Novel N -Acetyl Bioisosteres of Melatonin: Melatonergic Receptor Pharmacology, Physicochemical Studies, and Phenotypic Assessment of Their Neurogenic Potential

Article abstract of DOI:10.1021/acs.jmedchem.5b00245

Herein we present a new family of melatonin-based compounds, in which the acetamido group of melatonin has been bioisosterically replaced by a series of reversed amides and azoles, such as oxazole, 1,2,4-oxadiazole, and 1,3,4-oxadiazole, as well as other related five-membered heterocycles, namely, 1,3,4-oxadiazol(thio)ones, 1,3,4-triazol(thio)ones, and an 1,3,4-thiadiazole. New compounds were fully characterized at melatonin receptors (MT₁R and MT₂R), and results were rationalized by superimposition studies of their structures to the bioactive conformation of melatonin. We also found that several of these melatonin-based compounds promoted differentiation of rat neural stem cells to a neuronal phenotype in vitro, in some cases to a higher extent than melatonin. This unique profile constitutes the starting point for further pharmacological studies to assess the mechanistic pathways and the relevance of neurogenesis induced by melatonin-related structures.

Full text of DOI:10.1021/acs.jmedchem.5b00245

Subscriber access provided by NEW YORK UNIV
Article
Novel N-Acetyl Bioisosteres of Melatonin: Melatonergic
Receptor Pharmacology, Physicochemical Studies, and
Phenotypic Assessment of their Neurogenic Potential
Mario de la Fuente Revenga, Nerea Fernández-Sáez, Clara Herrera-
Arozamena, José A. Morales-García, Sandra Alonso-Gil, Ana Pérez-Castillo,
Daniel Henri Caignard, Silvia Rivara, and María Isabel Rodríguez-Franco
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.5b00245 • Publication Date (Web): 29 May 2015
Downloaded from http://pubs.acs.org on June 10, 2015
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 free 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 accessible to all
readers and 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.
Journal of Medicinal Chemistry 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 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Novel N-Acetyl Bioisosteres of Melatonin: Melatonergic
Receptor Pharmacology, Physicochemical Studies, and
Phenotypic Assessment of their Neurogenic Potential
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
†
†
†
Mario de la Fuente Revenga, Nerea FernándezꢀSáez, Clara HerreraꢀArozamena,
‡,§
‡,§
‡,§
José A. MoralesꢀGarcía, Sandra AlonsoꢀGil, Ana PérezꢀCastillo, DanielꢀHenri
#
⊥
,†
Caignard, Silvia Rivara, and María Isabel RodríguezꢀFranco*
†
Instituto de Química Médica, Consejo Superior de Investigaciones Científicas (IQMꢀ
CSIC), C/ Juan de la Cierva 3, 28006ꢀMadrid, Spain.
‡
Instituto de Investigaciones Biomédicas “Alberto Sols”, Consejo Superior de
Investigaciones Científicas (IIBꢀCSIC), C/Arturo Duperier 4, 28029ꢀMadrid, Spain.
§
Centro de Investigación Biomédica en Red sobre Enfermedades Neurodegenerativas
(CIBERNED), C/ Valderrebollo 5, 28031ꢀMadrid, Spain.
#
Institut de Recherches Servier, 125 Chemin de Ronde, 78290ꢀCroissy sur Seine,
France.
⊥Dipartimento di Farmacia, Università degli Studi di Parma, Parco Area delle Scienze
2
7/A, 43124ꢀParma (Italy).
th
Dedicated to our dear colleague Prof. José Elguero on the occasion of his 80 birthday
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 72
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ABSTRACT
Herein we present a new family of melatoninꢀbased compounds, in which the
acetamido group of melatonin has been bioisosterically replaced by a series of reversed
amides and azoles, such as oxazole, 1,2,4ꢀoxadiazole and 1,3,4ꢀoxadiazole, as well as
other related 5ꢀmembered heterocycles, namely 1,3,4ꢀoxadiazolꢀ(thio)ones, 1,3,4ꢀ
triazolꢀ(thio)ones and an 1,3,4ꢀaminothiadiazole. New compounds were fully
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1 2
characterized at melatonin receptors (MT R and MT R), and results were rationalized
by superimposition studies of their structures to the bioactive conformation of
melatonin. We also found that several of these melatoninꢀbased compounds promoted
differentiation of rat neural stem cells to a neuronal phenotype in vitro, in some cases to
a higher extent than melatonin. This unique profile constitutes the starting point for
further pharmacological studies to assess the mechanistic pathways, and the relevance
of neurogenesis induced by melatoninꢀrelated structures.
ACS Paragon Plus Environment

