Pharmacophore-guided drug discovery investigations leading to bioactive 5-aminotetrahydropyrazolopyridines. Implications for the binding mode of heterocyclic dopamine D3 receptor agonists

Article abstract of DOI:10.1021/jm0503805

Taking advantage of a 3D-QSAR based pharmacophore hypothesis, synthesis and biological evaluation of dopaminergic 5-aminotetrahydropyrazolo[1,5-a]pyridines are described. The data displayed substantial and selective D3 receptor affinity for the heterocycl

Full text of DOI:10.1021/jm0503805

J. Med. Chem. 2005, 48, 5771-5779
5771
Pharmacophore-Guided Drug Discovery Investigations Leading to Bioactive
5-Aminotetrahydropyrazolopyridines. Implications for the Binding Mode of
Heterocyclic Dopamine D3 Receptor Agonists§
Jan Elsner,† Frank Boeckler,† Frank W. Heinemann,‡ Harald Hu¨bner,† and Peter Gmeiner*,†
Department of Medicinal Chemistry, Emil Fischer Center, Friedrich Alexander University, Schuhstrasse 19,
D-91052 Erlangen, Germany, and Department of Inorganic Chemistry, Friedrich Alexander University,
Egerlandstrasse 1, D-91058 Erlangen, Germany
Received April 22, 2005
Taking advantage of a 3D-QSAR based pharmacophore hypothesis, synthesis and biological
evaluation of dopaminergic 5-aminotetrahydropyrazolo[1,5-a]pyridines are described. The data
displayed substantial and selective D3 receptor affinity for the heterocyclic test compound (()-1
when the enantiomer (S)-1 turned out to be responsible for the D3 binding (Ki high ) 4.0 nM).
(S)-1 exhibited binding affinity and ligand efficacy comparable to those of our previously
described D3 agonist FAUC 54, when subtype selectivity could be significantly improved. The
results indicate that the sp2 nitrogens of the pyrazole and thiazole rings of the dopaminergics
(S)-1 and pramipexole, respectively, are pharmacophoric elements of major importance. To
provide putative explanations for the high affinity of (S)-1, computational studies were
performed employing an active state D3 model.
Introduction
Dopamine receptors, which belong to the class A
rhodopsin-like G protein-coupled receptors, can be
divided phylogenetically into D1-like receptors compris-
ing the subtypes D1 and D5 and a D2-like group
including D2, D3, and D4.1 Interestingly, the D3 recep-
tor mRNA has a high abundance in limbic brain areas
associated with cognitive and emotional functions2 and,
therefore, can be considered as particularly related to
affective disorders.3 Thus, the D3 receptor has been
regarded as an interesting therapeutic target for the
treatment of schizophrenia,4 Parkinson’s disease,5 drug-
induced dyskinesia,6 and cocaine addiction.7
Since X-ray crystallographic or nuclear magnetic
resonance structural data are not available for any of
the aminergic G protein-coupled receptors, we estab-
lished extensively validated dopamine D2, D3, and D4
receptor models based on the crystal structure of bovine
rhodopsin when focusing on the similarity and diver-
gence of the D2-like dopamine receptor subtypes and
the respective ligand interactions.8,9 As a continuation
of these efforts, we were able to transform our homology
model of the D3 receptor from an inactive into an active
state by employing CoMFA/CoMSIA contribution maps
derived from a 3D-QSAR investigation on D3 agonists.10
Thus, combination of a ligand-based and a structure-
based modeling approach yielded a conceptual binding
model for D3 agonists, revealing novel insights into the
pharmacophoric requirements of potent D3 binding. As
depicted and exemplified in the conceptual Figure 1, the
essentials of our previous investigations are as follows:
(1) All ligands form a reinforced ionic bond (ionic
Figure 1. Outline of the ligand design strategy leading to
the 5-aminotetrahydropyrazolo[1,5-a]pyridine (S)-1 in order to
elucidate the significance of the hydrogen-bond acceptor func-
tion in pramipexole for D3 receptor recognition. A conceptual
pharmacophore model of the D3 receptor active site is il-
lustrated in its general form and exemplified by FAUC 54 and
(R)-7-OH-DPAT representing the two different classes of
stereoisomers. π1/π2 (orange indene) indicates the extended
aromatic system. Cyan and magenta spheres denote areas
where hydrogen-bond donors and acceptors can be located on
the ligand site, respectively. The single cyan sphere between
the yellow cones marks the hydrogen-bond donor feature of
the protonated amine, which is required in each agonist, as it
forms a reinforced ionic bond with the highly conserved
Asp3.32. The two yellow cones represent the hydrophobic side
chains.
* To whom correspondence should be addressed. Phone: +49(9131)-
8529383. Fax: +49(9131)8522585. E-mail: gmeiner@pharmazie.uni-
erlangen.de.
§ Dedicated to Professor Fritz Eiden on the occasion of his 80th
birthday.
interaction enforced by a hydrogen bond) between their
protonated amine (cyan sphere) and Asp3.32 within the
binding site. (2) The aromatic moieties of the ligands
† Department of Medicinal Chemistry.
‡ Department of Inorganic Chemistry.
10.1021/jm0503805 CCC: $30.25 © 2005 American Chemical Society
Published on Web 08/11/2005

5772 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 18
Elsner et al.
Scheme 1a
are capable of forming π-stacking interactions with
residues of the protein binding pocket within an ex-
tended aromatic area (π1/π2). The stereochemistry is
decisive for the placement of the aromatic moieties when
the (R)- and (S)-configurations lead to a displacement
close to π1 and π2, respectively. (3) Substituents at the
different aromatic moieties can occupy up to two hy-
drogen-bond donor (cyan spheres; d1 and d2) and hy-
drogen-bond acceptor positions (magenta spheres; a1
and a2) that are of varying importance. (4) All ligands
are anchored with at least one alkyl side chain (yellow
cones) in primarily hydrophobic parts of the binding
pocket when one of these forms a “propyl-specific” cavity.
Comparing the pKi values of the 3-formyl substituted
7-aminotetrahydroindolizine FAUC 54 with its 3-un-
substituted congener indicates that occupation of the
acceptor position a1 with the oxygen of the formyl
function of FAUC 54 is highly beneficial for D3 receptor
binding.11,12 Investigation of the D3 agonist (R)-7-OH-
DPAT facilitates positioning of the oxygen of its 7-hy-
droxy group onto the pharmacophoric area a1 when the
donor position d1 is occupied by the OH-hydrogen atom.
Based on the high D3 receptor affinities of (R)-7-OH-
DPAT and FAUC 54, we conclude that the occupation
of area a1 significantly contributes to D3 receptor
binding, which is supported by the importance of this
area shown in the respective CoMFA and CoMSIA
fields. However, as indicated by a simple atom-centered
alignment superimposing the core scaffolds of prami-
pexole13 and FAUC 54 (Figure 1), this apparently cannot
hold true for a main group of heterocyclic dopaminergics
including pramipexole, quinelorane, and quinpirole
when area a1 is not occupied by the endocyclic acceptor
functionality. Nevertheless, the exocyclic amino group
in position 2 of pramipexole can occupy the area d2 and,
thus, putatively provides positive contribution to D3
receptor binding. To learn more about the importance
of the endocyclic H-bond acceptor at the a1 site and to
exploit this structural element for the discovery of drug-
like D3 selective agonists, we were intrigued by the
question of whether an unsubstituted pyrazole is suf-
ficient as a bioisosteric subunit. Herein, we report on
synthesis, structural elucidation, and pharmacological
evaluation of a series of dipropylaminocyclohexene-fused
pyrazoles including the tetrahydropyrazolo[1,5-a]pyri-
dine (S)-1 and derivatives thereof.
a
Reagents and conditions: (a) methyl propiolate, K2CO3, air-
O2, DMF, rt, 16 h (30%); (b) 48% HBr, reflux, 1 h (80%); (c) H2,
Pd/C, EtOH, 20 bar, 80 °C, 16 h (79%); (d) (1) MsCl, Et3N, THF,
-30 °C, 1 h; (2) NaN3, DMSO, 35 °C, 24 h (45%); (e) H2, Pd/C,
MeOH, rt, 2.5 h (80%); (f) propionaldehyde, NaBH(OAc)3, 1,2-
dichloroethane, rt, 16 h (84-85%); (g) NCS, CHCl3, reflux, 16 h
(67%); (h) (1) (()-7, (-)-menthyl chloroformate, DIPEA, CH2Cl2,
0 °C to rt, 16 h (77%); (2) separation of diastereoisomers (31% for
each); (i) (1) (CH3)3SiI, CHCl3, reflux, 12 h; (2) MeOH, reflux, 12
h (89%).
