-Methyl-3-isoxazolyl)-3-(3-{[4-methyl-5-(1-methyl-1H-
pyrrol-2-yl)-4H-1 ,2,4-triazol-3-yl]thio}propyl)-2,3,4,5-tetrahy-
3
dro-1H-3-benzazepine (18). The compound was obtained as a dark
1
1
H NMR (CDCl
3
) δ: 8.13 (3H, m), 7.92 (1H, d), 7.55 (1H, s),
.49 (1H, d), 7.37 (1H, d) 7.16 (1H, d), 6.26 (1H, s), 3.69 (3H, s),
.40 (2H, t), 2.97 (4H, m), 2.79 (3H, s), 2.68 (6H, m), 2.47 (3H,
oil (83% yield). H NMR (DMSO-d
6
) δ: 2.22 (m, 2H), 2.47 (s,
7
3
3H), 3.1 (m, 4H), 3.27 (t, 2H), 3.3 (m, 2H), 3.4 (m, 2H), 3.6 (s,
3H), 3.71 (br s, 2H), 3.77 (s, 3H), 6.21 (dd, 1H), 6.56 (dd, 1H),
+
s), 2.06 (2H, m) MS m/z 525 [M + H] . Anal. (C30
C, H, N.
32 6
H N OS‚HCl)
6.75 (s, 1H), 7.05 (dd, 1H), 7.37 (d, 1H), 7.66 (dd, 1H), 7.73 (d,
+
1H), 10.57 (br s, 1H). MS. m/z 463 [M + H] . Anal. (C25
30 6
H N -
3-(3-{[5-(2,3-Dimethyl-6-quinoxalinyl)-4-methyl-4H-1,2,4-tria-
OS‚HCl) C, H, N.
zol-3-yl]thio}propyl)-7-(5-methyl-3-isoxazolyl)-2,3,4,5-tetrahy-
dro-1H-3-benzazepine (10). The compound was obtained as a light
3-(3-{[5-(2-Chlorophenyl)-4-methyl-4H-1,2,4-triazol-3-yl]thio}-
propyl)-7-(5-methyl-3-isoxazolyl)-2,3,4,5-tetrahydro-1H-3-ben-
yellow foam: 1H NMR (CDCl
) δ:8.19 (1H, s), 8.11 (2H, s), 7.55
1H, s), 7.50 (1H, d), 7.16 (1H, d), 6.26 (1H, s), 3.74 (3H, s), 3.41
2H, t), 2.96 (4H, m), 2.78 (3H, s), 2.77 (3H, s), 2.66 (6H, m),
zazepine (19). The compound was obtained as a colorless oil (40%
3
1
(
(
yield). H NMR (CD
3
OD) δ: 2.4 (m, 2H), 2.5 (s, 3H), 3.1-3.3
(m, 4H), 3.4-3.5 (m, 2H), 3.48 m (4H), 3.58 (s, 3H), 3.85 (m,
+
2
.47 (3H, s), 2.06 (2H, m) MS m/z 540 [M + H] . Anal. (C30
OS‚HCl) C, H, N.
-(5-Methyl-3-isoxazolyl)-3-{3-[(4-methyl-5-phenyl-4H-1,2,4-
H
33
N
7
-
2H), 6.58 (s, 1H), 7.38 (d, 1H), 7.55-7.75 (m, 6H). MS. m/z 494
+
[M + H] . Anal. (C26
5
H28ClN OS‚HCl) C, H, N.
7
3-(3-{[5-(3-Chlorophenyl)-4-methyl-4H-1,2,4-triazol-3-yl]thio}-
propyl)-7-(5-methyl-3-isoxazolyl)-2,3,4,5-tetrahydro-1H-3-ben-
triazol-3-yl)thio]propyl}-2,3,4,5-tetrahydro-1H-3-benzazepine (11).
1
The compound was obtained as an off-white foam (70% yield). H
zazepine (20). The compound was obtained as a colorless solid
1
NMR (DMSO-d
6
) δ: 2.23 (m, 2H), 2.47 (s, 3H), 3.10 (m, 4H),
(78% yield). H NMR (CD
3
OD) δ: 2.34 (m, 2H), 2.5 (s, 3H), 3.1-
3
2
7
.29 (t, 2H), 3.3 (m, 2H), 3.4 (m, 2H), 3.64 (s, 3H), 3.72 (br s,
3.3 (m, 4H), 3.4-3.5 (m, 2H), 3.41 (t, 2H), 3.47 (t, 2H), 3.74 (s,
H), 6.75 (s, 1H), 7.37 (d, 1H), 7.58 (m, 3H), 7.68 (dd, 1H), 7.73-
3H), 3.87 (m, 2H), 6.58 (m, 1H), 7.38 (d, 1H), 7.6-7.7 (m, 4H),
+
+
.75 (m, 3H), 10.55 (bs, 1H). MS. m/z 460 [M + H] . Anal.
7.73 (s, 1H), 7.8 (m, 1H). MS. m/z 494 [M + H] . Anal. (C26
H
28
-
(C
26
H
29
N
5
OS‚HCl) C, H, N.
5
ClN OS‚HCl) C, H, N.
4-[4-Methyl-5-({3-[7-(5-methyl-3-isoxazolyl)-1,2,4,5-tetrahy-
3-(3-{[5-(4-Chlorophenyl)-4-methyl-4H-1,2,4-triazol-3-yl]thio}-
propyl)-7-(5-methyl-3-isoxazolyl)-2,3,4,5-tetrahydro-1H-3-ben-
dro-3H-3-benzazepin-3-yl]propyl}thio)-4H-1,2,4-triazol-3-yl]-
benzonitrile (12). The compound was obtained as a colorless
zazepine (21). The compound was obtained as a colorless solid
foam: 1H NMR (CDCl
d), 7.15 (1H, d), 6.26 (1H, s), 3.64 (3H, s), 3.41 (2H, t), 2.95 (4H,
) δ: 7.81 (4H, s), 7.54 (1H, s), 7.49 (1H,
(48% yield). H NMR (CD
1
OD) δ: 2.34 (m, 2H), 2.5 (d, 3H), 3.1-
3
3
3.5 (br m, 6H), 3.38 (t, 2H), 3.46 (t, 2H), 3.72 (s, 3H), 3.9 (br m,
m), 2.65 (6H, m), 2.47 (3H, s), 2.05 (2H, m). MS. m/z 485 [M +
2H), 6.58 (m, 1H), 7.38 (d, 1H), 7.64 (m, 2H), 7.7 (dd, 1H), 7.74
+
+
H] . Anal. (C27
H
28
N
6
OS‚HCl) C, H, N.
