1-[3-(4-Butylpiperidin-1-yl)propyl]-1,2,3,4-tetrahydroquinolin-2-one (77-LH-28-1) as a Model for the Rational Design of a Novel Class of Brain Penetrant Ligands with High Affinity and Selectivity for Dopamine D4 Receptor

Article abstract of DOI:10.1021/acs.jmedchem.8b00265

In the present article, the M₁ mAChR bitopic agonist 1-[3-(4-butylpiperidin-1-yl)propyl]-1,2,3,4-tetrahydroquinolin-2-one (77-LH-28-1, 1) has been demonstrated to show unexpected D₄R selectivity over D₂R and D₃R and to behave as a D₄R antagonist. To better understand the structural features required for the selective interaction with the D₄R and to obtain compounds unable to activate mAChRs, the aliphatic butyl chain and the piperidine nucleus of 1 were modified, affording compounds 2-14. The 4-benzylpiperidine 9 and the 4-phenylpiperazine 12 showed high D₄R affinity and selectivity not only over the other D₂-like subtypes, but also over M₁-M₅ mAChRs. Derivative 12 was also highly selective over some selected off-targets. This compound showed biased behavior, potently and partially activating G_i protein and inhibiting β-arrestin2 recruitment in functional studies. Pharmacokinetic studies demonstrated that it was characterized by a relevant brain penetration. Therefore, 12 might be a useful tool to better clarify the role played by D₄R in disorders in which this subtype is involved.

Full text of DOI:10.1021/acs.jmedchem.8b00265

Subscriber access provided by UNIVERSITY OF TOLEDO LIBRARIES
1-[3-(4-Butylpiperidin-1-yl)propyl]-1,2,3,4-tetrahydroquinolin-2-one (77-
LH-28-1) as a Model for the Rational Design of a Novel Class of Brain
Penetrant Ligands with High Affinity and Selectivity for Dopamine D4 Receptor
Fabio Del Bello, Alessandro Bonifazi, Gianfabio Giorgioni, Carlo Cifani, Maria Vittoria
Micioni Di Bonaventura, Riccardo Petrelli, Alessandro Piergentili, Stefano Fontana,
Valerio Mammoli, Hideaki Yano, Rosanna Matucci, Giulio Vistoli, and Wilma Quaglia
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b00265 • Publication Date (Web): 28 Mar 2018
Downloaded from http://pubs.acs.org on March 28, 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 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
1-[3-(4-Butylpiperidin-1-yl)propyl]-1,2,3,4-tetrahydroquinolin-2-one (77-LH-28-1) as a
Model for the Rational Design of a Novel Class of Brain Penetrant Ligands with High
Affinity and Selectivity for Dopamine D4 Receptor
8
9
Fabio Del Bello,† Alessandro Bonifazi,∥ Gianfabio Giorgioni,† Carlo Cifani,† Maria Vittoria
Micioni Di Bonaventura,† Riccardo Petrelli,† Alessandro Piergentili*,† Stefano Fontana,§
Valerio Mammoli,§ Hideaki Yano,∥ Rosanna Matucci,‡ Giulio Vistoli,∇ and Wilma Quaglia†
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
†Scuola di Scienze del Farmaco e dei Prodotti della Salute, Università di Camerino, Via S.
Agostino 1, 62032 Camerino, Italy
∥Molecular Targets and Medications Discovery Branch, National Institute on Drug Abuse –
Intramural Research Program, National Institutes of Health, Baltimore, Maryland, 333
Cassell Drive, Baltimore, Maryland 21224
§Center for Drug Discovery and Development-DMPK, Aptuit, an Evotec Company, Via A.
Fleming, 4, 37135 Verona, Italy
∥Molecular Targets and Medications Discovery Branch, National Institute on Drug Abuse –
Intramural Research Program, National Institutes of Health, Baltimore, Maryland, 333
Cassell Drive, Baltimore, Maryland 21224
‡Dipartimento di Neuroscienze, Psicologia, Area del Farmaco e Salute del Bambino
(NEUROFARBA), Sezione di Farmacologia e Tossicologia, Università degli Studi di
Firenze, Viale Pieraccini 6, 50139 Firenze, Italy
∇Dipartimento di Scienze Farmaceutiche, Università degli Studi di Milano, Via Mangiagalli
25, 20133 Milano, Italy
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 46
1
2
Abstract
3
4
5
6
7
8
9
In the present paper, the M1 mAChR bitopic agonist 1ꢀ[3ꢀ(4ꢀbutylpiperidinꢀ1ꢀyl)propyl]ꢀ
1,2,3,4ꢀtetrahydroquinolinꢀ2ꢀone (77ꢀLHꢀ28ꢀ1, 1) has been demonstrated to show
unexpected D4R selectivity over D2R and D3R and to behave as a D4R antagonist. To better
understand the structural features required for the selective interaction with the D4R and
obtain compounds unable to activate mAChRs, the aliphatic butyl chain and the piperidine
nucleus of 1 were modified, affording compounds 2ꢀ14. The 4ꢀbenzylpiperidine 9 and the 4ꢀ
phenylpiperazine 12 showed high D4R affinity and selectivity not only over the other D2ꢀlike
subtypes, but also over M1ꢀM5 mAChRs. Derivative 12 was also highly selective over some
selected offꢀtargets. This compound showed biased behavior, potently and partially
activating Gi protein and inhibiting βꢀarrestin2 recruitment in functional studies.
Pharmacokinetic studies demonstrated that it was characterized by a relevant brain
penetration. Therefore, 12 might be a useful tool to better clarify the role played by D4R in
disorders in which this subtype is involved.
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
ACS Paragon Plus Environment

Page 3 of 46
Journal of Medicinal Chemistry
1
2
INTRODUCTION
3
4
5
6
7
8
9
Muscarinic acetylcholine receptors (mAChRs) are G proteinꢀcoupled receptors and include
five subtypes, namely M1ꢀM5.1 Among the five mAChRs, the M1 subtype is the most
abundantly expressed in the brain, including cortex, hippocampus, amygdala and striatum.2
The M1 subtype is believed to be important for cognition and attention, with potential benefit
for the treatment of cognitive symptoms exhibited by those suffering from Alzheimer’s
disease (AD) and schizophrenia.3 The search for a fully selective M1 agonist has been a
pharmaceutical quest for many decades.4 Among the several compounds published so far,
one of the first was 77ꢀLHꢀ28ꢀ1 (1ꢀ[3ꢀ(4ꢀbutylꢀ1ꢀpiperidinyl)propyl]ꢀ3,4ꢀdihydroꢀ2(1H)ꢀ
quinolinone, 1) (Figure 1) that selectively activates the M1 subtype over the other four M2ꢀ
M5 mAChR subtypes.5 Initially classified as an allosteric agonist, 1 is now considered as a
bitopic agonist, interacting concomitantly with the orthosteric site and part of an allosteric
site.6 The muscarinic agonist 1 extends down into the orthosteric site as well as up toward
extracellular receptor regions associated with an allosteric site. The piperidine basic center of
1 interacts with the Asp105 of the third transmembrane (TM) segment in the orthosteric
binding site. Mutational data support the requirement of such a residue for the receptor
activation by this muscarinic agonist.7 However, the 3,4ꢀdihydroquinolinꢀ2(1H)ꢀone
aromatic group of 1 has been predicted to occupy space between TM2, TM3, and TM7
domains, near the extracellular loops. In preꢀclinical models of schizophrenia and
Alzheimer’s disease, 1 displayed antipsychotic and cognitionꢀenhancing efficacy.5
Unfortunately, its efficacy was limited by nonselective effects on other receptors, including
dopamine D2 receptor subtype (D2R).8 The D2R belongs to the D2ꢀlike receptor family that
also includes D3 (D3R) and D4 (D4R) subtypes, which activate Gαi/o to inhibit adenylate
cyclase activity and cAMP production.9 The Gαi/o class of G proteins is inactivated by
pertussis toxinꢀcatalyzed ADPꢀribosylation. The DR signaling is now widely appreciated to
be more complex than simply changes in cAMP levels. Studies have shown that D2ꢀlike
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
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 46
1
2
3
4
receptors modulate numerous other signaling pathways, including phospholipases, ion
channels, mitogen activated protein (MAP) kinases, and the Na+/H+ exchanger.10
5
6
7
Several psychiatric and neurological disorders, in particular those involving the D2ꢀlike
8
9
subtypes,
including
Parkinson's
disease,
schizophrenia,
attention
deficit
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
disorder/hyperactivity disorder, Tourette’s disorder, neurolepticꢀinduced akathisia, postꢀ
traumatic stress disorder, drug abuse, bipolar disorder, and restless legs syndrome, have been
associated with hypoactivation or hyperactivation of dopamineꢀreleasing neurons.11
Furthermore unmet medical needs involve side effects (motor, metabolic, cardiovascular,
and emotional negative potential) associated with medications targeting D2ꢀlike receptors.
This may in part be due to the lack of drugs able to selectively discriminate among the D2ꢀ
like receptor subtypes. D2R shares 75% homology in the transmembrane regions with D3R
and 53% identity with D4R.12 Moreover, D2ꢀlike receptors share high homology with other
receptor families (muscarinic, adrenergic, serotoninergic receptors), which mediate different
physiological processes in distinct brain regions.13 Thus, the need for the identification of
new ligands with increased D2ꢀlike subtype selectivity and potentially reduced side effects
remains an important challenge in the design of therapeutically useful drugs.
Among the D2ꢀlike receptor subtypes, D4R aroused great interest in the early 1990s, when
the known atypical antipsychotic agent clozapine demonstrated to bind it with high affinity
and showed good selectivity over the other dopamine receptor subtypes.14 This observation
led to the development of novel D4R antagonists potentially useful for the treatment of
schizophrenia. However, due to the failure of such antagonists in the clinic, D4R has lost its
interest as a therapeutic target for decades. Recently, D4R has garnered a renewed attraction,
emerging as a potential therapeutic target for diseases such as cancer, drug addiction, as well
as Parkinson’s disease, in which D4R antagonists can attenuate LꢀDOPAꢀinduced
dyskinesias.15
4
ACS Paragon Plus Environment