Figure 2. ORTEP drawing of the molecular structure of (-)-
menthyl-derived tetrahydropyrazolo[1,5-a]pyridine (5R)-9, de-
termined with X-ray crystallography.
azide 6 in 45% yield. Reduction of the azide functionality
afforded the primary amine (()-7 followed by reductive
alkylation with propionaldehyde furnishing the desired
target compound (()-1 in racemic form. Treatment of
(()-1 with N-chlorosuccinimide (NCS) in refluxing
chloroform proceeded regioselectively, giving access to
the 3-chloro derivative 8.
To investigate the enantiospecifity of the receptor
recognition process, racemic (()-7 was resolved into its
enantiomers. Thus, treatment of (()-7 with (-)-menthyl
chloroformate gave a mixture of the diastereoisomeric
carbamates (5R)-9 and (5S)-9 that could be separated
by preparative HPLC. After successful attempts to grow
crystals suitable for X-ray analysis of (5R)-9, which was
eluted first, the assignment of the (R)-configuration to
the newly formed stereogenic center was based on the
stereochemical information from the (-)-menthyl group.
The X-ray crystal structure of (5R)-9 (Figure 2) clearly
indicated that the (-)-menthylcarbamoyl moiety adopted
an equatorial position. After cleavage of the carbamates
(5R)-9 and (5S)-9 by iodotrimethylsilane16 and subse-
Chemistry
The synthesis of target compounds was performed
starting from N-amino-4-benzyloxypyridinium mesity-
lenesulfonate 2 as a valuable building block that could
be prepared by electrophilic amination14 of 4-benzyloxy-
pyridine. Subsequent 1,3-dipolar cycloaddition with
methyl propiolate under oxidizing conditions afforded
the 5-benzyloxypyrazolo[1,5-a]pyridine derivative 3
(Scheme 1). Removal of the carboxylic ester group and
cleavage of the benzyl ether function under strong acidic
conditions furnished the 5-hydroxypyrazolopyridine 4
in 80% yield,15 which could be converted to the tetrahy-
dro derivative 5 by catalytic hydrogenation. Whereas
all attempts to oxidize the alcohol 5 and introduce the
amino function via a reductive amination reaction failed,
activation of the alcohol by MsCl followed by nucleo-
philic substitution with sodium azide gave rise to the

Heterocyclic Dopamine D3 Receptor Agonists
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 18 5773
Scheme 2a
complex of ligand, receptor, and G-protein thus indicat-
ing agonist properties were chosen for the comparison
of agonist affinities.
In most cases, the 5-aminotetrahydropyrazolo[1,5-a]-
pyridines investigated displayed only weak to moderate
D1 and D2 and D4 binding. However, substantial D3
binding affinity was observed for the heterocyclic target
compound (()-1. Interestingly, determination of the Ki
values of the pure enantiomers revealed that the
biological activity is almost exclusively due to (S)-1
(Ki high ) 4.0 nM). This finding is in accordance with
previously described dopaminergics22-25 and our recent
observations on aminoindolizines and aminotetrahy-
droindoles.11,26,27 Furthermore, (S)-1 showed substantial
D3 subtype selectivity over D2long and D2short and a
significant preference when compared to D4. Interest-
ingly, the 3-chloro derivative 8, which was investigated
in racemic form, also exhibited high binding affinity for
D3 (Ki high ) 6.1 nM) whereas compounds 12-15, all of
which bear larger substituents in position 3, displayed
only poor dopamine receptor recognition. To further
characterize the binding profiles of (()-1, (R)-1, (S)-1,
8, and 12-15, 5-HT1A and 5-HT2 affinities were inves-
tigated in a screening system with porcine cortical
membranes measuring the displacement of the radio-
ligands [3H]8-OH-DPAT and [3H]ketanserin, respec-
tively, by the test compounds at three representative
concentrations. All compounds except (S)-1 showed only
weak binding at the 5-HT receptors with a capability
to displace the radioligand at a concentration of 100 nM
with a rate of less than 15%. For (S)-1, a displacement
of 57% for 5-HT1A and 25% for 5-HT2 indicated a
moderate 5-HT affinity. For the most promising com-
pounds (()-1, (R)-1, (S)-1, and 8, receptor binding to the
porcine R1 receptor was investigated utilizing the selec-
tive radioligand [3H]prazosin when all compounds showed
affinities only in the micromolar range.
To investigate the intrinsic activity of the most
interesting test compounds (R)-1, (S)-1, and 8 compared
to FAUC 54 and the reference agonist quinpirole, a
mitogenesis assay was performed employing D3 receptor
expressing CHO dhfr- cells.28,29 Agonist activation of
dopamine receptors can be determined by measuring the
rate of [3H]thymidine incorporation into growing het-
erologously transfected cell lines. Intrinsic activities and
EC50 values are presented in Table 2 clearly indicating
substantial ligand efficacy and potency for (S)-1 (82%,
3.4 nM) and the 3-chloro derivative 8 (54%, 4.5 nM).
The obtained data reflect agonist and partial agonist
properties of the test compounds when the EC50 values
are in good consistence to the Ki high values derived from
the binding experiments.
a
Reagents and conditions: (a) H2, Pd/C, EtOH, 20 bar, 80 °C,
48 h (75%); (b) DMP, CH2Cl2, rt, 2 h (70%); (c) propylamine,
NaCNBH3, MeOH, rt, 16 h (60%); (d) propionaldehyde, NaB-
H(OAc)3, 1,2-dichloroethane, rt, 16 h (78%); (e) 13, LiAlH4, THF,
0 °C to rt, 1.5 h (82%); (f) MnO2, CH2Cl2, rt, 16 h (97%).
quent reductive alkylation of the obtained primary
amines (R)-7 and (S)-7, respectively, with propionalde-
hyde, the test compounds (R)-1 and (S)-1 could be
isolated in enantiomerically pure form.
Besides the 3-chloropyrazolopyridine derivative 8, we
were interested in synthesizing further 3-substituted
test compounds taking advantage of the carboxylate
function that we introduced in the beginning of the
synthesis. Thus, the building block 3 was subjected to
a chemoselective hydrogenolytic debenzylation giving
access to the carboxylic ester 10 in 75% yield (Scheme
2). In contrast to the findings observed with the unsub-
stituted alcohol 5, oxidation of 10 by using Dess-Martin
periodinane (DMP)17 afforded the ketone 11 in good
yield. Subsequent amination and N-alkylation in the
presence of NaBH3CN and NaBH(OAc)3, respectively,
gave access to the mono- and dipropylamines 12 and
13, respectively. The carboxylic ester function of 13 was
reduced with LiAlH4 to furnish the alcohol 14 and,
subsequently, oxidized by MnO2 to afford the final
product 15.
Results and Discussion
Biological Investigations. Receptor binding experi-
ments were established to evaluate the binding proper-
ties of the target compounds (()-1, (R)-1, (S)-1, 8, and
12-15 in comparison to the reference FAUC 54 (Table
1). D1 receptor affinities were determined utilizing
porcine striatal membranes and the D1 selective radio-
ligand [3H]SCH 23390.18 D2, D3, and D4 affinities were
investigated employing the cloned human dopamine
receptors D2long, D2short
19
,
D3,20 and D4.421 stably ex-
Theoretical Investigations. To understand the
high D3 affinity of the heterocyclic test compounds and
to gain more insights into the putative binding modes
of pramipexole and (S)-1, we employed the predictive
power of a very recently established 3D-QSAR model
for D3 agonists, which has also been used to derive the
pharmacophore hypothesis presented in the Introduc-
tion.10 This model was built based on a diverse training
set consisting of a series of 34 chiral 7-aminotetrahy-
droindolizines and 10 commercially available dopam-
inergics. The PLS analysis of the final CoMFA model
pressed in Chinese hamster ovary cells (CHO) and the
radioligand [3H]spiperone.18 The competition data were
analyzed according to a sigmoid model by nonlinear
regression. If the dose-response curves showed biphasic
properties and the calculated Hill coefficients (nH) were
between -0.50 and -0.75 with a better fit of equation
indicating a two-site model and if the amount of
calculated high affinity binding sites was greater than
15% of the total receptor population, Ki values for the
high and low affinity binding sites of the receptor were
derived. The Ki high values representing the ternary
revealed a good cross-validated q2 of 0.726 (sPRESS
)
cv

5774 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 18
Elsner et al.
