Selective Agonists for Dopamine/Neurotensin Receptor Heterodimers

Article abstract of DOI:10.1002/cmdc.201100499

The neuromodulatory peptide neurotensin has been described to functionally interact with dopaminergic pathways of the human brain. We employed radioligand binding studies to investigate the physical interaction between co-expressed dopamine D_2L or D₃ and neurotensin NTS₁ or NTS₂ receptors. Substantial cross-inhibitory effects of both receptor subtypes NTS₁ and NTS₂ on the agonist binding of D_2L or D₃ were detected in the presence of neurotensin. To identify ligand-specific modulation and subtype-dependent differences, the novel dopamine receptor agonists 5 and 6 bearing the 7-OH-DPAT pharmacophore were synthesized. Exceptional ligand specificity was observed for D₃-NTS₂ co-expression, which gave a 20-fold decrease in affinity for biphenylcarboxamide 5 in the presence of neurotensin. Comparing the binding properties of dopaminergic compounds in the presence of neurotensin, dopamine receptor subtype-selective profiles of the cross-inhibitory effect of neurotensin were observed.

Full text of DOI:10.1002/cmdc.201100499

DOI: 10.1002/cmdc.201100499
Selective Agonists for Dopamine/Neurotensin Receptor
Heterodimers
Susanne Koschatzky and Peter Gmeiner*[a]
The neuromodulatory peptide neurotensin has been described
to functionally interact with dopaminergic pathways of the
human brain. We employed radioligand binding studies to in-
vestigate the physical interaction between co-expressed dopa-
mine D2L or D3 and neurotensin NTS1 or NTS2 receptors. Sub-
stantial cross-inhibitory effects of both receptor subtypes NTS1
and NTS2 on the agonist binding of D2L or D3 were detected in
the presence of neurotensin. To identify ligand-specific modu-
lation and subtype-dependent differences, the novel dopamine
receptor agonists 5 and 6 bearing the 7-OH-DPAT pharmaco-
phore were synthesized. Exceptional ligand specificity was ob-
served for D3–NTS2 co-expression, which gave a 20-fold de-
crease in affinity for biphenylcarboxamide 5 in the presence of
neurotensin. Comparing the binding properties of dopaminer-
gic compounds in the presence of neurotensin, dopamine re-
ceptor subtype-selective profiles of the cross-inhibitory effect
of neurotensin were observed.
Introduction
The mechanistic understanding of G-protein-coupled receptors
(GPCRs) to function as monomers has been extended by their
ability to form homodimers, heterodimers, or higher-order olig-
omers.[1] Molecular interactions in these spatial complexes lead
to modulation of ligand pharmacology, signal transduction,
and cellular trafficking and thus may explain the observation
of tissue-selective GPCR activity in many cases.[2]
gics. To learn more about the tissue-selective modulation of
the dopaminergic system, NT-induced modulation of D3 recep-
tors co-expressed with NTS1 or NTS2 and the D2L–NTS2 interac-
tion should be investigated.
With the use of HEK293 cell lines co-expressing D3–NTS1, D3–
NTS2, and D2L–NTS2, herein we report the binding profiles of
dopamine and the dopamine receptor agonist 7-OH-DPAT and
their dimer-specific modulation in the presence of neuroten-
sin.[6] Binding-deficient NTS1 mutants were used to investigate
the role of the active receptor conformation of NTS1 as a pre-
requisite for the trans-modulatory effect between the D3 recep-
tors and NT receptors. Based on recent investigations revealing
that N-arylamidobutyl-substituted dopaminergics have excel-
lent D3 receptor binding and activating properties, we synthe-
sized the 7-OH-DPAT congeners of type A (Scheme 1).[6,7] Em-
ploying these novel test compounds, the influence of structur-
al changes on the NT-induced decrease in affinity was investi-
Such a functional interaction between the neuromodulatory
peptide neurotensin (NT, pE-L-Y-E-N-K-P-R-R-P-Y-I-L) and dopa-
minergic pathways in the CNS has been described to influence
the pathophysiology of brain diseases including schizophrenia
and Alzheimer’s disease.[3] Previous studies of the impact of NT
on the binding properties of dopamine and [3H]N-n-propylnor-
apomorphine in dopamine D2 receptor-rich brain areas report-
ed a negative cooperative effect of the neuropeptide on dopa-
mine receptor agonist properties.[4] Using the preferential dop-
amine D3 receptor agonist [3H]7-OH-DPAT (7-hydroxy-N,N-di-
propyl-2-aminotetralin) to label subcortical limbic areas of the
rat brain, NT was demonstrated to induce a 20% decrease in
binding affinity and a slight increase in the number of available
binding sites.[5] Given the enhanced influence of NT on D3 re-
ceptor-controlled mesolimbic rather than nigrostriatal dopa-
mine pathways, these findings indicate that the D3 receptor
subtype plays a crucial role in dopamine–neurotensin interac-
tions.
Scheme 1. Structural evolution of type A 7-OH-DPAT congeners.
Taking advantage of fluorescence-detected co-immunopreci-
pitation, we recently demonstrated a physical interaction and
the formation of heterodimers between D2L and the neuroten-
sin receptor NTS1.[6] By the use of a recombinant in vitro test
system, a cross-inhibitory effect on the agonist binding affinity
of D2 was observed in the presence of NT. Structurally diverse
ligands were investigated to determine the relationship be-
tween the intrinsic activity of dopaminergics and the modula-
tory effect of NTS1–NT binding on the affinity of dopaminer-
[a] S. Koschatzky, Prof. Dr. P. Gmeiner
Department of Chemistry and Pharmacy
Emil Fischer Center, Friedrich Alexander University
Schuhstr. 19, 91052 Erlangen (Germany)
E-mail: peter.gmeiner@medchem.uni-erlangen.de
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201100499.