Page 3 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
Melatonin (Figure 1) is a neurohormone produced in different tissues, primarily in
the pineal gland, but also in retina and gastrointestinal tract. Beyond its fundamental
role in circadian and seasonal rhythms’ regulation, melatonin is involved in a variety of
physiological processes like the modulation of immune and endogenous antioxidant
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
systems, among others. Due to this wide array of biologicallyꢀrelevant activities,
2
research efforts aiming to identify new melatonergic analogues increase continuously.
1 2 1
Two Gꢀproteinꢀcoupled receptors (GPCRs), named MT and MT (MT R and
MT R), mediate most of the melatonin physiological and pharmacological actions.
2
These melatonin receptors (MTRs) lack a crystalized structure that could allow the
characterization of their tertiary structure and the pattern of interactions undertaken with
their ligands. A number of homology models of the MT
1 2
R and MT R have been built,
although the low identity of their sequences with the used templates increases the
2
,3
speculation degree, thus limiting the utility of these models for drug design purposes.
In this regard, ligandꢀbased design of novel melatonergic agents is still widely
exploited. This approach has provided three approved melatonergic ligands, besides
melatonin itself: agomelatine for major depression, ramelteon for sleep disorders and
tasimelteon for nonꢀ24ꢀhour sleepꢀwake disorder (Figure 1), whilst others are in
4
developmental stages.
Subtle and major changes in the structure of melatonin have led to a plethora of
ligands, in many cases able to surpass the picomolar affinity of the endogenous agonist.
5ꢀ7
Based on these studies, several pharmacophoric models have been proposed. Among
these pharmacophores, there is a wide consensus about the necessity of an oxygen atom
attached to the aromatic or heteroaromatic core connected to the acetamidoꢀcontaining
side chain. Deletion of the methoxy group generally provokes a decrease in intrinsic
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 72
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
activity, like is the case of the antagonist luzindole (Figure 1). Regarding the indole
nucleus, there are numerous examples of aromatic scaffold hopping, like agomelatine
and other naphthalenic derivatives, which demonstrate that the presence of the indoleꢀ
8
ꢀ10
NH is not essential for activity.
Exploitation of a putative lipophilic pocket close to
Rꢀ
selective ligands, and has yielded key pharmacological tools, like luzindole or the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
the binding site of melatonin has led to the development of nonselective and MT
2
11
1
25
12ꢀ14
radioligand 2ꢀ[ I]iodomelatonin.
Both indole replacement and utilization of the
abovementioned nonꢀpolar cavity had proven to be successful approaches, although
usually resulted in an overall increase of the lipophilicity that generally undermines
15ꢀ17
drugꢀlike properties.
Therefore, from a drug discovery point of view, it is more
desirable to increase the enthalpic component of the binding freeꢀenergy (e.g., hydrogen
bonds, multipolar and cationꢀπ interactions) via optimization of specific polar
18
interactions.
Figure 1. Melatonergic drugs on the market and other selected melatonergic ligands.
ACS Paragon Plus Environment

Page 5 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Reported modifications on the acetamido group of melatonin were discouraging, as
generally resulted in a significant decrease, or total loss, of binding affinity and potency
1 2
at MT R and MT R. Indeed, there are few described cases of bioisosterism in which the
acetamido group has been replaced by other carbamoyl moieties whilst the 5ꢀ
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
9
methoxytryptamine skeleton was kept intact, such as the propyl urea 1 and
N
ꢀ(2ꢀ(5ꢀ
2
0,21
methoxyꢀ1Hꢀindolꢀ3ꢀyl)ethyl)ꢀ2,4ꢀdinitroaniline (MLꢀ23).
Piromelatine and its
hydroxyl analogue 2 represent examples of atypical acylation of the tryptamine core
22ꢀ24
with a 4
H
ꢀpyranꢀ4ꢀone fragment, able to retain significantly good affinities.
More
recently, Rami et al. described the use of 5ꢀmembered heterocycles replacing for either
the acyl or the whole acetamido group over the structure of agomelatine, generally
25
maintaining good binding affinities for MTRs.
The acetamido group of melatonin has been proposed to intervene in captodative
3
hydrogen bonds. Therefore, its bioisosteric replacement could represent a useful
strategy for the structural understanding of the not so wellꢀknown properties of the
binding pocket of this fragment, in terms of size, shape, electronic distribution and
lipophilicity. Exploitation of specific polar interactions occurring in this pocket with
acetamidoꢀlike moieties may result in a greater enthalpyꢀdriven binding of ligands.
Herein, we have replaced the
Nꢀacetyl fragment of melatonin with a series of reversed
unsaturated amides and some wellꢀestablished amide bioisosteres: oxazole, 1,2,4ꢀ
2
6
oxadiazole and 1,3,4ꢀoxadiazole; as well as other related acidic azoles bearing a
hydrogen that could intervene in hydrogen bonds: 1,3,4ꢀoxadiazolꢀ(thio)ones, 1,3,4ꢀ
triazolꢀ(thio)ones and 1,3,4ꢀaminothiazole (Figure 2). On another hand, the naturally
occurring alkaloid 5ꢀmethoxyꢀN,Nꢀdimethyltryptamine (5ꢀMeOꢀDMT) that shows high
2
7
affinity for several serotonin receptor populations, is also known to displace 2ꢀ
125
28
[
I]iodomelatonin from hamster brain membranes. This indoleamine was included in
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 72
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
the study to investigate the ability of its dimethylamino substituent to bind human
MTRs. The thermodynamic solubility at pH 7.4 of representative compounds of each
family showing significant binding to MTRs was determined. Considering the high
degree of similarity in the series, their relative hydrophilic nature was inferred from
solubility values. For those azole compounds capable of undergoing an acidꢀbase
equilibrium, the solubility was also determined at different pH values in order to study
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
29
their acid or basic character.
Beyond ligandꢀMTRs interactions, we also put our focus on one outstanding
property among the pleiotropic actions of melatonin: its ability to promote neurogenesis
30
in the adult brain. Neurogenesis involves the generation, maturation and integration of
new neurons in the neuron circuitry, which is a widespread phenomenon during
embryonic development. In the adult brain, neurogenesis remains vestigial and only two
areas retain a significant continuous neurogenic turnover, namely the subventricular
31
zone (SVZ) and the subgranular zone (SGZ) in the dentate gyrus of the hippocampus.
32
Neurogenesis appears to modulate learning and memory integration processes and is
33
sensitive to physiological, pathological and pharmacological stimuli. For instance,
ageing, neurodegenerative, and some mental diseases are associated with an exponential
3
4
decrease in hippocampal neurogenesis. Therefore, the controlled pharmacological
stimulation of the endogenous neural stem cells from adult brain neurogenic niches
might counteract the ageꢀrelated loss of memory and cognitive deterioration in some
35
pathological processes.
Melatonin plasma levels decline along with age in a similar manner as the
36
neurogenic rate does. Whether the two phenomena are related or not, remains unclear,
albeit melatonin positively modulates hippocampal neurogenesis by increasing both
precursor cell proliferation and survival, and delays the decline of neurogenesis in the
ACS Paragon Plus Environment