Page 5 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
Compounds belonging to different chemotypes, including arylꢀlinked piperazines, amideꢀ
linked piperazines, piperidines, morpholines and imidazolines, have been reported as
selective D4R ligands (15ꢀ19, Figure 2).15 Concerning piperidines and piperazines, a general
formula deduced from classical D3R and D4R ligands known in the literature has been
proposed. It consists of a lipophilic moiety linked by a spacer to a basic function and an
aromatic terminal, which seems to control intrinsic activity (Figure 3A).16,17 The aromatic
terminal can also be fused with the piperidine ring, affording 1,2,3,4ꢀtetrahydroisoquinoline
derivatives, acting as potent and selective D3R ligands.18 Another feature frequently observed
in D2ꢀlike ligands is the presence of an amide function in the lipophilic moiety or in the
spacer. Most of the ligands are characterized by the carbonyl function linked to the lipophilic
moiety and the NH connected through a spacer to the basic function. However, some D2ꢀlike
ligands bear an inverse amide function.16,17 While in some benzolactam derivatives the effect
of regioisomerism at the amide function on D2R and D3R affinity has been evaluated,19
concerning D4R ligands the optimal substitution pattern has not been clarified so far. The
structure of 1 (Figure 3B) fits the features of the pharmacophore reported in Figure 3A,
except for the presence of a butyl chain instead of the aromatic terminal. This structurally
intriguing difference prompted us to know more about the pharmacological dopaminergic
properties of 1, that was therefore evaluated for its affinity at D2R, D3R and D4R subtypes by
radioligand binding assays. Encouraged by its interesting affinity profile, according to which
1 showed high affinity for D4R and selectivity over D2R and D3R subtypes (Table 1), to
better understand the structural features required for the selective interaction with D4R and
obtain compounds unable to activate mAChRs, the aliphatic butyl chain of 1 was variously
modified and the novel compounds 2ꢀ4 and 8ꢀ10 (Figure 1) were prepared and evaluated by
radioligand binding studies at dopamine D2R, D3R and D4R and at M1ꢀM5 mAChRs.
Compounds 5ꢀ7, already synthesized and evaluated for their affinity at mAChRs,20 were
included in this study. Furthermore, since it is well known that not only piperidines but also
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
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 46
1
2
3
4
5
6
7
8
piperazines are able to bind DRs, the novel compounds 11, 13, and 14, as well as the known
compound 12, already evaluated for its analgetic activity,21 were prepared and studied
(Figure 1). Compounds 1, 9 and 12 were also evaluated by receptor binding assays at some
offꢀtargets (α1aꢀ, α1bꢀ and α1dꢀadrenoceptor (AR), β1ꢀ and β2ꢀAR, σ1 receptor, dopamine and
serotonin transporter (DAT and SERT, respectively)). The most relevant D4R selective
ligands were tested for their functional activities by bioluminescence resonance energy
transfer (BRET) assays that detect G protein activation and βꢀarrestin recruitment with D4R,
M1 and M4 mAChRs.
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
Moreover, considering that the lead 1 has been reported to be CNS penetrant,5 the
pharmacokinetic profiles of compounds 9 and 12 were determined by in vitro and in vivo
assays. Finally, docking simulations on the human D4R model based on the resolved
structure of the human dopamine D3R in complex with eticlopride as the template (PDB Id:
3PBL)22 were performed to rationalize the experimental observations and to elucidate the
binding mode of the proposed compounds.
RESULTS
Chemistry
The novel derivatives 2ꢀ4 and 8ꢀ14 were prepared according to the procedure reported in
Scheme 1. The intermediate 1ꢀ(3ꢀbromoꢀpropyl)ꢀ3,4ꢀdihydroꢀ1Hꢀquinolinꢀ2ꢀone23 was
reacted with the suitable amines in the presence of NaH in DMF to yield the desired
compounds 2ꢀ4 and 8ꢀ14. Alcohol derivatives 5ꢀ7 were prepared as reported in the
literature.20
Radioligand binding assays
The dopaminergic pharmacological profiles of 1, its analogues 2ꢀ14 and the known D4R
antagonist Lꢀ745,870 (15) for cloned human D2ꢀlike receptor subtypes were evaluated in in
6
ACS Paragon Plus Environment