ChemMedChem 2012, 7, 509 – 514
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
509

S. Koschatzky and P. Gmeiner
MED
gated, and subtype-selective differences for the negative coop-
erativity on dopaminergic binding affinities were identified.
To determine the modulatory effects of NT on dopamine re-
ceptor agonist binding of co-expressed D2L or D3 receptors, ho-
mologous competition experiments were carried out to deter-
mine the equilibrium dissociation constant (KD) values for
[3H]7-OH-DPAT in the presence and absence of NT. In analogy
to our previously described D2L–NTS1 system, agonist binding
of D2L in the presence of NTS2 was negatively modulated by
the addition of NT, leading to a threefold increase in KD from
6.9 to 21 nm (Table 1). Singly expressed D2L receptors or mixed
homogenates with singly expressed NTS2 receptors did not
show significant changes in affinity induced by NT.
Results and Discussion
Modulation of ligand binding affinity
Because of a broad overlap of protein expression in brain
tissue, dopamine D2 and D3 receptors and neurotensin recep-
tor subtypes NTS1 and NTS2 were chosen to be studied for
their cross-receptor interactions. Analogously to our previously
described in vitro test system co-expressing D2L and NTS1, we
established co-expressing systems of human dopamine D2L or
D3 receptors in combination with the human NT receptor sub-
types NTS1 or NTS2 by transient transfection in HEK293 cells
and performed radioligand binding studies to identify coopera-
tive effects on binding affinities.
By using the test system expressing D3 and NTS1, radioligand
binding revealed a fourfold decrease in dopamine agonist af-
finity to the high-affinity binding site of D3 receptors in the
presence of NT, whereas the KD values for singly expressed D3
receptors, as well as the mixture of singly expressed D3 and
To ensure approximately equi-
Table 1. Affinities of the co-expressing cell lines reveal various effects in the presence of neurotensin using
molar receptor expression levels,
transient HEK293 cell transfec-
tions were performed with equal
amounts of cDNA for co-expres-
sion of D3–NTS1 and D3–NTS2
and with a fourfold excess of
NTS2 cDNA for the D2L–NTS2
system.
[3H]7-OH-DPAT as radioligand.[a]
KD [nm][b]
Ki [nm]
7-OH-DPAT
Dopamine
5
6
ꢀNT
+NT
7.2ꢁ0.6[c]
62ꢁ11[c]
8
6.9ꢁ1.8
21ꢁ6.3[d]
3
8.4ꢁ0.6
8.9ꢁ11[d]
1
7.0ꢁ1.3
5.3ꢁ1.6
1
0.8ꢁ0.01
3.7ꢁ1.1[d]
4
0.47ꢁ0.016
0.75ꢁ0.14
1.6
0.54ꢁ0.08
1.9ꢁ0.21[d]
3
1.6ꢁ0.020
1.1ꢁ0.014
0.7
8.8ꢁ1.2[c]
3.3ꢁ1.0
1.4ꢁ0.33
Co-expression
D2L–NTS1
82ꢁ3.3[c]
29ꢁ8.0[d]
15ꢁ3.1[d]
Ki+NT/KiꢀNT
ꢀNT
9
9
11
13ꢁ1.5
2.5ꢁ0.42
0.57ꢁ0.08
Co-expression
D2L–NTS2
+NT
130ꢁ18[d]
19ꢁ1.9[d]
4.2ꢁ0.22[d]
Radioligand
displacement
Ki +NT/KiꢀNT
ꢀNT
9
–
–
–
–
–
–
7
–
–
–
–
–
–
7
–
–
–
–
–
–
studies were performed with
membrane preparations of cells
singly expressing D2L, D3, NTS1 or
NTS2, or a co-expression of a
dopamine receptor subtype with
NTS1 or NTS2. Orthosteric bind-
ing sites of NTS1 and NTS2 were
determined by the use of [3H]NT.
No influence on the binding af-
finity of NT onto co-expressed
neurotensin receptors was ob-
served in the presence of the
dopamine receptor agonist 7-
OH-DPAT (Supporting Informa-
tion). Hence, the agonist-bound
dopamine receptor conforma-
tions of D2L and D3 were unable
to alter the neurotensin receptor
affinity for NT. Saturation binding
experiments were performed to
investigate the influence of NT
on dopamine receptor antago-
nist binding of [3H]spiperone.
For all newly investigated co-ex-
pressing cell lines (D3–NTS1, D3–
NTS2, and D2L–NTS2), NT did not
induce changes in the binding
properties of the antagonist-
bound dopamine receptor con-
formation.
