Design of a True Bivalent Ligand with Picomolar Binding Affinity for a G Protein-Coupled Receptor Homodimer

Article abstract of DOI:10.1021/acs.jmedchem.8b01249

Bivalent ligands have emerged as chemical tools to study G protein-coupled receptor dimers. Using a combination of computational, chemical, and biochemical tools, here we describe the design of bivalent ligand 13 with high affinity (K_DB1 = 21 pM) for the dopamine D₂ receptor (D₂R) homodimer. Bivalent ligand 13 enhances the binding affinity relative to monovalent compound 15 by 37-fold, indicating simultaneous binding at both protomers. Using synthetic peptides with amino acid sequences of transmembrane (TM) domains of D₂R, we provide evidence that TM6 forms the interface of the homodimer. Notably, the disturber peptide TAT-TM6 decreased the binding of bivalent ligand 13 by 52-fold and had no effect on monovalent compound 15, confirming the D₂R homodimer through TM6 ex vivo. In conclusion, by using a versatile multivalent chemical platform, we have developed a precise strategy to generate a true bivalent ligand that simultaneously targets both orthosteric sites of the D₂R homodimer.

Full text of DOI:10.1021/acs.jmedchem.8b01249

Subscriber access provided by UNIVERSITY OF TOLEDO LIBRARIES
Article
Design of a true bivalent ligand with picomolar binding
affinity for a G protein-coupled receptor homodimer
Daniel Pulido, Veronica Casadó-Anguera, Laura Pérez-Benito, Estefania Moreno, Arnau Cordomí,
Laura López, Antoni Cortes, Sergi Ferré, Leonardo Pardo, Vicent Casadó, and Miriam Royo
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b01249 • Publication Date (Web): 26 Sep 2018
Downloaded from http://pubs.acs.org on September 27, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 41
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Design of a true bivalent ligand with picomolar
binding affinity for a G proteinꢀcoupled receptor
homodimer
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
⊥
Daniel Pulido,†,‡,□,# Verònica CasadóꢀAnguera,§,║, ,# Laura PérezꢀBenito,∇,# Estefanía
Moreno,§,║,⊥ Arnau Cordomí,∇ Laura López,∇ Antoni Cortés,§,║,⊥ Sergi Ferré,¶ Leonardo
Pardo,*,∇ Vicent Casadó*,§,║,⊥ and Miriam Royo*,†,‡,□
† Biomedical Research Networking Center in Bioengineering, Biomaterials and Nanomedicine
(CIBERꢀBBN), Barcelona Science Park, 08028 Barcelona, Spain
‡ Combinatorial Chemistry Unit, Barcelona Science Park, 08028 Barcelona, Spain
§ Department of Biochemistry and Molecular Biomedicine, Faculty of Biology, University of
Barcelona, 08028 Barcelona, Spain
║ Biomedical Research Networking Center in Neurodegenerative Diseases (CIBERNED), 08028
Barcelona, Spain
⊥ Institute of Biomedicine of the University of Barcelona (IBUB), 08028 Barcelona, Spain
∇ Laboratory of Computational Medicine, Biostatistics Unit, Faculty of Medicine, Universitat
Autònoma de Barcelona, 08193 Bellaterra, Spain
¶ Integrative Neurobiology Section, National Institute on Drug Abuse, Intramural Research
Program, National Institutes of Health, Baltimore, MD 21224
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 41
1
2
3
4
ABSTRACT
5
6
7
8
9
Bivalent ligands have emerged as chemical tools to study G proteinꢀcoupled receptor dimers.
Using a combination of computational, chemical, and biochemical tools, here we describe the
design of bivalent ligand 13 with high affinity (KDB1=21 pM) for the dopamine D2 receptor
(D2R) homodimer. Bivalent ligand 13 enhances the binding affinity relative to monovalent
compound 15 by 37ꢀfold, indicating simultaneous binding at both protomers. Using synthetic
peptides with amino acid sequences of transmembrane (TM) domains of D2R, we provide
evidence that TM6 forms the interface of the homodimer. Notably, the disturber peptide TATꢀ
TM6 decreased the binding of bivalent ligand 13 by 52ꢀfold and had no effect on monovalent
compound 15, confirming the D2R homodimer through TM6 ex vivo. In conclusion, using a
versatile multivalent chemical platform, we have developed a precise strategy to generate a true
bivalent ligand that simultaneously targets both orthosteric sites of the D2R homodimer.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
INTRODUCTION
It is now well accepted that many G proteinꢀcoupled receptors (GPCRs) form, in addition to
functional monomers,1 dimers and higherꢀorder oligomeric complexes constituted by a number
of equal (homo) or different (hetero) monomers.2 Oligomerization plays an important role in
terms of receptor function and structure, introducing changes in signaling pathways which are
due to the allosteric mechanisms of these complexes. Thus, these oligomers present functional
properties different from those of the constituent monomers (protomers), making oligomerization
a biological resource to generate pharmacological diversity.3 Considering the involvement of
GPCRs in the regulation of many physiological processes, these novel functional units have
recently received special attention as new targets for drug development.4 Besides the set of
2
ACS Paragon Plus Environment

Page 3 of 41
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
existing biochemical and biophysical tools,5 to gain insight into the mechanisms by which
oligomers signal, specific chemical tools can also contribute to evaluate their pharmacology and
to assess their potentiality as drug targets.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
One of these tools are bivalent ligands, defined as single chemical entities composed of two
pharmacophore units covalently linked by an appropriate spacer. These ligands are designed to
interact simultaneously with a (homo/hetero) GPCR dimer to enhance affinity and subtype
selectivity.6 Homobivalent ligands contain two copies of the same pharmacophore,7 whereas
heterobivalent ligands link two different pharmacophores.8 A requirement for bivalent ligands is
the simultaneous binding of the two pharmacophores at the orthosteric sites of the (homo/hetero)
dimer. Thus, the spacer length is a key factor in these ligands and depends on the dimer interface,
the structure of the pharmacophores, and the geometry of the attachment points.9 If the spacer
length is not suitable to cover the distance between the orthosteric sites of both GPCR dimer
protomers these ligands act in a nonꢀsimultaneous interaction mode, with a dualꢀacting profile.10
Other types of ligands composed by two pharmacophores connected by a spacer, but designed to
interact simultaneously with orthosteric and allosteric sites, are referred as bitopic ligands.11
Considering the aboveꢀmentioned diversity of interaction modes of these types of compounds,
the generation of a bivalent ligand requires not only a precise design, but also an accurate
validation of its type of interaction. Using a combination of computational, chemical and
biochemical tools, here we describe the design of a true bivalent ligand with high affinity for the
dopamine D2 receptor (D2R) homodimer. We have selected the D2R as a test case for two
reasons: a) because it forms homo/heterodimers12 and higherꢀorder oligomers13 implicated in
several neuropsychiatric disorders, such as Parkinson disease or schizophrenia;14 and b) due to
the existing controversy regarding the interaction mode of some of the described D2R
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 41
1
2
3
4
5
6
7
8
9
homodimer bivalent ligands.15 D2 receptors are present in several tissues and cell lines and, as
other class A GPCRs, exist in a dynamic equilibrium between monomers, dimers and higher
order oligomers. It has been suggested that bivalent ligands act stabilizing preexisting dimers;16
however, recent data shows that some of them can modulate the dynamics of oligomerization
shifting the equilibrium towards the dimeric state.17
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
RESULTS AND DISCUSSION
Design. The design of bivalent ligands requires the selection of: i. a scaffold that contains at least
two chemical functionalities that can be properly derivatized; ii. a ligand that binds the
orthosteric binding site with high affinity (pharmacophore unit); iii. an appropriate length spacer
to cover the distance between both protomers; and, finally, iv. if necessary, a linker between this
pharmacophore and the bivalent system, adequate in terms of both the topological position of the
attachment point and the chemistry used for the conjugation (Figure 1).9
4
ACS Paragon Plus Environment

