MT1-Selective Melatonin Receptor Ligands: Synthesis, Pharmacological Evaluation, and Molecular Dynamics Investigation of N-{[(3-O-Substituted)anilino]alkyl}amides

Article abstract of DOI:10.1002/cmdc.201200303

The design of compounds selective for the MT₁ melatonin receptor is still a challenging task owing to the limited knowledge of the structural features conferring selectivity for the MT₁ subtype, and only few selective compounds have been reported so far. N-(Anilinoalkyl)amides are a versatile class of melatonin receptor ligands that include nonselective MT₁/MT₂ agonists and MT₂-selective antagonists. We synthesized a new series of N-(anilinoalkyl)amides bearing 3-arylalkyloxy or 3-alkyloxy substituents at the aniline ring, looking for new potent and MT₁-selective ligands. To evaluate the effect of substituent size and shape on binding affinity and intrinsic activity, both flexible and conformationally constrained derivatives were prepared. The phenylbutyloxy substituent gave the best result, providing the partial agonist 4a, which was endowed with high MT₁ binding affinity (pK_i=8.93) and 78-fold selectivity for the MT₁ receptor. To investigate the molecular basis for agonist recognition, and to explain the role of the 3-arylalkyloxy substituent, we built a homology model of the MT₁ receptor based on the β₂ adrenergic receptor crystal structure in its activated state. A binding mode for MT₁ agonists is proposed, as well as a hypothesis regarding the receptor structural features responsible for MT₁ selectivity of compounds with lipophilic arylalkyloxy substituents.

Full text of DOI:10.1002/cmdc.201200303

MED
DOI: 10.1002/cmdc.201200303
MT1-Selective Melatonin Receptor Ligands: Synthesis,
Pharmacological Evaluation, and Molecular Dynamics
Investigation of N-{[(3-O-Substituted)anilino]alkyl}amides
Silvia Rivara,[a] Daniele Pala,[a] Alessio Lodola,[a] Marco Mor,[a] Valeria Lucini,[b]
Silvana Dugnani,[b] Francesco Scaglione,[b] Annalida Bedini,[c] Simone Lucarini,[c]
Giorgio Tarzia,[c] and Gilberto Spadoni*[c]
The design of compounds selective for the MT1 melatonin re-
ceptor is still a challenging task owing to the limited knowl-
edge of the structural features conferring selectivity for the
MT1 subtype, and only few selective compounds have been re-
ported so far. N-(Anilinoalkyl)amides are a versatile class of
melatonin receptor ligands that include nonselective MT1/MT2
agonists and MT2-selective antagonists. We synthesized a new
series of N-(anilinoalkyl)amides bearing 3-arylalkyloxy or 3-alky-
loxy substituents at the aniline ring, looking for new potent
and MT1-selective ligands. To evaluate the effect of substituent
size and shape on binding affinity and intrinsic activity, both
flexible and conformationally constrained derivatives were pre-
pared. The phenylbutyloxy substituent gave the best result,
providing the partial agonist 4a, which was endowed with
high MT1 binding affinity (pKi =8.93) and 78-fold selectivity for
the MT1 receptor. To investigate the molecular basis for agonist
recognition, and to explain the role of the 3-arylalkyloxy sub-
stituent, we built a homology model of the MT1 receptor
based on the b2 adrenergic receptor crystal structure in its acti-
vated state. A binding mode for MT1 agonists is proposed, as
well as a hypothesis regarding the receptor structural features
responsible for MT1 selectivity of compounds with lipophilic
arylalkyloxy substituents.