Page 7 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
vitro binding assays using the antagonist [3H]Nꢀmethylspiperone to label dopamine
receptors, similarly to previously described procedures.24,25 The affinity values, expressed as
pKi, were calculated according to the Cheng–Prusoff equation26 and are reported in Table 1.
The muscarinic pharmacological profiles of the novel compounds 2ꢀ4 and 8ꢀ14 were
evaluated by receptor binding assays at human recombinant hM1ꢀhM5 mAChR subtypes
stably expressed in Chinese hamster ovary (CHO) cell lines using the labelled antagonist
[3H]Nꢀmethylscopolamine ([3H]NMS), following previously described procedures.27,28 The
affinity values, expressed as pKi, are reported in Table 2. The affinity values of 1 and 5ꢀ7 at
M1ꢀM5 mAChRs, already reported in reference 20, are included in Table 2 for useful
comparison.
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
Finally, for the most interesting compounds 1, 9 and 12 the affinity values, expressed as pKi
and reported in Table 3, were determined by receptor binding assays at some offꢀtargets,
namely α1aꢀ, α1bꢀ and α1dꢀAR (HEK cells; [3H]prazosin), β1ꢀ and β2ꢀAR (HEK cells;
[3H]CGP12177), σ1 receptor (guinea pig brain cortices; [3H](+)ꢀpentazocine), DAT (rat brain
striata; [3H]WIN 35,428) and SERT (rat midbrains; [3H]citalopram), similarly to previously
described procedures.29ꢀ31 The experimental details are reported in Supporting Information.
Functional assays
Based on the radioligand binding results, compounds 1, 6, 9 and 12 were selected to be
evaluated for their functional activities in BRETꢀbased assays at D4R. Since compounds 1
and 6 had already been evaluated for their functional activities in BRETꢀbased assays at M1
and M4 mAChRs,20 the activities of compounds 9 and 12 were assessed in the same assays,
to make useful considerations. The potencies and efficacies, expressed as pEC50 (−log EC50)
and Emax (maximum efficacy), respectively, are reported in Table 4 along with those of
dopamine (D4R full agonist) and carbachol (M1 and M4 mAChR full agonist) as references.
In parallel, presence of antagonist as well as allosteric modulatory effects of the tested
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 46
1
2
3
4
compounds were studied using a fixed amount of dopamine (1 µM) at D4R and carbachol (1
µM) at M1 and M4 mAChRs (Table 4, pIC50 and Imax columns).
5
6
7
8
9
Since functionally selective compounds that exert preferential activation (and/or
inactivation) on G protein or βꢀarrestin are deemed to be therapeutically useful approaches,32
βꢀarrestin2 recruitment assays at D4R, M1 and M4 mAChRs were also performed to
characterize ligands’ agonist properties (Table 4, βꢀarrestin2 recruitment, pEC50 and Emax
columns). In parallel, presence of antagonist and allosteric modulatory effects were assessed
using a fixed amount of dopamine (1 µM) at D4R and carbachol (1 µM) at M1 and M4
mAChRs (Table 4, βꢀarrestin2 recruitment, pIC50 and Imax columns).
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
In vitro and in vivo pharmacokinetic studies
Taken together all the biological results, compounds 9 and 12 were selected to be evaluated
for their in vitro ADME profile. For both compounds, kinetic solubility at pH 7.4,33 protein
binding at human plasma and at mouse plasma and brain tissue binding determined by using
equilibrium dialysis,34 hepatic intrinsic clearance determined in mouse liver microsomes35
and permeability studies in MDCKIIꢀMdr1 cell lines36 were assessed. The results are
reported in Table 5.
Moreover, compound 12 was evaluated for in vivo pharmacokinetic and brain penetration
profile following subcutaneous (s.c.) administration in male CDꢀ1 mice at the dose of 3
mg/kg. The results are reported in Figure 4 and Table 6.
DISCUSSION
Affinity for dopamine D2-like receptor subtypes
The data reported in Table 1 show that the M1 bitopic compound 1 binds with high affinity to
D4R (pKi = 9.01), and with significantly lower affinity to the other two D2ꢀlike receptor
subtypes (pKi D2R = 6.17 and pKi D3R = 6.21). Consequently, 1 is highly selective for D4R
8
ACS Paragon Plus Environment

Page 9 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
over D2R (selectivity ratio D4/D2 = 691) and D3R (selectivity ratio D4/D2 = 631). This
selectivity is comparable to those of already known D4R selective ligands belonging to
different chemotypes (Figure 2).15 Interestingly, for the first time, it has been demonstrated
that the aromatic terminal present in all the known piperidine and piperazine ligands
selective for D4R can be replaced by a butyl chain (Figure 3). The D4/D3 selectivity ratio
shown by 1 is about 3.7 times higher than that of the known selective D4R antagonist 15,
tested under the same experimental conditions used for the compounds of the present paper.
Similarly to 1, all its analogues 2ꢀ14 show affinities for D4R higher than those for D2R and
D3R, a novel class of D4R selective derivatives being obtained. Within this series, 1 is the
most selective compound for D4R over D2R and D3R, together with compound 9, which even
shows increased selectivity ratios (D4/D2 = 724; D4/D3 = 1445).
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
The butyl chain of 1 plays an important role in the affinity for all the dopamine D2ꢀlike
receptors subtypes. Its shortening or lengthening, affording 3 and 4, respectively, slightly
reduces both affinity and selectivity for D4R, while its replacement with a methyl group
(compound 2) is strongly detrimental. Instead, the introduction of a phenyl ring directly
linked to the position 4 of the piperidine nucleus (compound 8) or spaced by one or two
methylene groups (compounds 9 and 10, respectively) is compatible with high affinity and
selectivity for D4R. Indeed, compounds 8ꢀ10 show D4R affinity values similar to that of 1.
The benzyl group confers to the ligand the best affinity for D4R and selectivity over the other
D2ꢀlike receptor subtypes, as demonstrated by 9, which shows selectivity values even higher
than those of 15.
The introduction of a polar hydroxy group in the chain of compounds 3 and 4, affording 5
and 6, respectively, causes a decrease in both affinity and selectivity for D4R. However,
while the affinity of compound 5 for D4R subtype decreases 14 times compared to 3, that of
6 decreases only 3 times compared to 4. The further elongation of the hydroxyalkyl chain,
affording 7, is detrimental for D4R affinity and selectivity.
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 46
1
2
3
4
5
6
7
8
The replacement of the piperidine ring of 1, 9, and 10 with a piperazine ring, affording 11,
13, and 14, respectively, causes a strong decrease in affinity, especially for D4R, suggesting
that the introduction of a second basic function is detrimental for D4R affinity. However,
when the terminal substituent is a phenyl ring, such a modification does not affect D4R
affinity, piperazine derivative 12 showing a pKi value for D4R and selectivity over D2R and
D3R comparable to those of its piperidine analog 8. The observation that the basicity of N4
of 12 is lower than that of N1 allows us to hypothesize that probably the N4 doesn’t interfere
with the interaction with the corresponding receptor site.
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
Affinity for hM1-hM5 mAChR subtypes
From the results reported in Table 2, all the tested compounds show micromolar affinities for
hM1ꢀhM5 mAChR subtypes. Among the piperidine derivatives, the lowest affinities are those
of compound 2, bearing a methyl group in 4ꢀposition of the piperidine nucleus, while the
highest affinities are shown by compound 10, bearing a phenethyl group in the same
position. These results suggest that the reduction in the length of the alkyl chain of 1 (2 and
3) causes a decrease in affinity, while the presence of a pentyl chain (4) or an aromatic ring
(9 and 10), unless it is directly linked to the piperidine ring (8), results in affinity values for
all the hM1ꢀhM5 mAChR subtypes similar to those of 1. Compounds 4ꢀ6 and 9 show
selectivity profiles similar to that of 1, with higher affinity values for hM1, hM2 and hM4
with respect to the other two subtypes. The piperazine derivatives 11ꢀ14 exhibit affinity
values for the mAChR subtypes lower than those of the corresponding piperidine analogs 1
and 8ꢀ10.
Off-target binding profiles.
Due to their interesting affinity and selectivity profiles for D4R, compounds 1, 9, and 12
were selected to be evaluated for their binding affinities at some offꢀtargets, namely α1aꢀ, α1bꢀ
10
ACS Paragon Plus Environment