Page 5 of 41
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. A) Components for the design of bivalent ligands. B) Bivalent ligands (12ꢀ13) and
their corresponding monovalent counterparts (14ꢀ15).
Herein, the selected scaffold (i) is the nitrilotriacetic acid (NTA), which contains three
symmetric carboxylic acids and permits the controlled desymmetrization of each of these
functional groups.18 This multivalent platform allows not only the attachment of two
pharmacophore units, but also the introduction of a reporter molecule for imaging studies or
another pharmacophore unit to study higherꢀorder oligomers, such as trimers.19
As a proof of concept, a neutral antagonist has been selected as pharmacophore with the aim to
design bivalent ligands whose potential simultaneous interaction with the D2R homodimer would
result only in increased affinity values, avoiding cooperative mechanisms that could encumber
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 41
1
2
3
4
5
6
7
8
9
the evaluation of the binding interaction. The selected pharmacophore unit (ii) is a derivative of
the D2R antagonist spiperone, namely the Nꢀ(pꢀaminophenethyl)spiperone 6 (NAPS),20 which
was functionalized with an extra succinic acid linker (iv) to facilitate its incorporation to the
bivalent system (Figure 1). The resulting pharmacophoreꢀlinker derivative 7 was docked into a
D3ꢀbased homology model of D2R, and its stability was assessed by 1ꢁs of unbiased molecular
dynamic (MD) simulations. (Figure S1). Results showed that the pharmacophore unit (red)
remains highly stable at the binding site during the simulation, whereas the linker moiety (green)
is very flexible and achieves diverse conformations between extracellular loops (ECL) 2 and 3,
always at the extracellular aqueous environment, which makes the selected attachment point
(purple) adequate to link the spacer moieties.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The selected spacers (iii) were different length oligoethylene glycol (OEG) moieties with the
aim to increase water solubility of the final bivalent ligands. A key factor in the design of
bivalent ligands is the spacer length, which depends on the dimer interface. Crystal structures of
GPCRs display several dimerization interfaces21 that can be grouped into three clusters,
depending on the transmembrane helices (TMs) involved: TMs 1 and 2 (TM1/2 interface), TMs
4 and 5 (TM4/5 interface), and TMs 5 and 6 (TM5/6 interface). Using a computational tool,9b
developed in house for the Molecular Operating Environment (MOE) software (Chemical
computing group Inc., Montreal QC, Canada), we calculated the preferred spacer length for the
different interfaces via the shortest pathway along the D2R homodimer van der Waals surface
(Table S1). This surface represents a favorable interaction between the dimer and the
spacer/linker/pharmacophore moieties of the bivalent ligand. We constructed sets of molecules
formed by the pharmacophore (starting at the N atom of the amide bond of the triazaspiro
moiety) / linker / spacer (ꢀOCH2CH2ꢀ)n=2ꢀ4 / scaffold / spacer (ꢀOCH2CH2ꢀ)n=2ꢀ4 / linker /
6
ACS Paragon Plus Environment

Page 7 of 41
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
pharmacophore (ending at the N atom of the amide bond of the triazaspiro moiety) as inputs for
our MOEꢀbased tool. The tool predicts the favorable conformation of these input molecules, and
the interaction energy between these atoms and the rest of the system, calculated for each of the
input molecules after a stepped energy minimization protocol.9b The lengths of energetically
favorable spacers were used as recommendations for synthesis.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The TM5/6 dimerization interface led to the ligand with the shortest spacer/scaffold/spacer
(25ꢀatoms, calculated between attachment atoms shown in purple in Figure 1), because this
interface has the shortest distance between orthosteric sites (33 Å), and also the attachment point
directs toward TMs 5 and 6 (Table S1). The TM4/5 interface gave the largest distance (43ꢀatoms,
43 Å), whereas the TM1/2 interface is in between (31 atoms, 36 Å) (Table S1). Based on these
data, we designed two bivalent ligands: 13 (25ꢀatoms between both attachment atoms),
representing the shortest possible bivalent interaction (via TM5/6), and a longer alternative, 12
(35ꢀatoms), which could also interact at other dimer interaction interfaces, excluding TM4/5,
which is on the opposite side to the direction of the linker elongation, and therefore implausible
to reach it.
Chemical synthesis. The NTAꢀbased core 2 was prepared according to literature procedures22
starting from glycine methyl ester hydrochloride, which was dialkylated with benzyl
bromoacetate, and then hydrogenated to remove the benzyl protecting groups, affording desired
compound in 83% yield (Scheme S1).
The pharmacophoreꢀlinker derivative 7 was prepared following the described methodology
with minor modifications.20a Briefly, the Nꢀalkylation of spiperone with 4ꢀ(Nꢀtertꢀ
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 41
1
2
3
4
5
6
7
8
butyloxycarbonyl)aminophenethyl bromide afforded 5 in 65% yield. Removal of Boc group
using HCl (2 M in dioxane) provided NAPS (6, 66% yield) which was subsequently acylated
with succinic anhydride to afford compound 7 (76% yield).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
OEGꢀbased precursors 8, 10 and 9, 11 were prepared from commercially available OEGs in
good yields (88% for 8, 87% for 9, and 57% for 10, 56% for 11) (Scheme S3). The final bivalent
ligands 12 and 13 were synthesized by acylation of two carboxylic acids of the NTA scaffold (2)
with compounds 10 or 11, respectively (Scheme 1). Finally, the synthesis of monovalent ligands
14 and 15 required differentiation between the free carboxylic acids of 2. This desymmetrization
was accomplished by means of the favored formation of a sixꢀmembered cyclic anhydride
between the pair of carboxylic acids, which reacted selectively with only one equivalent of 8 or 9
to form the amide. Then, the resulting free carboxylic acid could be acylated in a further step
using the corresponding compounds 10 or 11 respectively (Scheme 1).
Scheme 1. a Synthesis of the bivalent ligands 12 and 13 and the monovalent ligands 14 and 15.
8
ACS Paragon Plus Environment

Page 9 of 41
Journal of Medicinal Chemistry
1
2
3
a
Reagents and conditions: (a) 10, EDC·HCl, HOBt·H2O, DIEA, DMF, rt, 16 h (49%); (b) 11,
4
5
6
7
8
9
EDC·HCl, HOBt·H2O, DIEA, DMF, rt, 16 h (64%); (c) EDC·HCl, dry DMF, rt, 2 h, then 8,
DIEA, dry DMF, rt, 90 min (79%); (d) 10, EDC·HCl, HOBt·H2O, DIEA, DMF, rt, 16 h (25%);
(e) EDC·HCl, dry DMF, rt, 2 h, then 9, DIEA, dry DMF, rt, 90 min (80%); (f) 11, EDC·HCl,
HOBt·H2O, DIEA, DMF, rt, 16 h (24%).
Biological assays. In vitro binding affinities of the bivalent ligands (12 and 13) and the
corresponding monovalent counterparts (14 and 15) were obtained from [3H]YMꢀ09151ꢀ2
radioligand competitionꢀbinding assays using membranes from sheep brain striatum that
naturally express D2R. Data were analyzed according to a ‘twoꢀstate dimer model’ (Table 1).23
The model assumes GPCR dimers as a main functional unit and provides a more robust analysis
of parameters obtained from saturation and competition experiments with orthosteric ligands, as
compared with the commonly used ‘twoꢀindependentꢀsite model’.23,24 In competition
experiments the model analyzes the interactions of the radioligand with a competing ligand and it
provides the affinity of the competing ligand for the first protomer in the unoccupied dimer
(KDB1) and the affinity of the competing ligand for the second protomer when the first protomer
is already occupied by the competing ligand (KDB2). All studied compounds show monophasic
nonꢀcooperative curves, as expected for an antagonist with a nonꢀcooperative binding to D2R
dimer. In these conditions, KDB1 is enough to characterize the binding of these compounds.
Table 1. Affinity constants (KDB1) of the D2R ligands 7, 12ꢀ15 with or without TM6 peptides.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Compound
K
DB1 (nM)
+ TM6 D2R
+ TM6 A2AR
0.70±0.06
0.07±0.03*###
0.021±0.003**###
1.5±0.6*
7
12
13
14
15
1.1±0.3^^
0.8±0.2
0.05±0.01
0.8±0.2
0.77±0.04
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 41
1
2
3
4
5
6
Values are mean±SEM from 3ꢀ10 determinations. Statistical significance was calculated by
oneꢀway ANOVA followed by Bonferroni’s post hoc test. *p<0.05, **p<0.01 compared with 7.
###p<0.001 compared with the corresponding monovalent ligand. ^^p<0.01 compared with the
respective control without TM peptides.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Compound 7 has high affinity for D2R (KDB1=0.70 nM). Monovalent compound 15 (25ꢀatoms,
K
DB1=0.77 nM) has similar affinity for D2R than compound 7, whereas monovalent compound
14 (35ꢀatoms, KDB1=1.5 nM) shows a slightly less favorable binding affinity. These results are
remarkable since attachment of the spacer should decrease binding affinity. This suggests that
the OEG spacer favorably interacts with residues at the groove connecting both protomers.
Notably, bivalent ligands 12 (35ꢀatoms, KDB1=0.07 nM) and 13 (25ꢀatoms, KDB1=0.021 nM)
significantly enhance the binding affinity relative to monovalent counterparts 14 and 15 (21ꢀfold
and 37ꢀfold, respectively). Clearly, addition of the second pharmacophore unit increases binding
affinity due to its higher local concentration in a close radius above the second protomer. Thus,
compounds 12 and 13 seem to act as bivalent ligands, that is, both pharmacophores
simultaneously target both orthosteric sites of the homodimer.
To further test that the antagonistic nature of these compounds on D2R signaling remains
unaltered, we resolved the realꢀtime signaling signature by using a labelꢀfree method (DMR)25 in
CHO cells stably coꢀexpressing A2AR and D2R to mimic the pattern receptor expression of brain
striatum, where a high proportion of D2R form heteromers with A2AR.13 This approach detects
changes in local optical density due to cellular mass movements induced upon receptor
activation (see Experimental Section). The magnitude of the signaling by sumanirole, a highly
selective D2R full agonist, significantly decreased in the presence of both bivalent ligands 12 and
13 as much as when adding spiperone (Figures 2A and 2B). Because the affinity of compound 13
is 3.5ꢀfold higher than 12, additional biochemical experiments were carried out with 13 and its
10
ACS Paragon Plus Environment