Page 7 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
7,38
hippocampi of aged mice.
This behaviour is in agreement with previous
observations showing that exogenous melatonin administration, in addition to a much
better retained cognitive performance, ameliorates the process of ageing by delaying the
39
onset of deteriorated health stages in senescent mice. Given the abovementioned
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
7
properties of melatonin and our interest in developing new melatoninꢀbased
4
0ꢀ42
43,44
compounds
with potential brainꢀrepairing actions,
we conducted a phenotypic
screening in vitro to assess the neurogenic potential of new derivatives using rat neural
stem cells.
Figure 2. Illustration showing the replacement of the acetamido moiety of melatonin
carried out in this work.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 72
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
RESULTS
Synthesis of New Melatonin-Based Compounds. The synthesis of all compounds
here described is depicted in Scheme 1.
Synthesis of retroamides and oxazoles. NꢀPropargylꢀ and Nꢀallylꢀindolylamides 3ꢀ6
were obtained in good yields from commercial 2ꢀ(5ꢀmethoxyꢀ1Hꢀindolꢀ3ꢀyl)alkanoic
acids that were activated with carbonyldiimidazole (CDI) and then treated with
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
propargylamine or allylamine. Subsequently, propargyl amides 3 and 4 underwent a
45,46
gold(III)ꢀcatalysed cycloisomerization,
in moderate yields.
Synthesis of 1,2,4- and 1,3,4-oxadiazoles. Intermediate ester 9 was synthesised by
affording the corresponding oxazoles 7 and
8
acidꢀcatalysed esterification from 2ꢀ(5ꢀmethoxyꢀ1Hꢀindolꢀ3ꢀyl)acetic acid. Ester 10 was
obtained in good yield via Michael addition of the electronꢀrich 5ꢀmethoxyindole with
47
ethyl acrylate catalysed by anhydrous zirconium (IV) chloride. 1,2,4ꢀOxadiazoles 11
and 12 were prepared directly from the corresponding esters 9 and 10 by
48
cyclocondensation with acetamidoxime in fair yields. To obtain 1,3,4ꢀoxadiazole
isomers, esters 9 and 10 were subjected to a microwaveꢀpromoted hydrazinolysis giving
intermediate hydrazides 13 and 14 in quantitative yields. Then, these hydrazides
underwent a condensation with the corresponding orthoester in acidic medium to afford
1,3,4ꢀoxadiazoles 15ꢀ17, in moderate to fair yields.
Synthesis of 1,3,4-oxadiazol-2-(thio)ones. The microwaveꢀheating of hydrazides 13
and 14 with CDI and triethylamine (TEA) afforded the corresponding 1,3,4ꢀoxadiazolꢀ
2ꢀones 18 and 19. Similarly, the use of carbon disulfide in basic medium afforded 1,3,4ꢀ
oxadiazolꢀ2ꢀthiones 20 and 21.
Synthesis of 4-alkyl-1,2,4-triazol-5-(thio)ones and 2-amino-1,3,4-thiadiazole.
Reaction of hydrazides 13 and 14 with methylisocyanate or ethylisothiocyanate afforded
ACS Paragon Plus Environment

Page 9 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
the corresponding acylsemicarbazides 22 and 23, or acylthiosemicarbazides 24 and 25,
which underwent microwaveꢀinduced cyclocondensation in basic medium to give 4ꢀ
methylꢀ1,2,4ꢀtriazolꢀ5ꢀones 26 and 27, and 4ꢀethylꢀ1,2,4ꢀtriazolꢀ5ꢀthiones 28 and 29 in
good and very good yields, respectively. There are several methods in the literature for
the synthesis of 2ꢀaminoꢀ1,3,4ꢀoxadiazoles and 2ꢀaminoꢀ1,3,4ꢀthiadiazoles from
acylsemicarbazides and acylthiosemicarbazides. After some failed attempts with sulfuric
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
49
v
26
acid and P ꢀreagent, only in the case of acylthiosemicarbazide 24 successful
cyclization to 2ꢀaminoꢀ1,3,4ꢀthiadiazole 30 was achieved with phosphorus oxychloride at
room temperature in fair yield.
All new melatoninꢀbased derivatives were purified by chromatographic techniques
1
and structurally characterized by their analytical and spectroscopic data (HPLC, H
13
NMR, C NMR and HRMS). Compounds 18ꢀ21, 26ꢀ30 can exist in two possible
tautomers; for each compound, the major species (which is depicted in Scheme 1) was
determined by a combination of NMR bidimensional experiments, such as COSY
(homonuclear correlation spectroscopy), HSQC (heteronuclear single quantum
coherence) and HMBC (heteronuclear multiple bond correlation) diagrams (see
Supporting Information for further details).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 72
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 1. Reagents and conditions: (a) CDI, CH
b) AuCl , CH Cl , N , rt; (c) H SO (cat.), EtOH, reflux; (d) ethylacrylate, ZrCl
CH Cl , N , rt; (e) acetamidoxime, NaH, THF, mol. sieves, 80 ºC; (f) hydrazine hydrate,
150 °C (mw), 45 min; (g) orthoester, AcOH (cat.), 125 °C (mw), 1 h; (h) compounds
with X = O: CDI, TEA, THF, 100 ºC (mw), 10 min; (i) compounds with X = S: CS
2
Cl
2
, propargylamine or allylamine, rt;
(
3
2
2
2
2
4
4
(anh.),
2
2
2
2
,
EtOH, KOH (aq.), 150 ºC (mw), 10 min; (j) intermediates with X = O and R = Me:
MeNCO, EtOH, rt; (k) intermediates with X = S and R = Et: EtSCN, EtOH, rt; (l)
3
NaOH (aq.), EtOH, 100 ºC (mw), 15 min; (m) POCl , rt.
ACS Paragon Plus Environment