Table 1. Receptor Binding Data of the 5-Amino-4,5,6,7-tetrahydropyrazolo[1,5-a]pyridines 1, 8, and 12-15 for Human and Porcine
Dopamine Receptorsa
Ki (nM)a
[3H]SCH23990
pD1
[3H]spiperone
hD2short hD3
160 ( 45
compd
(()-1
R1
Pr
R2
hD2long
220 ( 49
hD4.4
58 ( 18
1600 ( 910
13 000 ( 3200
58 ( 10
1700 ( 400
200 ( 45
6400 ( 1400
90 000 ( 15 000
>20 000
H
highb
lowb
6.7 ( 1.6
200 ( 21
>20 000
>20 000
18 000 ( 5400
>20 000
8500 ( 1300
>20 000
(R)-1
(S)-1
Pr
Pr
H
H
2700 ( 760
4.0 ( 1.3
high
low
high
low
180 ( 31
250 ( 39
>20 000
15 000 ( 1500
120 ( 66
10 000 ( 660
210 ( 78
110 ( 28
8
Pr
Cl
6.1 ( 1.5
230 ( 32
12 000 ( 500
17 000 ( 2 500
4000 ( 0
12 000 ( 0
5.3 ( 1.1
150 ( 16
0.88 ( 0.15
38 ( 7.0
>20 000
>20 000
>20 000
>20 000
>20 000
>20 000
12 000 ( 2600
>100 000
>20 000
>20 000
4 000 ( 0
41 ( 7.0
3000 ( 170
21 ( 5.1
1900 ( 560
35 ( 6.0
12
H
CO2Me
CO2Me
CH2OH
CHO
>100 000
>20 000
13
Pr
Pr
Pr
14
>20 000
16 000 ( 1500
14 000 ( 2000
32 ( 8.5
15
5900 ( 850
52 ( 13
FAUC 54
high
low
high
low
high
low
>20 000c
>20 000c
>20 000c
6900 ( 1800
21 ( 4.9
2500 ( 220
8.1 ( 1.1
pramipexole
quinpirole
6300 ( 1300
63 ( 17
130 ( 22
24 ( 6.3
420 ( 140
1.8 ( 0.093
53 ( 5.9
3100 ( 800
3000 ( 590
a Ki values in nM ( SEM are based on the means of 2-7 experiments each done in triplicate. b Ki values of the high and low affinity
binding state of the receptor when analysis of the binding data resulted in a biphasic dose-response curve. c Determined by using bovine
D1 receptors.
Table 2. Intrinsic Activity of the
a putative binding mode that yields a satisfying expla-
nation of the corresponding D3 affinity.
Tetrahydropyrazolo[1,5-a]pyridines (R)-1, (S)-1, 8, and FAUC
54 Relative to the Reference Compound Quinpirole at the
Dopamine D3 Receptor Established by Measuring the
Stimulation of Mitogenesis
Application of a similarity-based algorithm led to an
alternative alignment of pramipexole displacing the
thiazole ring from π2 toward π1 (Figure 3)10 when the
heterocyclic sp2 nitrogen is closely approaching a1 and
the amino hydrogens can occupy at least one of the
donor areas d1 and d2. However, this modified alignment
is formed at the cost of a less precise match of the
protonated amine and the propyl side chain with their
pharmacophoric areas. Interestingly, the D3 affinity of
(S)-1 was still underestimated (1.10 pKi units) when the
QSAR-derived π1 alignment of pramipexole was trans-
ferred to (S)-1. Considering the inability to predict (S)-1
correctly and the disputable issue concerning the ori-
entation and placement of the side chain within the
propyl-specific receptor cavity, we continued to search
for a binding mode that allows reasonably good predic-
tions for both investigated ligands. We found such an
alternative explanation of the binding properties by
taking into consideration that the ligands may poten-
tially bind to the receptor through a mediated interac-
tion. We were intrigued by the question of whether we
could find indications supporting that water participates
in the receptor binding of (S)-1 and pramipexole. Our
hypothesis was inspired by recent investigations,30
reporting a paradigm of a drug design-relevant media-
tion of ligand-protein interactions between the enzyme
tRNA-guanine transglycosylase and a novel inhibitor by
a water molecule. It is interesting to note that ligands
with a pyrazole partial structure are described to stably
interact with a protein through an associated water
(PDB code: 1UDT, sildenafil-PDE5 complex).31 More-
over, several crystal structures of bovine rhodopsin
(PDB code: 1L9H, 1GZM, and 1U19),32-34 which is the
only G protein-coupled receptor successfully crystallized
test compounds
FAUC
(S)-1 (R)-1
8
54
quinpirole pramipexole
ligand efficacya 82% n.e.b 54% 86%
100%
2.6
93%
1.5
EC50 (nM)c
3.4 4.5 1.1
a Rate of incorporation of [3H]thymidine (in %) as evidence for
a mitogenetic activity relative to the full agonist effect of quinpirole
(100%) as the mean value of quadruplicates from 2 to 8 experi-
ments. b No effect. c EC50 values are derived from the mean curves
of the experiments.
Table 3. Prediction of Binding Affinities for Pramipexole and
(S)-1 in Three Different Alignments Employing an Established
CoMFA Model of D3 Agonists
pramipexole
(S)-1
∆pKi
alignment
pKi
∆pKi
pKi
experiment
prediction
prediction
prediction
9.06
7.45
8.90
8.60
8.40
6.52
7.30
7.68
π2
-1.61
-0.16
-0.46
-1.88
-1.10
-0.72
π1
π2 + H2O
0.719) at an optimum of five components. To obtain
completely unbiased predictions, we removed prami-
pexole from the training set, decreasing the cross-
validated q2 only marginally (q2 ) 0.700).
cv
cv
Following our initial alignment defined by an atom-
based superimposition of the core scaffold of pramipex-
ole or (S)-1 on FAUC 54, we discovered a substantial
underestimation of the predicted binding affinities by
1.61 and 1.88 pKi units for pramipexole and (S)-1,
respectively (Table 3). These prediction errors exceeded
by far the values as expected on the basis of the sPRESS
value of the CoMFA model reflecting that the initial π2
alignment of neither pramipexole nor (S)-1 represents

Heterocyclic Dopamine D3 Receptor Agonists
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 18 5775
Figure 4. Depiction of the water-mediated ligand-receptor
complex of (S)-1 and the “active-state” D3 receptor. The
transmembrane helices (TM3 to TM7) are shown for better
orientation as uniquely colored transparent tubes. The picture
was prepared with VMD1.8.2.
Based on these theoretical calculations, a water-
mediated ligand-receptor interaction appears to be
feasible for (S)-1. Thus, we decided to include a ligand-
associated water molecule in the CoMFA predictions
applying the geometry of the bound water as obtained
from DFT calculations. As given in Table 3, the pKi
value of (S)-1 for this was improved by 1.16 units and
0.38 unit compared to the pure π2 and π1 alignments,
respectively, and deviates by 0.72 unit from the experi-
mental value. Additionally, a reasonable precision of the
pKi prediction (deviation: 0.46 pKi unit) was achieved
for the corresponding complex of pramipexole with
water. Obviously, the relative disposition of the oxygen
function and the electrostatic and steric requirements
are crucial to the binding energy, since (S)-3-hydroxy-
methyl-7-dipropylamino-5,6,7,8-tetrahydroindolizine dis-
played only a pKi of 6.38,11 reflecting that a hydroxy-
methyl group is not able to simulate the water media-
tion. Interestingly, this could be also predicted by our
CoMFA displaying a calculated pKi of 6.49 when em-
ploying a minimal energy conformation with an orienta-
tion of the hydroxymethyl group similar to the π2 + H2O
complex of (S)-1. Consequently, the concept of a water-
mediated ligand-receptor interaction seems to be a
plausible alternative, as it yields satisfying predictions
in the range of the CoMFA sPRESS value for both
compounds. However, it should be recalled that various
enthalpy or entropy related effects contribute to the
total binding energy and, thus, other explanations which
may be not determinable with the CoMFA-methodology
could also be responsible for the high D3 binding of (S)-1
or pramipexole. For instance, the modified π-stacking
interaction of the pyrazole moiety of (S)-1 as compared
to the formyl substituted pyrrole partial structure of
FAUC 54 could be relevant for the binding energy, as
similarly suggested for a series of heteroaromatic hSST4
receptor ligands.36
Figure 3. Depiction of the three different alignment modes
of pramipexole and (S)-1, which were evaluated by the predic-
tion of their corresponding pKi values employing the previously
established 3D-QSAR model. The π2 alignment is generated
by a simple atom-based superimposition of the aromatic
moieties on π2, while the π1 alignment evolved from a similar-
ity based repositioning of pramipexole on pergolide as per-
formed in the original QSAR investigation. The π2 + H2O
alignment reflects the addition of a water molecule mediating
the hydrogen-bonding interactions between the ligand and the
receptor.
up to date, display water in hydrophilic regions of the
binding pocket.
To characterize the bonding geometry and quantify
the energy of water attached to the 5-aminotetrahydro-
pyrazolo[1,5-a]pyridine substructure of (S)-1 relative to
other standard ligands, we utilized a validated quantum
mechanical approach established by Rablen et al.35 Our
calculations yield a distance of 1.94 Å between the sp2
hybridized pyrazole nitrogen and one of the water
hydrogens, which is in the typical range for a hydrogen-
bonding interaction. The related bonding angle toward
the water oxygen R(N-HO) only deviated by 14.3° from
the ideal linear arrangement. The interaction energy
of -∆E ) 7.24 kcal/mol lies in the top range of all the
comparative values published in the original investiga-
tion, in general, and of the interaction energies with sp3/
sp2 nitrogens as acceptors, in particular. For instance,
the bonding energy for an imidazole system (sp2 N as
acceptor) falls short by approximately 0.6 kcal/mol (-∆E
) 6.68 kcal/mol) and for pyridine by almost 1 kcal/mol
(-∆E ) 6.26 kcal/mol).