Page 11 of 41
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
corresponding monovalent counterpart 15. Compound 13 had a better inhibitory potency
antagonizing sumanirole signal (15±3 nM) than the corresponding monovalent ligand 15
(280±70 nM) due to its higher affinity (Figure 2C).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Antagonistic effect of the studied compounds on global cellular response induced by
sumanirole. Dynamic mass redistribution (DMR) assays were performed in CHO cells stably
expressing D2R and A2AR. A) Quantification of the antagonist effect of all D2R ligands on DMR.
Values are mean±SEM from 3 determinations carried out in triplicates. Statistical significance
was calculated by oneꢀway ANOVA followed by Dunnett’s post hoc test. **p<0.01 compared to
sumanirole alone. B) Representative DMR curves from one of these experiments in which cells
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 41
1
2
3
4
5
6
7
8
9
were treated with medium (control) (black), with 1ꢁM of spiperone (grey), with 1ꢁM of
monovalent compounds 14 (light blue) or 15 (dark blue) or with 500 nM of bivalent compounds
12 (orange) or 13 (red) for 30 minutes. After that, cells were treated with 100 nM of sumanirole.
Each curve is the mean of a representative optical trace experiment carried out in triplicates. The
resulting shifts of reflected light wavelength (pm) were monitored over time. C) Doseꢀresponse
of the antagonistic effect of bivalent compound 13 (black) (IC50=15±3 nM) and monovalent 15
(red) (IC50=280±70 nM) on the DMR induced by 100 nM sumanirole. Data are mean±SEM from
3ꢀ8 experiments and are presented as percentage of the maximal effect of sumanirole.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Because of the higher affinity of 13, we predicted the TM5/6 interface for homodimerization
of D2R. To validate this hypothesis with a bimolecular fluorescence complementation (BiFC)
assay in HEKꢀ293T cells, we used synthetic peptides with the amino acid sequence of TMs 5 and
6 and TM7 (negative control) of D2R fused to the cellꢀpenetrating HIV transactivator of
transcription (TAT) peptide to alter interꢀprotomer interactions.13,26 In this assay, two
complementary halves of YFP (Venus variant; cYFP and nYFP) are separately fused to the D2
receptor and the fluorescence is obtained after reconstitution of the functional YFP when the D2
receptors homodimerize. Only the transmembrane peptide TATꢀTM6 bound to the receptor and
disturbed the quaternary structure of the homodimer, causing a significant fluorescence decrease
(Figure 3), indicating that only TM6 forms the interface of the D2R homodimer, according to the
recently reported results.26 We also tested the ability of compounds 13 and 15 to modulate the
dynamics of oligomerization, as it has been suggested for other bivalent compounds using TIRF
microscopy.17 With this aim, we treated the cells transfected with the two complementary halves
of YFP with 100 nM of compounds 13 or 15 for 10 minutes before reading the fluorescence.
12
ACS Paragon Plus Environment

Page 13 of 41
Journal of Medicinal Chemistry
1
2
3
4
Under our conditions, neither bivalent compound 13 nor monovalent 15 significantly altered the
5
6
dimerization state in fluorescence complementation assays (p > 0.793) (Figure S2).
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3500
3000
2500
***
2000
1500
1000
500
0
Figure 3. Effect of TATꢀTM peptides on disturbance of the D2R homodimer, determined by
BiFC experiments in HEKꢀ293T cells transfected with D2RꢀnYFP and D2RꢀcYFP cDNA. Values
are mean±SEM from 6ꢀ9 determinations. Statistical significance was calculated by oneꢀway
ANOVA followed by Bonferroni’s post hoc test. ***p<0.001 compared to non TATꢀTM treated
complementation.
Because we have identified the TATꢀTM6 peptide as a disturber of the interꢀprotomer
interaction, we tested the binding affinity of compounds 13 and 15 in the presence of TATꢀTM6
peptides of D2R and adenosine A2R (negative control) in native tissue (Figure 4). Neither TATꢀ
TM6 peptide of D2R nor A2R influenced the binding of monovalent compound 15. In contrast,
TATꢀTM6 peptide of D2R, but not TATꢀTM6 peptide of A2R, decreased the binding of the
bivalent ligand 13 (KDB1 (13)=0.021nM vs. KDB1 (13+TM6)=1.1nM). Remarkably, in the
presence of the TATꢀTM6 peptide of D2R, bivalent compound 13 performed as the monovalent
compound 15 (KDB1 (13+TM6)=1.1nM vs. KDB1 (15)=0.77nM).
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 41
1
2
3
4
A
B
5
6
7
8
120
100
80
60
40
20
0
120
100
80
60
40
20
0
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
-10
-8
-6
-4
-10
-8
-6
-4
log [compound 15] (M)
log [compound 13] (M)
Figure 4. Effect of TATꢀTM6 peptides of D2R and A2AR on competition experiments of
[3H]YMꢀ09151ꢀ2 vs. D2R ligands. Competition curves with increasing concentrations of
monovalent 15 (in A) or bivalent 13 (in B) D2R ligands in the absence (black) or in the presence
of TATꢀTM6 of A2AR (red) or TATꢀTM6 of D2R (blue), using membranes from sheep brain
striatum. Data are mean±SEM from 3 experiments performed in triplicate.
This suggests that the TATꢀTM6 peptide alters the homodimer in such a way that compound
13 binds the orthosteric binding site of the first protomer without reaching the second one. These
results show the importance of the simultaneous binding of the two pharmacophore units at both
orthosteric sites of the homodimer for obtaining an improvement in affinity, and confirm the
interꢀprotomer interaction of D2R homodimer through TM6. These results also ratify the bivalent
interaction mode of compound 13, validating it as a true bivalent ligand. Interestingly, in brain
striatal tissue, the effect caused by the TATꢀTM6 peptide seems enough to avoid the
simultaneous occupancy of both orthosteric binding sites of the homodimer by the bivalent
ligand but, in the BiFC assay performed with HEKꢀ293T cells, the homodimer is also bound by
the complemented fluorescent protein and this may hamper the disruption caused by the
14
ACS Paragon Plus Environment