D2L
+NT
Ki+NT/KiꢀNT
ꢀNT
Mixture of Singly
Expressed D2L and NTS2
+NT
Ki+NT/KiꢀNT
ꢀNT
5.8ꢁ0.9
0.78ꢁ0.15
0.72ꢁ0.10
Co-expression
D3–NTS1
+NT
46ꢁ18[d]
8.7ꢁ1.1[d]
3.7ꢁ0.5[d]
Ki+NT/KiꢀNT
ꢀNT
8
–
–
–
11
5
–
–
–
–
–
–
Mixture of Singly
Expressed D3 and NTS1
+NT
Ki+NT/KiꢀNT
ꢀNT
4.1ꢁ0.93
0.15ꢁ0.04
0.68ꢁ0.20
Co-expression
D3–NTS2
+NT
27ꢁ6.1[d]
3.1ꢁ0.83[d]
2.4ꢁ0.30[d]
Ki+NT/KiꢀNT
ꢀNT
6
–
–
–
–
–
–
–
–
–
–
–
–
20
–
–
–
–
–
–
–
–
–
–
–
–
3
–
–
–
–
–
–
–
–
–
–
–
–
D3
+NT
K
i+NT/KiꢀNT
ꢀNT
1.0ꢁ0.0010
1.0ꢁ0.11
1
0.30ꢁ0.010
0.52ꢁ0.05
1.7
0.47ꢁ0.01
0.41ꢁ0.03
0.8
Mixture of Singly
Expressed D3 and NTS2
+NT
K
i+NT/KiꢀNT
ꢀNT
Co-expression
D3–NTS1W129A
+NT
Ki+NT/KiꢀNT
ꢀNT
Co-expression
D3–NTS1 W134A
+NT
Ki+NT/KiꢀNT
[a] The affinities of investigated substances were determined on membrane preparations of transiently trans-
fected HEK293 cells co-expressing D2L or D3 and NTS1 or NTS2 using [3H]7-OH-DPAT for displacement experi-
ments; final NT concentration was 100 nm for all +NT rows. Data represent the mean ꢁSEM and are derived
from three to ten individual experiments, each done in triplicate. Hill slopes were between ꢀ0.80 and ꢀ1.20.
The ratio Ki+NT/KiꢀNT indicates the change in affinity. [b] KD values were derived from three to six individual satu-
ration experiments for D3–NTS1 and for D3–NTS2, or homologous displacement experiments for D2–NTS2, each
done in triplicate; they were determined on membrane preparations of transiently transfected HEK293 cells
using the radioligand [3H]7-OH-DPAT. [c] Binding data according to Koschatzky et al.[6] [d] Significance was cal-
culated in an unpaired t test relative to control experiments in the absence of NT: p<0.023.
510
www.chemmedchem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2012, 7, 509 – 514

Heterodimer-Selective Agonists
NTS1, remained unchanged in the presence of NT. Finally, satu-
ration experiments with a D3–NTS2 co-expressing cell line also
demonstrated a negative cooperative effect, as indicated by an
increase in the KD value of 7-OH-DPAT from 0.54 to 1.9 nm by
NT, relative to wild-type D3 receptors. Agonist-bound NTS1 and
NTS2 also exerted substantial negative cooperativity on the D2
and D3 binding of the endogenous ligand dopamine, as indi-
cated by the Ki+NT/KiꢀNT ratios between 6 and 9.
The unchanged binding affinities of the singly expressed D3
receptors for NT-induced modulations underscore the require-
ment of co-expression and co-processing of both receptor sub-
types in the cell membrane to facilitate an intramembrane re-
ceptor–receptor interaction. To confirm that the allosteric
effect across the receptor–receptor interface is exerted by spe-
cific agonist binding, further control experiments were per-
formed. In earlier studies, we described two mutations in the
extracellular loop of NTS1 (W129A and W134A), which show a
complete loss in agonist binding of [3H]NT(8–13), but which ex-
hibit retention of antagonist binding of [3H]SR4869.[8] We pro-
duced transiently co-expressing cell lines of D3 and
NTS1W129A or NTS1W134A. As expected, no specific binding of
[3H]NT was detected. On the other hand, the binding affinity of
the dopamine receptor agonist 7-OH-DPAT remained un-
changed in the presence of NT (100 nm). Thus, agonist binding
of the neurotensin receptor and subsequent transition into the
active receptor conformation seems necessary for the coopera-
tive effects within the heterodimer.
Scheme 2. Reagents and conditions: a) 4-bromobutyronitrile, K2CO3, NaI,
CH3CN, 24 h, reflux, (59%); b) LiAlH4 in Et2O, 6 h, RT, (92%); c) BBr3 in CH2Cl2,
2 h, ꢀ788C, 16 h, RT; d) 4-biphenylcarboxylic acid chloride, NEt3, CH2Cl2,
18 h, RT, (8%); e) pyrazolo[1,5-a]pyridine-3-carboxylic acid, TBTU, DIPEA,
CH2Cl2, 4 h, RT, (15%).
for their binding behavior and heterodimer-selective suscepti-
bilities toward NT and the co-expressing cell lines D2L–NTS1,
D2L–NTS2, D3–NTS1, and D3–NTS2.
K
i+NT/KiꢀNT ratios, which indicate the potency of a coopera-
tive, NT-dependent effect for test compounds 5 and 6 relative
to dopamine and our reference D2/D3 agonist 7-OH-DPAT, are
depicted in Figure 1. For all four co-expressing cell lines, the
endogenous ligand dopamine exhibits similar but marked de-
creases in dopamine binding affinities (between six- and nine-
fold) induced by NT bound to its receptor. The synthetic ago-
nist 7-OH-DPAT displays a substantial (eightfold) decrease in af-
finity for the D2L–NTS1 system, whereas binding affinities for
Synthesis of 7-OH-DPAT analogues
Very recent SAR studies indicate that the replacement of a
propyl substituent in the dopaminergic agent N,N-dipropylami-
noindane with a biphenylcarboxamidobutyl group or a 7a-
azaindole analogue thereof leads to highly potent dopamine
D2 and D3 receptor agonists.[6,9] We planned to take advantage
of this strategy to determine whether variously substituted 7-
OH-DPAT analogues of type A show distinct susceptibility to
the formation of heterodimers and, putatively, subtype-specific
modulation profiles. Therefore, this required the preparation
and investigation of molecular probes 5 and 6 for dimer-spe-
cific binding properties.[10] Synthesis of the new dopaminergics
5 and 6 was performed by starting from racemic 7-methoxy-2-
(N-propylamino)tetralin (1).[11] Reduction of the carbonitrile
function by lithium aluminum hydride afforded primary amine
3. Subsequent ether cleavage yielded the hydroxytetralin de-
rivative 4, which served as a central intermediate. For the
transformation of 4 into the final products 5 and 6, N-acylation
was carried out with 4-biphenylcarboxylic acid chloride and a
TBTU-promoted coupling of pyrazolo[1,5-a]pyridine-3-carboxyl-
ic acid, respectively (Scheme 2).