Based on the alignment of the water-(S)-1 complex
with the QSAR dataset, as used for the prediction of
the binding affinity, we visualized and examined the
putative interactions between (S)-1, water, and the D3
receptor on a molecular level in this active-state model
(Figure 4). The ligand binds as expected through a
reinforced ionic bond with its protonated amine to
Asp3.32, which is completely conserved among biogenic

5776 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 18
Elsner et al.
g; 99 mmol) and methyl propiolate (6.1 g; 72.5 mmol), the
reaction mixture was stirred vigorously at room temperature
overnight. The precipitate was filtered, and the filtrate was
concentrated. The residue was treated with a saturated
NaHCO3 solution and extracted with CH2Cl2. The organic layer
was washed with 1 N HCl and H2O, dried, and evaporated.
Flash chromatography (hexanes-EtOAc 9:1) yielded 3 (6.3 g;
30%) as a pale yellow oil: 1H NMR δ 3.90 (s, 3H), 5.17 (s, 2H),
6.68 (dd, J ) 7.5 Hz, 2.1 Hz, 1H), 7.35-7.50 (m, 5H), 7.53 (d,
J ) 2.1 Hz, 1H), 8.28 (s, 1H), 8.34 (br d, J ) 7.5 Hz, 1H); EIMS
282 (M+). Anal. (C16H14N2O3) C,H,N.
5-Hydroxypyrazolo[1,5-a]pyridine (4). A mixture of 3
(5.00 g, 17.7 mmol) and 48% HBr (60 mL) was refluxed for 1
h. The cooled solution was neutralized (2 N NaOH, NaHCO3)
and extracted with Et2O. The organic layer was dried and
evaporated, and the residue was purified by flash chromatog-
raphy (hexanes-EtOAc 1:1) to give 4 (1.9 g, 80%) as a pale
yellow solid: for experimental data, see ref 15.
4,5,6,7-Tetrahydropyrazolo[1,5-a]pyridin-5-ol (5). To a
solution of 4 (1 g; 7.46 mmol) in EtOH (20 mL) was added 10%
Pd/C (220 mg). The mixture was stirred under 20 bar of
hydrogen and at 80 °C for 16 h. The mixture was filtered, and
the filtrate was evaporated. The residue was purified by flash
chromatography (hexanes-EtOAc 2:3) to afford 5 (814 mg
(79%) as a colorless oil: 1H NMR δ 2.10-2.25 (m, 2H), 2.83
(dd, J ) 16.3 Hz, 6.4 Hz, 1H), 3.10 (dd, J ) 16.3 Hz, 4.6 Hz,
1H), 4.13-4.20 (m, 1H), 4.29-4.37 (m, 2H), 6.01 (d, J ) 1.8
Hz, 1H), 7.44 (d, J ) 1.8 Hz, 1H); EIMS 138 (M+). Anal.
(C7H10N2O) C,H,N.
5-Azido-4,5,6,7-tetrahydropyrazolo[1,5-a]pyridine (6).
Et3N (327 mg, 3.24 mmol) and MsCl (399 mg, 3.51 mmol) were
added to a solution of 5 (404 mg, 2.93 mmol) in THF (25 mL)
at -30 °C. After 1 h, the precipitate was filtered and the
filtrate was evaporated without heating. The residue was
dissolved in DMSO (25 mL), NaN3 (981 mg, 15.1 mmol) was
added, and the reaction mixture was stirred at 35 °C for 24 h.
After the addition of saturated NaHCO3 solution, the mixture
was extracted with Et2O and washed thoroughly with brine.
The organic layer was dried (Na2SO4) and evaporated, and the
residue was purified by flash chromatography (hexanes-
EtOAc 4:1) to give 6 (214 mg, 45%) as a colorless oil: 1H NMR
δ 2.14-2.33 (m, 2H), 2.90 (dd, J ) 16.7 Hz, 6.9 Hz, 1H), 3.14
(dd, J ) 16.7 Hz, 5.0 Hz, 1H), 4.07-4.01 (m, 1H), 4.15-4.23
(m, 1H), 4.27-4.37 (m, 1H), 6.04 (d, J ) 1.8 Hz, 1H), 7.47 (d,
J ) 1.8 Hz, 1H); EIMS 163 (M+). Anal. (C7H11N5) C,H,N.
amine GPCRs. Beyond its hydrogen bond to the pyrazole
nitrogen, the interstitial water can putatively form two
other hydrogen bonds. On one hand, the oxygen can
accept a hydrogen bond from His6.55, and on the other
hand, the free hydrogen can donate one to Ser5.42.
Thus, we can assume that the interstitial water is able
to mediate the binding between the ligand and up to
two residues in TM5 and TM6, alternatively or simul-
taneously. Consequently, it appears to be also reason-
able on the level of molecular interactions that append-
ing a single water molecule to the ligand-receptor
complex can significantly contribute to the high D3
binding affinity of (S)-1, as the lack of interaction
feasibility of the regionally restricted acceptor system
in (S)-1 is compensated.
In accordance with earlier studies,37 residues that
differ between the D3 and D2 receptors being also
located closely to the putative agonist binding site and,
thus, possibly responsible for inducing the D3 receptor
preference of (S)-1 could not be identified.
Conclusions
Employing a 3D-QSAR derived pharmacophore hy-
pothesis for the design of dopamine D3 receptor ago-
nists, a series of novel 5-aminotetrahydropyrazolo[1,5-
a]pyridines was synthesized and evaluated in vitro for
dopamine receptor affinity and ligand efficacy. The
results showed substantial and selective D3 receptor
affinity for the heterocyclic test compound (()-1 when
the enantiomer (S)-1 turned out to be exclusively
responsible for the D3 binding. The low nanomolar
Ki high value for the D3 affinity of (S)-1 (4.0 nM) indicates
that the sp2 nitrogens of the pyrazole and thiazole rings
of the dopaminergics (S)-1 and pramipexole, respec-
tively, are pharmacophoric elements of major impor-
tance. Based on our computational investigations, the
interactions between the endocyclic nitrogen of the
heteroarene subunit and the receptor may be mediated
by an interstitial water molecule.
Experimental Section
5-Amino-4,5,6,7-tetrahydropyrazolo[1,5-a]pyridine ((()-
7). A mixture of 6 (200 mg, 1.23 mmol) and 10% Pd/C (100
mg) in MeOH was stirred under a balloon of H2 at room
temperature for 2.5 h. The mixture was filtered through Celite,
the filtrate was evaporated carefully, and the residue was
purified by flash chromatography (CH2Cl2-MeOH 4:1) to give
(()-7 (134 mg, 80%) as a colorless oil: 1H NMR δ 1.88-1.99
(m, 1H), 2.13-2.21 (m, 1H), 2.55 (dd, J ) 16.1 Hz, 8.8 Hz,
1H), 3.08 (dd, J ) 16.1 Hz, 5.1 Hz, 1H), 3.35 (dddd, J ) 9.7
Hz, 8.8 Hz, 5.1 Hz, 3.0 Hz, 1H), 4.12 (ddd, J ) 12.9 Hz, 10.0
Hz, 5.2 Hz, 1H), 4.34 (ddd, J ) 12.9 Hz, 5.6 Hz, 4.5 Hz, 1H),
5.99 (br d, J ) 1.8 Hz, 1H), 7.45 (d, J ) 1.8 Hz, 1H); EIMS
137 (M+). Anal. (C7H11N3) C,H,N.