Page 15 of 41
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
transmembrane peptide. In fact, at the beginning of the usage of TATꢀTM peptides to disturb
oligomers, some authors hypothesized that the complementation was irreversible. However,
more recently, different papers have reported that it is actually reversible.26ꢀ28
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Molecular modelling of bivalent ligand 13 into D2R homodimer model. Accordingly, we
constructed a computational model of the D2R homodimer, using exclusively TM6 as the
molecular interface (see Experimental Section), and performed 1ꢁs of unbiased MD simulations
to evaluate the stability of compound 13 in the model (Figures 5 and S3ꢀS4). This TM6 interface
predicts similar distances between orthosteric binding sites than the TM5/6 interface, thus,
leading to the same number of atoms for the spacer. The MD simulations showed that compound
13 comfortably fulfills and maintains simultaneous binding of the two pharmacophoric units at
both orthosteric sites throughout the simulation (Figure S3), thus, providing further confidence in
the bivalent interaction and the picomolar binding affinity.
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 41
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5. Evolution of bivalent ligand 13 in the D2R homodimer (TM1 and TM4, ECL1 and part
of ECL2 are omitted for clarity), constructed via the TM6 interface, as devised from MD
simulations. The structures of 13 (the color code of the atoms is as in Figure 1) are extracted
from the simulations (20 structures collected every 50 ns), whereas the structure of the D2R
homodimer corresponds to the initial model. A detailed analysis of the simulation (Figure S3)
confirms that the designed bivalent ligand 13 remains stable at the orthosteric binding cavities
through the unbiased 1 ꢁs MD simulation.
CONCLUSIONS
We have developed a precise strategy to create bivalent ligands of GPCR (homo/hetero)
dimers based on a versatile multivalent chemical platform. The use of computational tools that
consider the TM interfaces, distances between orthosteric binding sites and mode of interaction
of the pharmacophore units, allows a reduction in the number of synthesized bivalent ligands, yet
a high success in the affinity results. Bivalent ligand 13 showed picomolar binding affinity, and
the use of different TATꢀTM6 disturber peptides allowed the confirmation, in native tissue, of its
16
ACS Paragon Plus Environment

Page 17 of 41
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
simultaneous interaction with both orthosteric sites of the D2R homodimer, constituted through
TM6. Furthermore, our results confirm the recently described interface interaction of D2R
homodimer through TM6.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
This strategy can be applied to other GPCR oligomers, thus allowing the generation and
validation of novel ligands with a clear bivalent interaction mode. These ligands can be used as
pharmacological tools in combination with disturber TATꢀTM peptides to validate interꢀ
protomer GPCR interactions, both in vitro and in native tissue, and this information could be
potentially used for the design of new therapeutic compounds targeting GPCR oligomers.
EXPERIMENTAL SECTION
General Methods. Reagents and solvents were purchased from commercial sources and were
used without further purification. TLC was performed on Merck 60F254 silica plates were
visualized by UV light (254 nm), or by potassium permanganate stains. Flash chromatography
on silica was carried out on a Teledyne Isco Combiflash Rf instrument using Redisep Rf silica
1
columns. HꢀNMR (400 MHz) and 13CꢀNMR (101 MHz) spectroscopy was performed on a
Varian Mercury 400 MHz instrument at the NMR unit of the Scientific and Technological
Centers of the University of Barcelona (CCiTUB). Chemical shifts (δ) are expressed in ppm
relative to tetramethylsilane (TMS). Coupling constants (J) are expressed in Hertz (Hz). The
following abbreviations are used to indicate multiplicity: s: singlet; d: doublet, t: triplet, m:
multiplet, and br: broad signal. Analytical RPꢀHPLC and mass spectra were performed on a
Waters Alliance 2795 with an automated injector and a photodiode array detector Waters 2996
coupled to an electrospray ion source (ESIꢀMS) Micromass ZQ mass detector, using a XSelectTM
C18 reversedꢀphase analytical column (4.6 mm×50 mm, 3.5 ꢁm), and the MassLynx 4.1 software.
The instrument was operated in the positive ESI (+) ion mode. Analyses were carried out with
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 41
1
2
3
4
5
6
7
8
9
several elution systems. System A: a linear gradient 5–100% CH3CN (0.07% HCOOH) in H2O
(0.1% HCOOH) over 4.5 min at a flow rate of 2 mL/min; and System B: a linear gradient 5–
100% CH3CN (0.07% HCOOH) in H2O (0.1% HCOOH) over 3.5 min at a flow rate of
1.6 mL/min. Purity of all test compounds was determined by HPLC analysis to be ≥95 %. Highꢀ
Resolution Mass Spectroscopy (HRMS) was carried out using an LC/MSDꢀTOF spectrometer
from Agilent Technologies, at the molecular characterization mass spectrometry unit of the
Scientific and Technological Centers of the University of Barcelona (CCiTUB). Semiꢀ
preparative RPꢀHPLC purification was performed on a Waters system with a 2545 binary
gradient module, a 2767 manager collector and a 2489 UV detector, coupled to an electrospray
ion source (ESIꢀMS) Micromass ZQ mass detector, and the MassLynx 4.1 software. Gradients
and columns used are detailed in each case.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Synthesis.
4-((4-(2-(8-(4-(4-fluorophenyl)-4-oxobutyl)-4-oxo-1-phenyl-1,3,8-triazaspiro[4.5]decan-3-
yl)ethyl)phenyl)amino)-4-oxobutanoic acid (7). To a solution of 6 (267 mg, 519 ꢀmol, 1.0 eq)
in CH3CN (15 mL) was added succinic anhydride (62.3 mg, 623 ꢀmol, 1.2 eq) and the mixture
was stirred at room temperature overnight (16 h). After completion of the reaction the mixture
was evaporated to dryness. The resulting crude was dissolved in CH2Cl2 (20 mL) and
immediately washed with brine (2×20 mL). The resulting organic phase was dried over MgSO4
and evaporated. The crude was purified by flash chromatography on silica using CH2Cl2 and
MeOH as solvents (0 to 15% MeOH in CH2Cl2) to afford compound 7 as a white solid (244 mg,
1
1
397 ꢀmol, 76%). H NMR (400 MHz, DMSOꢀd6, 298 K) δ H NMR (400 MHz, DMSOꢀd6) δ
9.89 (s, NH), 8.10 – 8.01 (m, 2H), 7.52 – 7.45 (m, 2H), 7.39 – 7.30 (m, 2H), 7.22 – 7.12 (m, 4H),
6.80 – 6.71 (m, 3H), 4.57 (s, 2H), 3.54 (t, J = 7.2 Hz, 2H), 3.02 (t, J = 6.9 Hz, 2H), 2.83 (t, J =
18
ACS Paragon Plus Environment