Page 11 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
and α1dꢀAR, β1ꢀ and β2ꢀAR, σ1 receptor, DAT and SERT. Compared to D4R, all the
compounds show significantly lower affinities for all the investigated offꢀtargets (Table 3).
The observation that the piperidine analogues 1 and 9 bind σ1 receptor with high affinity
values (pKi = 8.14 and 8.00, respectively) is not surprising, considering that the piperidine
scaffold characterizes several potent σ1 receptor ligands37 and numerous dopamine D4
receptor compounds have been reported to bind σ1 receptor with high affinity.38
Interestingly, compared to 1 and 9, the piperazine derivative 12 shows significantly lower
affinity for σ1 receptor (pKi = 6.57) and, consequently, its selectivity binding profile for D4R
is the best among the compounds of this series.
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
Functional activities at D4R and M1 and M4 mAChRs
Due to their higher selectivity for D4R over D2R and D3R subtypes, compounds 1, 9, and 12
were selected to be evaluated for their functional activity at D4R. Compound 6 was included
in this study because, unlike 1, it had previously been demonstrated to behave as a weak M1
and M4 mAChR antagonist.20
Gi protein activation BRET shows that, compounds 1, 6, and 9 don’t possess a partial agonist
property for D4R (Table 4). When tested for an antagonist activity in D4RꢀGi protein
activation with a fixed amount (1 µM) dopamine, compounds 1, 6, and 9 exhibited pIC50 and
Imax values comparable to those of the selective D4R antagonist 15. On the contrary,
compound 12 behaves as a D4R partial agonist (Emax = 48%) with a potency value higher
than that of dopamine (pEC50 = 9.80) and as a partial antagonist (Imax = 41%) when tested in
the presence of dopamine (pIC50 = 8.01).
D4R βꢀarrestin2 recruitment result shows that compounds 1, 6, 9, and 12 have no agonist
activity in agreement with the Gi protein activation. While they all show antagonist activity
as expected, 1 and 12 have a higher pIC50 compared to 15. Of note, compounds 1, 6, 9, and
12 show Imax lower than 15 indicating good efficacies as antagonists in D4R mediated βꢀ
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 46
1
2
3
4
5
6
7
8
arrestin2 recruitment. In summary, for D4R activities, compounds 1, 6, and 9 behave as
antagonists in both Gi activation and βꢀarrestin2 recruitment assays, while, interestingly,
compound 12 shows a biased behavior, partially activating Gi protein and inhibiting βꢀ
arrestin2 recruitment.
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
Compounds 1 and 6 had previously been reported to show functionally different behaviors at
M1 and M4 mAChRs.20 Therefore, the activity profiles of 9 and 12 were assessed at the same
receptors for useful comparison. For G protein activation, analogously to 6,20 compounds 9
and 12 show weak partial agonist activities. In the presence of carbachol (1 µM), they
behave as partial antagonists. On the contrary, as previously reported,20 1 behaves as a
positive allosteric modulator in M1 mAChR Gq activation while it exhibits partial agonism
for M4 mAChR Gi activation. In βꢀarrestin2 recruitment assay, compounds 6, 9, and 12 show
no agonist activity. In the presence of carbachol, they all show a nearly full to full antagonist
activity albeit a significantly lower potency compared to scopolamine. Therefore, in M1 and
M4 mAChR activity, 6, 9, and 12 are all weak partial agonists with a low potency in G
protein activation and antagonists with a low potency in βꢀarrestin2 recruitment.
In vitro ADME and in vivo pharmacokinetic studies
Compounds 9 and 12, selected for an in vitro ADME characterization, showed similar
ADME profiles (Table 5). Both compounds are characterized by high kinetic solubility at pH
7.4 (477 and 520 µM for 9 and 12, respectively), moderate protein binding either in mouse
plasma or in mouse brain tissue homogenates, and high apparent permeability without
relevant efflux ratios (1.92 and 1.22 for 9 and 12, respectively) when tested in MDCKIIꢀ
Mdr1 cell lines, therefore resulting no Pꢀgp substrates.
Based on the overall results from the ADME in vitro assays and due to its interesting D4
selectivity profile, compound 12 was selected to investigate its brain penetration and
systemic exposure following s.c. administration at 3 mg/kg in male CDꢀ1 mice. From the
12
ACS Paragon Plus Environment

Page 13 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
results reported in Table 6 it emerges that compound 12 shows a relevant brain penetration
(Kp = 4.35) and, based on a Kpuu,brain of 4.05, no impact of efflux transporters but even
potential for active brain influx.39,40 After 0.25 h from the treatment, exposures reach the
mean maximum concentration either in plasma or brain (481 ng/mL and 2046 ng/g,
respectively) (Figure 4). The exposures (AUClast) obtained were 518 and 2384 h*ng/mL in
plasma and brain, respectively. Free concentration exposures (AUCulast), calculated by
correcting total concentration (AUClast) for the free fraction in plasma or brain, resulted at
50.5 and 204 h*ng/mL, respectively.
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
To rationalize the high clearance of the tested compound, a similarityꢀbased predictive
analysis was performed as recently published.41 Briefly, the method is based on the
MetaQSAR resource which is a curated database of xenobiotic metabolism reactions
collected by a metaꢀanalysis of the recent specialized literature. The possible metabolites for
12 were then derived by considering the reported metabolites for the three most similar
compounds included in the database: prilocaine (Tanimoto distance = 0.7), imipramine
(0.67) and lidocaine (0.67). From their metabolic profile, three putative metabolic reactions
can be figure out for 12: (a) hydroxylation of a Csp3 atom carrying a heteroatom; (b)
oxidation of aryl moieties; (c) Nꢀglucuronidation of cyclic amines. Even though prilocaine
and lidocaine include the metabolites deriving from the hydrolysis of their amide function,
the quinolinone ring should be stable, since lactams larger than four membered rings are
resistant to metabolic hydrolysis.42
Molecular modeling
To rationalize the observed results, docking studies were performed using a D4R homology
model based on the structure of the human dopamine D3R in complex with eticlopride as the
template. Though a novel D2 structure in complex with risperidone has been recently
released (PDB Id: 6c38),43 the here used D3 structure in complex with eticlopride still
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 46
1
2
3
4
5
6
7
8
remains the more suitable template to generate the D4 homology model both for the
markedly greater identity (78.39% vs 55.22% for D3/D4 and D2/D4 alignments, respectively)
and for the more extended coverage (100% vs 91% for D3/D4 and D2/D4 alignments,
respectively).
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 5A shows the key interactions stabilizing the putative complex between D4R and 1
which is vastly dominated by the pivotal ionꢀpair that the charged ammonium head elicits
with Asp115 reinforced by the Hꢀbond with Tyr438. Again, the quinolinone ring elicits Hꢀ
bonds with Cys119, Ser196 and Ser200 plus πꢀπ stacking with Phe410, Phe411, His414 as
well as apolar contacts with Val116, Leu187 and Val193 which also contact the linker
between the piperazine and quinolinone rings. The carbon skeleton of the piperazine ring and
the alkyl tail stabilize an extended set of hydrophobic contacts with surrounding apolar
residues such as Leu90, Phe91, Leu111 and Met112. The reported interactions can clearly
rationalize the observed affinity values. In detail, the beneficial role of the alkyl chain is
easily understood when considering the surrounding apolar residues which are also wellꢀ
suited to accommodate aromatic moieties as exemplified by 9, the benzyl group of which
elicits a clear πꢀπ stacking with Phe91 (see Figure 5B). The crucial role of these hydrophobic
interactions is further confirmed by the negative effect exerted by the introduction of a
hydroxyl function. Even though the binding site reveals some accessible Hꢀbonding residues
which might contact the hydroxyl function, such a polar contact would hamper the proper
approach of the tail to the mentioned apolar residues with an overall negative impact. For
example, Figure 5C shows that the hydroxyl function of 5 approaches Thr434 while inducing
the complete loss of the mentioned apolar contacts. This finding emphasizes that the addition
of a polar group can be deleterious, regardless of the elicited contacts, if it destabilizes a set
of hydrophobic contacts involving extended lipophilic regions of the ligands.
With a view to explaining the different affinity of the piperazine derivatives, the mentioned
PM7ꢀbased semiꢀempirical calculations were utilized to calculate the pKa values of the two
14
ACS Paragon Plus Environment