5-Dipropylamino-4,5,6,7-tetrahydropyrazolo[1,5-a]py-
ridine ((()-1). To a solution of (()-7 (20 mg, 0.15 mmol) in
1,2-dichloroethane (3 mL) were added propionaldehyde (17.6
mg, 0.31 mmol) and Na(OAc)3BH (81.4 mg, 0.39 mmol), and
the reaction mixture was stirred at room temperature for 16
h. After the addition of saturated NaHCO3 solution, the organic
layer was separated and the aqueous layer was extracted twice
with CH2Cl2. The combined organic layers were dried (Na2-
SO4) and evaporated, and the residue was purified by flash
chromatography (CH2Cl2-MeOH 19:1) to give (()-1 (27 mg,
84%) as a colorless oil: 1H NMR δ 0.89 (t, J ) 7.4 Hz, 6H),
1.41-1.51 (m, 4H), 1.91-2.03 (m, 1H), 2.11-2.18 (m, 1H), 2.46
(t, J ) 7.4 Hz, 4H), 2.67 (dd, J ) 16.0 Hz, 11.2 Hz, 1H), 2.96
(dd, J ) 16.0 Hz, 4.9 Hz, 1H), 3.03-3.12 (m, 1H), 4.04 (ddd, J
) 12.7 Hz, 12.4 Hz, 4.7 Hz, 1H), 4.38 (ddd, J ) 12.7 Hz, 5.7
Chemistry. All reactions were carried out under nitrogen
atmosphere except 1,3-dipolar cycloadditions, ester hydrolyses,
and catalytic hydrogenations. Solvents were purified and dried
by standard procedures. All reagents were of commercial
quality and used as purchased. MS were run on a Finnegan
MAT TSQ 70 spectrometer by EI (70 eV) with solid inlet. R20
D
values were measured on a Perkin-Elmer 241 instrument. The
1H NMR spectra were obtained on a Bruker AM 360 (360 MHz)
spectrometer, if not otherwise stated in CDCl3 relative to TMS
(J values in Hz). IR spectra were performed on a Jasco FT/IR
410 spectrometer. Purification by chromatography was per-
formed using Silica Gel 60; TLC analyses were performed
using Merck 60 F254 aluminum sheets and analyzed by UV
light (254 nm) or in the presence of iodine. Preparative and
analytical HPLC was performed on an Agilent 1100 prepara-
tive HPLC system employing a VWL detector. CHN elemen-
tary analyses were performed at the Department of Organic
Chemistry of the Friedrich Alexander University.
Methyl 5-Benzyloxypyrazolo[1,5-a]pyridine-3-carbox-
ylate (3). To an ice-cooled solution of 4-benzyloxypyridine (13.8
g; 74.5 mmol) in CH2Cl2 (75 mL) was added dropwise a solution
of O-mesitylenesulfonylhydroxylamine (90% purity; 17.8 g;
74.5 mmol) in CH2Cl2 (75 mL). The reaction mixture was
stirred at 0 °C for 1 h and at room temperature for 1 h. After
the addition of Et2O, the separated oil was washed with Et2O
and dried under reduced pressure. The crude product (2) was
dissolved in DMF (100 mL). After the addition of K2CO3 (13.7

Heterocyclic Dopamine D3 Receptor Agonists
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 18 5777
Hz, 2.3 Hz, 1H), 5.96 (br d, J ) 1.8 Hz, 1H), 7.42 (d, J ) 1.8
Methyl 5-Oxo-4,5,6,7-tetrahydropyrazolo[1,5-a]pyridine-
3-carboxylate (11). To a solution of 10 (161 mg, 0.82 mmol)
in CH2Cl2 (10 mL) was added a solution of DMP (Dess-Martin
periodinane; 452 mg; 1.07 mmol) at room temperature. After
2 h, an aqueous NaHCO3/NaS2O3 solution (1:1) and Et2O (100
mL) were added and stirring was continued for 10 min. The
organic layer was separated, and the aqueous layer was
extracted twice with Et2O. The combined organic layers were
dried (Na2SO4) and evaporated, and the residue was purified
by flash chromatography (hexanes-EtOAc 9:1) to give 11 (112
mg, 70%) as a pale reddish oil: 1H NMR δ 2.91 (t, J ) 6.4 Hz,
2H), 3.83 (s, 3H), 4.00 (s, 2H), 4.55 (t, J ) 6.4 Hz, 2H), 7.92 (s,
1H); EIMS 194 (M+). Anal. (C9H10N2O3) C,H,N.
Methyl 5-Propylamino-4,5,6,7-tetrahydropyrazolo[1,5-
a]pyridine-3-carboxylate (12). To a solution of 11 (60 mg,
0.31 mmol) in MeOH (6 mL) were added propylamine (21.6
mg, 0.37 mmol) and NaCNBH3 (38.8 mg, 0.62 mmol), and the
reaction mixture was stirred at room temperature for 16 h.
After the addition of 1 N HCl, saturated NaHCO3 solution and
Et2O were added. The organic layer was dried (Na2SO4) and
evaporated, and the residue was purified by flash chromatog-
raphy (CH2Cl2-MeOH 19:1) to give 12 (44 mg, 60%) as a
colorless oil: 1H NMR δ 0.94 (t, J ) 7.5 Hz, 3H), 1.47-1.58
(m, 2H), 1.91-2.01 (m, 1H), 2.15-2.23 (m, 1H), 2.67 (t, J )
6.7 Hz, 2H), 2.79 (dd, J ) 17.4 Hz, 7.8 Hz, 1H), 3.09-3.16 (m,
1H), 3.40 (dd, J ) 17.4 Hz, 5.0 Hz, 1H), 3.81 (s, 3H), 4.12 (ddd,
J ) 13.0 Hz, 8.8 Hz, 4.9 Hz, 1H), 4.32 (ddd, J ) 13.0 Hz, 5.4
Hz, 5.2 Hz, 1H), 7.86 (s, 1H); EIMS 237 (M+). Anal. (C12H19N3O2)
C,H,N.
Methyl 5-Dipropylamino-4,5,6,7-tetrahydropyrazolo-
[1,5-a]pyridine-3-carboxylate (13). To a solution of 12 (29.9
mg, 0.13 mmol) in 1,2-dichloroethane (3 mL) were added
propionaldehyde (8.7 mg, 0.15 mmol) and Na(OAc)3BH (48 mg,
0.23 mmol), and the reaction mixture was stirred at room
temperature for 16 h. After the addition of saturated NaHCO3
solution, the organic layer was separated and the aqueous
layer was extracted twice with CH2Cl2. The combined organic
layers were dried (Na2SO4) and evaporated, and the residue
was purified by flash chromatography (CH2Cl2-MeOH 19:1)
to give 13 (27.5 mg, 78%) as a colorless oil: 1H NMR δ 0.89 (t,
J ) 7.5 Hz, 6H), 1.41-1.51 (m, 4H), 1.93-2.05 (m, 1H), 2.12-
2.19 (m, 1H), 2.48 (t, J ) 7.3 Hz, 4H), 2.81 (dd, J ) 17.4 Hz,
11.7 Hz, 1H), 3.06-3.14 (m, 1H), 3.35 (dd, J ) 17.4 Hz, 5.0
Hz, 1H), 3.81 (s, 3H), 4.08 (ddd, J ) 13.0 Hz, 12.3 Hz, 4.7 Hz,
1H), 4.37 (ddd, J ) 13.0 Hz, 5.5 Hz, 2.5 Hz, 1H), 7.85 (s, 1H);
EIMS 279 (M+). Anal. (C15H25N3O2) C,H,N.
(5-Dipropylamino-4,5,6,7-tetrahydropyrazolo[1,5-a]py-
rid-3-yl)-methanol (14). A solution of 13 (20 mg; 0.072 mmol)
in THF (2 mL) was treated with LiAlH4 (1 M in THF, 72 µL,
0.072 mmol) at room temperature and stirred for 1.5 h. After
the addition of 0.2 mL of H2O, the suspension was filtered and
washed with Et2O. The organic layer was dried (Na2SO4) and
evaporated, and the residue was purified by flash chromatog-
raphy (CH2Cl2-MeOH 9:1) to give 14 (15 mg, 82%) as a
colorless oil: 1H NMR δ 0.89 (t, J ) 7.5 Hz, 6H), 1.41-1.51
(m, 4H), 1.91-2.03 (m, 1H), 2.11-2.18 (m, 1H), 2.47 (t, J )
7.5 Hz, 4H), 2.64 (dd, J ) 16.0 Hz, 11.4 Hz, 1H), 3.00 (dd, J )
16.0 Hz, 5.0 Hz, 1H), 3.05-3.13 (m, 1H), 4.04 (ddd, J ) 12.8
Hz, 12.4 Hz, 4.6 Hz, 1H), 4.36 (ddd, J ) 12.8 Hz, 5.7 Hz, 2.1
Hz, 1H), 4.51 (d, J ) 3.2 Hz, 2H), 7.44 (s, 1H); EIMS 251 (M+).
Anal. (C14H25N3O) C,H,N.
Hz, 1H); EIMS 221 (M+). Anal. (C13H23N3) C,H,N.