Page 19 of 41
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
7.2 Hz, 2H), 2.78 – 2.60 (m, 4H), 2.56 – 2.45 (m, 4H), 2.45 – 2.30 (m, 4H), 1.87 – 1.76 (m, 2H),
1.45 – 1.36 (m, 2H); 13C NMR (101 MHz, DMSOꢀd6, 298 K) δ 198.5, 173.8, 173.0, 169.9, 164.8
(d, J = 251.1 Hz), 143.0, 137.6, 133.8 (d, J = 2.7 Hz), 132.9, 130.8 (d, J = 9.3 Hz), 129.0, 128.8,
118.9, 118.0, 115.6 (d, J = 21.9 Hz), 114.6, 62.8, 59.5, 57.0, 49.0, 41.2, 35.8, 32.0, 31.0, 28.9,
28.4, 21.3; HPLC: System A, tR: 2.02 min, 99% (214 nm), 99% (240 nm); LRMS: calculated
mass for C35H40FN4O5: 615.3 [M+H]+, found by HPLCꢀMS (ESI): 615.2.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2-oxo-7,10,13-trioxa-3-azahexadecan-16-aminium chloride (8). To a solution of 1ꢀ(tertꢀ
butoxycarbonylꢀamino)ꢀ4,7,10ꢀtrioxaꢀ13ꢀtridecanamine (401 mg, 1.25 mmol, 1.0 eq) in CH2Cl2
(4 mL) was added acetic anhydride (130 ꢀL, 1.38 mmol, 1.1 eq) and DIEA (470 ꢀL, 2.76 mmol,
2.2 eq). The resulting mixture was stirred at room temperature for 3 h. After this time, the crude
was washed with saturated NaHCO3 (2×5 mL) and brine (1×5 mL). The organic phase was dried
over MgSO4 and evaporated. Subsequent treatment of the crude with a 2 M solution of HCl in
dioxane (10 mL, 20 mmol, 16 eq) at room temperature for 1 h, followed by the evaporation of
dioxane and HCl to dryness, afforded compound 8 as a pale yellow oil (352 mg, 1.10 mmol,
88%). 1H NMR (400 MHz, D2O, 298 K) δ 3.73 – 3.64 (m, 10H), 3.58 (t, J = 6.4 Hz, 2H), 3.25 (t,
13
J = 6.8 Hz, 2H), 3.12 (t, J = 7.2 Hz, 2H), 2.01 – 1.92 (m, 5H), 1.83 – 1.75 (m, 2H); C NMR
(101 MHz, D2O, 298 K) δ 173.9, 69.5, 69.4, 69.3, 69.2, 68.3, 68.2, 37.6, 36.4, 28.1, 26.4, 21.8;
LRMS: calculated mass for C12H27ClN2O4 (hydrochloride): 298.2, calculated mass for
C12H27N2O4 (amine): 263.2 [M+H]+, found by HPLCꢀMS (ESI): 263.0.
2-(2-(2-acetamidoethoxy)ethoxy)ethan-1-aminium chloride (9). To a solution of 1ꢀ(tertꢀ
butoxycarbonylꢀamino)ꢀ3,6ꢀdioxaꢀ8ꢀoctanamine (120 mg, 0.48 mmol, 1.0 eq) in CH2Cl2 (4 mL)
was added acetic anhydride (50 ꢀL, 0.53 mmol, 1.1 eq) and DIEA (181 ꢀL, 1.06 mmol, 2.2 eq).
The resulting mixture was stirred at room temperature for 3 h. After this time, the crude was
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 41
1
2
3
4
5
6
7
8
9
washed with saturated NaHCO3 (2×5 mL) and brine (1×5 mL). The organic phase was dried over
MgSO4 and evaporated. Subsequent treatment of the crude with a 2 M solution of HCl in
dioxane (10 mL, 20 mmol, 16 eq) at room temperature for 1 h, followed by the evaporation of
dioxane and HCl to dryness, afforded compound 9 as a pale yellow oil (95.3 mg, 0.42 mmol,
87%). 1H NMR (400 MHz, D2O, 298 K) δ 3.76 (t, J = 5.0 Hz, 2H), 3.70 (s, 4H), 3.63 (t, J = 5.4
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
13
Hz, 2H), 3.38 (t, J = 5.4 Hz, 2H), 3.21 (br t, J = 5.1 Hz, 2H), 1.99 (s, 3H); C NMR (101 MHz,
D2O, 298 K) δ 174.3, 69.5, 69.4, 68.7, 66.3, 39.0, 38.9, 21.8; LRMS: calculated mass for
C8H19ClN2O3: 226.1 (hydrochloride), calculated mass for C8H19N2O3 (amine): 191.1 [M+H]+,
found by HPLCꢀMS (ESI): 190.9.
18-((4-(2-(8-(4-(4-fluorophenyl)-4-oxobutyl)-4-oxo-1-phenyl-1,3,8-triazaspiro[4.5]decan-
3-yl)ethyl)phenyl)amino)-15,18-dioxo-4,7,10-trioxa-14-azaoctadecan-1-aminium
chloride
(10). To a mixture of 7 (50.7 mg, 82.5 ꢀmol, 1.0 eq), EDC·HCl (23.8 mg, 0.12 mmol, 1.5 eq)
and HOBt·H2O (19.0 mg, 0.12 mmol, 1.5 eq) was added a solution of 1ꢀ(tertꢀbutoxycarbonylꢀ
amino)ꢀ4,7,10ꢀtrioxaꢀ13ꢀtridecanamine (39.7 mg, 0.12 mmol, 1.5 eq) in DMF (5 mL). The
resulting mixture was stirred at room temperature overnight (18 h). After this time the solvent
was evaporated to dryness. The crude was dissolved in AcOEt (15 mL) and washed with
saturated NaHCO3 (3×15 mL), 0.5% w/v citric acid (3×15 mL) and brine (1×15 mL). The
organic phase was dried over MgSO4 and evaporated to obtain the Bocꢀprotected compound
(44.3 mg, 48.3 ꢀmol). This compound was dissolved in dioxane (1 mL) and a 4 M solution of
HCl in dioxane (0.5 mL, 2.0 mmol, 41 eq) was added. The mixture was stirred at room
temperature for 1 h. Then, the dioxane and HCl were evaporated to dryness. Finally, the crude
was dissolved in H2O (1 mL) and lyophilized to afford compound 10 (40.1 mg, 47.0 ꢀmol, 57%).
1H NMR (400 MHz, D2O, 298 K) δ 8.05 – 7.97 (m, 2H), 7.43 – 7.32 (m, 4H), 7.32 – 7.19 (m,
20
ACS Paragon Plus Environment

Page 21 of 41
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
4H), 7.10 – 7.02 (m, 1H), 7.02 – 6.93 (m, 2H), 5.45 (s, NH), 4.64 (s, 2H), 3.76 – 3.67 (m, 2H),
3.68 – 3.57 (m, 8H), 3.57 – 3.50 (m, 2H), 3.52 – 3.34 (m, 6H), 3.28 – 3.02 (m, 8H), 2.93 (br t, J
= 6.5 Hz, 2H), 2.70 – 2.55 (m, 2H), 2.57 – 2.38 (m, 4H), 2.13 – 1.96 (m, 2H), 1.96 – 1.89 (m,
2H), 1.83 – 1.60 (m, 4H); 13C NMR (101 MHz, D2O, 298 K) δ 200.7, 174.3, 173.2, 172.7, 165.9
(d, J = 253.6 Hz), 141.7, 135.8, 134.9, 132.4, 131.0 (d, J = 9.6 Hz), 129.6, 129.6, 121.9, 121.2,
118.4, 115.8 (d, J = 22.0 Hz), 69.4, 69.3, 69.2, 68.2, 63.5, 59.1, 56.0, 48.8, 41.5, 37.6, 36.2, 34.9,
32.0, 31.8, 31.0, 28.2, 27.0, 26.4, 18.0; HPLC: System B, tR: 1.67 min, 98% (214 nm), 97% (240
nm); LRMS: calculated mass for C45H62ClFN6O7: 852.4 (hydrochloride), calculated mass for
C45H62FN6O7 (amine): 817.5 [M+H]+, found by HPLCꢀMS (ESI): 817.3.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2-(2-(2-(4-((4-(2-(8-(4-(4-fluorophenyl)-4-oxobutyl)-4-oxo-1-phenyl-1,3,8-
triazaspiro[4.5]decan-3-yl)ethyl)phenyl)amino)-4-oxobutanamido)ethoxy)ethoxy)ethan-1-
aminium chloride (11). To a mixture of 7 (50.6 mg, 82.3 ꢀmol, 1.0 eq), EDC·HCl (23.6 mg,
0.12 mmol, 1.5 eq) and HOBt·H2O (18.8 mg, 0.12 mmol, 1.5 eq) was added a solution of 1ꢀ(tertꢀ
butoxycarbonylꢀamino)ꢀ3,6ꢀdioxaꢀ8ꢀoctanamine (30.5 mg, 0.12 mmol, 1.5 eq) in DMF (5 mL).
The resulting mixture was stirred at room temperature overnight (18 h). After this time the
solvent was evaporated to dryness. The crude was dissolved in AcOEt (15 mL) and washed with
saturated NaHCO3 (3×15 mL), 0.5% w/v citric acid (3×15 mL) and brine (1×15 mL). The
organic phase was dried over MgSO4 and evaporated to obtain the Bocꢀprotected compound
(50.9 mg, 60.2 ꢀmol). This compound was dissolved in dioxane (1 mL) and a 4 M solution of
HCl in dioxane (0.5 mL, 2.0 mmol, 39 eq) was added. The mixture was stirred at room
temperature for 1 h. Then, the dioxane and HCl were evaporated to dryness. Finally, the crude
was dissolved in H2O (1 mL) and lyophilized to afford compound 11 (36.0 mg, 46.1 ꢀmol, 56%).
1H NMR (400 MHz, D2O, 298 K) δ 8.04 – 7.94 (m, 2H), 7.42 – 7.29 (m, 4H), 7.31 – 7.17 (m,
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 41
1
2
3
4
5
6
7
8
9
4H), 7.09 – 7.02 (m, 1H), 7.00 – 6.92 (m, 2H), 4.63 (s, 2H), 3.77 – 3.65 (m, 4H), 3.66 – 3.58 (m,
4H), 3.54 (t, J = 5.5 Hz, 2H), 3.47 – 3.33 (m, 4H), 3.32 (t, J = 5.5 Hz, 2H), 3.25 – 3.04 (m, 6H),
2.92 (br t, J = 6.6 Hz, 2H), 2.69 – 2.57 (m, 2H), 2.57 – 2.50 (m, 2H), 2.52 – 2.38 (m, 2H), 2.11 –
1.93 (m, 2H), 1.65 (d, J = 14.5 Hz, 2H); 13C NMR (101 MHz, D2O, 298 K) δ 200.7, 174.6,
173.2, 172.8, 165.9 (d, J = 253.8 Hz), 141.6, 135.7, 134.9, 132.4, 131.0 (d, J = 9.9 Hz), 129.6,
129.6, 121.8, 121.2, 118.4, 115.8 (d, J = 22.2 Hz), 69.5, 69.4, 68.8, 66.3, 63.4, 59.1, 56.0, 48.7,
41.5, 39.0, 38.8, 34.9, 32.0, 31.6, 30.7, 27.0, 18.0; HPLC: System B, tR: 1.60 min, 99% (214
nm), 97% (240 nm); LRMS: calculated mass for C41H54ClFN6O6: 780.4 (hydrochloride),
calculated mass for C41H54FN6O6 (amine): 745.4 [M+H]+, found by HPLCꢀMS (ESI): 745.2.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Methyl
24-((4-(2-(8-(4-(4-fluorophenyl)-4-oxobutyl)-4-oxo-1-phenyl-1,3,8-
triazaspiro[4.5]decan-3-yl)ethyl)phenyl)amino)-3-(21-((4-(2-(8-(4-(4-fluorophenyl)-4-
oxobutyl)-4-oxo-1-phenyl-1,3,8-triazaspiro[4.5]decan-3-yl)ethyl)phenyl)amino)-2,18,21-
trioxo-7,10,13-trioxa-3,17-diazahenicosyl)-5,21,24-trioxo-10,13,16-trioxa-3,6,20-
triazatetracosanoate (12). Compound 2 (2.6 mg, 12.7 ꢀmol, 1.0 eq), compound 10 (21.6 mg,
25.4 ꢀmol, 2.0 eq), EDC·HCl (7.3 mg, 38.0 ꢀmol, 3.0 eq) and HOBt·H2O (5.8 mg, 38.0 ꢀmol,
3.0 eq) were dissolved in DMF (2 mL) and DIEA (7.0 ꢀL, 41.1 ꢀmol, 3.2 eq) was added. The
resulting mixture was stirred at room temperature overnight (16 h). After this time the solvent
was evaporated to dryness, and the crude was purified by semiꢀpreparative reversedꢀphase HPLC
(45 to 72% acetonitrile in aqueous 10 mM NH4HCO3 in 8 min, XBridge C18 19×150 mm 5ꢀm)
1
affording compound 12 (11.3 mg, 6.27 ꢀmol, 49%). H NMR (400 MHz, CDCl3, 298 K) δ 9.21
(s, 2 NH), 8.05 – 7.94 (m, 4H), 7.76 – 7.65 (m, 2 NH), 7.52 – 7.42 (m, 4H), 7.33 – 7.21 (m, 4H),
7.20 – 7.08 (m, 8H), 7.05 – 6.97 (m, 2 NH), 6.91 – 6.78 (m, 6H), 4.58 (s, 4H), 3.77 – 3.65 (m,
7H), 3.65 – 3.42 (m, 30H), 3.42 – 3.23 (m, 16H), 3.12 (t, J = 6.7 Hz, 4H), 3.09 – 2.95 (m, 8H),
22
ACS Paragon Plus Environment