(m, 3H). MS. m/z 494 [M + H] . Anal. (C26
H
28ClN
5
OS‚HCl) C,
7
-(5-Methyl-3-isoxazolyl)-3-[3-({4-methyl-5-[4-(trifluorometh-
yl)phenyl]-4H-1,2,4-triazol-3-yl}thio)propyl]-2,3,4,5-tetrahydro-
H-3-benzazepine (13). The compound was obtained as a colorless
H, N.
3-(3-{[5-(3,4-Dichlorophenyl)-4-methyl-4H-1,2,4-triazol-3-yl]-
thio}propyl)-7-(5-methyl-3-isoxazolyl)-2,3,4,5-tetrahydro-1H-3-
1
1
solid (50% yield). H NMR (CD
3
OD) δ: 2.37 (m, 2H), 2.5 (d,
benzazepine (22). The compound was obtained as a colorless oil
1
3
3
1
H), 3.1-3.3 (m, 4H), 3.4-3.5 (m, 2H), 3.39 (t, 2H), 3.47 (t, 2H),
(68% yield). H NMR (CD
3
OD) δ: 2.3 (m, 2H), 2.44 (s, 3H), 3.0-
.76 (s, 3H), 3.87 (m, 2H), 6.58 (m, 1H), 7.39 (d, 1H), 7.7 (dd,
3.25 (m, 4H), 3.3-3.5 (m, 6H), 3.72 (s, 3H), 3.8 (m, 2H), 6.51 (s,
1H), 7.32 (d, 1H), 7.63 (dd, 1H), 7.68 (m, 2H), 7.81 (d, 1H), 7.95
(d, 1H). MS. m/z 528 [M + H] . Anal. (C26
+
H), 7.73 (d, 1H), 7.96 (m, 4H). MS. m/z 528 [M + H] . Anal.
+
(C
27
H
28
F N
3
5
OS‚HCl) C, H, N.
H27Cl
2 5
N OS‚HCl) C,
7
-(5-Methyl-3-isoxazolyl)-3-(3-{[4-methyl-5-(4-methylphenyl)-
H, N.
4
H-1,2,4-triazol-3-yl]thio}propyl)-2,3,4,5-tetrahydro-1H-3-ben-
3-(3-{[5-(3-Fluorophenyl)-4-methyl-4H-1,2,4-triazol-3-yl]thio}-
propyl)-7-(5-methyl-3-isoxazolyl)-2,3,4,5-tetrahydro-1H-3-ben-
zazepine (14). The compound was obtained as a glassy solid (70%
1
yield). H NMR (CD
H), 3.1-3.3 (m, 4H), 3.4-3.54 (m, 6H), 3.82 (s, 3H), 3.8-3.9
br m, 2H), 6.58 (m, 1H), 7.38 (d, 1H), 7.57 (d, 2H), 7.69 (dd,
3
OD) δ: 2.45 (m, 2H), 2.50 (s, 3H), 2.53 (d,
zazepine (23). The compound was obtained as a colorless oil (38%
1
3
(
yield). H NMR (CD
3
OD) δ: 2.38 (m, 2H), 2.5 (s, 3H), 3.1-3.3
(m, 4H), 3.4-3.5 (m, 2H), 3.44 (m, 4H), 3.77 (s, 3H), 3.86 (m,
+
1
H), 7.75 (m, 3H). MS. m/z 474 [M + H] . Anal. (C27
HCl) C, H, N.
-(5-Methyl-3-isoxazolyl)-3-(3-{[4-methyl-5-(2-pyridinyl)-4H-
,2,4-triazol-3-yl]thio}propyl)-2,3,4,5-tetrahydro-1H-3-benza-
zepine (15). The compound was obtained as a colorless oil (50%
H
31
5
N OS‚
2H), 6.58 (s, 1H), 7.38 (d, 1H), 7.47 (m, 1H), 7.6 (m, 2H), 7.65-
+
7.75 (m, 3H). MS. m/z 478 [M + H] . Anal. (C26
5
H28FN OS‚HCl)
7
C, H, N.
1
3-{3-[(4,5-Dimethyl-4H-1,2,4-triazol-3-yl)thio]propyl}-7-(5-
methyl-3-isoxazolyl)-2,3,4,5-tetrahydro-1H-3-benzazepine (24).
1
1
yield). H NMR (DMSO-d
6
) δ: 2.22 (m, 2H), 2.47 (s, 3H), 3.10
The compound was obtained as a colorless oil (52% yield). H NMR
(
3
m, 2H), 3.31 (m, 4H), 3.32-3.5 (m, 4H), 3.71 (m, 2H), 3.96 (s,
6
(DMSO-d ) δ: 2.15 (m, 2H), 2.41 (s, 3H), 2.47 (s, 3H), 3.2 (t,
H), 6.75 (m, 1H), 7.37 (d, 1H), 7.54 (m, 1H), 7.67 (dd, 1H), 7.73
2H), 3.51 (s, 3H), 3.0-3.8 (br m, 10H), 6.75 (m, 1H), 7.36 (d,
(d, 1H), 8.02 (m, 1H), 8.14 (d, 1H), 8.74 (m, 1H), 10.4 (br s, 1H).
1H), 7.66 (dd, 1H), 7.73 (d, 1H), 10.58 (br s, 1H). MS. m/z 398
+
+
MS. m/ z 461 [M + H] . Anal. (C25
-(5-Methyl-3-isoxazolyl)-3-(3-{[4-methyl-5-(4-pyridinyl)-4H-
,2,4-triazol-3-yl]thio}propyl)-2,3,4,5-tetrahydro-1H-3-benza-
zepine (16). The compound was obtained as light foam (70% yield).
H N
28 6
OS‚HCl) C, H, N.
[M + H] . Anal. (C21
27 5
H N OS‚HCl) C, H, N.