Trans-modulatory effect on the binding affinities of the
long-chain 7-OH-DPAT derivatives
To determine a correlation between structural modifications
and the magnitude of the NT-induced cooperativity via cross-
receptor interaction, test compounds 5 and 6 were evaluated
Figure 1. Subtype-dependent differences of the negative cooperativity on
dopamine receptor agonist binding: &, D2L–NTS1; &, D2L–NTS2; &, D3–NTS1;
&, D3–NTS2.
ChemMedChem 2012, 7, 509 – 514
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
511

S. Koschatzky and P. Gmeiner
MED
D3–NTS1, D3–NTS2, and D2L–NTS2 were less influenced (three- to
fourfold). The azaindole 6 induced stronger susceptibilities of
binding affinities for the two D2L-based heterodimers than to
the D3–NTS1/2 systems. Interestingly, a significant decrease in
affinity was observed for biphenylcarboxamide 5, in the pres-
ence of NT, toward D3–NTS1 and D3–NTS2 co-expressing cell
lines, with Ki+NT/KiꢀNT values of 11 and 20, respectively.
Cell culture
Human embryonic kidney cells (HEK293) were cultured in DMEM–
Ham’s F12 medium (1:1), supplemented with 10% fetal calf serum,
penicillin G (100 UmLꢀ1), streptomycin (100 mgmLꢀ1), and gluta-
mine (2 mm). All cells were grown at 378C under a humidified at-
mosphere with 5% CO2.
Membrane preparations
Conclusions
HEK293 cells were transiently transfected with 24 mg DNA per Petri
dish (145ꢁ25 mm) of the cDNA encoding the proteins of dopa-
mine or neurotensin receptor constructs, or a mixture of the
cDNAs, by using TransIT-293 transfection reagent according to the
protocol given by the manufacturer. Transfected cells were cultivat-
ed for 48 h. For the membrane preparations, the cell medium was
removed, and cells were washed once with phosphate-buffered
saline. The cell material was then abraded with a cell scraper and
resuspended in 10 mL harvest buffer (10 mm Tris·HCl, 0.5 mm
EDTA, 5.4 mm KCl, and 140 mm NaCl, pH 7.4) into a centrifuge
tube. After centrifugation at 220 g for 8 min, the pellet was resus-
pended in 5 mL homogenate buffer (50 mm Tris·HCl, 5 mm EDTA,
1.5 mm CaCl2, 5 mm MgCl2, 5 mm KCl, and 120 mm NaCl, pH 7.4).
Cells were used directly or stored at ꢀ808C. After thawing or di-
rectly, the cells were homogenized using a Polytron (20000 rpm,
5 times for 5 s each in an ice bath) and pelleted at 50000 g for
18 min. The supernatant was discarded, and the membrane pellet
was resuspended in binding buffer (50 mm Tris, 1 mm EDTA, 5 mm
MgCl2, 100 mgmLꢀ1 bacitracin, 5 mgmLꢀ1 soybean trypsin inhibitor,
pH 7.4) and homogenized with a Potter–Elvehjem homogenizer.
Membrane preparations were stored at ꢀ808C in small aliquots.
Protein concentration was determined by the method of Lowry
et al. using bovine serum albumin as a standard.[14]
Radioligand binding studies of the physical interaction be-
tween co-expressed dopamine D2L or D3 and neurotensin NTS1
or NTS2 receptors indicate substantial cross-inhibitory effects of
both receptor subtypes NTS1 and NTS2 on the agonist binding
of D2L or D3 in the presence of neurotensin. To identify ligand-
specific modulation and subtype-dependent differences, the
novel dopamine receptor agonists 5 and 6 bearing the 7-OH-
DPAT pharmacophore were synthesized. Exceptional ligand
specificity was observed for D3–NTS2 co-expression, which
gave a 20-fold decrease in affinity for biphenylcarboxamide 5
in the presence of neurotensin. Dopamine receptor subtype-
selective profiles of the cross-inhibitory effect of neurotensin
were observed by comparing the binding properties of the
dopaminergic agents in the presence of neurotensin. Subtype-
selective differences of modulatory effects between dopamine
and neurotensin receptor binding sites might support the de-
velopment of tissue-selective dopaminergic drugs based on
the concept of GPCR heteromers as biological targets of future
drugs.
Saturation experiments with [3H]7-OH-DPAT
Experimental Section
Dopamine receptor agonist binding site saturation experiments
with [3H]7-OH-DPAT (specific activity 157–163 Cimmolꢀ1) as radioli-
gand were performed in 24-well plates with a total assay volume
of 500 mL using six different concentrations of radioligand (0.1–
5 nm). Total and unspecific binding were measured in the presence
of buffer or 7-OH-DPAT (10 mm), respectively. To investigate the in-
fluence of neurotensin receptor agonist NT on dopaminergic bind-
ing, either substance or buffer was added to the reaction mixture.
After the addition of the membrane homogenates (90 mg(mL pro-
tein)ꢀ1), the mixture was incubated for 1 h at 378C. The assay was
stopped by rapid filtration through GF/B filters precoated with
0.3% polyethylenimine. Filters were washed five times with ice-
cold Tris/EDTA buffer (50 mm Tris, 1 mm EDTA, pH 7.4), dried at
508C, sealed with MeltiLex solid scintillator (PerkinElmer), and radi-
oactivity counted in a MicroBeta Trilux instrument (PerkinElmer).