Introduction
Melatonin (N-acetyl-5-methoxytryptamine (1), Figure 1) is a neu-
rohormone primarily secreted by the pineal gland at night in
all species.[1] Melatonin has been claimed to have an effect in
almost all of the main physiological functions of the body,[2]
particularly in circadian rhythms and sleep regulation,[3] control
of mood and behavior,[4,5] neuroimmunomodulation,[6] blood
pressure control,[7] and pain perception.[8] Moreover, anti-
inflammatory,[9] antioxidant,[10] neuroprotectant,[11] and anti-
tumor[12,13] activities have been ascribed to melatonin. Recently,
the melatonin signaling pathway has been implicated in the
events leading to the development of type 2 diabetes.[14,15] Al-
though the endogenous role of melatonin and the mecha-
nisms by which it regulates the physiology of mammalian and
non-mammalian species have not been fully elucidated, three
therapeutic agents targeting the melatonin membrane recep-
tors (Circadin,[16] Rozerem,[17] and Valdoxan[18]) are already in
use, and melatonin receptor agonists are now appearing as
new promising therapeutic options in the treatment of circadi-
an rhythm sleep disorders and depression.[19,20] Melatonin inter-
acts with high-affinity G protein-coupled receptors (GPCRs),
two of which, MT1 and MT2, have been found in mammals, in-
cluding humans, and subsequently cloned.[21] Both MT1 and
MT2 receptors exhibit sub-nanomolar binding affinity for mela-
tonin and are negatively coupled to adenylate cyclase, al-
though they can also interact with other intracellular signaling
pathways.[22] A third melatonin receptor subtype, termed Mel1c,
has been cloned from Xenopus laevis, chicken, and zebrafish,
but it is not expressed in mammals.[23] In addition to these
high-affinity GPCRs, a distinct binding site, termed MT3 and dis-
playing lower affinity for melatonin (Ki =10–60 nm), was puri-
fied and characterized as the hamster homologue of the
human enzyme quinone reductase 2.[24] In mammals, melato-
nin receptors are mainly expressed in the brain, with consider-
able variation in the location and density of expression among
species and also in some peripheral tissues.[25,26] The differential
role of MT1 and MT2 subtypes has been only partially elucidat-
ed. Experimental evidence supports that activation of MT1 re-
ceptors in mice inhibits neuronal firing within the SCN,[27] in-
hibits prolactin secretion in photoperiodic species,[28] modu-
lates visual function in mouse retina,[29] and induces vasocon-
striction of rat caudal arteries.[30] On the other hand, activation
[a] Prof. S. Rivara, Dr. D. Pala, Dr. A. Lodola, Prof. M. Mor
Dipartimento di Farmacia, Universitꢀ degli Studi di Parma
V.le G. P. Usberti 27A, 43124 Parma (Italy)
[b] Dr. V. Lucini, Dr. S. Dugnani, Prof. F. Scaglione
Dipartimento di Farmacologia, Chemioterapia e Tossicologia Medica
Universitꢀ degli Studi di Milano
Via Vanvitelli 32, 20129 Milano (Italy)
[c] Dr. A. Bedini, Dr. S. Lucarini, Prof. G. Tarzia, Prof. G. Spadoni
Dipartimento di Scienze Biomolecolari
Universitꢀ degli Studi di Urbino “Carlo Bo”
Piazza Rinascimento 6, 61029 Urbino (Italy)
E-mail: gilberto.spadoni@uniurb.it
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201200303.
ChemMedChem 0000, 00, 1 – 12
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

G. Spadoni et al.
MED
of MT2 receptors induces vasodilation,[31] enhances immune
function,[32] inhibits dopamine release in rabbit retina,[33] produ-
ces a phase shift in circadian rhythms,[34] and promotes non-
rapid eye movement sleep in rats.[35]
oxygen atom in position 5 of melatonin, or in the topologically
equivalent position in ligands with different scaffolds. However,
the identification and fine-tuning of suitable steric and physi-
cochemical properties of the substituents able to confer MT1
subtype selectivity are still a challenge, as only sparse informa-
tion is available, and no conclusive structure–activity relation-
ships (SARs) have been reported. The symmetrical structure of
dimeric compounds previously suggested that the two outer
units of melatonin bioisosteres could bind two orthosteric
binding sites in two units of receptor homodimers,[39] even
though longer spacers are reported to be required by bivalent
ligands.[47] On the other hand, one of the two outer portions
could be replaced by a simple lipophilic group at no cost of
MT1 affinity,[39] and allosteric binding sites have recently been
characterized in the extracellular regions of different GPCRs.