7
7-(5-Methyl-3-isoxazolyl)-3-(3-{[4-methyl-5-(trifluoromethyl)-
4H-1,2,4-triazol-3-yl]thio}propyl)-2,3,4,5-tetrahydro-1H-3-ben-
1
zazepine (25). The compound was obtained as a colorless oil (79%
1
1
H NMR (DMSO-d
6
) δ: 2.23 (m, 2H), 2.47 (s, 3H), 3.1 (m, 4H),
yield). H NMR (DMSO-d
6
) δ: 2.22 (m, 2H), 2.47 (s, 3H), 3.0-
3
7
2
.3 (m, 4H), 3.4 (m, 2H), 3.7 (m, 2H), 3.72 (s, 3H), 6.75 (m, 1H),
3.2 (m, 4H), 3.2-3.5 (m, 6H), 3.7 (m, 2H), 3.69 (s, 3H), 6.75 (m,
1H), 7.36 (d, 1H), 7.66 (dd, 1H), 7.73 (d, 1H), 10.59 (br s, 1H).
MS. m/z 452 [M + H] . Anal. (C21
.37 (d, 1H), 7.66 (dd, 1H), 7.73 (d, 1H), 7.82 (d, 2H), 8.81 (d,
+
+
H), 10.56 (br s, 1H). MS. m/z 461 [M + H] . Anal. (C25
H
28
6
N -
H
24
3 5
F N OS‚HCl) C, H, N.
OS‚HCl) C, H, N.
-(5-Methyl-3-isoxazolyl)-3-(3-{[4-methyl-5-(2-thienyl)-4H-
,2,4-triazol-3-yl ]thio}propyl)-2,3,4,5-tetrahydro-1H-3-benza-
zepine (17). The compound was obtained as an oil (73% yield).
7-(5-Methyl-3-isoxazolyl)-3-(3-{[4-methyl-5-(1-methylethyl)-
4H-1,2,4-triazol-3-yl]thio}propyl)-2,3,4,5-tetrahydro-1H-3-ben-
zazepine (26). The compound was obtained as a colorless foam
(49% yield). H NMR (CD3OD) δ: 1.47 (m, 6H), 2.38(m, 2H),
2.5 (s, 3H), 3.1-3.33 (m, 4H), 3.4-3.5 (m, 2H), 3.48 (m, 1H),
7
1
1
1
H NMR (DMSO-d
6
) δ: 2.2 (m, 2H), 2.46 (s, 3H), 3.1 (m, 4H),

Dopamine D
3
Receptor Antagonists
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 21 5087
3
1
.44 (m, 4H), 3.77 (s, 3H), 3.84 (m, 2H), 6.58 (s, 1H), 7.38 (d,
(s, 3H), 3.65-3.75 (m, 2H), 3.7 (s, 3H), 6.09 (s, 1H), 7.29 (m,
+
H), 7.68 (dd, 1H), 7.72 (s, 1H). MS. m/z 426 [M + H] . Anal.
OS‚HCl) C, H, N.
-(3-{[5-(1,1-Dimethylethyl)-4-methyl-4H-1,2,4-triazol-3-yl]-
s3H), 7.33 (br s, 1H), 7.48 (d, 1H), 7.74 (d, 1H), 7.86 (t, 1H), 8.12
+
(C
H
23 31
N
5
(d, 1H), 8.18 (d, 1H), 10.6 (br s, 1H). MS. m/z 538 [M + H] .
3
35 7
Anal. (C31H N S‚HCl) C, H, N.
thio}propyl)-7-(5-methyl-3-isoxazolyl)-2,3,4,5-tetrahydro-1H-3-
benzazepine (27). The compound was obtained as a colorless oil
3-(3-{[5-(3,4-Difluorophenyl)-4-methyl-4H-1,2,4-triazol-3-yl]-
thio}propyl)-7-(1,3-dimethyl-1H-pyrazol-5-yl)-2,3,4,5-tetrahydro-
1
(92% yield). H NMR (DMSO-d
6
) δ: 1.44 (s, 9H), 2.22 (m, 2H),
1H-3-benzazepine (36). The compound was obtained as a colorless
1
2
1
.52 (s, 3H), 3.25 (t, 2H), 3.71 (s, 3H), 3.0-3.9 (m, 10H), 6.8 (s,
H), 7.42 (d, 1H), 7.72 (dd, 1H), 7.79 (d, 1H), 10.2 (br s, 1H).
oil (55% yield). H NMR (DMSO-d
6
) δ: 10.60 (bs,1H), 7.85 (dt,
1H), 7.70-7.76 (m, 2H), 7.35 (m, 3H), 6.15 (s, 1H), 3.76 (s, 3H),
+
MS. m/z 440 [M + H] . Anal. (C24
33 5
H N OS‚HCl) C, H, N.
3.71 (bm, 2H), 3.65 (s, 3H), 3.40-3.20 (m, 6H), 3.09 (bm, 4H),
+
3
-{3-[(5-cyclopentyl-4-methyl-4H-1,2,4-triazol-3-yl)thio]pro-
2.23 (m, 2H), 2.17 (s, 3H). MS. m/z 541 [M + H] . Anal.
pyl}-7-(5-methyl-3-isoxazolyl)-2,3,4,5-tetrahydro-1H-3-benza-
zepine (28). The compound was obtained as a colorless foam (52%
yield). H NMR (CD
31 3 6
(C28H F N S‚HCl) C, H, N.
7-(1,3-Dimethyl-1H-pyrazol-5-yl)-3-[3-({4-methyl-5-[4-(trif-
luoromethyl)phenyl]-4H-1,2,4-triazol-3-yl}thio)propyl]-2,3,4,5-
1
3
OD) δ: 1.90 (m, 6H), 2.29 (m, 2H), 2.38
(
(
7
m, 2H), 2.5 (s, 3H), 3.05-3.25 (m, 4H), 3.4-3.5 (m, 2H), 3.53
m, 1H), 3.4 (m, 4H), 3.76 (s, 3H), 3.84 (m, 2H), 6.58 (s, 1H),
.38 (d, 1H), 7.68 (dd, 1H), 7.72 (br s, 1H). MS. m/z 452 [M +
tetrahydro-1H-3-benzazepine (37). The compound was obtained
1
as a colorless foam (59% yield). H NMR (DMSO-d
6
) δ: 10.48
(bs,1H), 8.00 (m, 4H), 7.30 (m, 3H), 6.2 (s, 1H), 3.76-3.68 (2s,
+
H] . Anal. (C25
-(5-Methyl-3-isoxazolyl)-3-(3-{[4-methyl-5-(4-methylcyclo-
hexyl)-4H-1,2,4-triazol-3-yl]thio}propyl)-2,3,4,5-tetrahydro-1H-
-benzazepine (29). The compound was obtained as a colorless
H
33
5
N OS‚HCl) C, H, N.