2-Chloro-5-dipropylamino-4,5,6,7-tetrahydropyrazolo-
[1,5-a]pyridine (8). A mixture of (()-1 (12.7 mg, 0.06 mmol)
and N-chlorosuccinimide (15,3 mg, 0.12 mmol) in CHCl3 (4 mL)
was refluxed for 16 h. The solvent was evaporated, and the
residue was purified by flash chromatography (CH2Cl2-MeOH
19:1) to give 8 (10 mg, 67%) as a pale yellow oil: 1H NMR δ
0.89 (t, J ) 7.4 Hz, 6H), 1.41-1.51 (m, 4H), 1.91-2.02 (m, 1H),
2.10-2.17 (m, 1H), 2.47 (t, J ) 7.4 Hz, 4H), 2.54 (dd, J ) 16.3
Hz, 10.7 Hz, 1H), 2.90 (dd, J ) 16.3 Hz, 5.0 Hz, 1H), 3.04-
3.12 (dddd, m, 1H), 4.02 (ddd, J ) 12.7 Hz, 12.3 Hz, 4.6 Hz,
1H), 4.02 (ddd, J ) 12.7 Hz, 5.4 Hz, 2.3 Hz, 1H), 7.37 (s, 1H);
EIMS 255 (M+). Anal. (C13H22N3) C,H,N.
(1R,2S,5R)-Menthyl N-(4,5,6,7-tetrahydropyrazolo[1,5-
a]pyrid-5-yl)-carbamate (9). To a solution of (()-7 (110 mg,
0.80 mmol) and diisopropylethylamine (156 mg, 1.21 mmol)
in CH2Cl2 (3 mL) was added slowly (1R,2S,5R)-(-)-menthyl
chloroformate (265 mg, 1.21 mmol) at 0 °C. The mixture was
stirred at room temperature for 16 h. After the addition of
saturated NaHCO3 solution, the organic layer was separated
and the aqueous layer was extracted thrice with CH2Cl2. The
combined organic layers were dried (Na2SO4) and evaporated,
and the residue was purified by flash chromatography (CH2-
Cl2-MeOH 97:3) to afford a diastereomeric mixture of 9 (197
mg, 77%) as colorless crystals.
The diastereoisomers were separated by preparative HPLC
using a normal phase column (Luna 5 µm silica). The mobile
phase used was 15% acetonitrile in diisopropyl ether with a
flow rate of 21.2 mL/min. The two fractions were eluted with
retention times of 17.20 min for the (5R)-isomer ((5R)-9) and
18.30 min for the (5S)-isomer ((5S)-9). Final purity of the
separated diastereoisomers was checked by an analytical
normal phase column (Luna 5 µm silica) using the same mobile
phase with a flow rate of 1 mL/min. Pure diastereomers (5R)-9
and (5S)-9 were eluted at 15.87 and 16.96 min, respectively.
Compound (5R)-9 was isolated (80 mg, 31%) as colorless
crystals: 1H NMR δ 0.79-0.97 (m, 10 H), 1.03-1.10 (m, 1H),
1.27-1.69 (m, 5H), 1.86-1.91 (m, 1H), 2.02-2.06 (m, 1H),
2.09-2.16 (m, 1H), 2.24-2.29 (m, 1H), 2.70 (dd, J ) 16.2 Hz,
7.6 Hz, 1H), 3.19 (dd, J ) 16.2 Hz, 4.9 Hz, 1H), 4.14 (br s,
1H), 4.19-4.28 (m, 2H), 4.57-4.66 (m, 2H), 6.01 (br d, J )
1.9 Hz, 1H), 7.46 (d, J ) 1.9 Hz, 1H); EIMS 319 (M+). Anal.
(C18H29N3O2) C,H,N. Compound (5S)-9 was isolated (80 mg,
31%) as colorless crystals.
(R)-5-Amino-4,5,6,7-tetrahydropyrazolo[1,5-a]pyri-
dine ((R)-7). To a solution of (5R)-9 (70 mg, 0.22 mmol) in
CHCl3 (3 mL) was added iodotrimethylsilane (188 mg, 0.94
mmol), and the reaction mixture was refluxed for 12 h. After
cooling to room temperature, the mixture was treated with
MeOH (3 mL) and refluxed for an additional 12 h. The solvent
was evaporated, and the residue was purified by flash chro-
matography (CH2Cl2-MeOH 4:1) to give (R)-7 (27 mg, 89%)
as a colorless oil: [R]20 ) +55.7° (c ) 0.5, CHCl3). (S)-7 (26
D
mg, 87%) ([R]20D ) -55.4° (c ) 0.5, CHCl3)) was prepared from
(5S)-9 as described above for (R)-7.
(R)-5-Dipropylamino-4,5,6,7-tetrahydropyrazolo[1,5-a]-
pyridine ((R)-1). (R)-1 (20.6 mg, 85%) was prepared as
described for (()-1 using (R)-7 (15 mg, 0.11 mmol) in 2 mL of
1,2-dichloroethane, propionaldehyde (13.2 mg, 0.23 mmol), and
Na(OAc)3BH (61.1 mg, 0.29 mmol): [R] ) +25.2° (c ) 0.3,
CHCl3). (S)-1 (19.4 mg, 80%) ([R]20 ) -25° (c ) 0.3, CHCl3))
D
5-Dipropylamino-4,5,6,7-tetrehydropyrazolo[1,5-a]py-
ridine-3-carbaldehyde (15). To a solution of 14 (12 mg, 0.048
mmol) in CH2Cl2 (2 mL) was added MnO2 (42.5 mg, 0.49
mmol), and the reaction mixture was stirred at room temper-
ature for 16 h. The suspension was filtered, and the filtrate
was evaporated. Flash chromatography (CH2Cl2-MeOH 95:
5) yielded 15 (11.5 mg, 97%) as a pale yellow oil: 1H NMR δ
0.89 (t, J ) 7.5 Hz, 6H), 1.41-1.51 (m, 4H), 1.98-2.09 (m, 1H),
2.14-2.21 (m, 1H), 2.47 (t, J ) 7.5 Hz, 4H), 2.88 (dd, J ) 17.4
Hz, 10.6 Hz, 1H), 3.09-3.17 (m, 1H), 3.34 (dd, J ) 17.4 Hz,
5.0 Hz, 1H), 4.11 (ddd, J ) 13.3 Hz, 12.5 Hz, 4.5 Hz, 1H), 4.39
(ddd, J ) 13.3 Hz, 5.5 Hz, 2.5 Hz, 1H), 7.88 (s, 1H), 9.85 (s,
1H); EIMS 249 (M+). Anal. (C14H23N3O) C,H,N.
was prepared from (S)-7 as described above for (R)-1.
Methyl 5-Hydroxy-4,5,6,7-tetrahydropyrazolo[1,5-a]-
pyridine-3-carboxylate (10). To a solution of 3 (850 mg; 3.01
mmol) in EtOH (12 mL) was added 10% Pd/C (200 mg). The
mixture was hydrogenated in an autoclave under 20 bar of
hydrogen at 80 °C for 48 h. The mixture was filtered, and the
filtrate was evaporated. The residue was purified by flash
chromatography (hexanes-EtOAc 1:1) to afford 10 (443 mg,
75%) as a pale yellow oil: 1H NMR δ 2.29-2.42 (m, 2H), 2.82
(dd, J ) 17.0 Hz, 9.6 Hz, 1H), 3.66 (dd, J ) 17.0 Hz, 5.3 Hz,
1H), 3.81 (s, 3H), 4.05-4.34 (m, 3H), 7.89 (s, 1H); EIMS 196
(M+). Anal. (C9H12N2O3) C,H,N.

5778 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 18
Elsner et al.
Dopamine Receptor Binding Studies. Receptor binding
studies were carried out as described.18 In brief, the dopamine
D1 receptor assay was done with porcine striatal membranes
at a final protein concentration of 40 µg/assay tube and the
radioligand [3H]SCH 23390 at 0.3 nM (Kd ) 0.62-0.95 nM).
Competition experiments with the human D2long, D2short, D3,
and D4.4 receptors were run with preparations of membranes
from CHO cells stably expressing the corresponding receptor
and [3H]spiperone at a final concentration of 0.5 nM. The
assays were carried out with a protein concentration of 5-20
µg/assay tube and Kd values of 0.06-0.14 nM, 0.09-0.15 nM,
0.08-0.35 nM, and 0.15-0.40 nM for the D2long, D2short, D3,
and D4.4 receptors, respectively. Protein concentration was
established by the method of Lowry using bovine serum
albumin as standard.38
5-HT and r1 Receptor Binding Studies. Receptor bind-
ing experiments were performed with homogenates prepared
from porcine cerebral cortex as described.39 Assays were run
with membranes at a protein concentration per each assay
tube of 100, 80, and 60 µg/mL for 5-HT1A, 5-HT2, and R1
receptor, respectively, and Kd values of 2.3 nM for 5-HT1A, 1.7
nM for 5-HT2, and 0.05-0.10 nM for the R1 receptor. A
screening system was established to investigate serotonin
receptor binding when measuring the displacement of the
radioligands [3H]8-OH-DPAT and [3H]ketanserin (each at a
final concentration of 0.5 nM) from 5-HT1A and 5-HT2 carrying
membranes by the test compounds at final concentrations of
10 µM, 100 nM, and 1 nM in triplicate. Determination of R1
affinity was done in a competition experiment with the
radioligand [3H]prazosin at a final concentration of 0.4 nM.