Page 11 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Pharmacology. Characterization at human MT R and MT R. Binding and
1
2
functional activity studies of new compounds were carried out at human MT R or
1
MT
2
R stably transfected in Chinese hamster ovary cells (CHO), using 2ꢀ
125
50
[
I]iodomelatonin as radioligand and following described protocols. The binding
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
affinities are gathered in Table 1. Retroamides with unsaturated side chains (3ꢀ6)
showed the highest affinities, with K s in the low nanoꢀ and subꢀnanomolar range. The
i
alkenyl or alkynyl substitution does not significantly affect the potency of compounds.
Conversely, the length of the spacer does; ethylene spacer is preferred in both cases (4
and 6). Compound 6 showed the highest affinities towards MTRs of all the series,
2
displaying only 3ꢀfold lower value than melatonin at MT R.
Binding affinities of oxadiazoleꢀbearing compounds are more favoured by the
presence of a methylene (11 and 15) than an ethylene spacer (12, 16). Compared to 16,
absence of the methyl substituent in the 1,3,4ꢀoxadiazole 17 slightly increased affinity.
The heteroaromatic isomerism, that is 1,2,4ꢀ and 1,3,4ꢀoxadiazole, does not seem
relevant in the case of methyleneꢀoxadiazoles (11, 15). Conversely, in the case of
ethyleneꢀcounterparts, the 1,3,4ꢀoxadiazole (16) showed higher affinity than the 1,2,4ꢀ
oxadiazole (12). 1,3ꢀOxazoles 7 and 8 show lower binding affinities than oxadiazole
derivatives.
1,3,4ꢀOxadiazolones and ꢀthiones (18ꢀ21), together with 1,3,4ꢀtriazolꢀones and ꢀ
thiones (26ꢀ29) display a dramatic drop in binding affinity, with the remarkable
exception of 1,3,4ꢀoxadiazolone 19, which was the compound with the highest affinity
for both MT R (K = 35 nM) and MT R (K = 4 nM) of all synthesized azole derivatives.
1
i
2
i
The compound with the most marked selectivity profile [(K
500], together with rather good binding affinity at the MT subtype (K
ꢀaminoꢀ1,3,4ꢀthiadiazole derivative 30. Unfortunately, the synthesis of other aminoꢀ
i
1 i 2
(MT ) / (K MT ) >
2
i
= 17 nM) is the
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 72
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
azoles was not accomplished. Compared to 30, the dimethylamino derivative 5ꢀMeOꢀ
DMT is equipotent at MT R (K = 16 nM), but shows much milder selectivity (around
2
i
10ꢀfold).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment

Page 13 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1 2 i
Table 1. Radioligand displacement binding studies at human MT R and MT R (K ±
SEM, nM).
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
Ratio
Compd.
n
R
MT
1
R
MT
2
R
a
(
MT /MT )
1
2
3
4
1
2
14 ± 4
3.8 ± 0.5
1.0 ± 0.6
3.6
5
5.0 ± 0.9
5
6
7
8
1
2
1
2
1
2
1
2
17 ± 0.7
2.5 ± 0.2
280 ± 20
>1000
4.9 ± 0.4
3.5
5.5
ꢀ
0.45 ± 0.08
>1000
498 ± 11
44 ± 3
ꢀ
1
1
1
1
1
2
5
6
170 ± 15
>1000
3.9
ꢀ
>1000
210 ± 10
709 ± 10
16 ± 3
10
3.6
O
Me
N
N
190 ± 8
1
1
7
8
2
570 ± 10
68 ± 15
8.6
1
2
1
2
1
2
1
2
>1000
35 ± 1
>1000
>1000
>1000
>1000
>1000
>1000
160 ± 50
4 ± 0.5
ꢀ
10
ꢀ
19
20
535 ± 30
530 ± 50
560 ± 60
>1000
2
2
2
2
2
1
6
7
8
9
ꢀ
ꢀ
ꢀ
220 ± 50
>1000
ꢀ
ꢀ
3
0
1
2
2
>1000
210 ± 10
17 ± 3
16 ± 0.5
>500
10
5ꢀMeOꢀDMT
Melatonin
0.091 ± 0.005
0.15 ± 0.07
0.6
a
i
Ratio calculated only for compounds with K < 100 nM for either of the subtypes.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 72
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Only compounds showing significant affinity for either receptor (K < 1000 nM)
i
3
5
were functionally characterized in the [ S]GTPγS binding assay, and results are
summarized in Table 2. The most potent compounds were retroamides, where the spacer
length determined the intrinsic activity and potency in each pair of derivatives.
Compounds 3 and 5 with a methylene linker were partial agonists, whereas those
bearing an ethylene, 4 and 6, were full agonists. Azoleꢀcontaining compounds showed a
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
very similar profile, weak partial agonism at MT
compounds 15 and 19 displayed EC50 values in the 10 M range at MT
DMT was nearly equipotent at both receptor subtypes but showed higher efficacy at the
MT R than at the MT R.
2 1
R and little or no effect at MT R. Only
ꢀ8
2
R. 5ꢀMeOꢀ
1
2
ACS Paragon Plus Environment