6H), 3.70 (bm, 2H), 3.50-3.20, 3.10 (bm, 10H), 2.24 (quint., 2H),
+
7
2.17 (s, 3H). MS. m/z 509 [M + H] . Anal. (C27
H
30
F
N
2 6
S‚HCl)
C, H, N.
3
7-(1,3-Dimethyl-1H-pyrazol-5-yl)-3-(3-{[4-methyl-5-(4-meth-
yl-1,3-oxazol-5-yl)-4H-1,2,4-triazol-3-yl]thio}propyl)-2,3,4,5-tet-
1
oil (87% yield). H NMR (DMSO-d
6
) δ: 0.95 (d, 3H), 1.4-1.9
(m, 9H), 2.17 (m, 2H), 2.47 (s, 3H), 2.99 (m, 1H), 3.21 (t, 2H),
rahydro-1H-3-benzazepine (38). The compound was obtained as
1
3
.0-3.5 (m, 8H),3.52 (s, 3H), 3.69 (br s, 2H), 6.75 (s, 1H), 7.37
a light yellow oil (78% yield). H NMR (DMSO-d
6
) δ : 10.25
(d, 1H), 7.66 (dd, 1H), 7.73 (br s, 1H), 10.44 (br s, 1H). MS. m/z
(bs,1H), 8.59 (s, 1H), 7.35 (m, 3H), 6.15 (s, 1H), 3.76 (s, 3H),
+
4
80 [M + H] . Anal. (C27
-(5-Methyl-3-isoxazolyl)-3-(3-{[4-methyl-5-(tetrahydro-2H-
pyran-4-yl)-4H-1,2,4-triazol-3-yl]thio}propyl)-2,3,4,5-tetrahydro-
H-3-benzazepine (30). The compound was obtained as a off-white
H
37
N
5
OS‚HCl) C, H, N.
3.71 (s, 3H), 3.80-3.70 (bm, 2H), 3.40-3.20 (m, 6H), 3.09 (bm,
7
4H), 2.40 (s, 3H), 2.21 (m, 2H), 2.17 (s, 3H). MS. m/z 478 [M +
+
H] . Anal. (C25
H
31
7
N OS‚HCl) C, H, N.
1
7-(1,3-Dimethyl-1H-pyrazol-5-yl)-3-(2-{[4-methyl-5-(2-methyl-
5-quinolinyl)-4H-1,2,4-triazol-3-yl]-thio}ethyl)-2,3,4,5-tetrahy-
1
foam (52% yield). H NMR (CD
3
OD) δ: 1.87 (m, 2H), 1.93 (m,
H), 2.03 (m, 2H), 2.44 (s, 3H), 3.0-3.7 (m, 13H), 3.55 (m, 2H),
.71 (s, 3H), 4.05 (m, 2H), 6.52 (m, 1H), 7.32 (d, 1H), 7.62 (dd,
2
3
1
dro-1H-3-benzazepine (39). The compound was obtained as a
1
yellowish foam (69% yield). H NMR (DMSO-d
6
) δ: 10.82 (bs,
+
H), 7.66 (br s, 1H). MS. m/z 468 [M + H] . Anal. (C25
33 5 2
H N O S‚
1H), 8.26 (d, 1H), 8.19 (d, 1H), 7.92 (t, 1H), 7.80 (d, 1H), 7.56 (d,
1H), 7.37 (m, 3H), 6.16 (m, 1H), 3.90-3.80 (bm, 2H), 3.77 (s,
3H), 3.70 (m, 2H), 3.65 (m, 2H), 3.46 (s, 3H), 3.50-3.10 (bm,
HCl) C, H, N.
-(5-Methyl-3-isoxazolyl)-3-[3-({4-methyl-5-[2-(methyloxy)-
phenyl]-4H-1,2,4-triazol-3-yl}thio)propyl]-2,3,4,5-tetrahydro-
H-3-benzazepine (31). The compound was obtained as a colorless
7
+
6H), 2.74 (s, 3H), 2.17 (s, 3H). MS. m/z 524 [M + H] . Anal.
1
33 7
(C30H N S‚HCl) C, H, N.
1
solid (72% yield). H NMR (CD
3
OD) δ: 2.41 (m, 2H), 2.0 (s, 3H),
7-(1,3-Dimethyl-1H-pyrazol-5-yl)-3-(2-{[4-methyl-5-(5-methyl-
2-pyrazinyl)-4H-1,2,4-triazol-3-yl]thio}ethyl)-2,3,4,5-tetrahydro-
3
3
1
.1-3.3 (m, 4H), 3.4-3.5 (m, 2H), 3.47 (m, 4H), 3.6 (s, 3H), 3.85-
.95 (m, 2H), 3.95 (s, 3H), 6.58 (m, 1H), 7.24 (t, 1H), 7.32 (d,
1H-3-benzazepine (40). The compound was obtained as a brownish
1
H), 7.38 (d, 1H), 7.7 (dd, 1H), 7.56 (dd, 1H), 7.68-7.75 (m, 3H).
oil (69% yield). H NMR (DMSO-d
6
) δ: 10.65 (bs, 1H), 9.18 (d,
+
MS. m/z 490 [M + H] . Anal. (C27
H
31
N
5
2
O S‚HCl) C, H, N.
1H), 8.71 (d, 1H), 7.35 (m, 3H), 6.15 (s, 1H), 3.91 (s, 3H), 3.80-
7
-(5-Methyl-3-isoxazolyl)-3-[3-({4-methyl-5-[3-(methyloxy)-
phenyl]-4H-1,2,4-triazol-3-yl}thio)propyl]-2,3,4,5-tetrahydro-
H-3-benzazepine (32). The compound was obtained as a colorless
3.70 (bm, 2H), 3.76 (s, 3H), 3.66 (m, 2H), 3.58 (m, 2H), 3.40-
+
3.30. MS. m/z 475 [M + H] . Anal. (C25
30 8
H N S‚HCl) C, H, N.