Page 15 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
nitrogen atoms by using a specific script included in VEGA. The obtained results reveal that
in the alkyl derivative 11 the N4 atom is clearly more basic than the N1 atom [9.22 (N1) vs
10.22 (N4)], while the two nitrogen atoms possess comparable basicity in benzyl and
phenethyl analogues [10.38 (N1) vs 10.55(N4) and 10.17 (N1) vs 10.11 (N4) for 13 and 14,
respectively]. These results may indicate that the N4 protonated species is the only involved
in receptor recognition for 11, while both protonated forms can be involved for the benzyl
and phenethyl analogues 13 and 14. What is more, docking results reveal that the shift of the
protonated center on the N4 atom is detrimental for binding since it worsens the interactions
involving the protonated N4 atom and impacts on the pose of the quinolinone ring, thus
destabilizing the key contacts involving it (results not shown). Thus, these results can
rationalize the marked drop in affinity of 11 and the less pronounced decrease in affinity of
13 and 14 due to the possible simultaneous involvement of both protonated forms. Similar
semiꢀempirical calculations confirmed the vastly greater basicity of the N1 atom of 12 [10.19
(N1) vs 7.35 (N4)] while docking simulations reveal that the Nꢀphenyl tail of 12 approaches
some key residues in ECL2 such as Cys185 and Arg186 (Figure 5D). Even though other
simulated ligands can also contact such residues, one may argue that the greater rigidity of
the 4ꢀphenylpiperazine moiety of 12 has a greater impact on the local folding of ECL2
explaining its biased behavior due to the role of the extracellular loop ECL2 in influencing
the coupling with the signaling proteins as already reviewed.44
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
To explain the observed selectivity of these derivatives towards D4R, Figure 6 reports the
complex between D4R and 1 (as seen in Figure 5), in which the interacting residues are
colored by conservation degree as derived from multiple alignments of D2R, D3R and D4R
computed by Clustal X.45 The greatest differences among the three subtypes are detected in
the region of TM3 surrounding the key and conserved aspartate residues. In detail, a key role
should be played by the Leu111 and Met112 residues which are replaced by Phe and Val
residues in D2R and D3R. (i.e., Phe110 and Val 111 as well as Phe106 and Val107 in D2R
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 46
1
2
3
4
5
6
7
8
and D3R, respectively). Another relevant difference is seen in Leu90 which is replaced by a
tryptophan residue in both D2R (Trp90) and D3R (Trp85). Collectively, Figure 6 underlines
that the subpockets contacting the charged moieties and the quinolinone ring are highly
conserved while that harboring the apolar tail shows relevant differences among the
monitored subtypes. In detail, these different residues should render the subpocket
accommodating the alkyl tail in D2R and D3R narrower and more rigid compared to the
corresponding D4R cavity and so unable to conveniently harbor bulky apolar groups which,
instead, beneficially contribute to the D4R affinity as before discussed. These findings can
explain why the selectivity is roughly proportional to the steric hindrance of this
hydrophobic tail and why the D2R vs. D3R selectivity is almost always close to 1.0 since the
relevant residues are conserved in these two subtypes.
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
With a view to corroborating the obtained docking results, a correlative study was performed
by using the docking scores computed by ReScore+ as well as a small set of representative
physicochemical descriptors. For each ligand, the score averages as computed by considering
all the 10 generated poses were utilized in order to account for the possible role of multiple
binding modes as recently proposed.46 The so obtained best equation (eq. 1) emphasizes the
role of an optimal balance between lipophilic and polar moieties as encoded by lipole as well
as the relevance of ionic contacts exerted by the protonated group as parameterized by the
Electrostatic scores.
pKi = 2.77 + 1.12 Lipole ꢀ 0.038 Elect_Mean
N = 14; r2 = 0.80 q2 = 0.71 SE = 0.435 F = 22.43
Notably, the score averages perform better than the corresponding best values thus
suggesting that for rather flexible ligands such as those involved in the study the possibility
of assuming slightly different binding modes beneficially contributes to the observed affinity
reasonably due to positive entropic effects.
16
ACS Paragon Plus Environment

Page 17 of 46
Journal of Medicinal Chemistry
1
2
CONCLUSION
3
4
5
6
7
8
9
In summary, the bitopic M1 mAChR agonist 1 shows unexpectedly high D4R affinity and
selectivity over D2R and D3R and behaves as an antagonist in both Gi activation and βꢀ
arrestin2 recruitment assays for D4R. Interestingly, the structure of 1 differs from those of
other known selective D4R ligands, that are characterized by the presence of a properly
substituted aromatic group at position 4 of the piperidine ring (Figure 3). On the contrary, 1
bears a butyl aliphatic chain, indicating that not only πꢀπ interactions, but also hydrophobic
interactions might be favorable for D4R affinity. The aliphatic chain of 1 can be successfully
replaced by an aryl or alkylaryl chain (8ꢀ10) without affecting high D4R affinity and
selectivity over D2R and D3R subtypes. From functional studies, it emerges that 9 behaves as
a D4R antagonist and, unlike 1, is a weak partial antagonist at M1 mAChR. Compound 9
shows high D4R affinity and selectivity not only over the other D2ꢀlike subtypes, but also
over M1ꢀM5 mAChRs. The substitution of the piperidine nucleus with a piperazine ring
causes a sharp decrease in affinity, except for compound 12, which shows affinity for D4R
and selectivity over D2R and D3R comparable to its piperidine analog 8. Compound 12 also
displays high selectivity over M1ꢀM5 mAChRs and other selected offꢀtargets, namely α1aꢀ,
α1bꢀ and α1dꢀAR, β1ꢀ and β2ꢀAR, σ1 receptor, DAT and SERT. Interestingly, in in vivo
pharmacokinetic studies, a relevant brain penetration characterizes this compound. From
functional studies, it shows a biased behavior, potently and partially activating Gi protein and
inhibiting βꢀarrestin2 recruitment. Future studies with 12 might reveal mechanistically
related behaviors and the interplay between Gꢀproteinꢀ and βꢀarrestinꢀmediated signaling in
D4Rꢀrelated physiological effects.
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
Docking simulations clearly confirm the key role of hydrophobic interactions, elicited by the
alkyl chain, which are also responsible for the observed D4R selectivity. Moreover, they
suggest that an affinitive ligand should possess an optimal balance between hydrophobic and
hydrophilic portions as also confirmed by correlative studies. Finally, the biased behavior of
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 46
1
2
3
4
12 which contacts Cys185 and Arg186 emphasizes the role of the extracellular loop ECL2 in
modulating the receptor activation.44
5
6
7
8
9
Considering the renewed interest garnered by D4R, recently emerged as a potential
therapeutic target for diseases such as cancer, drug addiction, as well as Parkinson’s disease,
in which D4R antagonists can attenuate LꢀDOPAꢀinduced dyskinesias,15 the selective D4R
compounds 9 and especially 12 might help to better clarify the role played by this subtype in
the above disorders and are good candidates for further evaluation in in vivo animal models
for D4Rꢀmediated pathologies.
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
EXPERIMENTAL SECTION
Chemistry
General
Melting points were taken in glass capillary tubes on a Büchi SMPꢀ20 apparatus and are
uncorrected. IR and NMR spectra were recorded on PerkinꢀElmer 297 and Varian Mercury
AS400 instruments, respectively. Chemical shifts are reported in parts per million (ppm)
relative to tetramethylsilane (TMS), and spin multiplicities are given as s (singlet), d
(doublet), dd (double doublet), t (triplet), or m (multiplet). IR spectral data (not shown
because of the lack of unusual features) were obtained for all compounds reported and are
consistent with the assigned structures. Mass spectra were obtained using a Hewlett Packard
1100 MSD instrument utilizing electronꢀspray ionization (ESI). The microanalyses were
recorded on FLASH 2000 instrument (ThermoFisher Scientific). The elemental composition
of the compounds agreed to within ±0.4% of the calculated value. Chromatographic
separations were performed on silica gel columns (Kieselgel 40, 0.040ꢀ0.063 mm, Merck) by
flash chromatography. Compounds were named following IUPAC rules as applied by
18
ACS Paragon Plus Environment