Page 15 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 2. Functional studies (GTPγS) at MT R and MT R of selected compounds
1
2
MT
EC50 (nM)
1
R
MT
2
R
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
a
Compd.
n
1
2
1
2
1
2
R
E
max (%)
EC50 (nM)
Emax (%)
3
4
5
6
7
8
289 ± 13
38 ± 3
67 ± 8
106 ± 9
64 ± 4
92 ± 2
23 ± 4
nd
81 ± 3
3 ± 0.3
132 ± 70
11 ± 9
nd
82 ± 5
99 ± 9
70 ± 8
88 ± 4
nd
525 ± 31
204 ± 54
989 ± 23
nd
503 ± 5
32 ± 1
1
1
1
711 ± 23
16 ± 2
295 ± 13
48 ± 2
15
16
1
2
>1000
nd
71 ± 28
24 ± 3
32 ± 4
961 ± 16
19 ± 6
881 ± 23
17
18
2
>1000
nd
457 ± 141
25 ± 2
1
2
1
2
nd
326 ± 30
nd
nd
26 ± 4
nd
>1000
29 ± 3
nd
1
2
2
9
0
1
30 ± 3
nd
>1000
nd
nd
1096 ± 316
32 ± 9
26
28
1
1
nd
nd
nd
nd
>1000
>1000
nd
nd
N
N
Et
3
0
1
2
nd
nd
110 ± 25
35 ± 7
28 ± 3
N
S
H
5ꢀMeOꢀDMT
Melatonin
257 ± 98
62 ± 3
112 ± 110
2
0.098 ± 0.003
100
0.069 ± 0.003
100
a
nd: not determined. Full agonist: Emax > 80%; Partial agonist: 20% < Emax < 80%;
50
Antagonist: Emax < 20%.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 72
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Thermodynamic solubility studies. The thermodynamic solubility of a selection
of azoles was determined in buffer at pH 7.4 (Table 3), following described
51,52
protocols.
For comparison purposes, agomelatine and melatonin were also included.
Melatonin was the most soluble compound at physiologic pH. No significant
differences were found between 1,2,3ꢀoxadiazole 11 and 1,3,4ꢀoxadiazole 15 at pH 7.4;
both compounds showed a solubility similar to that of agomelatine. Additionally, for
those compounds bearing acidic or basic azoles, solubility was also determined at pH
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
.5 and pH 9.3 as an indirect estimation of their acidꢀbase properties. The solubility of
1
9, 21, 27, and 29 showed a clear dependence on the pH, peaking at basic pH values. It
is especially notable in the case of 19 and 21, demonstrating the acidic nature of these
compounds. The presence of the thione group in 21, compared to the ketone in 19,
increases the pH dependence in solubility. 2ꢀAminoꢀ1,3,4ꢀthiadiazole derivative 30 also
showed a mild pH difference of its solubility, higher at acidic pH, according to a weak
53
basic nature.
a
Table 3. Thermodynamic solubility (µM) of selected compounds.
Compd.
pH 5.5
pH 7.4
pH 9.3
Melatonin
nd
nd
6640 ± 25
804 ± 24
nd
nd
Agomelatine
11
nd
1165 ± 23
970 ± 8
nd
nd
1
5
9
nd
1
1406 ± 1
907 ± 5
860 ± 13
131 ± 1
1718 ± 85
2100 ± 20
3460 ± 200
1104 ± 17
137 ± 1
7920 ± 30
18660 ± 40
1408 ± 30
208 ± 2
21
27
29
30
1417 ± 77
1437 ± 78
a
Results are the mean ± SD of three independent experiments; nd: not determined
ACS Paragon Plus Environment

Page 17 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
In vitro neural differentiation and maturation in the presence of various
melatonin analogues. A selection of new melatoninꢀbased compounds covering
different structural features, namely a propargyl derivative (4), an 1,3,4ꢀoxadiazole (16),
an 1,3,4ꢀoxadiazolꢀ2ꢀone (19) and the 2ꢀaminoꢀ1,3,4ꢀthiadiazole (30), was screened at
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
0 ꢁM in primary cultures of neural stem cells. Melatonin and luzindole were also
tested for comparison purposes. Neural stem cells were obtained from the SGZ of the
dentate gyrus of the hippocampus of adult Wistar rats and induced to proliferate
54,55
forming neurospheres (NS), as previously described.
The neurogenic potential of
each compound was determined by direct observation of the expression of two
immunostained markers, namely human βꢀtubulin III (TuJ1) and microtubuleꢀassociated
protein 2 (MAPꢀ2) and cell nuclei were identified by staining with 4',6ꢀdiamidinoꢀ2ꢀ
phenylindole (DAPI). TuJ1 is expressed in neural stem cells in early stages of their
differentiation, whereas the expression of MAPꢀ2 indicates a consolidated neuronal
31
stage. After incubation of SGZꢀNS in the presence of the compounds (10 ꢁM) for 7
days, they were allowed to differentiate for 48 h and finally, after immunostaining, the
expression of TuJ1 and MAPꢀ2 was visualized by fluorescenceꢀconfocal microscopy.
Figure 3 shows the neurogenic effects of new melatoninꢀbased compounds on
neural stem cell cultures. As expected, melatonin induced both early neurogenesis and
3
0
cell maturation. All melatonin analogues (4, 16, 19, and 30) promoted the expression
of TuJ1, in most cases in a larger extent than melatonin, showing also the typical
neuronal morphology. Greater differences could be observed in the expression of MAPꢀ
2
, 1,3,4ꢀoxadiazole derivatives 16 and 19 being the most effective agents at stimulating
cell maturation, apparently even more than melatonin itself. Surprisingly, the
propargylic derivative 4, a full agonist at both MT R and MT R and one of the most
1
2
potent melatonergic ligands in the series, did not promote MAPꢀ2 expression.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 72
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 3. Expression of TuJ1 (green) and MAPꢀ2 (red) in cultured SGZꢀderived NS in
the presence of different compounds at 10 ꢁM. DAPI (blue) was used as a nuclear
marker. Scale bar, 200 µm.
ACS Paragon Plus Environment