Materials
All cell culture material was purchased from Invitrogen LifeTechnol-
ogies (Karlsruhe, Germany). The radioligands [3H]spiperone (102–
114 Cimmolꢀ1) and [3H]7-OH-DPAT (157–163 Cimmolꢀ1) were pur-
chased from GE Healthcare (Freiburg, Germany), and
[3H]neurotensin (100–112 Cimmolꢀ1) was purchased from Perki-
nElmer (Rodgau, Germany). Dopamine (3,4-dihydroxyphenethyla-
mine), spiperone, haloperidol, 7-OH-DPAT [(R)-(+)-7-hydroxy-2-(N,N-
di-n-propylamino)tetralin hydrobromide], and other substances
were purchased from Sigma (Steinheim, Germany), unless other-
wise stated.
Biology
Expression vectors
Homologous and non-homologous displacement
experiments with [3H]7-OH-DPAT
Wild-type hNTS1, hNTS2, and hD2L cDNA was purchased from the
UMR cDNA Resource Center and subcloned into a pcDNA 3.1(+)
eukaryotic expression vector (Invitrogen, Karlsruhe, Germany) using
EcoR1/Xba1 restriction sites.[8,12,13] The pcDNA3.1(+) of the hDRD3
receptor was used as described previously.[9a] Oligonucleotide pri-
mers were purchased from Biomers.net (http://www.biomers.net/
index.html; Ulm, Germany). Restriction enzymes were purchased
from New England Biolabs (Frankfurt am Main, Germany). The fidel-
ity of PCR amplification and introduction of the fluorescence pro-
tein tags in the receptor cDNAs were confirmed by sequencing at
LGC Genomics (Berlin, Germany).
To determine binding at the high-affinity binding site of the dopa-
mine receptors, the agonist [3H]7-OH-DPAT (specific activity 157–
163 Cimmolꢀ1) was used as tritiated radioligand to determine the
KD and Bmax values in homologous competition experiments, or Ki
values of various test compounds in non-homologous competition
experiments. Competition assays were performed in 96-well plates
at a total volume of 200 mL. A solution of the [3H]7-OH-DPAT radio-
ligand (1.0 nm) was used for labeling of protein (90 mgmLꢀ1 per
well). Varying concentrations of unlabeled test compound (0.01–
512
www.chemmedchem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2012, 7, 509 – 514

Heterodimer-Selective Agonists
100000 nm) were added to the radioligand. To determine unspecif-
ic binding, 10 mm 7-OH-DPAT was used. Total binding was deter-
mined in the absence of test compound. After addition of the
membrane homogenates, the mixture was incubated for 1 h at
378C. The assay was stopped by rapid filtration as described
above.
TLC was performed with Merck 60 F254 aluminum sheets, and analy-
sis was by UV light (l 254 nm). Analytical HPLC was performed on
Agilent 1100 HPLC systems equipped with a VWL detector and a
Zorbax Eclipse XDB-C8 column (4.6 mmꢁ150 mm, 5 mm). Purity
was determined by using the aforementioned binary solvent
system: A/B (10:90 v/v) to 100% A over 21 min, isocratic 100% A
for 3 min; flow rate: 1.0 mLminꢀ1; l=254 nm.
4-[(7-methoxy-1,2,3,4-tetrahydronaphthalen-2-yl)-
Saturation experiments using [3H]spiperone and
[3H]neurotensin
(propyl)amino]butanenitrile, 2: 4-Bromobutyronitrile (0.857 mL,
8.54 mmol) was added dropwise to a suspension of compound 1
(781 mg, 3.56 mmol), KI (533 mg, 3.2 mmol), and K2CO3 (2.76 g,
8.9 mmol) in CH3CN (60 mL). After being held at reflux for 24 h, the
mixture was allowed to cool to room temperature, and the solvent
was evaporated. The residue was dissolved in H2O, acidified with
an excess of 2n HCl, and extracted with Et2O. The aqueous solu-
tion was basified with an excess of 2n NaOH and extracted with
Et2O. The combined organic layers were dried (MgSO4) and evapo-
rated to give 2 as a yellow liquid (591 mg, 59%). 1H NMR
(360 MHz, CDCl3): d=0.92 (t, J=7.4 Hz, 3H), 1.44–1.55 (m, 2H),
1.57–1.70 (m, 1H), 1.75–1.85 (m, 2H), 1.96–2.04 (m, 1H), 2.43–2.52
(m, 4H), 2.66 (t, J=6.44 Hz, 2H), 2.71–3.00 (m, 5H), 3.80 (s, 3H),
6.65 (d, J=2.6, 1H), 6.71 (dd, J=8.3, 2.7 Hz, 1H), 7.01 ppm (d, J=
8.3 Hz, 1H); 13C NMR (90 MHz, CDCl3): d=11.85, 14.57, 22.25, 24.91,
26.04, 29.04, 32.31, 48.38, 52.35, 55.30, 56.38, 112.22, 113.94,
120.17, 128.49, 129.47, 137.55, 157.63 ppm; IR n˜ =2956s, 2931s,
2244w, 1503s, 1262m cmꢀ1; MS (EI) m/z 286; purity 99% (HPLC).