This prompts an alternative hypothesis: that lipophilic groups
emerging from the aromatic portion of melatonin receptor li-
gands would be accommodated in a channel pointing toward
the extracellular loops. Further pursuit of this hypothesis re-
quires that isosteric replacement of one moiety in dimeric li-
gands by non-melatonin-like groups be conserved in different
classes of melatonin receptor ligands with some MT1 selectivity,
and that a channel exists that can accommodate such groups
within the structure of the MT1 receptor, as far as this can be
postulated from the known three-dimensional structures of
crystallized GPCRs.
Studies aimed at characterizing the pathophysiological role
of melatonin should take advantage of the availability of com-
pounds selective for either the MT1 or MT2 receptor subtype.
Whereas numerous MT2-selective ligands have been described
(for reviews, see references [36], [37], and [38]), the literature
only reports a few examples of MT1-selective compounds. MT1-
selective ligands (Figure 1) comprise different series of dimeric
compounds. Homo- and heterodimers of the agomelatine scaf-
fold exhibited nanomolar binding affinities for MT1.[39,40] Dimers
based on the indole,[41] 4-azaindole,[42] or aniline[43] skeleton
were also reported, but they exhibited weaker affinity and
MT1-selectivity than the parent naphthalene dimers. Of the di-
meric compounds, a three methylene linker conferred the
highest MT1 selectivity.[39,43] Compounds with sub-nanomolar
affinity but modest selectivity for the MT1 subtype were ob-
tained introducing a 4-phenylbutyl substituent in position 2 of
the benzoxazole or benzofuran ring in different series of mela-
toninergic ligands.[44] The presence of a linear hexyloxy chain
in position 6 of chromane derivative S25567 conferred some
selectivity for the MT1 receptor.[45] Recently, a series of melato-
nin indole derivatives bearing different 5-arylalkyloxy substitu-
ents has been described, with some compounds having ap-
proximate 10-fold selectivity for the MT1 receptor when
a phenyl ring is connected to the oxygen atom in position 5
with a three methylene or a three-methyl-oxy linker.[46] MT1-
selective ligands reported so far share a common structural
feature, that is, a bulky lipophilic substituent connected to the
We recently discovered a class of high affinity melatonin re-
ceptor ligands having a N-(anilinoethyl)amide scaffold. This
class was very versatile, as it could be properly functionalized
to provide non-selective MT1/MT2 agonists or MT2-selective
partial agonists and antagonists.[48] Moreover, metabolically
protected derivatives could be obtained which main-
tained the same receptor profile as the parent com-
pounds.[49] We report here the synthesis and pharma-
cological characterization of new N-(3-O-substituted-
anilinoalkyl)amides (4a–n) in which lipophilic sub-
stituents of different size and shape were introduced
at the oxygen atom of the aniline core. We investi-
gated arylalkyl and alkyl substituents looking for opti-
mal chain length. We also prepared some conforma-
tionally constrained derivatives with the aim of defin-
ing the conformational preference of the flexible
alkyl chain. Moreover,
a three-dimensional (3D)
model of the active form of the MT1 receptor was
built by homology modeling, and a binding scheme
for agonist compounds is hypothesized. Based on
the different amino acid composition of MT1 and MT2
receptors, an explanation for the MT1 selectivity of
compounds carrying arylalkyloxy substituents is pro-
posed.
Results and Discussion
Chemistry
The new melatonin receptor ligands 4a–n were syn-
Figure 1. Structures of melatonin, agomelatine, and representative MT1-selective ligands.
thesized as described in Scheme 1. Phenol derivatives
&2
&
www.chemmedchem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 12
ÝÝ
These are not the final page numbers!