Mitogenesis Assay. The mitogenesis experiments were
done with a CHO dhfr- cell line stably transfected with the
human dopamine D3 receptor according to literature.20,40,41 In
brief, cells were grown in DMEM supplemented with dialyzed
fetal calf serum, amino acid supplement, L-glutamine, penicil-
lin G, and streptomycin at 37 °C under a humidified atmo-
sphere of 5% CO2-95% air at a density of 10 000 cells/well.
After 72 h the growth medium was removed and the cells were
rinsed twice with serum free medium. Incubation was started
by adding eight different concentrations of the test compounds
as quadruplicates (with a final concentration of 0.01-10 000
nM) diluted in 10 µL of sterile water to each well containing
90 µL serum free medium. After incubation for 16-18 h, 0.02
µCi of [3H]thymidine (specific activity 25 Ci/mmol, Amersham
Biosciences) in 10 µL of serum free medium was added to each
well for 4 h at 37 °C. Finally, cells were trypsinized and
harvested onto GF/C filters using an automated cell harvester.
Filters were washed five times with ice-cold PBS buffer and
counted in a microplate scintillation counter.
Data Analysis. The resulting competition curves of the
receptor binding experiments were analyzed by nonlinear
regression using the algorithms in PRISM 3.0 (GraphPad
Software, San Diego, CA). The data were initially fit using a
sigmoid model to provide a slope coefficient (nH) and an IC50
value, representing the concentration corresponding to 50%
of maximal inhibition. Data were then calculated for a one-
site (nH ∼ 1) or a two-site model (nH < 1) depending on the
slope factor. IC50 values were transformed to Ki values or to
Ki high and Ki low in the fact when a two-site model was preferred
after fitting mono- and biphasic curves according to the
equation of Cheng and Prusoff.42
tions.35 Treatment of the electron correlation with either
density functional theory (B3LYP) or second-order Møller-
Plesset theory (MP2) was demonstrated to yield similar and
reliable interaction energies, when using an appropriate basis
set. Thus, due to the more efficient use of computing resources
and the smaller basis set superposition error (BSSE) reported
for the DFT method, we applied Becke’s three parameter
hybrid method using the correlation functional of Lee, Yang,
and Parr (B3LYP). Moreover, Rablen et al. demonstrated that
structure optimization with a 6-31+G(d(X+),p) basis set and
subsequent single point calculation using a 6-31++G(2d-
(X+),p) basis set achieves the desired level of accuracy in dipole
moment and dimerization energy of water. Thereby, an ad-
ditional set of diffuse d polarization functions, referred to as
“X+”, are included only on atoms having lone pairs. The
starting geometry of water in complex with (S)-1 was derived
from the optimized structure of the pyridine-water complex
reported in the original publication. To facilitate the compari-
son of our results with the geometry and energy data previ-
ously reported, all calculations were performed in exact
compliance with the original method and the homogeneity with
the literature data was proven by sample calculations.
Acknowledgment. The authors wish to thank Dr.
H. H. M. Van Tol (Clarke Institute of Psychiatry,
Toronto), Dr. J.-C. Schwartz, and Dr. P. Sokoloff (IN-
SERM, Paris) as well as J. Shine (The Garvan Institute
of Medical Research, Sydney) for providing dopamine
D4, D3, and D2 receptor expressing cell lines, respec-
tively. This work was supported by the BMBF and
Fonds der Chemischen Industrie.
Supporting Information Available: X-ray crystallo-
graphic data of (5R)-9, elemental analyses, and IR spectro-
scopic data. This material is available free of charge via the
Internet at http://pubs.acs.org.
References
(1) Sibley, D. R.; Monsma, F. J., Jr. Molecular biology of dopamine
receptors. Trends Pharmacol. Sci. 1992, 13, 61-69.
(2) Sokoloff, P.; Giros, B.; Martres, M. P.; Bouthenet, M. L.;
Schwartz, J. C. Molecular cloning and characterization of a novel
dopamine receptor (D3) as a target for neuroleptics. Nature 1990,
347, 146-151.
(3) Dikeos, D. G.; Papadimitriou, G. N.; Avramopoulos, D.; Karadi-
ma, G.; Daskalopoulou, E. G.; Souery, D.; Mendlewicz, J.;
Vassilopoulos, D.; Stefanis, C. N. Association between the
dopamine D3 receptor gene locus (DRD3) and unipolar affective
disorder. Psychiatr. Genet. 1999, 9, 189-195.
(4) Richtand, N. M.; Woods, S. C.; Berger, S. P.; Strakowski, S. M.
D3 dopamine receptor, behavioral sensitization, and psychosis.
Neurosci. Biobehav. Rev. 2001, 25, 427-443.
(5) Joyce, J. N.; Ryoo, H. L.; Beach, T. B.; Caviness, J. N.; Stacy,
M.; Gurevich, E. V.; Reiser, M.; Adler, C. H. Loss of response to
levodopa in Parkinson’s disease and co-occurrence with demen-
tia: role of D3 and not D2 receptors. Brain Res. 2002, 955, 138-
152.
(6) Bezard, E.; Ferry, S.; Mach, U.; Stark, H.; Leriche, L.; Boraud,
T.; Gross, C.; Sokoloff, P. Attenuation of levodopa-induced
dyskinesia by normalizing dopamine D3 receptor function. Nat.
Med. 2003, 9, 762-767.
(7) Pilla, M.; Perachon, S.; Sautel, F.; Garrido, F.; Mann, A.;
Wermuth, C. G.; Schwartz, J. C.; Everitt, B. J.; Sokoloff, P.
Selective inhibition of cocaine-seeking behaviour by a partial
dopamine D3 receptor agonist. Nature 1999, 400, 371-375.
(8) Palczewski, K.; Kumasaka, T.; Hori, T.; Behnke, C. A.; Mo-
toshima, H.; Fox, B. A.; Le Trong, I.; Teller, D. C.; Okada, T.;
Stenkamp, R. E.; Yamamoto, M.; Miyano, M. Crystal structure
of rhodopsin: A G protein-coupled receptor. Science 2000, 289,
739-745.
(9) Boeckler, F.; Lanig, H.; Gmeiner, P. Modeling the Similarity and
Divergence of Dopamine D2-like Receptors and Identification
of Validated Ligand-Receptor Complexes. J. Med. Chem. 2005,
48, 694-709.
(10) Boeckler, F.; Ohnmacht, U.; Lehmann, T.; Utz, W.; Hu¨bner, H.;
Gmeiner, P. CoMFA and CoMSIA Investigations Revealing
Novel Insights into the Binding Modes of Dopamine D3 Receptor
Agonists. J. Med. Chem. 2005, 48, 2493-2508.
Experimental data resulting from the mitogenesis assay
were each normalized and then combined to get a mean curve.
Nonlinear regression analysis of this curve provided an EC50
value indicating the concentration of the test compound which
induced half of the maximal uptake of [3H]thymidine. The top
value of the curve represented the maximal intrinsic activity
which could be correlated to the effect of the full agonist
quinpirole (100%).
Modeling. To characterize the geometry and energy of the
hydrogen-bond interaction of the water-(S)-1 complex and to
evaluate the strength of the interaction in comparison to other
hydrogen-bond complexes, we employed a validated quantum-
chemical method reported by Rablen et al. for our calcula-

Heterocyclic Dopamine D3 Receptor Agonists
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 18 5779
(11) Lehmann, T.; Hu¨bner, H.; Gmeiner, P. Dopaminergic 7-amino-
tetrahydroindolizines: ex-chiral pool synthesis and preferential
D3 receptor binding. Bioorg. Med. Chem. Lett. 2001, 11, 2863-
2866.
(12) Gmeiner, P.; Lerche, H. New and efficient synthesis of 5,6,7,8-
tetrahydroindolizidines. Application to the synthesis of phar-
macologically relevant chiral amino derivatives from L-aspar-
agine. Heterocycles 1990, 31, 9-12.
(13) Dijkstra, D.; Rodenhuis, N.; Vermeulen, E. S.; Pugsley, T. A.;
Wise, L. D.; Wikstrom, H. V. Further characterization of
structural requirements for ligands at the dopamine D(2) and
D(3) receptor: exploring the thiophene moiety. J. Med. Chem.
2002, 45, 3022-3031.
(14) Tamura, Y.; Minamikawa, J.; Ikeda, M. O-mesitylenesulfonyl-
hydroxylamine and related compoundsspowerful aminating
reagents. Synthesis 1977, 1-17.
(15) Brown, R. F. C.; Eastwood, F. W.; Fallon, G. D.; Lee, S. C.;
McGeary, R. P. The pyrolytic rearrangement of 1-alkynoyl-3-
methylpyrazoles: synthesis of pyrazolo[1,5-a]pyridin-5-ols and
related compounds. Aust. J. Chem. 1994, 47, 991-1007.