Page 19 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Melatonin, compounds 16 (the most effective at stimulating the neurogenic effect in
vitro) and 19 (the most potent partial agonist among azoles) were further tested in the
presence of the nonꢀselective antagonist luzindole (10 µM) to evaluate the implication
1 2
of MT R and MT R in the expression of the neurogenic markers (Figure 4). In all three
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
cases, preincubation with luzindole blocked the expression of both TuJ1 and MAPꢀ2
promoted by the tested compounds, although residual expression of both TuJ1 and
MAPꢀ2 could be observed. Luzindole apparently reduced the basal expression of TuJ1
compared to the compoundꢀfree control.
According to the in vitro parallel artificial membrane permeability assay for the
bloodꢀbrain barrier (PAMPAꢀBBB), the 1,3,4ꢀoxadiazole derivative 16 showed higher
passive CNSꢀpermeability than 19 (data not shown). This nonꢀcellular assay mimics the
BBB and allows a facile highꢀthroughput evaluation of compounds in a simplified and
4
3,56,57
validated model.
Based on the results obtained in the NS studies and its positive
predicted permeability in the CNS, the neurogenic profile of 16 was studied at lower
concentrations (1 and 0.1 µM). Statistical analysis showed that compound 16 stimulated
the expression of TuJ1 and MAPꢀ2 at all tested concentrations in SGZꢀderived NS, in a
doseꢀdepend manner (Figure 5).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 72
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. Effect of luzindole (10 ꢁM) on the expression of TuJ1 (green) and MAPꢀ2
(red) promoted by melatonin, 16, and 19 at 10 ꢁM in SGZꢀderived NS. DAPI (blue) was
used as a nuclear marker. Scale bar, 200 µm.
ACS Paragon Plus Environment

Page 21 of 72
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 5. (Top) Doseꢀresponse (µM) effect of compound 16 on the expression of TuJ1
(green) and MAPꢀ2 (red) in SGZꢀderived NS. DAPI (blue) was used as a nuclear
+
+
marker. Scale bar, 200 µm. (Bottom) Number of TuJ1 and MAPꢀ2 expressing cells in
**
***
NS is shown and measured as the mean ± SD. p ≤ 0.01; p ≤ 0.001.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 72
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
DISCUSSION
Reversed amides have been already reported as a class of MTRs ligands, for
instance, on the naphthalene series of agomelatine, but not yet on the classical 5ꢀ
19,58
methoxyindole scaffold.
Binding data showing clear preference for the ethylene
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
spacer (4 and 6) are consistent with literature ones. Both compounds retain good
binding affinity, possibly due to the ability to arrange their pharmacophoric groups in a
6,59
way resembling that of melatonin in its bioactive conformation (Figure 6).
hypothesis is also supported by their full agonistic properties.
This
Figure 6. Superposition of melatonin (yellow carbons) in its putative bioactive
6
conformation and 6 (green carbons).
Azole derivatives binding potency was also affected by the length of the spacer; 10ꢀ
fold or higher differences were found in the case of compounds bearing the same
heterocycle. The preferred spacer length depends on the nature of the azole ring. In the
case of the oxazoles (7, 8) the spacer seems to favour the binding to a receptor subtype,
but the low affinity displayed does not allow drawing any conclusions. In the case of
1,2,4ꢀ and 1,3,4ꢀoxadiazole matching pairs, the methylene spacer is clearly favoured (11
ACS Paragon Plus Environment

Page 23 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
/ 12, 15 / 16). These oxadiazole derivatives lack a hydrogenꢀdonor centre compared to
the melatonin acetamido group. Nevertheless, the hydrogenꢀacceptor point can be
located in a nitrogen atom of the oxadiazole ring, whose relative position within the
heterocycle depends on the chain length (Figure 7a).
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 7. Compounds 15 (orange carbons), 16 (gray carbons) (panel a) and compound
19 (green carbons) (panel b) superposed on melatonin (yellow carbons).
Compound 17, the norꢀisomer of 16 shows the same selectivity and potency as 15,
whereas compound 16 binding is penalized in the MT R. This penalty is probably
2
attributable to steric clashes in the region where the amide of melatonin is
accommodated, worsened by the rigid coplanar arrangement of the methyl group in the
oxadiazole ring (Figure 7a).
With regard to the oxadiazole ring isomerism, the comparison between the isomer
pairs 11 / 15 and 12 / 16, assuming that each pair would bind to MTRs in a similar pose,
demonstrates that the 1,3,4ꢀoxadiazole ring is slightly preferred in terms of potency. The
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 72
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
isomerism of the oxadiazole does not seem to cause a change in the selectivity; the same
subtle MT R preference was observed for the azole compounds described by Rami et
2
25
al. The carbonyl group of melatonin is putatively involved in a hydrogen bond, and
this moiety is necessary for binding to both receptor subtypes. On another hand, the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
60,61
presence of the NH from the amide group is not essential in the MT
2
subtype.
Thus,
the contribution of the melatonin carbonyl group to the MT R binding should be of
2
greater relevance. Assuming that imineꢀlike nitrogen within the oxadiazole ring
substitutes for the carbonyl group of melatonin, the hydrogen bond acceptor properties
of the heterocycle shall be crucial for MT
2
R recognition. Previous studies on both
oxadiazole isomers revealed that calculated hydrogen bond acceptor properties of
2
6
nitrogen atoms of 1,3,4ꢀoxadiazole are greater than those of 1,2,4ꢀoxadiazoles. Even if
the contribution of this calculated difference is limited, it could explain the preference
of 1,3,4ꢀoxadiazoles for MT
2
R.
In the case of acidic azoles, the affinity constants of the compounds with a
ꢀ7
ꢀ7
methylene linker (18, 20, 26, and 28) are all within the 2·10 ꢀ 6·10 M range for
MT R, regardless of the nature of the azole ring or the bulk of the substituents on it.
2
Conversely, their ethylene counterparts (19, 21, 27, and 29) showed little or no affinity
for either receptor subtype, with the remarkable exception of 19 that turned out to be the
most potent compound in the whole series of azoles. The 1,3,4ꢀoxadiazolone ring of 19
can adopt two tautomeric forms, although its NMRꢀcharacterization in DMSO showed
the keto form as the sole species. However, the acidic nature of the heterocycle, further
confirmed by its differential solubility at several pHs, could also imply the existence of
equilibrium with the conjugated base in aqueous media. Melatonin solubility in water is
62,63
reported to be in a range between 0.4 ꢀ 0.5 mM.
In our experiment, the solubility of
melatonin in buffer solution at pH 7.4 was found to be slightly above that range (6.64
ACS Paragon Plus Environment