Page 23 of 41
Journal of Medicinal Chemistry
1
2
3
4
2.91 (t, J = 7.1 Hz, 4H), 2.71 – 2.50 (m, 8H), 2.17 (br s, 4H), 1.82 – 1.66 (m, 8H), 1.54 (d, J =
5
6
13
14.4 Hz, 4H); C NMR (101 MHz, CDCl3, 298 K) δ 172.6, 172.2, 171.1, 170.8, 166.0 (d, J =
7
8
9
255.4 Hz), 142.2, 137.5, 133.0, 132.9, 130.8 (d, J = 9.1 Hz), 129.7, 129.3, 120.2, 119.7, 115.9 (d,
J = 21.9 Hz), 114.9, 77.2, 70.6, 70.2, 70.1, 69.9, 69.8, 69.4, 63.6, 58.7, 56.6, 56.0, 52.0, 48.8,
41.8, 38.0, 37.2, 35.6, 33.1, 31.7, 29.4, 29.0, 27.4; HPLC: System A, tR: 2.18 min, >99% (214
nm), >99% (240 nm); LRMS: calculated mass for C97H130F2N13O18: 1803.0 [M+H]+, found by
HPLCꢀMS (ESI): 1803.0, 902.3 [M+2H]2+, 601.9 [M+3H]3+. HRMS (ESI): calculated exact
mass for C97H130F2N13O18 [M+H]+: 1802.9619, found 1802.9618.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Methyl
19-((4-(2-(8-(4-(4-fluorophenyl)-4-oxobutyl)-4-oxo-1-phenyl-1,3,8-
triazaspiro[4.5]decan-3-yl)ethyl)phenyl)amino)-3-(16-((4-(2-(8-(4-(4-fluorophenyl)-4-
oxobutyl)-4-oxo-1-phenyl-1,3,8-triazaspiro[4.5]decan-3-yl)ethyl)phenyl)amino)-2,13,16-
trioxo-6,9-dioxa-3,12-diazahexadecyl)-5,16,19-trioxo-9,12-dioxa-3,6,15-
triazanonadecanoate (13). Compound 2 (2.7 mg, 13.1 ꢀmol, 1.0 eq), compound 11 (20.5 mg,
26.2 ꢀmol, 2.0 eq), EDC·HCl (7.5 mg, 39.3 ꢀmol, 3.0 eq) and HOBt·H2O (6.0 mg, 39.3 ꢀmol,
3.0 eq) were dissolved in DMF (2 mL) and DIEA (7.0 ꢀL, 41.1 ꢀmol, 3.1 eq) was added. The
resulting mixture was stirred at room temperature overnight (16 h). After this time the solvent
was evaporated to dryness, and the crude was purified by semiꢀpreparative reversedꢀphase HPLC
(45 to 67% acetonitrile in aqueous 10 mM NH4HCO3 in 8 min, XBridge C18 19×150 mm 5ꢀm)
affording compound 13 (14.0 mg, 8.44 ꢀmol, 64 %).
1H NMR (400 MHz, CDCl3, 298 K) δ 9.11 (s, 2 NH), 8.04 – 7.93 (m, 4H), 7.81 (br t, J = 5.7
Hz, 2 NH), 7.51 – 7.41 (m, 4H), 7.30 – 7.23 (m, 4H), 7.22 – 7.16 (m, 2 NH), 7.16 – 7.09 (m,
8H), 6.93 – 6.87 (m, 4H), 6.87 – 6.81 (m, 2H), 4.57 (s, 4H), 3.74 – 3.63 (m, 7H), 3.58 – 3.31 (m,
38H), 3.12 (t, J = 6.7 Hz, 4H), 3.05 – 2.94 (m, 8H), 2.91 (t, J = 6.9 Hz, 4H), 2.70 – 2.54 (m, 8H),
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 41
1
2
3
4
13
2.24 – 2.11 (m, 4H), 1.50 (d, J = 14.2 Hz, 4H); C NMR (101 MHz, CDCl3, 298 K) δ 197.1,
5
6
7
8
173.2, 172.9, 172.1, 171.1, 171.0, 166.0 (d, J = 255.0 Hz), 142.3, 137.4, 133.1, 133.0, 130.8 (d, J
= 9.3 Hz), 129.7, 129.3, 120.3, 119.6, 115.9 (d, J = 21.9 Hz), 115.1, 70.4, 69.8, 63.6, 59.0, 58.5,
56.3, 55.9, 52.0, 48.4, 41.7, 39.5, 39.2, 35.7, 33.0, 31.6, 27.2, 18.8; HPLC: System A, tR: 2.08
min, 98% (214 nm), >99% (240 nm); LRMS: calculated mass for C89H114F2N13O16: 1658.8
[M+H]+, found by HPLCꢀMS (ESI): 1658.9, 830.3 [M+2H]2+, 553.9 [M+3H]3+. HRMS (ESI):
calculated exact mass for C89H114F2N13O16 [M+H]+: 1658.8469, found: 1658.8454.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Methyl
3-(2,18-dioxo-7,10,13-trioxa-3,17-diazanonadecyl)-24-((4-(2-(8-(4-(4-
fluorophenyl)-4-oxobutyl)-4-oxo-1-phenyl-1,3,8-triazaspiro[4.5]decan-3-
yl)ethyl)phenyl)amino)-5,21,24-trioxo-10,13,16-trioxa-3,6,20-triazatetracosanoate
(14).
Compound 2 (45.0 mg, 219 ꢀmol, 1.0 eq) and EDC·HCl (42.0 mg, 219 ꢀmol, 1.0 eq) were
dissolved in dry DMF (1 mL) and the mixture was stirred at room temperature for 2 h under Ar
atmosphere. Then, a solution of compound 8 (65.5 mg, 219 ꢀmol, 1.0 eq) and DIEA (75 ꢀL, 441
ꢀmol, 2.0 eq) in dry DMF (1 mL) was added and the resulting mixture was stirred at room
temperature for 90 min. After this time the solvent was evaporated to dryness, and the crude was
purified by Waters Porapak™ Rxn RP column (aqueous 10 mM NH4HCO3) to afford the
intermediate 3ꢀ(2ꢀmethoxyꢀ2ꢀoxoethyl)ꢀ5,21ꢀdioxoꢀ10,13,16ꢀtrioxaꢀ3,6,20ꢀtriazadocosanoic acid
1
(78.4 mg, 174 ꢀmol, 79%). H NMR (400 MHz, D2O, 298 K) δ 3.70 (s, 3H), 3.68 – 3.59 (m,
8H), 3.59 – 3.50 (m, 6H), 3.37 (s, 2H), 3.34 – 3.24 (m, 4H), 3.21 (t, J = 6.8 Hz, 2H), 1.95 (s,
13
3H), 1.85 – 1.70 (m, 4H); C NMR (101 MHz, D2O, 298 K) δ 178.7, 174.0, 173.9, 173.8, 69.5,
69.3, 69.2, 68.3, 68.2, 58.2, 58.2, 55.4, 52.0, 36.4, 35.9, 28.2, 28.1, 21.8; MS: calculated exact
mass for C19H36N3O9: 450.2 [M+H]+, found by HPLCꢀMS (ESI): 450.2. This intermediate (8.0
mg, 17.8 ꢀmol, 1.0 eq), compound 10 (16.7 mg, 19.6 ꢀmol, 1.1 eq), EDC·HCl (5.1 mg, 26.7
24
ACS Paragon Plus Environment