(16) Olah, G. A.; Narang, S. C. Iodotrimethylsilane, a versatile
synthetic reagent. Tetrahedron 1982, 38, 2225-2277.
(17) Dess, D. B.; Martin, J. C. A useful 12-I-5 triacetoxyperiodinane
(the Dess-Martin periodinane) for the selective oxidation of
primary or secondary alcohols and a variety of related 12-I-5
species. J. Am. Chem. Soc. 1991, 113, 7277-7287.
(18) Hu¨bner, H.; Haubmann, C.; Utz, W.; Gmeiner, P. Conjugated
enynes as nonaromatic catechol bioisosteres: synthesis, binding
experiments, and computational studies of novel dopamine
receptor agonists recognizing preferentially the D(3) subtype.
J. Med. Chem. 2000, 43, 756-762.
(19) Hayes, G.; Biden, T. J.; Selbie, L. A.; Shine, J. Structural
subtypes of the dopamine D2 receptor are functionally distinct:
expression of the cloned D2A and D2B subtypes in a heterologous
cell line. Mol. Endocrinol. 1992, 6, 920-926.
(20) Sokoloff, P.; Andrieux, M.; Besancon, R.; Pilon, C.; Martres, M.
P.; Giros, B.; Schwartz, J. C. Pharmacology of human dopamine
D3 receptor expressed in a mammalian cell line: comparison
with D2 receptor. Eur. J. Pharmacol. 1992, 225, 331-337.
(21) Asghari, V.; Sanyal, S.; Buchwaldt, S.; Paterson, A.; Jovanovic,
V.; Van Tol, H. H. Modulation of intracellular cyclic AMP levels
by different human dopamine D4 receptor variants. J. Neuro-
chem. 1995, 65, 1157-1165.
(22) Schneider, C. S.; Mierau, J. Dopamine autoreceptor agonists:
resolution and pharmacological activity of 2,6-diaminotetrahy-
drobenzothiazole and an aminothiazole analogue of apomor-
phine. J. Med. Chem. 1987, 30, 494-498.
(23) van Vliet, L. A.; Rodenhuis, N.; Dijkstra, D.; Wikstrom, H.;
Pugsley, T. A.; Serpa, K. A.; Meltzer, L. T.; Heffner, T. G.; Wise,
L. D.; Lajiness, M. E.; Huff, R. M.; Svensson, K.; Sundell, S.;
Lundmark, M. Synthesis and pharmacological evaluation of
thiopyran analogues of the dopamine D3 receptor-selective
agonist (4aR,10bR)-(+)-trans-3,4,4a,10b-tetrahydro-4-n-propyl-
2H,5H-[1]benzopyrano[4,3-b]-1,4-oxazin-9-ol (PD 128907). J.
Med. Chem. 2000, 43, 2871-2882.
(24) Wikstrom, H.; Andersson, B.; Sanchez, D.; Lindberg, P.; Arvids-
son, L. E.; Johansson, A. M.; Nilsson, J. L.; Svensson, K.; Hjorth,
S.; Carlsson, A. Resolved monophenolic 2-aminotetralins and
1,2,3,4,4a,5,6,10b-octahydrobenzo[f]quinolines: structural and
stereochemical considerations for centrally acting pre- and
postsynaptic dopamine-receptor agonists. J. Med. Chem. 1985,
28, 215-225.
(26) Gmeiner, P.; Mierau, J.; Hofner, G. Enantiomerically pure
aminoindolizines: bicyclic ergoline analogues with dopamine
autoreceptor activity. Arch. Pharm. (Weinheim) 1992, 325, 57-
60.
(27) Bergauer, M.; Hu¨bner, H.; Gmeiner, P. Practical ex-chiral-pool
methodology for the synthesis of dopaminergic tetrahydroin-
doles. Tetrahedron 2004, 60, 1197-1204.
(28) Mierau, J.; Schneider, F. J.; Ensinger, H. A.; Chio, C. L.; Lajiness,
M. E.; Huff, R. M. Pramipexole binding and activation of cloned
and expressed dopamine D2, D3 and D4 receptors. Eur. J.
Pharmacol. 1995, 290, 29-36.
(29) Chio, C. L.; Lajiness, M. E.; Huff, R. M. Activation of heterolo-
gously expressed D3 dopamine receptors: comparison with D2
dopamine receptors. Mol. Pharmacol. 1994, 45, 51-60.
(30) Brenk, R.; Naerum, L.; Gradler, U.; Gerber, H. D.; Garcia, G.
A.; Reuter, K.; Stubbs, M. T.; Klebe, G. Virtual screening for
submicromolar leads of tRNA-guanine transglycosylase based
on a new unexpected binding mode detected by crystal structure
analysis. J. Med. Chem. 2003, 46, 1133-1143.
(31) Sung, B. J.; Hwang, K. Y.; Jeon, Y. H.; Lee, J. I.; Heo, Y. S.;
Kim, J. H.; Moon, J.; Yoon, J. M.; Hyun, Y. L.; Kim, E.; Eum, S.
J.; Park, S. Y.; Lee, J. O.; Lee, T. G.; Ro, S.; Cho, J. M. Structure
of the catalytic domain of human phosphodiesterase 5 with
bound drug molecules. Nature 2003, 425, 98-102.
(32) Okada, T.; Fujiyoshi, Y.; Silow, M.; Navarro, J.; Landau, E. M.;
Shichida, Y. Functional role of internal water molecules in
rhodopsin revealed by X-ray crystallography. Proc. Natl. Acad.
Sci. U.S.A. 2002, 99, 5982-5987.
(33) Li, J.; Edwards, P. C.; Burghammer, M.; Villa, C.; Schertler, G.
F. Structure of bovine rhodopsin in a trigonal crystal form. J.
Mol. Biol. 2004, 343, 1409-1438.
(34) Okada, T.; Sugihara, M.; Bondar, A. N.; Elstner, M.; Entel, P.;
Buss, V. The retinal conformation and its environment in
rhodopsin in light of a new 2.2 Å crystal structure. J. Mol. Biol.
2004, 342, 571-583.
(35) Rablen, P. R.; Lockman, J. W.; Jorgensen, W. L. Ab Initio Study
of Hydrogen-Bonded Complexes of Small Organic Molecules with
Water. J. Phys. Chem. A 1998, 102, 3782-3797.
(36) Angeles, A. R.; Neagu, I.; Birzin, E. T.; Thornton, E. R.; Smith,
A. B., 3rd; Hirschmann, R. Synthesis and binding affinities of
novel SRIF-mimicking â-D-glucosides satisfying the requirement
for a π-cloud at C1. Org. Lett. 2005, 7, 1121-1124.
(37) Lundstrom, K.; Turpin, M. P.; Large, C.; Robertson, G.; Thomas,
P.; Lewell, X. Q. Mapping of dopamine D3 receptor binding site
by pharmacological characterization of mutants expressed in
CHO cells with the Semliki Forest virus system. J. Recept.
Signal Transduction Res. 1998, 18, 133-150.
(38) Lowry, O. H.; Rosebrough, N. J.; Farr, A. L.; Randall, R. J.
Protein measurement with the Folin phenol reagent. J. Biol.
Chem. 1951, 193, 265-275.
(39) Heindl, C.; Hu¨bner, H.; Gmeiner, P. Ex-chiral pool synthesis and
receptor binding studies of 4-substituted prolinol derivatives.
Tetrahedron: Asymmetry 2003, 14, 3141-3152.
(40) Hu¨bner, H.; Kraxner, J.; Gmeiner, P. Cyanoindole derivatives
as highly selective dopamine D(4) receptor partial agonists:
solid-phase synthesis, binding assays, and functional experi-
ments. J. Med. Chem. 2000, 43, 4563-4569.
(41) Bettinetti, L.; Schlotter, K.; Hu¨bner, H.; Gmeiner, P. Interactive
SAR studies: rational discovery of super-potent and highly
selective dopamine D3 receptor antagonists and partial agonists.
J. Med. Chem. 2002, 45, 4594-4597.
(25) Titus, R. D.; Kornfeld, E. C.; Jones, N. D.; Clemens, J. A.;
Smalstig, E. B.; Fuller, R. W.; Hahn, R. A.; Hynes, M. D.; Mason,
N. R.; Wong, D. T.; Foreman, M. M. Resolution and absolute
configuration of an ergoline-related dopamine agonist, trans-4,-
4a,5,6,7,8,8a,9-Octahydro-5-propyl-1H(or 2H)-pyrazolo[3,4-g]-
quinoline. J. Med. Chem. 1983, 26, 1112-1116.
(42) Cheng, Y.; Prusoff, W. H. Relationship between the inhibition
constant (K1) and the concentration of inhibitor which causes
50 per cent inhibition (I50) of an enzymatic reaction. Biochem.
Pharmacol. 1973, 22, 3099-3108.
JM0503805