Page 25 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
mM, Table 3). The solubility of 19 at pH 7.4 was lower (2.10 mM), but within the same
order of magnitude as melatonin. The bioisosteric replacement of the acetamido group
of melatonin for 1,3,4ꢀoxadiazolone produced a lower drop (~3ꢀfold) in solubility than
the indoleꢀnaphthalene scaffold hopping (8ꢀfold), suggesting a comparable degree of
hydrophilicity for the 1,3,4ꢀoxadiazolone ring and the acetamide at physiological pH. It
is not possible to determine the exact contribution of hydrophilicity, acidity, and
plausible tautomerism to the binding affinity of compound 19 since this combination of
physicochemical properties adds a great degree of complexity to the interpretation of
ligandꢀreceptor interactions within an already elusive binding pocket. Nevertheless,
superimposition of 19 to the bioactive conformation of melatonin suggests that the
heterocyclic nitrogen closer to the ethylene spacer can mimic the carbonyl of the amide
group of melatonin (Figure 7b). Compounds 21, 27, and 29 could be superimposed to
the structure of melatonin in a similar manner; however, none of them displayed any
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 2
affinity for either MT R or MT R. In the case of 21, the sulfur atom that replaces for the
carbonylꢀlike oxygen of 19 would reasonably affect both the acidity and the probable
tautomeric equilibrium in aqueous media. Moreover, this bioisosteric replacement
represents an additional bulk compared to the carbonyl oxygen of 19. All three factors
could be responsible for the affinity drop of 21. The same applies to compounds 27 and
29, in which not only the nature of the azole is different, but also the presence of a
substituent attached to it generates a greater bulk, more likely to be hindered by steric
clashes within the receptor binding site.
The binding affinity constant of 30 for the MT R is similar to those of methyleneꢀ
2
oxadiazoles 11 and 15, with virtually no affinity for the MT
1
R, showing the highest
selectivity towards the MT R of all compounds included in this work. This marked
2
selectivity could be attributed to a differential ability of either receptor subtype to
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 72
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
accommodate the slightly basic ethylamino chain of 30 within their binding pockets.
The selectivity profile of 30 encouraged us to increase the family of aminoꢀthiadiazoles
and aminoꢀoxadiazoles, but unfortunately the synthetic approaches failed to provide
additional analogues.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5ꢀMeOꢀDMT shows a milder selectivity for the MT
2 a
R than 30. The pK = 8.68 of
its nonꢀmethoxylated analogue, dimethyltryptamine (DMT), suggests a predominantly
64
protonated state of 5ꢀMeOꢀDMT at physiological pH. Current knowledge of MTRs’
structure is not sufficient to devise a binding mode for basic tryptamine derivatives
125
based on experimental data. Displacement of 2ꢀ[ I]iodomelatonin by 5ꢀMeOꢀDMT
28
from hamster brain membranes was found to be 134ꢀfold less than melatonin. In our
case, a similar decrease in affinity was found at the human MT R subtype (100ꢀfold).
2
65,66
Besides its longꢀknown agonistic properties at 5ꢀHT receptors,
this methoxylated
tryptamine is also able to partially activate both MTRs, being 2ꢀfold more efficient at
MT R. This suggests an atypical case of bioisosterism in which the protonated
dimethylamino group of 5ꢀMeOꢀDMT can partially substitute for the uncharged amide
2
67
of melatonin. Considering the hypothermic effect of melatonin, and the similar
affinities shown by 5ꢀMeOꢀDMT for serotonin receptors (5ꢀHT R, K = 11 nM; 5ꢀ
1
A
i
68
HT2AR, K
i
= 92 nM) and human MTRs (MT
1
R, K
i
= 210 nM; MT
2
i
R, K = 16 nM),
these findings arise questions on whether the hypothermia observed in rats treated
consecutively with a 5ꢀHT2A antagonist and 5ꢀMeOꢀDMT could be mediated not only
27
via 5ꢀHTRs, but also via MTRs.
In contrast to the differences found in the binding affinities of all tested compounds
bearing a fiveꢀmembered ring, their functional profile is markedly similar and appears
unaffected by the nature of the azole employed to replace the acetamido group of
melatonin. While their potencies vary depending on the compound, they all retain the
ACS Paragon Plus Environment