Page 25 of 41
Journal of Medicinal Chemistry
1
2
3
4
ꢀmol, 1.5 eq) and HOBt·H2O (4.1 mg, 26.7 ꢀmol, 1.5 eq) were dissolved in DMF (1.5 mL) and
5
6
7
8
DIEA (7.0 ꢀL, 41.1 ꢀmol, 2.2 eq) was added. The resulting mixture was stirred at room
temperature overnight (15 h). After this time the solvent was evaporated to dryness, and the
crude was purified by semiꢀpreparative reversedꢀphase HPLC (37 to 45% acetonitrile in aqueous
10 mM NH4HCO3 in 8 min, XBridge C18 19×150 mm 5ꢀm) affording compound 14 (5.5 mg,
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
4.40 ꢀmol, 25%). H NMR (400 MHz, CDCl3, 298 K) δ 9.19 (br s, NH), 8.03 – 7.94 (m, 2H),
7.68 (br s, 2 NH), 7.51 – 7.43 (m, 2H), 7.31 – 7.22 (m, 2H), 7.20 – 7.09 (m, 4H), 7.00 (br s, NH),
6.90 – 6.78 (m, 3H), 6.56 (br s, NH), 4.58 (s, 2H), 3.76 – 3.66 (m, 5H), 3.66 – 3.42 (m, 28H),
3.42 – 3.26 (m, 14H), 3.22 – 2.97 (m, 6H), 2.93 (t, J = 7.0 Hz, 2H), 2.70 – 2.54 (m, 4H), 2.27 –
2.11 (m, 2H), 1.95 (s, 3H), 1.82 – 1.67 (m, 8H), 1.52 (d, J = 14.8 Hz, 2H); 13C NMR (101 MHz,
CDCl3, 298 K) δ 172.7, 172.1, 171.1, 170.8, 170.5, 137.5, 132.8, 130.8 (d, J = 9.4 Hz), 129.7,
129.3, 120.2, 119.7, 116.0 (d, J = 21.8 Hz), 114.8, 70.6, 70.2, 70.1, 70.0, 69.8, 69.6, 69.5, 63.6,
58.8, 58.7, 56.0, 52.0, 48.8, 41.7, 38.1, 38.0, 37.3, 37.3, 35.5, 33.2, 33.1, 32.1, 31.8, 29.9, 29.8,
29.4, 29.1, 29.0, 27.1, 23.4, 22.8; HPLC: System A, tR: 1.98 min, 99% (214 nm), 98% (240 nm);
LRMS: calculated mass for C64H95FN9O15: 1248.7 [M+H]+, found by HPLCꢀMS (ESI): 1248.5,
624.9 [M+2H]2+. HRMS (ESI): calculated exact mass for C64H95FN9O15: 1248.6926 [M+H]+,
found: 1248.6936.
Methyl 3-(2,13-dioxo-6,9-dioxa-3,12-diazatetradecyl)-19-((4-(2-(8-(4-(4-fluorophenyl)-4-
oxobutyl)-4-oxo-1-phenyl-1,3,8-triazaspiro[4.5]decan-3-yl)ethyl)phenyl)amino)-5,16,19-
trioxo-9,12-dioxa-3,6,15-triazanonadecanoate (15). Compound 2 (42.1 mg, 205 ꢀmol, 1.0 eq)
and EDC·HCl (39.3 mg, 205 ꢀmol, 1.0 eq) were dissolved in dry DMF (1 mL) and the mixture
was stirred at room temperature for 2 h under Ar atmosphere. Then, a solution of compound 9
(46.5 mg, 205 ꢀmol, 1.0 eq) and DIEA (70 ꢀL, 412 ꢀmol, 2.0 eq) in dry DMF (1 mL) was added
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 41
1
2
3
4
5
6
7
8
and the resulting mixture was stirred at room temperature for 90 min. After this time the solvent
was evaporated to dryness, and the crude was purified by Waters Porapak™ Rxn RP column
(aqueous 10 mM NH4HCO3) to afford the intermediate 3ꢀ(2ꢀmethoxyꢀ2ꢀoxoethyl)ꢀ5,16ꢀdioxoꢀ
9
1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
9,12ꢀdioxaꢀ3,6,15ꢀtriazaheptadecanoic acid (62.1 mg, 164 ꢀmol, 80%). H NMR (400 MHz,
D2O, 298 K) δ 3.71 (s, 3H), 3.68 – 3.56 (m, 10H), 3.44 (t, J = 5.5 Hz, 2H), 3.41 (s, 2H), 3.36 (t, J
= 5.3 Hz, 2H), 3.30 (s, 2H), 1.98 (s, 3H); 13C NMR (101 MHz, D2O, 298 K) δ 178.6, 174.3,
174.1, 174.0, 69.4, 68.7, 68.7, 58.1, 58.1, 55.2, 52.0, 38.9, 38.5, 21.7; LRMS: calculated mass for
C15H28N3O8: 378.2 [M+H]+, found by HPLCꢀMS (ESI): 378.1. This intermediate (7.25 mg, 19.2
ꢀmol, 1.0 eq), compound 11 (16.5 mg, 21.1 ꢀmol, 1.1 eq), EDC·HCl (5.5 mg, 28.8 ꢀmol, 1.5 eq)
and HOBt·H2O (4.4 mg, 28.8 ꢀmol, 1.5 eq) were dissolved in DMF (1.5 mL) and DIEA (7.5 ꢀL,
44.1 ꢀmol, 2.3 eq) was added. The resulting mixture was stirred at room temperature overnight
(15 h). After this time the solvent was evaporated to dryness, and the crude was purified by semiꢀ
preparative reversedꢀphase HPLC (35 to 43% acetonitrile in aqueous 10 mM NH4HCO3 in 8
1
min, XBridge C18 19×150 mm 5ꢀm) affording compound 15 (5.0 mg, 4.53 ꢀmol, 24%). H
NMR (400 MHz, CDCl3, 298 K) δ 8.87 (br s, NH), 8.04 – 7.95 (m, 2H), 7.69 – 7.57 (m, 2 NH),
7.50 – 7.41 (m, 2H), 7.31 – 7.20 (m, 2H), 7.19 – 7.10 (m, 4H), 6.97 (br s, NH), 6.90 – 6.79 (m,
3H), 6.55 (br s, NH), 4.57 (s, 2H), 3.75 – 3.65 (m, 5H), 3.66 – 3.27 (m, 34H), 3.10 (br s, 6H),
2.92 (t, J = 6.8 Hz, 2H), 2.70 – 2.57 (m, 4H), 2.14 (br s, 2H), 1.98 (s, 3H), 1.51 (d, J = 14.3 Hz,
13
2H); C NMR (101 MHz, CDCl3, 298 K) δ 172.8, 172.2, 170.9, 170.9, 170.8, 170.7, 166.0 (d,
J= 254.9 Hz), 137.3, 130.8 (d, J = 9.4 Hz), 129.6, 129.3, 120.2, 119.6, 115.9 (d, J = 21.9 Hz) ,
115.1, 70.4, 70.3, 70.1, 69.9, 69.8, 63.7, 58.7, 56.0, 52.1, 49.0, 39.6, 39.5, 39.2, 39.2, 33.1, 32.1,
31.7, 29.8, 23.3, 22.8; HPLC: System A, tR: 1.90 min, 96% (214 nm), 96% (240 nm); LRMS:
calculated mass for C56H79FN9O13: 1104.6 [M+H]+, found by HPLCꢀMS (ESI): 1104.8, 552.9
26
ACS Paragon Plus Environment