1
3-(2-{[5-(3,4-Difluorophenyl)-4-methyl-4H-1,2,4-triazol-3-yl]-
thio}ethyl)-7-(1,3-dimethyl-1H-pyrazol-5-yl)-2,3,4,5-tetrahydro-
1
3
foam (80% yield). H NMR (CD OD) δ: 2.35 (m, 2H), 2.5 (s,
3
3
7
H), 3.1-3.3 (m, 4H), 3.4-3.5 (m, 2H), 3.37 (t, 2H), 3.47 (t, 2H),
.73 (s, 3H), 3.9 (m, 2H), 3.9 (s, 3H), 6.58 (s, 1H), 7.18 (m, 1H),
1H-3-benzazepine (41). The compound was obtained as a colorless
1
oil (85% yield). H NMR (DMSO-d
6
) δ: 10.65 (bs, 1H), 7.85 (ddd,
.27 (m, 2H), 7.38 (d, 1H), 7.53 (t, 1H), 7.7 (dd, 1H), 7.73 (s, 1H).
1H), 7.63 (m, 1H), 7.35 (m, 3H), 6.15 (s, 1H), 3.80-3.70 (bm,
+
MS. m/z 490 [M + H] . Anal. (C27
31 5 2
H N O S‚HCl) C, H, N.
2H), 3.77 (s, 3H), 3.65 (s, 3H), 3.63 (m, 2H), 3.58 (m, 2H), 3.40-
7
-(5-Methyl-3-isoxazolyl)-3-[3-({4-methyl-5-[4-(methyloxy)-
phenyl]-4H-1,2,4-triazol-3-yl}thio)propyl]-2,3,4,5-tetrahydro-
H-3-benzazepine (33). The compound was obtained as a colorless
3.30 (bm, 2H), 3.15 (bm, 4H), 2.17 (s, 3H). MS. m/z 495 [M +
+
H] . Anal. (C26
H
28
F N
2 6
S‚HCl) C, H, N.
1
7-(1,3-Dimethyl-1H-pyrazol-5-yl)-3-(1-methyl-3-{[4-methyl-
5-(2-methyl-5-quinolinyl)-4H-1,2,4-triazol-3-yl]thio}propyl)-
1
3
foam (69% yield). H NMR (CD OD) δ: 2.4 (m, 2H), 2.5 (s, 3H),
3
3
7
.1-3.3 (m, 4H), 3.4-3.5 (m, 2H), 3.46 (m, 4H), 3.79 (s, 3H),
2,3,4,5-tetrahydro-1H-3-benzazepine (42). The compound was
1
.87 (m, 2H), 3.94 (s, 3H), 6.58 (s, 1H), 7.24 d, 2H), 7.38 (d, 1H),
obtained as a yellowish foam (49% yield). H NMR (DMSO-d
6
)
+
.7 (dd, 1H), 7.71 (d, 1H), 7.75 (d, 2H). MS. m/z 490 [M + H] .
δ: 10.58 (bs, 1H), 8.50 (bs, 1H), 8.27 (d, 1H), 8.00 (t, 1H), 7.89
(d, 1H), 7.67 (d, 1H), 7.28 (m, 3H), 6.09 (s, 2H), 3.7 (s, 3H), 3.60
(bm, 2H), 3.60-3.50 (m, 2H), 3.55 (m, 1H), 3.41 (s, 3H), 3.40 (m,
1H), 3.22 (m, 1H), 3.15-2.95 (m, 4H), 2.79 (s, 3H), 2.45 (m, 1H),
31 5 2
Anal. (C27H N O S‚HCl) C, H, N.
7
-(5-Methyl-3-isoxazolyl)-3-[3-({4-methyl-5-[6-(methyloxy)-3-
pyridinyl]-4H-1,2,4-triazol-3-yl}thio)propyl]-2,3,4,5-tetrahydro-
H-3-benzazepine (34). The compound was obtained as a light
+
1
2.11 (s, 3H), 2.00 (m, 1H), 1.32 (d, 3H). MS. m/z 552 [M + H] .
1
hazel foam (59% yield). H NMR (DMSO-d
6
) δ: 2.2 (m, 2H), 2.44
37 7
Anal. (C32H N S‚HCl) C, H, N.
(
7
1
s, 3H), 3.0-3.7 (m, 12H), 3.61 (s, 3H), 3.92 (s, 3H), 6.72 (s, 1H),
Computational Modeling. Construction of the hERG Models.
The closed state of the hERG channel was modeled based on its
homology with the bacterial Kcsa structure from Strep. liVidans,
solved in 1998 by McKinnon et al.43 Alignment of the pore region
and the S6 helix is obvious from the key GYG motif (GFG in
hERG) and the key hinge glycine in S6. The alignment of the S5
helices of hERG and Kcsa, however, was not obvious. This was
based finally on the observation of a highly conserved glutamate
at the extracellular side of S5 which is found in most potassium
.02 (d, 1H), 7.35 (d, 1H), 7.63 (dd, 1H), 7.71 (s, 1H), 8.04 (dd,
+
H), 8.52 (d, 1H), 10.47 (br s, 1H). MS. m/z 491 [M + H] . Anal.
26 30 6 2
(C H N O S‚HCl) C, H, N.
7-(1,3-Dimethyl-1H-pyrazol-5-yl)-3-(3-{[4-methyl-5-(2-meth-
yl-5-quinolinyl)-4H-1,2,4-triazol-3-yl]thio}propyl)-2,3,4,5-tet-
rahydro-1H-3-benzazepine (35). The compound was obtained as
a yellowish oil (45% yield). H NMR (DMSO-d
2
1
6
) δ: 2.11 (s, 3H),
.24 (m, 2H), 2.67 (s, 3H), 3.03 (m, 4H), 3.2-3.4 (m, 6H), 3.40

5
088 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 21
Micheli et al.
channels. An NMR structure of the S5-SS1 loop peptide has been
published44 which suggests that this region is largely helical. This
was added to the model as a helix-turn-â-strand motif. The
tetrameric ensemble was fully minimized using the CHARMm
program,45 with an initial 500 steps of Steepest Descent, followed
by 5000 steps of ABNR, and a constant dielectric of 5. Helical
H-bonding distance constraints were used to maintain the overall
fold of the bundle.