Page 19 of 46
Journal of Medicinal Chemistry
1
2
ChemBioDraw Ultra (version 11.0) software for systematically naming organic chemicals.
3
4
The purity of the novel compounds was determined by combustion analysis and was ≥95%.
5
6
7
8
9
General Procedure to Synthesize Compounds 2-4 and 8-14
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
NaH (0.018 mol) was added portionwise to a solution of amine (0.005 mol) in DMF (20 mL)
at 0 °C. After the solution was stirred for 15 min, 1ꢀ(3ꢀbromoꢀpropyl)ꢀ3,4ꢀdihydroꢀ1Hꢀ
quinolinꢀ2ꢀone23 (0.005 mol) was added portionwise. The reaction was stirred overnight at
room temperature. The mixture was diluted with water and extracted with EtOAc. The
organic layer was washed with water and brine, dried over anhydrous Na2SO4, filtered, and
concentrated under vacuum to afford colorless oils that were purified by column
chromatography eluting with EtOAc/MeOH (8:2).
1-(3-(4-Methylpiperidin-1-yl)propyl)-3,4-dihydroquinolin-2(1H)-one (2)
This was prepared starting from 4ꢀmethylpiperidine: 53% yield. 1H NMR (CDCl3) δ 0.98 (d,
J = 6.3 Hz, 3H), 1.45ꢀ2.44 (m, 9H), 2.60 (m, 2H), 2.79 (m, 2H), 2.88 (m, 2H), 3.24 (m, 2H),
3.99 (t, J = 7.2 Hz, 2H), 7.00 (t, J = 7.3 Hz, 1H), 7.08 (d, J = 8.1 Hz, 1H), 7.15 (d, J = 7.3
Hz, 1H), 7.26 (m, 1H). The free base was transformed into the oxalate salt, which was
crystallized from 2ꢀPrOH: mp 138ꢀ139 °C, ESI/MS: m/z 287.2 (M + H+). Anal.
(C18H26N2O.H2C2O4) C, H, N.
1-(3-(4-Propylpiperidin-1-yl)propyl)-3,4-dihydroquinolin-2(1H)-one (3)
1
This was prepared starting from 4ꢀpropylpiperidine: 46% yield. HꢀNMR (CDCl3) δ 0.87 (t,
J = 7.1 Hz, 3H), 1.06ꢀ2.09 (m, 13H), 2.50 (m, 2H), 2.63 (m, 2H), 2.88 (m, 2H), 3.01 (m, 2H),
3.97 (t, J = 7.5 Hz, 2H), 6.99 (t, J = 7.5 Hz, 1H), 7.08 (d, J = 8.1 Hz, 1H), 7.15 (d, J = 7.5
Hz, 1H), 7.24 (m, 1H). The free base was transformed into the oxalate salt, which was
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 46
1
2
3
4
crystallized from 2ꢀPrOH: mp 142ꢀ143 °C, ESI/MS: m/z 315.2 (M + H+). Anal.
(C20H30N2O.H2C2O4) C, H, N.
5
6
7
8
9
1-(3-(4-Pentylpiperidin-1-yl)propyl)-3,4-dihydroquinolin-2(1H)-one (4)
This was prepared starting from 4ꢀpentylpiperidine: 43% yield. 1HꢀNMR (CDCl3) δ 0.90 (t, J
= 7.2 Hz, 3H), 1.16ꢀ2.02 (m, 17H), 2.45 (m, 2H), 2.63 (m, 2H), 2.88 (m, 2H), 2.95 (m, 2H),
3.96 (t, J = 7.5 Hz, 2H), 6.99 (t, J = 7.4 Hz, 1H), 7.08 (d, J = 8.1 Hz, 1H), 7.15 (d, J = 7.4
Hz, 1H), 7.23 (m, 1H). The free base was transformed into the oxalate salt, which was
crystallized from 2ꢀPrOH: mp 118ꢀ119 °C, ESI/MS: m/z 343.3 (M + H+). Anal.
(C22H34N2O.H2C2O4) C, H, N.
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-(3-(4-Phenylpiperidin-1-yl)propyl)-3,4-dihydroquinolin-2(1H)-one (8)
This was prepared starting from 4ꢀphenylpiperidine: 47% yield. 1HꢀNMR (CDCl3) δ 1.59 (m,
2H), 1.99 (m, 2H), 2.40 (m, 2H), 2.56ꢀ2.79 (m, 5H), 2.90 (m, 2H), 3.08 (m, 2H), 3.65 (m,
2H), 4.05 (t, J = 6.9 Hz, 2H), 7.03 (t, J = 7.4 Hz, 1H), 7.09 (d, J = 8.0 Hz, 1H), 7.17 (d, J =
7.4 Hz, 1H), 7.28 (m, 6H). The free base was transformed into the oxalate salt, which was
crystallized from 2ꢀPrOH: mp 150ꢀ151 °C, ESI/MS: m/z 349.2 (M + H+). Anal.
(C23H28N2O.H2C2O4) C, H, N.
1-(3-(4-Benzylpiperidin-1-yl)propyl)-3,4-dihydroquinolin-2(1H)-one (9)
1
This was prepared starting from 4ꢀbenzylpiperidine: 44% yield. HꢀNMR (CDCl3) δ 1.31ꢀ
2.03 (m, 9H), 2.45 (m, 2H), 2.53 (d, J = 6.9 Hz, 2H), 2.61 (m, 2H), 2.87 (m, 2H), 2.95 (m,
2H), 3.95 (t, J = 7.4 Hz, 2H), 6.98 (t, J = 7.5 Hz, 1H), 7.07 (d, J = 8.1 Hz, 1H), 7.09ꢀ7.30 (m,
7H). The free base was transformed into the oxalate salt, which was crystallized from 2ꢀ
20
ACS Paragon Plus Environment

Page 21 of 46
Journal of Medicinal Chemistry
1
2
3
4
PrOH: mp 164ꢀ166 °C, ESI/MS: m/z 363.2 (M + H+), 385.1 (M + Na+). Anal.
(C24H30N2O.H2C2O4) C, H, N.
5
6
7
8
9
1-(3-(4-Phenethylpiperidin-1-yl)propyl)-3,4-dihydroquinolin-2(1H)-one (10)
This was prepared starting from 4ꢀphenethylpiperidine: 42% yield. 1HꢀNMR (CDCl3) δ 1.21ꢀ
1.97 (m, 11H), 2.40 (m, 2H), 2.63 (m, 4H), 2.89 (m, 4H), 3.97 (t, J = 7.4 Hz, 2H), 6.99 (t, J =
7.4 Hz, 1H), 7.08 (d, J = 8.1 Hz, 1H), 7.12ꢀ7.32 (m, 7H). The free base was transformed into
the oxalate salt, which was crystallized from 2ꢀPrOH: mp 140ꢀ142 °C, ESI/MS: m/z 377.3
(M + H+). Anal. (C25H32N2O.H2C2O4) C, H, N.
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-(3-(4-Butylpiperazin-1-yl)propyl)-3,4-dihydroquinolin-2(1H)-one (11)
This was prepared starting from 4ꢀbutylpiperazine: 55% yield. 1HꢀNMR (CDCl3) δ 0.90 (t, J
= 7.1 Hz, 3H), 1.17ꢀ1.83 (m, 6H), 2.21 (m, 2H), 2.62 (m, 2H), 2.84 (m, 2H), 3.02 (m, 2H),
3.19 (m, 2H), 3.48ꢀ3.94 (m, 6H), 4.07 (t, J = 7.4 Hz, 2H), 6.97ꢀ7.22 (m, 4H). The free base
was transformed into the oxalate salt, which was crystallized from EtOH: mp 225ꢀ226 °C,
ESI/MS: m/z 330.2 (M + H+), 352.2 (M + Na+). Anal. (C20H31N3O.H2C2O4) C, H, N.
1-(3-(4-Phenylpiperazin-1-yl)propyl)-3,4-dihydroquinolin-2(1H)-one (12)
1
This was prepared starting from 4ꢀphenylpiperazine: 59% yield. HꢀNMR (CDCl3) δ 1.99
(m, 2H), 2.58ꢀ2.93 (m, 10H), 3.30 (m, 4H), 4.02 (t, J = 7.4 Hz, 2H), 6.98ꢀ7.29 (m, 9H). The
free base was transformed into the oxalate salt, which was crystallized from 2ꢀPrOH: mp
212ꢀ213 °C, ESI/MS: m/z 350.2 (M + H+). Anal. (C22H27N3O.H2C2O4) C, H, N.
1-(3-(4-Benzylpiperazin-1-yl)propyl)-3,4-dihydroquinolin-2(1H)-one (13)
This was prepared starting from 4ꢀbenzylpiperazine: 57% yield. 1HꢀNMR (CDCl3) δ 1.92 (m,
2H), 2.40ꢀ2.75 (m, 12H), 2.87 (m, 2H), 3.57 (s, 2H), 3.98 (t, J = 7.5 Hz, 2H), 6.99 (t, J = 7.4
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 46
1
2
3
4
5
6
7
Hz, 1H), 7.06 (d, J = 8.0 Hz, 1H), 7.15 (d, J = 7.4 Hz, 1H), 7.19ꢀ7.37 (m, 6H). The free base
was transformed into the oxalate salt, which was crystallized from 2ꢀPrOH: mp 222ꢀ223 °C,
ESI/MS: m/z 364.2 (M + H+). Anal. (C23H29N3O.H2C2O4) C, H, N.
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
1-(3-(4-Phenethylpiperazin-1-yl)propyl)-3,4-dihydroquinolin-2(1H)-one (14)
1
This was prepared starting from 4ꢀphenethylpiperazine: 50% yield. HꢀNMR (CDCl3) δ 1.90
(m, 2H), 2.36ꢀ2.87 (m, 18H), 3.97 (t, J = 7.5 Hz, 2H), 6.99 (t, J = 7.4 Hz, 1H), 7.07 (d, J =
8.1 Hz, 1H), 7.14ꢀ7.35 (m, 7H). The free base was transformed into the hydrochloride salt,
which was crystallized from 2ꢀPrOH: ESI/MS: m/z 378.2 (M + H+). Anal. (C24H31N3O.HCl)
C, H, N.
Biology
D2-like radioligand binding assays
Membranes were prepared from HEK293 cells stably expressing human D2, D3 or D4, grown
in a 50:50 mix of DMEM and Ham’s F12 culture media, supplemented with 20 mM HEPES,
2 mM Lꢀglutamine, 0.1 mM nonꢀessential amino acids, 1X antibiotic/antimycotic, 10% heatꢀ
inactivated fetal bovine serum, and 200 µg/ml hygromycin (Life Technologies, Grand Island,
NY) and kept in an incubator at 37 °C and 5% CO2. Upon reaching 80ꢀ90% confluence, cells
were harvested using preꢀmixed Earle’s Balanced Salt Solution (EBSS) with 5 mM EDTA
(Life Technologies) and centrifuged at 3,000 rpm for 10 min at 21 °C. The supernatant was
removed and the pellet was resuspended in 10 ml hypotonic lysis buffer (5 mM MgCl2, 5
mM Tris, pH 7.4 at 4 °C) and centrifuged at 20,000 rpm for 30 min at 4 °C. The pellet was
then resuspended in fresh EBSS buffer made from 8.7 g/L Earle’s Balanced Salts without
phenol red (US Biological, Salem, MA), 2.2 g/L sodium bicarbonate, pH to 7.4. A Bradford
protein assay (BioꢀRad, Hercules, CA) was used to determine the protein concentration and
membranes were diluted to 500 µg/ml and stored at ꢀ80 °C for later use.
22
ACS Paragon Plus Environment