Page 27 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
mild MT R subtype preference and are partial agonists with an Emax that barely goes
2
above the 30% of melatonin maximum response at either receptor subtype. These
2
compounds, upon binding to MT R, presumably fit their heterocyclic substituent in the
pocket where the amide of melatonin is placed, as depicted above (exemplified for 15
and 19 in Figure 7). Nevertheless, their relatively low binding affinity suggests that
either the accommodation of the azole ring in the region occupied by the acetamido
group of melatonin is hampered by plausible steric clashes, or that such accommodation
is secondary to some alteration in the proper arrangement of the 5ꢀmethoxyindolꢀ3ꢀylꢀ
alkyl portion. Anyway, conformational changes that lead to receptor activation would
be impeded, thus resulting in the observed weak partial agonism.
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 the present study, we analysed the neurogenic effects of melatonin and some
selected synthetic analogues in vitro (Figure 3), and observed that the degree of
differentiation and maturation of primary neural stem cells did not correlate with their
potency at MTRs (Tables 1 and 2). Compound 4 is a full agonist, weaker than melatonin
1 2
at both MT R and MT R, yet the most potent full agonist in the series of analogues here
presented. Compound 4 induces a moderate cell differentiation, but fails to stimulate
neural maturation in vitro. In this regard, partial melatonergic agonists 16, 19 and 30
seem to perform much better at promoting neuronal development (Figure 3). Despite its
poor intrinsic activity and weak potency at MTRs, compound 16 turned out to be the
most potent compound at stimulating neurogenesis in vitro in a doseꢀdependent manner
(Figure 5), apparently in a greater extent than the reference compound melatonin.
In order to evaluate whether other biological pathways could be involved in the
neurogenic effect of 16, the compound was screened at 10 ꢁM concentration on several
receptors reportedly related to neurogenesis, namely: glycogen synthase kinase 3β
6
9
70
(
1 2
GSKꢀ3β); cannabinoid receptors 1 and 2 (CB , CB ); serotonin receptors subtype
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 72
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
7
1
72
1A, 2A, 2B, 2C (5ꢀHT , 5ꢀHT , 5ꢀHT , 5ꢀHT ); serotonin transporter (SERT);
1
A
2A
2B
2C
retinoic acid receptorꢀalpha (RARα); and peroxisome proliferatorꢀactivated receptorꢀ
73
gamma (PPARγ). No significant interaction with the aboveꢀmentioned receptors was
observed.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The fact that luzindole was unable to fully block the neural differentiation in vitro
3
0
elicited by melatonin has already been observed by RamírezꢀRodríguez et al. The
residual expression of TuJ1 stimulated by melatonin, 16, and 19 in the presence of
luzindole (Figure 4) could be related to the interaction of these compounds with other
systems. Indeed, melatonin has also been reported to interact with nuclear receptors
1 2
from the retinoic acid superfamily, not related to MT R and MT R signalling
74
pathways.
Although luzindole is unable to counteract totally the effects of melatonin, 16, or
9, interestingly it blocks the basal neural differentiation below its constitutive levels as
1
observed in the TuJ1 expression in vitro (Figure 4). This effect has already been
described and could be related to the nature of luzindole, a pure antagonist at MT R and
Both receptor subtypes can be found in the dentate
2
3
0,75
an inverse agonist at MT R.
1
76
gyrus of rat hippocampus. Luzindole ability to decrease melatonergic intrinsic activity
below its constitutive level could block, or partially block, other intracellular pathways
via signalling crossꢀtalk, reducing basal neural differentiation. The magnitude of this
observed effect is rather low at basal levels, although it could be of relevance in the case
of neural differentiation elicited by compound 16. Taking into account the low potency
of 16 at MTRs, a hypothetical neurogenic contribution beyond MT R and/or MT R
1
2
could be partially masked by a negative crossꢀtalk.
The blockade exerted by luzindole on the neurogenic actions of melatonin, 16, and
9 in vitro suggests the involvement of MT R and/or MT R. It is surprising how weak
1
1
2
ACS Paragon Plus Environment

Page 29 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
partial agonists like 16 and 19, barely able to stimulate the MTRs activation (≤30%
35
relative intrinsic activity in the [ S]GTPγS assay, see Table 2), are potent neurogenic
agents in vitro. To gain insights at this point, additional functional assays were carried
out with 16 and 19; their intrinsic activities at MTRs were measured as impedance in
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
7
7,78
the case of MT
1 2
R and as decrease of cAMP levels in the case of MT R.
Summarizing the results, 16 was too weak to elicit any measurable stimulation of the
MT R, whereas it reached a full agonist response at the MT R (EC = 1.6 ꢁM; E
=
=
1
2
50
max
9
1
2%). Compound 19 behaved as a full agonist at both MT R (EC50 = 48 nM; Emax
1
2
00%) and MT R (EC50 = 5.5 nM; Emax = 97%). A plausible hypothesis to explain these
7
9
observed amplification effects is functional selectivity. This phenomenon is quite
ubiquitous in GPCRs, and MTRs are known to couple to several Gꢀproteins, with
special preference for G proteins, that upon dissociation could differentially activate
i
74
35
parallel pathways. The [ S]GTPγS assay is unable to discriminate between different
Gꢀproteins but the preferential activation of certain MT RꢀGꢀprotein populations would
n
provoke the activation of their coupled signalling pathways, and thus only their
mediated effects would be observed (Figure 8). These findings are insufficient to
establish a plausible hypothesis for the observed neurogenic effect; however, they
appear to suggest a distinctive activation mechanism of MTRs by azole derivatives.
ACS Paragon Plus Environment