Membrane preparations of co-expressed or singly expressed dopa-
mine or neurotensin receptors in HEK293 cells were incubated in
96-well plates with 10 different concentrations (0.005–2 nm) of the
tritiated dopamine receptor antagonist [3H]spiperone (specific ac-
tivity 102–114 Cimmolꢀ1
)
or [3H]NT (specific activity 100–
112 Cimmolꢀ1). Nonspecific binding was defined in the presence of
10 mm haloperidol or NT, respectively, and total binding was mea-
sured in the absence of any competing drug. To investigate the in-
fluence of various NTS1 or D3 receptor ligands, either substance or
buffer was added to the reaction mixture. After addition of mem-
brane preparations with protein concentrations of 40 mgmLꢀ1, the
assay mixture was incubated for 30–60 min at 378C and stopped
by rapid filtration and further processed as described above for
[3H]7-OH-DPAT.
N-(4-aminobutyl)-7-methoxy-N-propyl-1,2,3,4-tetrahydronaph-
thalen-2-amine, 3: A solution of LiAlH4 (1m) in Et2O (6.3 mL,
6.3 mmol) was added dropwise to a cooled solution of 2 (360 mg,
1.26 mmol) in THF (15 mL). After stirring at room temperature for
1 h, the mixture was cooled to 08C, quenched with aqueous
NaHCO3, filtered over Celite/Mg2SO4/Celite, and washed several
times with CH2Cl2 and EtOAc. The solvent was evaporated to give
3 as a yellow liquid (336.6 mg, 92% yield). 1H NMR (360 MHz,
CDCl3): d=0.92 (t, J=7.3 Hz, 3H), 1.47–1.74 (m, 7H), 2.00–2.07 (m,
1H), 2.11–2.39 (m, 2H), 2.47–2.65 (m, 4H) 2.71–2.92 (m, 5H), 2.95–
3.08 (m, 1H), 3.80 (s, 3H), 6.66 (d, J=2.7, 1H), 6.71 (dd, J=8.3,
2.7 Hz, 1H), 7.01 ppm (d, J=8.3 Hz, 1H); 13C NMR (90 MHz, CDCl3):
d=11.93, 22.15, 26.09, 26.46, 28.97, 29.06, 32.37, 42.06, 50.49,
52.60, 55.29, 56.73, 112.10, 113.96, 128.71, 129.46, 137.81,
157.53 ppm; IR n˜ =2930s, 2868m, 2361m, 1503s, 1262m cmꢀ1; MS
(EI) m/z 290; purity 95% (HPLC).
Data analysis
Analyses of the saturation experiments were performed by nonlin-
ear regression data analysis for the determination of KD and Bmax
values using Prism (GraphPad Software, San Diego, CA, USA). The
resulting competition curves were analyzed by nonlinear regres-
sion using the algorithms in Prism. The data were initially fitted
using a sigmoid model and an IC50 value, representing the concen-
tration corresponding to 50% maximal inhibition. Data were then
calculated for a one-site model using Prism. IC50 values were trans-
formed into Ki values according to the equation for the calculation
of competition curves as described by Cheng and Prusoff.[15] Data
are presented as mean ꢁSEM. Significances were calculated in an
unpaired t test relative to control experiments in the absence of
NT.
7-[(4-aminobutyl)(propyl)amino]-5,6,7,8-tetrahydronaphthalen-2-
ol, 4: Compound 3 (720 mg, 2.5 mmol) was added to a solution of
BBr3 (1m; 11.3 mL, 11.3 mmol) in CH2Cl2, which was pre-cooled at
ꢀ788C, and stirring was continued at ꢀ788C for 2 h. The mixture
was then stirred for 23 h at room temperature. After quenching
with an excess of aqueous NaHCO3 for 30 min, the aqueous layer
was adjusted to mild basic, and was extracted three times with
CH2Cl2. The combined organic layers were dried (MgSO4) and
evaporated. The crude product of 4 was not further purified
(101.5 mg). LC–MS (APCI) m/z 276.
Chemistry
General
Reagents and dry solvents were of commercial quality and were
used as purchased. All reactions were carried out under nitrogen
atmosphere. MS was performed on a JEOL JMS-GC Mate II spec-
trometer by EI (70 eV) with solid inlet or a Bruker Esquire 2000 by
APC or ionization. HR-EIMS analyses were run on a JEOL JMS-GC
Mate II using Peak-Matching (M/DM>5000). NMR spectra were col-
lected on a Bruker Avance 360 or a Bruker Avance 600 spectrome-
ter relative to TMS in the solvents indicated (J values in Hz). IR
spectra were run on a Jasco FT/IR 410 spectrometer. Melting
points were determined with a MEL-TEMP II melting-point appara-
tus (Laboratory Devices, USA) in open capillaries and are uncorrect-
ed. Preparative RP-HPLC (Agilent 1100 preparative series) was per-
formed under the following conditions: column: Zorbax Eclipse
XDB-C8, 21.2ꢁ250 mm, 5 mm particle size; eluent: CH3OH (A) and
0.1% TFA in H2O (B); flow rate: 20 mLminꢀ1; UV detection at
l 254 nm.
N-{4-[(7-hydroxy-1,2,3,4-tetrahydronaphthalen-2-yl)-
(propyl)amino]butyl}-4-phenylbenzamide, 5: Compound
4
(53.3 mg, 0.19 mmol) was dissolved in 6 mL dry CH2Cl2, and
0.074 mL Et3N (0.57 mmol) was added. The mixture was cooled to
08C. A solution of 4-biphenylcarboxylic acid chloride (39.7 mg,
0.18 mmol) in dry CH2Cl2 (6 mL) was then added dropwise. The
mixture was stirred for 18 h at room temperature before aqueous
NaHCO3 was added. The aqueous layer was extracted with CH2Cl2,
and the combined organic layers were dried (MgSO4) and evapo-
ChemMedChem 2012, 7, 509 – 514
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
513

S. Koschatzky and P. Gmeiner
MED
[1] a) S. R. George, B. F. O’Dowd, S. P. Lee, Nat. Rev. Drug Discovery 2002, 1,
808–820; b) G. Milligan, Mol. Pharmacol. 2004, 66, 1–7; c) G. Milligan,
Biochim. Biophys. Acta Biomembr. 2007, 1768, 825–835.
rated. The residue was purified by preparative RP-HPLC (gradient:
10!90% A over 16 min, then 90% A for 2 min, 90!10% A over
4 min, tR: 16.0 min) to give 5 as a gray–white solid (7 mg, 8% refer-
[2] a) N. J. Birdsall, Trends Pharmacol. Sci. 2010, 31, 499–508; b) J. P. Vilarda-
ga, V. O. Nikolaev, K. Lorenz, S. Ferrandon, Z. Zhuang, M. J. Lohse, Nat.