Supporting Information Available: Cross-selectivity profile
for compound 35 and additional chemistry information with
synthetic details for the preparation of the intermediates are hereby
reported. This material is available free of charge via the Internet
at http://pubs.acs.org.
References
(1) Sokoloff, P.; Giros, B.; Martres, M. P.; Bouthenet, M. L.; Schwartz,
J. C. Molecular cloning and characterization of a novel dopamine
An open state model was also built using the rat Kv1.2 structure
as a template.46 The same S5 glutamate, pore GYG, and S6 glycine
receptor (D ) as a target for neuroleptics. Nature 1990, 347, 146-
3
151.
residues were used to generate the alignment. Use was made of
the S5-SS1 loop model constructed for the closed state structure,
and the same CHARMm conditions were used for final refinement.
(
2) Diaz, J.; L e´ vesque, D.; Lammers, C. H.; Griffon, N.; Martres, M.
P.; Schwartz, J. C.; Sokoloff, P. Phenotypical characterization of
neurons expressing the dopamine D3 receptor in the rat brain.
Neuroscience 1995, 65, 731-745.
3
Construction of the Dopamine D Receptor Models. The
(
3) Diaz, J.; Pilon, C.; Le Foll, B.; Gros, C.; Triller, A.; Schwartz, J. C.;
Sokoloff, P. Dopamine D3 receptors expressed by all mesencephalic
dopamine neurons. J. Neurosci. 2000, 20, 8677-8684.
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.
3
model of the dopamine D receptor was built using our “in house”
program GPCR_Builder.47 This is an automated 7TM receptor
homology modeling program which incorporates a database of all
known human and rodent Family A receptor sequences and a
prealigned set of transmembrane domains for these. The models
(
are based on the TM domain taken from the crystal structure of
bovine rhodopsin.48 A special set of optimized 3D templates are
(5) Landwehrmeyer, B.; Mengod, G.; Palacios, J. M. Dopamine D3
receptor mRNA and binding sites in human brain. Brain Res. Mol.
Brain Res. 1993, 18, 187-192.
used to add the extracellular loops. An internal rotamer library is
then applied to all residue side chains except those conserved
residues known to be involved in an important functional role (e.g.,
the asparagines in TM1 and TM7, the aspartate on TM2, and the
DRY motif on TM3). The program automatically generates a
CHARMm script which can be used for model optimization and
ligand docking. Typically, the same set of conditions as those
described for the hERG models above were used during the
minimization stage. A strict alignment between rhodopsin and
(
6) Suzuki, M.; Hurd, Y. L.; Sokoloff, P.; Schwartz, J. C.; Sedvall, G.
D dopamine receptor mRNA is widely expressed in the human brain.
3
Brain Res. 1998, 779, 58-74.
(
7) Heidbreder, C. A.; Gardner, E. L.; Xi, Z-X.; Thanos, P. K.; Mugnaini,
M.; Hagan, J. J.; Ashby, C. R., Jr. The role of central dopamine D3
receptors in drug addiction: a review of pharmacological evidence.
Brain Res. ReV. 2005, 49, 77-105.
(
8) Joyce, J. N.; Millan, M. J. Dopamine D3 receptor antagonists as
therapeutic agents. Drug DiscoVery Today 2005, 10, 917-25.
(9) Micheli, F.; Heidbreder, C. Selective dopamine D3 receptor antago-
dopamine D
3
gives rise to a very short loop length (two residues)
nists: A review 2001-2005. Recent Pat. CNS Drug DiscoVery 2006,
between the disulfide-bonded cysteine in the second extracellular
loop and the top of TM5. During the optimization, some unfolding
of the extracellular side of TM5 is therefore allowed by relaxing
the helical distance constraints in this region. This is in keeping
with the reported SCAM results carried out by Javitch et al. on the
1
, 271-288.
(
10) Newman, A. H.; Grundt, P.; Nader, M. A. Dopamine D3 receptor
partial agonists and antagonists as potential drug abuse therapeutic
agents. J. Med. Chem. 2005, 48, 3663-79.
(11) Heidbreder, C. A.; Andreoli, M.; Marcon, C.; Thanos, P. K.; Ashby,
C. R., Jr.; Gardner, E. L. Role of dopamine D3 receptors in the
addictive properties of ethanol. Drugs Today 2004, 40, 355-65.
12) Reavill, C.; Taylor, S. G.; Wood, M. D.; Ashmeade, T.; Austin, N.
E.; Avenell, K. Y.; Boyfield, I.; Branch, C. L.; Cilia, J.; Coldwell,
M. C.; Hadley, M. S.; Hunter, A. J.; Jeffrey, P.; Jewitt, F.; Johnson,
C. N.; Jones, D. N.; Medhurst, A. D.; Middlemiss, D. N.; Nash, D.
J.; Riley, G. J.; Routledge, C.; Stemp, G.; Thewlis, K. M.; Trail, B.;
Vong, A. K.; Hagan, J. J. Pharmacological actions of a novel, high-
affinity, and selective human dopamine D3 receptor antagonist, SB-
receptor.49
D
2
Manual Docking of Ligands in the Receptor and Channel
(
Models. The ligands used in this study were built in an extended
conformation using the Spartan program.50 Following initial
optimization with molecular mechanics, further refinement was
carried out at the semiempirical level using the AM1 Hamiltonian.
Natural atomic orbital atom-centered charges were then derived
with a single point calculation at the ab initio Hartree Fock level
using a 3-21G* basis set. The placement of the ligand in the protein
2
77011-A. J. Pharmacol. Exp. Ther. 2000, 294, 1154-65.