Page 23 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
Radioligand competition binding experiments were conducted using thawed membranes.
Test compounds were freshly dissolved in 30% DMSO and 70% H2O to a stock
concentration of 10 mM or 100 µM. Each test compound was then diluted into 10 halfꢀlog
serial dilutions using 30% DMSO vehicle. Previouslyꢀfrozen membranes were diluted in
fresh EBSS to a 200 µg/ml (for D2 or D3) or 300 µg/ml (D4) stock for binding. Radioligand
competition experiments were conducted in 96ꢀwell plates containing 300 µl fresh EBSS
buffer, 50 µl of diluted test compound, 100 µl of membranes (20 µg total protein for D2 or
D3, 30 µg total protein for D4), and 50 µl of [3H]Nꢀmethylspiperone (0.4 nM final
concentration; Perkin Elmer). Nonspecific binding was determined using 10 µM (+)ꢀ
butaclamol (SigmaꢀAldrich, St. Louis, MO) and total binding was determined with 30%
DMSO vehicle. All compound dilutions were tested in duplicate and the reaction incubated
for one hour at room temperature. The reaction was terminated by filtration through a
PerkinElmer UniꢀFilterꢀ96 GF/B, presoaked for 60 min in 0.5% polyethylenimine, using a
Brandel 96ꢀwell plates Harvester manifold (Brandel Instruments, Gaithersburg, MD). The
filters were washed 3 times with 3 ml (3 × 1 ml/well) of iceꢀcold binding buffer. Then 65 µL
of PerkinElmer MicroScint 20 scintillation cocktail was added to each well, and filters were
counted using a PerkinElmer MicroBeta microplate counter (calculated efficiency: 56%).
IC50 values for each compound were determined from dose−response curves, and Ki values
were calculated using the Cheng−Prusoff equation; Kd values for [3H]Nꢀmethylspiperone
were determined via separate homologous competitive binding experiments. These analyses
were performed using GraphPad Prism version 6.00 for Macintosh (GraphPad Software, San
Diego, CA). Ki values were determined from at least three independent experiments and are
reported as mean ± SEM.
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
Muscarinic radioligand binding assays
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 46
1
2
3
4
5
6
7
8
CHO cells stably transfected with the human muscarinic receptors (hM1ꢀ5) were grown in
Dulbecco’s modified Eagle’s medium containing 10% fetal bovine serum and 100 units/ml
each of penicillin G and streptomycin (Sigma–Aldrich, Milano, Italy), 4 mM glutamine
(Sigma–Aldrich, Milano, Italy), nonꢀessential aminoacid solution 100x (Sigma–Aldrich,
Milano, Italy) and 50 ꢁg/ml of Gꢀ418 (Gibco, Grand Iland, NY) at 37°C under 5% CO2.28
Cells were grown to confluence, tripsinized and harvested with PBS buffer (25 mM sodium
phosphate, containing 5 mM MgCl2 at pH 7.4). Cells were suspended in iceꢀcold PBS buffer
and homogenized for 30 s using an UltraꢀTurrax (setting 5) (IKAꢀWerke GmbH & Co. KG,
Staufen, Germany). The homogenate was then pelleted by centrifugation (17,000g for 15
min at 4 °C) and the membranes were resuspended in the same buffer, reꢀhomogenized in
PBS with UltraꢀTurrax and stored at 80 °C. The protein content was measured according to
the method of Bradford47 using the BioꢀRad protein assay reagent (BioꢀRad Laboratories,
Munchen, Germany).
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
All binding assays were conducted in a 250 µL mixture of binding buffer (25 mM sodium
phosphate pH 7.4 containing 5 mM MgCl2) in 96 well plate (Sarstedt, Verona, Italy) and
thawed membranes (25–70 ꢁg/ml).27 In ligand inhibition binding assays, 0.2 nM of [3H]Nꢀ
methylscopolamine chloride ([3H]NMS, PerkinꢀElmer Life and Analytical Science, Monza,
Italy) and the test ligands (concentrations ranging from 0.1 nM to 1 mM) were used. The
samples were incubated for 2 h at room temperature. Nonspecific binding was determined in
the presence of 10 µM atropine. In all assays, the total binding was always less than 10% of
the total counts added. Incubation was terminated by rapid filtration through UniFilter GF/B
plates (PerkinꢀElmer Life and Analytical Science, Monza, Italy) using a FilterMate Cell
Harvester (PerkinꢀElmer Life and Analytical Science, Monza, Italy). Filters were than
washed with iceꢀcold water and dried before the addition of 25 ꢁL of scintillation liquid
Microscint 20 (PerkinꢀElmer Life and Analytical Science, Monza, Italy). The quantity of
membraneꢀbound radioligand was measured by TopCount NXT Microplate Scintillation
24
ACS Paragon Plus Environment