MT1-Selective Melatonin Receptor Ligands
MED
Structure–activity relationships
Binding affinities and intrinsic activities at the human MT1 and
MT2 receptors of the newly synthesized N-(anilinoethyl)acet-
amides with 3-alkyloxy- or 3-arylalkyloxy substituents (4a–n)
were assessed as described in the Experimental Section. Re-
sults are reported in Table 1, along with data for a number of
reference compounds. A total of 12 new compounds (4a and
4d–n) were examined to investigate the effect of length, size,
and shape of the aniline-3-O-substituent on MT1/MT2 subtype
selectivity. Combination with arylation of the aniline nitrogen
(4c) and lengthening of the ethyl spacer (4b) were also evalu-
ated. Replacement of the ether methyl group of non-selective
agonist 2a with a phenylbutyl substituent (4a) gave the ex-
pected favorable effect on MT1/MT2 selectivity. Phenylbutyloxy
derivative 4a showed high binding affinity for MT1 receptors
(pKi =8.93) and considerably lower affinity for MT2 receptors
(pKi =7.04), thus leading to a MT1/MT2 selectivity ratio of 78.
Compound 4a is one of the most MT1-selective compounds re-
ported in the literature, although it could not reach the 100-
200-fold selectivity measured for some agomelatine homodim-
ers;[39] this high selectivity, however, could not be reproduced
by other authors, who observed threefold selectivity for the
reference derivative.[46] Interestingly, introduction of the phe-
nylbutyloxy substituent on the melatonin indole ring did not
provide any selectivity,[45] suggesting that the effect of specific
lipophilic substituents may be class-dependent. The presence
Scheme 1. Reagents and conditions: a) BBr3, CH2Cl2, RT, 18 h; b) NaH, R1X,
DMF, RT, 8–16 h (or 2-(2-bromoethyl)naphthalene, K2CO3, CH3CN, RT, 16 h for
4h); c) LiAlH4, THF, 08C, 1 h.
3a–c were obtained by ether cleavage of the corresponding
methoxy derivatives 2a–c[48] using boron tribromide. Subse-
quent O-alkylation of phenols 3a–c with a suitable alkyl- or ar-
ylalkylhalide in the presence of a base (NaH, or K2CO3 for 4h)
afforded the O-alkylated compounds 4a–m and 4o. Hydroxy-
ethoxy derivative 4n was prepared by LiAlH4 reduction of the
corresponding ester 4o.
Table 1. Binding affinity (pKi) and intrinsic activity (IAr) of N-{[(3-O-substituted)anilino]alkyl}amide derivatives at human MT1 and MT2 melatonin receptors
stably expressed in NIH3T3 cells.