Page 27 of 41
Journal of Medicinal Chemistry
1
2
3
4
5
[M+2H]2+. HRMS (ESI): calculated exact mass for C56H79FN9O13 [M+H]+: 1104.5776, found:
1104.5799.
6
7
8
9
TM with TAT peptides. A peptide derived from the HIV transactivator of transcription, HIV
TAT (YGRKKRRQRRR), was fused to peptides with the amino acid sequences of human A2AR
or D2R TM domain 6, human D2R TM domain 5 and human D2R TM domain 7 (Genemed
Synthesis), to promote integration of the TM domains in the plasma membrane. Because HIV
TAT binds to the phosphatidylinositolꢀ(4, 5)ꢀbisphosphate found on the inner surface of the
membrane, HIV TAT peptide was fused to the Nꢀterminus of TM6 and to the Cꢀterminus of TM5
and TM7 to obtain the right orientation of the inserted peptide. The amino acid sequences were:
TATꢀTM6 of D2R: YGRKKRRQRRR M374LAIVLGVFIICWLPFFITHIL395;
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
TATꢀTM6 of A2AR: YGRKKRRQRRRL235AIIVGLFALCWLPLHIINCFTFF258;
TM5ꢀTAT of D2R: F189VVYSSIVSFYVPFIVTLLVYIKIY213YGRKKRRQRRR;
TM7ꢀTAT of D2R: A410FTWLGYVNSAVNPIIYTTFNI431YGRKKRRQRRR.
Radioligand binding experiments. Brains of male and female sheep of 4ꢀ6 months old were
freshly obtained from the local slaughterhouse. Striatal brain tissues were disrupted with a
Polytron homogenizer (PTA 20 TS rotor, setting 3; Kinematica, Basel, Switzerland) for two 5 sꢀ
periods in 10 volumes of 50 mM TrisꢀHCl buffer at pH 7.4, containing a proteinase inhibitor
cocktail (Sigma, St. Louis, MO, USA). Membranes were obtained by centrifugation, twice at
105000 g for 45 min at 4°C. The pellet was stored at −80°C, washed once more as described
above and resuspended in 50 mM TrisꢀHCl buffer for immediate use. Membrane protein was
quantified by the bicinchoninic acid method (Pierce Chemical Co., Rockford, IL, USA) using
bovine serum albumin dilutions as standard. Binding experiments were performed with
membrane suspensions at room temperature in 50 mM TrisꢀHCl buffer at pH 7.4, containing 10
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 41
1
2
3
4
5
6
7
8
9
mM MgCl2. For D2R competitionꢀbinding assays, membrane suspensions (0.2 mg of protein/mL)
were incubated for 2 h with a constant free concentration of 0.8 nM of the D2R antagonist
[3H]YMꢀ09151ꢀ2 (KDA1 = 0.15 nM) and increasing concentrations of each tested ligand. Nonꢀ
specific binding was determined in the presence of 30 µM of dopamine, because at this
concentration dopamine does not displace the radioligand from sigma receptors. Competitionꢀ
binding assays using TATꢀTM peptides were performed as described previously, but
preincubating peptides and membranes for 1 h before the addition of the ligands and the
radioligand. In all cases, free and membraneꢀbound ligands were separated by rapid filtration of
500 ꢀL aliquots in a cell harvester (Brandel, Gaithersburg, MD, USA) through Whatman GF/C
filters embedded in 0.3% polyethylenimine, that were subsequently washed for 5 s with 5 mL of
iceꢀcold 50 mM TrisꢀHCl buffer. The filters were incubated with 10 mL of Ecoscint H
scintillation cocktail (National Diagnostics, Atlanta, GA, USA) overnight at room temperature
and radioactivity counts were determined using a TriꢀCarb 2800 TR scintillation counter
(PerkinElmer) with an efficiency of 62%. Radioligand competition curves were analyzed by
nonlinear regression using the commercial Grafit curveꢀfitting software (Erithacus Software,
Surrey, UK) by fitting the binding data to the mechanistic twoꢀstate dimer receptor model.23 The
macroscopic equilibrium dissociation constants from competition experiments were determined
applying the following general equation:
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ꢁDA2
ꢀꢁDA2 A + 2 A2 +
A Bꢂ ꢃꢄ
ꢁDAB
ꢁDA2 A B
Abound
=
2
ꢁDA1 ꢁDA2
ꢁDB1
B
ꢁDA1 ꢁDA2 B
ꢁDA1 ꢁDA2 + ꢁDA2 A + A2 +
+
+
ꢁDAB
ꢁDB1 ꢁDB2
where A represents the radioligand concentration, B the assayed competing compound
concentration and KDAB the hybrid allosteric modulation between A and B. For A and B nonꢀ
28
ACS Paragon Plus Environment

Page 29 of 41
Journal of Medicinal Chemistry
1
2
3
4
cooperative and nonꢀallosteric modulation between A and B, the equation can be simplified due
5
6
to the fact that KDA2 = 4KDA1, KDB2 = 4KDB1 and KDAB = 2KDB1
;
7
8
9
2 ꢁDA1 A B
ꢁDB1
ꢀ4 ꢁDA1 A + 2 Aꢅ +
ꢂ ꢃꢆ
Abound
=
2 ꢁDA1 A B 4 ꢁDA1ꢅ B
ꢁDA1ꢅ Bꢅ
ꢁDB1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4ꢁDA1ꢅ + 4ꢁDA1 A + Aꢅ +
+
+
ꢅ
ꢁDB1
ꢁDB1
Cell culture. CHO cells stably coꢀexpressing the human cDNAs of A2AR and D2R were
obtained and tested as described in Orru et al. (2011).29 This clone was grown in Minimum
Essential Medium (MEMα; Gibco) supplemented with 2 mM Lꢀglutamine, 100 ꢁg/mL sodium
pyruvate, MEM nonessential amino acid solution (1/100), 100U/mL penicillin/streptomycin, 5%
(vol/vol) of heatꢀinactivated FBS (all supplements from Invitrogen) and with 600 mg/mL
Geneticin (G 418 Sulfate, Calbiochem) and 300 mg/mL Hygromycin B (Invitrogen). HEKꢀ293T
cells were grown in Dulbecco’s modified Eagle’s medium supplemented with 2 mM Lꢀ
glutamine, 100 ꢁg/mL sodium pyruvate, MEM nonessential amino acid solution (1/100),
100U/mL penicillin/streptomycin, 5% (vol/vol) of heatꢀinactivated FBS. All cells were cultured
at 37ºC and 5% CO2.
Dynamic Mass Redistribution (DMR) Assay. The global cell signaling profile was measured
using an EnSpire Multimode Plate Reader (PerkinElmer, Waltham, Massachusetts, US). This
labelꢀfree approach uses refractive waveguide grating optical biosensors, integrated into 384ꢀ
well microplates. Changes in local optical density are measured in a detection zone up to 150 nm
above the surface of the sensor. Cellular mass movements induced upon receptor activation are
detected by illuminating the underside of the biosensor with polychromatic light and measured as
changes in the wavelength of the reflected monochromatic light. These changes are a function of
the refraction index. The magnitude of this wavelength shift (in picometers) is directly
29
ACS Paragon Plus Environment