Page 25 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
Counter (PerkinꢀElmer Life and Analytical Science, Monza, Italy). Stock solutions of tested
compounds were made in DMSO, and dilutions were usually made in the incubation buffer.
DMSO at the highest concentration used (10%) had no effect on binding. Data (specific
counts per minute) were analyzed using a fourꢀparameter logistic equation in GraphPad
Prism 5.02 (GraphPad Software, San Diego, CA). Kd values for [3H]NMS were determined
via separate saturation binding experiments and the affinity estimates for test compounds
(pKI) were calculated from the IC50 values using the Cheng and Prusoff correction. Data are
presented as mean ± S.E.M. of at least three experiments, each one performed in duplicate.
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
BRET assays
BRET assays were performed in transiently transfected HEK293T cells with the constructs
described later. For G protein activation assays, a pair between RLuc8 (renilla luciferase
variant, donor) fused Gα subunit−mVenus (yellow fluorescent protein variant, acceptor)
fused Gγ subunit was used as described.20 For βꢀarrestin engagement assays, a pair between
RLuc8 fused receptor and mVenus fused effector was used as described.20 Briefly, a constant
amount of plasmid cDNA (15 ꢁg total) was transfected into HEK293T cells using
polyethylenimine (PEI) (SigmaꢀAldrich) in a 1:2 ratio (30 ꢁg PEI) in 10 cm dishes. Cells
were maintained in culture with DMEM supplemented with 10% FBS. Experiments were
performed ∼48 h postꢀtransfection. Cells were harvested, washed, and resuspended in PBS.
Approximately 200,000 cells/well were distributed in 96ꢀwell plates and 5 ꢁM
coelenterazine H was added to each well. Two min after addition of luciferase substrate
ligands were added to each well. Typically, in antagonist mode, a ligand was added 15 min
before the addition of agonist. The fluorescence (excitation at 500 nm and emission at 540
nm for 1 s recording in BRET and luminescence (no filters, 1 s recording) were quantified 2
min after a ligand was added (Mithras LB 940, Berthold). In parallel, the BRET signal from
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 46
1
2
3
4
5
6
7
8
the same batch of cells was determined by quantifying and calculating the ratio of the light
emitted by mVenus (510−540 nm) over that emitted by RLuc8 (485 nm). The net BRET
values were obtained by subtracting the background determined in cells expressing RLuc8
alone. Results are expressed as the BRET change produced by the corresponding drug. Data
and statistical analysis were performed with Prism 5.01 (GraphPad Software). For Gi
activation assays, receptor, Gαi1−RLuc8 (RLuc8 inserted at 91st amino acid), β1, and
γ2−mVenus (mVenus fused at Nterminus) constracts were transfected. For Gq activation
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
−mVenus, and
assays, receptor, Gαq−Rluc8 (RLuc8 inserted at 97th amino acid), β1, γ2
Ric8A constracts were transfected. For βꢀ arrestin engagement assays, receptor and βꢀ
arrestin2−mVenus (mVenus fused at Nꢀterminus) constracts were transfected. The cDNAs
for human D4R, M1, and M4 mAChRs were obtained from www.cdna.org and tagged with
signal peptide (SP),48 followed by a Myc epitope tag (SM hereafter) for surface receptor
expression detection. SM receptor constracts were used for the engagement assays. For the
receptor−Rluc8 fusion constructs, RLuc8 cDNA sequence was fused in frame to the Cꢀ
terminus of D4R, M1, and M4 mAChRs.
Pharmacokinetic studies
In vivo pharmacokinetic profile assays
All studies described in this paper using mice were done in compliance with national
legislation, with Aptuit policy on the Care and Use of Animals and with related codes of
practice. The pharmacokinetic parameters of 12 were determined following s.c. (bolus)
administration in male CDꢀ1 Mice of 3 mg free base/kg. Blood and brain samples (n = 3)
were withdrawn after 0.083, 0.25, 0.5, 1, 2, 4, 6 h from administration. The blood was
withdrawn in animals anaesthetized by isofluorane, by cardiac puncture; K2EDTA was used
as anticoagulant and stored in wet ice. Within 30 min from withdrawal, blood samples were
26
ACS Paragon Plus Environment

Page 27 of 46
Journal of Medicinal Chemistry
1
2
3
4
centrifuged (3000 g, 10 min, 4 °C) and two aliquots of 50 µL were diluted with 150 µL of
HEPES 0.1 N. The samples were stored at ꢀ20 °C up to the day of analysis.
5
6
7
8
9
Brain samples were collected after decapitation of anaesthetized animals and removal from
skull. The samples were stored at ꢀ20°C up the day of analysis day. After defrosting, each
brain was weight, diluted 1:5 with HEPES 0.1 N and homogenized with Precellys 24
homogenizer, Bertin Instruments (2 homogenization at 6100 rpm for 30” with 15” of pause).
The calibration curves were prepared the same day of study using blank matrix from naïve
mice. The working solution, obtained by serial dilution in 50% of acetonitrile in water from a
stock solution in DMSO, and spiking in matrix were performed by Hamilton Microlab
STARlet. The calibration curves were stored at ꢀ20 °C until to the analysis day.
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 samples were prepared by protein precipitation (2 volumes of acetonitrile with 20 ng/mL
of Rolipram used as internal standard (IS)) and vortexꢀmixed for 5 minutes. After
centrifugation (3000 g, 10 min, 4 °C), the supernatant were transferred and diluted (1:1.8)
with water using Hamilton Microlab STARlet.
The chromatography was performed by using UPLC (ACQUITY UPLC IꢀClass System ꢀ
Waters) on a Acquity UPLC BEH C18 1.7µm (2.1x30 mm) using a fast chromatography
gradient (1.5 min; phase A 0.1% of formic acid in water, phase B 0.1% of formic acid in
acetonitrile; gradient from 95% to 5% of A). The mass spectrometer was an AB Sciex triple
quadrupole, API4000. The analytes were monitored in MRM using as quantifier channels
350.2/160.2 and 276.1/208.2 for 12 and IS, respectively. The calibration standard and quality
control report for both matrixes are reported in supporting information.
Computational details
The primary sequence of the human D4R was retrieved from Uniprot (Id: P21917,
DRD4_HUMAN). The corresponding homology model was built by using Modeller9.1249
based on the resolved structure of the human dopamine D3 receptor in complex with
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 46
1
2
3
4
5
6
7
8
eticlopride as the template (PDB Id: 3PBL). Among the 10 generated structures, the best
models was selected accordingly to both the scores computed by Modeller9.12 (i.e. DOPE
and GA341 scores) and the percentages of residues falling in the allowed regions of the
Ramachandran plot. The so selected model was then minimized by keeping fixed the
backbone atoms to preserve the predicted folding and underwent a set of structural checks to
avoid the occurrence of wrong configurations, unpredicted gaps, cis peptide bonds or
colliding residues. The so refined model reveals a percentage of residues falling in the
allowed regions of the Ramachandran plot equal to 98.8% (90.3% within the core regions)
while the residues within the allowed regions of the chi space correspond to 95.1%. The
ligands were simulated in the protonated state and their conformational space was explored
by MonteCarlo simulations which generated 1000 minimized conformers for each ligand.
The so obtained lowest energy structure was then optimized by PM7 semiꢀempirical
calculations50 and finally underwent to docking simulations. Docking simulations were
performed by PLANTS51 focusing the search on a 13 Å radius sphere around Asp115 which
is involved in a key ionꢀpair with the protonated ligand. For each simulated molecule, 10
poses were generated and ranked by the ChemPLP scoring function with a speed equal to 1.
All computed results were minimized and rescored by using ReScore+ as previously
described.52
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
ASSOCIATED CONTENT
Supporting Information
Elemental analysis results for compounds 2ꢀ4 and 8ꢀ14. Experimental details of radioligand
binding assays at α1aꢀ, α1bꢀ and α1dꢀAR, β1ꢀ and β2ꢀAR, σ1 receptor, DAT and SERT.
Calibration standard and quality control report for 12. Molecular Formula Strings.
AUTHOR INFORMATION
Corresponding Author
28
ACS Paragon Plus Environment

Page 29 of 46
Journal of Medicinal Chemistry
1
2
3
4
*Phone: +390737402235. Fax: +390737637345. Eꢀmail: alessandro.piergentili@unicam.it.
Notes
5
6
7
The authors declare no competing financial interest.
8
9
ACKNOWLEDGMENTS
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 work was supported by grants from the University of Camerino (Fondo di Ateneo per la
Ricerca 2014ꢀ2015) and by the Intramural Research Program of the National Institute on
Drug Abuse (NIDAꢀIRP) (A.B. and H.Y.). We express our gratitude to Dr. Amy H. Newman
and Prof. Mario Giannella for their suggestions and comments for this manuscript, to
Michael P. Ellenberger and Adrian M. Guerrero for their technical assistance in the
radioligand binding studies as well as to Aptuit and the staff that performed the
pharmacokinetic assays.
ABBREVIATIONS USED
mAChRs, muscarinic acetylcholine receptors; AD, Alzheimer’s disease; TM,
transmembrane; MAP, mitogen activated protein; BRET, bioluminescence resonance energy
transfer; NMS, Nꢀmethylscopolamine; TMS, tetramethylsilane; EBSS, Earle’s Balanced Salt
Solution; CLi = intrinsic clearance; Cmax, maximum mean concentration; AUC, area under
the curve; IS, internal standard.
29
ACS Paragon Plus Environment