Chem. Biol. 2008, 4, 126–131; c) D. J. Daniels, N. R. Lenard, C. L. Etienne,
P.-Y. Law, S. C. Roerig, P. S. Portoghese, Proc. Natl. Acad. Sci. USA 2005,
102, 19208–19213; d) J. Hillion, M. Canals, M. Torvinen, V. Casadꢂ, R.
Scott, A. Terasmaa, A. Hansson, S. Watson, M. E. Olah, J. Mallol, E. I.
Canela, M. Zoli, L. F. Agnati, C. F. Ibꢃnez, C. Lluis, R. Franco, S. Ferrꢄ, K.
Fuxe, J. Biol. Chem. 2002, 277, 18091–18097; e) R. Rozenfeld, L. A. Devi,
Trends Pharmacol. Sci. 2010, 31, 124–130; f) J.-P. Pin, R. Neubig, M.
Bouvier, L. Devi, M. Filizola, J. A. Javitch, M. J. Lohse, G. Milligan, K. Palc-
zewski, M. Parmentier, M. Spedding, Pharmacol. Rev. 2007, 59, 5–13;
g) A. J. Rashid, B. F. O’Dowd, S. R. George, Endocrinology 2004, 145,
2645–2652; h) J. Kꢅhhorn, H. Hꢅbner, P. Gmeiner, J. Med. Chem. 2011,
54, 4896–4903; i) W. Guo, E. Urizar, M. Kralikova, J. C. Mobarec, L. Shi, M.
Filizola, J. A. Javitch, EMBO J. 2008, 27, 2293–2304; j) C. Jomphe, P.-L.
Lemelin, H. Okano, K. Kobayashi, L.-E. Trudeau, Eur. J. Neurosci. 2006, 24,
2789–2800.
1
enced to crude product 5); mp: 1068C; H NMR (360 MHz, CDCl3):
d=0.89 (t, J=7.3 Hz, 3H), 1.50 (tq, J=7.5, 7.5 Hz, 2H), 1.55–1.73
(m, 5H), 1.98 ꢀ2.04 (m, 1H), 2.51 (t, J=7.6 Hz, 2H), 2.60 (t, J=7.0,
2H), 2.68–2.84 (m, 4H), 2.97–3.04 (m, 1H), 3.51 (dt, J=6.4, 6.5 Hz,
2H), 6.55 (d, J=2.5 Hz, 1H), 6.62 (dd, J=8.2, 2.6 Hz, 1H), 6.92 (d,
J=8.2 Hz, 1H), 7.40 (tt, J=7.4, 1.1 Hz, 1H), 7.47 (tt, J=7.5, 1.5 Hz,
2H), 7.60–7.62 (m, 2H), 7.65 (tt, J=8.5, 1.5 Hz, 2H), 7.85 ppm (tt,
J=8.2, 1.0 Hz, 2H); 13C (90 MHz, CDCl3): d=11.93, 21.68, 25.80,
26.14, 27.52, 28.94, 31.98, 40.06, 50.01, 52.58, 56.83, 113.36, 115.65,
127.19, 127.46, 127.90, 128.90, 129.54, 133.46, 137.53, 140.04,
153.83, 167.51 ppm; IR n˜ =3299s, 2930s, 1635s, 1541s, 1506s,
1485s, 1263m cmꢀ1; HRMS (EI) m/z calcd for C30H36N2O2: 456.2777,
found 456.2777; purity 98% (HPLC).
N-{4-[(7-hydroxy-1,2,3,4-tetrahydronaphthalen-2-yl)-
(propyl)amino]butyl}pyrazolo[1,5-a]pyridine-3-carboxamide, 6:
DIPEA (109 mL, 0.66 mmol) was added to a solution of pyrazolo[1,5-
a]pyridine-3-carboxylic acid (34.5 mg, 0.21 mmol) in CH2Cl2 (6 mL).