(
13) Macdonald, G. J.; Branch, C. L.; Hadley, M. S.; Johnson, C. N.;
Nash, D. J.; Smith, A. B.;, Stemp, G.; Thewlis, K. M.; Vong, A. K.;
Austin, N. E.; Jeffrey, P.; Winborn, K. Y.; Boyfield, I.; Hagan, J. J.;
Middlemiss, D. N.; Reavill, C.; Riley, G. J.; Watson, J. M. Wood,
M.; Parker, S. G.; Ashby, C. R., Jr. Design and synthesis of trans-
4
5
was carried out manually with the Quanta program. Multiple
starting poses were used, ensuring each time that the ligand was
always in an energetically acceptable conformation. For each pose
the cavity residue side chains were adjusted, where necessary, by
application of the rotamer library from Karplus and Dunbrack.28
The complex was then optimized using the CHARMm program
with 500 steps of SD followed by 5000 steps of ABNR. In addition
to helical distance constraints on the protein, any expected ligand-
protein hydrogen bonds were also constrained. Where appropriate,
dihedral constraints were applied to the ligand’s planar aromatic
centers. Visualization of the results was carried out using the Quanta
program, from which the pictures in this paper were derived.
3
-(2-(4-((3-(3-(5-methyl-1,2,4-oxadiazolyl))-phenyl)carboxamido)cy-
clohexyl)ethyl)-7-methylsulfonyl-2,3,4,5-tetrahydro-1H-3-benza-
zepine (SB-414796): a potent and selective dopamine D3 receptor
antagonist. J. Med. Chem. 2003, 46, 4952-64.
(14) Hadley, M. S.; Lightfoot, A. P.; MacDonald, G. J.; Stemp, G.
Preparation of tetrahydrobenzazepines as modulators of dopamine
D3 receptors useful as antipsychotic agents. WO 2002040471, 2001.
(
15) Bonanomi, G.; Damiani, F.; Gentile, G.; Hamprecht, D. W.; Micheli,
F.; Tarsi, L.; Tedesco, G.; Terreni, S. Preparation of fused benza-
zepines having affinity for dopamine D3 receptor. WO 2005118549,
2
005.
(
16) Hamprecht, D.; Micheli, F.; Tarsi, L.; Tedesco, G. Preparation of
Acknowledgment. We would like to thank all the colleagues
who helped in generating the data reported in this manuscript,
in particular, Stefania Beato, Antonio Felici, Laurie Gordon,
Phil Green, Alfredo Paio, Daniele Andreotti, Matteo Sartori,
Kevin Read, Stefano Fontana, Monica Rigo, Serenella Zambon,
Maria Anna Fedrigo, Alberto Ruffo, Claudia Mundi, Carla
Marchioro, Silvia Davalli, and Mahmoud Hamdan.
7
2
-pyrazolylbenzazepines having affinity for D3 receptor. WO
005123717, 2005.
(17) Bonanomi, G.; Cardullo, F.; Damiani, F.; Gentile, G.; Hamprecht,
D.; Micheli, F.; Tarsi, L. Preparation of fused benzazepines having
affinity for the dopamine D3 receptor. WO 2006002928, 2005.
18) Arista, L.; Bonanomi, G.; Damiani, F.; Hamprecht, D.; Micheli, F.;
Tarsi, L.; Tedesco, G. Preparation of tetrahydrobenzazepine deriva-
tives as modulators of dopamine D3 receptors. WO 2005087764,
(
2
005.
Note Added after ASAP Publication. This manuscript was
released ASAP on September 15, 2007. Due to an oversight,
an incomplete list of authors was received by the Journal. The
corrected version published ASAP on September 20, 2007.
(19) ACD Labs suite is available from http://www.acdlabs.com.
(
20) Ertl, P.; Rohde, B.; Selzer, P. Fast Calculation of Molecular Polar
Surface Area as a Sum of Fragment-Based Contributions and Its
Application to the Prediction of Drug Transport Properties. J. Med.
Chem. 2000, 43, 3714-3717.

Dopamine D
3
Receptor Antagonists
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 21 5089
(
(
21) The research complied with national legislation and with company
(39) Andreoli, M.; Tessari, M.; Pilla, M.; Valerio, E.; Hagan, J. J.;
policy on the Care and Use of Animals and with related codes of
practice.
22) Aptuit Limited, 3 Simpson Parkway, Livingston, West Lothian, EH54
Heidbreder, C. A. Selective antagonism at dopamine D receptors
3
prevents nicotine-triggered relapse to nicotine-seeking behavior.
Neuropsychopharmacology 2003, 28, 1272-1280.
7BH, Scotland. www.aptuit.uk.
(
40) Heidbreder, C. A.; Andreoli, M.; Marcon, C.; Gardner, E. L.; Ashby,
C. R., Jr. Evidence for the role of dopamine D3 receptors in oral
operant ethanol reinforcement and reinstatement of ethanol seeking
responding in mice. Addict. Biol. 2007, 12, 35-50.
41) Schwarz, A. J.; Reese, T.; Gozzi, A.; Bifone, A. Functional MRI
using intravascular contrast agents: detrending of the relative
cerebrovascular (rCBV) time course. Magn. Reson. Imaging 2003,
21, 1191-1200.
(
(
23) CERB, Chemin de Montifault, 18800 Baugy, France. www.cerb.fr.
24) Jamieson, C.; Moir, E. M.; Rankovic, Z.; Wishart, G. Medicinal
chemistry of hERG optimizations: Highlights and hang-ups. J. Med.
Chem. 2006, 49, 5029-5046.
(
(
(
(
(
(
25) Mitcheson, J. S.; Chen, J.; Lin, M.; Culberson, C.; Sanguinetti, M.
C. A structural basis for drug-induced long QT syndrome. Proc. Natl.
Acad. Sci. U.S.A. 2000, 97, 12329-333.
26) Sanguinetti, M. C.; Mitcheson, J. S. Predicting drug-hERG channel
interactions that cause acquired long QT syndrome. Trends Phar-
macol. Sci. 2005, 26, 119-124.
(42) Schwarz, A.; Whitcher, B.; Reese, T.; Gozzi, A.; Bifone, A. In Group-
leVel data-driVen phMRI analysis. Proceedings of the International
Society for Magnetic Resonance in Medicine, 2005, 13, 157.
27) Oesterberg, F.; Aqvist, J. Exploring blocker binding to a homology
model of the open hERG K+ channel using docking and molecular
dynamics methods. FEBS Lett. 2005, 579, 2939-2944.
28) Dunbrack, R. L., Jr.; Karplus, M. Backbone-dependent rotamer library
for proteins. Application to side-chain prediction. J. Mol. Biol. 1993,
(
43) Kaab, S.; Dixon, J.; Duc, J.; Ashen, D.; Nabauer, M.; Beuckelmann,
D. J.; Steinbeck, G.; McKinnon, D.; Tomaselli, G. F. Molecular basis
of transient outward potassium current downregulation in human heart
failure: a decrease in Kv4.3 mRNA correlates with a reduction in
current density. Circulation 1998, 98, 1383-1393.