The mixture was cooled to 08C before a solution of TBTU (95.3 mg,
0.297 mmol) in DMF (1 mL) was added. A solution of 4 (76.2 mg,
0.27 mmol) in CH2Cl2 (6 mL) was added dropwise. The mixture was
stirred at room temperature for 3 h, before aqueous NaHCO3 was
added. The aqueous layer was extracted with CH2Cl2, and the com-
bined organic layers were dried (MgSO4) and evaporated. The resi-
due was purified by preparative RP-HPLC (gradient: 10!80% A
over 15 min, then 80% A for 2 min, 80!10% A over 3 min, tR:
12.4 min) to give 6 as a gray–white solid (17 mg, 15% referenced
to crude product 5); mp: 1568C; 1H NMR (360 MHz, CDCl3): d=
0.88 (t, J=7.4 Hz, 3H), 1.59 (tq, J=7.6, 7.6 Hz, 2H), 1.52–1.70 (m,
5H), 1.99–2.04 (m, 1H), 2.54 (t, J=7.7 Hz, 2H), 2.63 (t, J=6.9 Hz,
2H), 2.65–2.81 (m, 4H), 3.00–3.01 (m, 1H), 3.49 (dq, J=6.5, 6.5 Hz,
2H), 6.54 (d, J=2.5 Hz, 1H), 6.55–6.59 (m, 1H), 6.63 (dd, J=8.2,
2.64 Hz, 1H), 6.88 (d, J=8.2 Hz, 1H), 6.91 (ddd, J=6.9, 6.9, 1.4 Hz,
1H), 7.33 (ddd, J=8.9, 6.8, 1.1 Hz, 1H), 8.23 (s, 1H), 8.31 (dt, J=9.0,
1.2 Hz, 1H), 8.47 ppm (ddd, J=7.0, 1.0, 1.0 Hz, 1H); 13C (90 MHz,
CDCl3): d=11.85, 21.31, 25.63, 25.71, 27.62, 28.77, 31.86, 39.27,
50.09, 52.59, 57.13, 106.90, 113.60, 113.65, 115.70, 119.62, 126.49,
127.69, 128.78, 129.51, 140.52, 140.66, 154.27, 163.68 ppm; IR n˜ =
3307s, 2932s, 1637s, 1557s, 1503s, 1273m cmꢀ1; HRMS (EI) m/z
calcd for C25H32N4O2: 420.2525, found 420.2526; purity 99% (HPLC).
[3] a) E. B. Binder, B. Kinkead, M. J. Owens, C. B. Nemeroff, Pharmacol. Rev.
2001, 53, 453–486; b) M. Boules, A. Shaw, P. Fredrickson, E. Richelson,
CNS Drugs 2007, 21, 13–23; c) K. Fuxe, D. Marcellino, A. S. Woods, L.
Giuseppina, T. Antonelli, L. Ferraro, S. Tanganelli, L. F. Agnai, J. Neural
Transm. 2009, 116, 923–939.
[4] a) G. von Euler, P. Mailleux, J.-J. Vanderhaeghen, K. Fuxe, Neurosci. Lett.
1990, 109, 325–330; b) S. Tanganelli, X.-M. Li, L. Ferraro, G. von Euler,
W. T. O’Connor, C. Bianchi, L. Beani, K. Fuxe, Eur. J. Pharmacol. 1993, 230,
159–166; c) X.-M. Li, L. Ferraro, S. Tanganelli, W. T. O’Connor, U. Hassel-
rot, U. Ungerstedt, K. Fuxe, J. Neural Transm. 1995, 102, 125–137;
d) L. F. Agnati, K. Fuxe, F. Benfenati, N. Basttistini, Acta Physiol. Scand.
1983, 119, 459–461; e) G. von Euler, K. Fuxe, Acta Physiol. Scand. 1987,
131, 625–626.
[5] Y. Liu, M. Hillefors-Berglund, G. von Euler, Brain Res. 1994, 643, 343–
348.
[6] S. Koschatzky, N. Tschammer, P. Gmeiner, ACS Chem. Neurosci. 2011, 2,
308–316.
[7] a) I. Boyfield, M. C. Coldwell, M. S. Hadley, C. N. Johnson, G. J. Riley, E. E.
Scott, R. Stacey, G. Stemp, K. M. Thewlis, Bioorg. Med. Chem. Lett. 1997,
7, 1995–1998; b) N. Tschammer, M. Dçrfler, H. Hꢅbner, P. Gmeiner, ACS
Chem. Neurosci. 2010, 1, 25–35.
[8] S. Hꢆrterich, S. Koschatzky, J. Einsiedel, P. Gmeiner, Bioorg. Med. Chem.
2008, 16, 9359–9368.
[9] a) M. Dçrfler, N. Tschammer, K. Hamperl, H. Hꢅbner, P. Gmeiner, J. Med.
Chem. 2008, 51, 6829–6838; b) N. Tschammer, J. Elsner, A. Goetz, K. Ehr-
lich, S. Schuster, M. Ruberg, J. Kꢅhhorn, D. Thompson, J. Whistler, H.
Hꢅbner, P. Gmeiner, J. Med. Chem. 2011, 54, 2477–2491.
[10] G. Stemp, C. N. Johnson, Patent, 1998, EP0817767.
[11] L. A. van Vliet, P. G. Tepper, D. Dijkstra, G. Damsma, H. Wikstroem, T. A.
Pugsley, H. C. Akunne, T. G. Heffner, S. A. Glase, L. D. Wise, J. Med. Chem.
1996, 39, 4233–4237.
Acknowledgements
[12] K. Ehrlich, A. Gçtz, S. Bollinger, N. Tschammer, L. Bettinetti, S. Hꢆrterich,
H. Hꢅbner, H. Lanig, P. Gmeiner, J. Med. Chem. 2009, 52, 4923–4935.
[13] J. Einsiedel, C. Held, M. Hervet, M. Plomer, N. Tschammer, H. Hꢅbner, P.
Gmeiner, J. Med. Chem. 2011, 54, 2915–2923.
[14] O. H. Lowry, N. R. Rosebrough, A. L. Farr, R. J. Randall, J. Biol. Chem.
1951, 193, 265–275.
We gratefully acknowledge the Deutsche Forschungsgemein-
schaft for financial support (DFG 13/8). Manuel Plomer is ac-
knowledged for providing the cDNA of hNTS2, and Dr. Harald
Hꢀbner, for reference binding data.
[15] Y.-C. Cheng, W. H. Prusoff, Biochem. Pharmacol. 1973, 22, 3099–3108.
Keywords: dopamine · GPCRs · heterodimers · neurotensin ·
receptors
Received: October 27, 2011
Revised: December 5, 2011
Published online on December 23, 2011
514
www.chemmedchem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2012, 7, 509 – 514