230, 543-74.
29) Thanos, P. K.; Katana, J. M.; Ashby, C. R., Jr.; Michaelides, M.;
Gardner, E. L.; Heidbreder, C. A.; Volkow, N. D. A novel dopamine
D3 receptor antagonist (SB-2770110-A) attenuates ethanol consump-
tion in ethanol preferring (P) and non-preferring (NP) rats. Pharmacol.
Biochem. BehaV. 2005, 81, 190-197.
(44) Torres, A. M.; Bansal, P. S.; Sunde, M.; Clarke, C. E.; Bursill, J.
A.; Smith, D. J.; Bauskin, A.; Breit, S. N.; Campbell, T. J.; Alewood,
P. F.; Kuchel, P. W.; Vandenberg, J. I. Structure of the HERG K+
Channel S5P Extracellular Linker: Role of an Amphipathic .alpha.-
Helix in C-Type Inactivation. J. Biol. Chem. 2003, 278, 42136-
(
30) Laszy, J.; Laszlovszky, I.; Gyerty a´ n, I. Dopamine D3 receptor
antagonists improve the learning performance of memory-impaired
rats. Psychopharmacology 2005, 179, 567-575.
42148.
(45) www.accelrys.com.
(
31) Lacroix, L. P.; Ceolin, L.; Zocchi, A.; Varnier, G.; Garzotti, M.;
Curcuruto, O.; Heidbreder, C. A. Selective dopamine D(3) receptor
antagonists enhance cortical acetylcholine levels measured with high-
performance liquid chromatography/tandem mass spectrometry with-
out anti-cholinesterases. J. Neurosci. Methods 2006, 157, 25-31.
32) Schwarz, A.; Gozzi, A.; Reese, T.; Bertani, S.; Crestan, V.; Hagan,
J.; Heidbreder, C.; Bifone, A. Selective dopamine D3 receptor
antagonist SB-277011-A potentiates phMRI response to acute
amphetamine challenge in the rat brain. Synapse 2004, 54, 1-10.
33) Okauchi, T.; Suhara, T.; Maeda, J.; Kawabe, K.; Obayashi, S.; Suzuki,
K. Effect of endogenous dopamine on endogenous dopamine on
extrastriated [(11)C]FLB 457 binding measured by PET. Synapse
(
46) Magis, C.; Gasparini, D.; Lecoq, A.; Le Du, M. H.; Stura, E.;
Charbonnier, J. B.; Mourier, G.; Boulain, J.-C.; Pardo, L.; Caruana,
A.; Joly, A.; Lefranc, M.; Masella, M.; Menez, A.; Cuniasse, P.
Structure-Based Secondary Structure-Independent Approach To
Design Protein Ligands: Application to the Design of Kv1.2
Potassium Channel Blockers. J. Am. Chem. Soc. 2006, 128, 16190-
16205.
(
(
(
(
47) Blaney, F. E.; Tennant, M. Computational tools and results in the
construction of G protein-coupled receptor models. In Membrane
Protein Models; Findlay, J. B. C., Ed.; Bios Scientific Publishers
Ltd., Oxford, UK, 1996; pp 161-176.
2001, 41, 87-95.
(
48) Polczewski, K.; Kumasaka, T.; Hori, T.; Behnke, C. A.; Motoshima,
H.; Fox, B. A.; Le Trong, I.; Teller, D. C.; Okada, T.; Stenkamp, R.
E.; Yamamoto, M.; Miyano, M. 2000. Crystal structure of rhodop-
sin: A G protein-coupled receptor. Science 2000, 289, 739-745.
34) Burris, K. D.; Filtz, T. M.; Chumpradit, S.; Kung, M. P.; Foulon,
C.; Hensler, J. G.; Kung, H. F.; Molinoff, P. B. Characterization of
[
125I](R)-trans-7-hydroxy-2-[N-propyl-N-(3′-iodo-2′-propenyl)ami-
no]tetralin binding to dopamine D3 receptors in rat olfactory tubercle.
J. Pharmacol. Exp. Ther. 1994, 268, 935-942.
(49) Shi, L.; Simpson, M. M.; Ballesteros, J. A.; Javitch, J. A. The First
Transmembrane Segment of the Dopamine D2 Receptor: Acces-
sibility in the Binding-Site Crevice and Position in the Transmem-
brane Bundle. Biochemistry 2001, 40, 12339-12348.
(
35) Cheng, Y.; Prusoff, W. H. Relationship between the inhibition
constant (Ki) and the concentration of inhibitor which causes 50 per
cent inhibition (I50) of an enzymatic reaction. Biochem. Pharmacol.
1973, 22, 3099-3108.
(50) www.wavefun.com.
(
(
(
36) Dewey, S. L.; Brodie, J. D.; Gerasimov, M.; Horan, B.; Gardner, E.
L.; Ashby, C. R., Jr. A pharmacologic strategy for the treatment of
nicotine addiction. Synapse 1999, 31, 76-86.
(
51) An undesirable feature of some drugs is their ability to delay cardiac
repolarization, an effect that can be measured as prolongation of the
QT interval on the surface electrocardiogram (ECG). Because of its
inverse relationship to heart rate, the QT interval is routinely
transformed (normalized) by means of various formulas into a heart
rate independent “corrected” value known as the QTc interval. The
QTc interval is intended to represent the QT interval at a standardized
heart rate of 60 bpm.
37) Horan, B.; Smith, M.; Gardner, E. L.; Lepore, M.; Ashby, C. R., Jr.
(
-)-Nicotine produces conditioned place preference in Lewis, but not
Fischer 344 rats. Synapse 1997, 26, 93-94.
38) Horan, B.; Gardner, E. L.; Dewey, S. L.; Brodie, J. D.; Ashby, C.
R., Jr. The selective sigma(1) receptor agonist, 1-(3,4-dimethox-
yphenethyl)-4-(phenylpropyl)piperazine (SA4503), blocks the acqui-
sition of the conditioned place preference response to (-)-nicotine in
rats. Eur. J. Pharmacol. 2001, 426, R1-2.
JM0705612