Compd
n
R
R1
hMT1
hMT2
pKi[a]
IAr[b]
pKi[a]
IAr[b]
melatonin (1)
9.39Æ0.08
9.09
1.00Æ0.01
0.95
9.47Æ0.13
9.19
1.00Æ0.02
1.06
2a[47]
4a
4b
4c
1
1
2
1
1
1
1
1
1
1
1
1
1
1
1
2
1
CH3
CH3
CH3
Ph
CH3
(CH2)4Ph
(CH2)4Ph
(CH2)4Ph
(CH2)2-O-CH2Ph
CH2-C6H4-m-Ph
CH2-C6H4-p-Ph
CH2Ph
(CH2)2-b-naphthyl
(CH2)6Ph
8.93Æ0.17
8.07Æ0.23
7.45Æ0.01
8.60Æ0.01
7.42Æ0.27
7.42Æ0.14
6.98Æ0.07
7.32Æ0.16
7.81Æ0.23
7.60Æ0.17
8.56Æ0.11
9.01Æ0.13
7.14Æ0.05
7.08Æ0.12
9.08
0.68Æ0.09
0.23Æ0.04
0.12Æ0.02
0.81Æ0.05
À0.07Æ0.03
0.47Æ0.02
0.15Æ0.05
0.41Æ0.03
0.61Æ0.02
0.60Æ0.02
0.73Æ0.01
0.80Æ0.01
0.61Æ0.07
0.42Æ0.14
0.87
7.04Æ0.10
6.22Æ0.12
6.48Æ0.29
7.63Æ0.03
6.42Æ0.10
6.53Æ0.21
6.74Æ0.18
6.55Æ0.17
6.56Æ0.10
6.71Æ0.04
8.49Æ0.47
8.62Æ0.19
6.54Æ0.05
6.44Æ0.26
8.70
0.61Æ0.05
0.28Æ0.06
0.01Æ0.01
0.98Æ0.05
0.08Æ0.03
0.79Æ0.27
0.54Æ0.05
0.66Æ0.05
0.23Æ0.03
0.44Æ0.10
0.80Æ0.07
1.05Æ0.01
0.31Æ0.02
0.11Æ0.07
1.07
4d
4e
4 f
4g
4h
4i
4j
4k
4l
4m
4n
5[47]
6[47]
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
Ph
(CH2)8Ph
(CH2)3CH3
(CH2)5CH3
(CH2)7CH3
(CH2)2OH
CH3
CH3
8.38
0.79
10.18
0.61
[a] pKi values are calculated from IC50 values that were obtained from competition curves by the method of Cheng and Prusoff,[53] and are the meanÆSEM
of at least three determinations performed in duplicate. [b] Relative intrinsic activity values were obtained by dividing the maximum analogue-induced
G protein activation by that of melatonin. Data represent the meanÆSEM of at least three determinations performed in duplicate.
ChemMedChem 0000, 00, 1 – 12
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
&
3
&
ÞÞ
These are not the final page numbers!

G. Spadoni et al.
MED
of the phenylbutyl chain also affected intrinsic activity, as com-
pound 4a behaved as a partial agonist at both receptor sub-
types. Similarly, MT1-selective agomelatine homo- and hetero-
dimers showed a decreased ability to activate melatonin recep-
tors.[40,41]
substituent on the N-anilinoethylamide structure. 5-HEAT (N-{2-
[5-(2-hydroxyethoxy)-1H-indol-3-yl]ethyl}acetamide) is a func-
tional MT1-selective compound, having similar binding affinities
for MT1 and MT2 receptors but showing MT1 agonist and MT2
antagonist behavior (relative intrinsic activities: MT1 =0.92,
MT2 =0.16).[52] Compound 4n had similar low MT1 and MT2
intrinsic activities and low binding affinities.
Elongation of the ethylamido side chain in 4a by one meth-
ylene group (4b) led to a slight decrease in MT1 and MT2 bind-
ing affinities, maintaining the same selectivity ratio (MT2/MT1 Ki
=71). Relative to its methoxy analogue 5, compound 4b
showed enhanced MT1 selectivity. The presence of a phenyl
ring on the aniline nitrogen (4c) had a negative effect on MT1
binding affinity without significantly affecting potency at MT2
receptors and resulting in a loss in MT1 selectivity (MT2/MT1 Ki
ꢀ10). Indeed, an N-phenyl substituent is a structural element
conferring selectivity for the MT2 receptor, as can be seen for
partial agonist 6.[48] It has been proposed that the phenyl sub-
stituent can occupy a region of the melatonin receptors that is
larger and more accessible in the MT2 receptor than in the
MT1, thus leading to the observed MT2 selectivity.[50] Combina-
tion of two substituents conferring opposite subtype selectivi-
ty provided low binding affinity at both receptor subtypes
(4c). Replacement of one methylene group of the butyl chain
with an oxygen atom in compound 4d led to a modest de-
crease in MT1 binding affinity and to an increased affinity for
the MT2 receptor, resulting in a loss of selectivity (MT2/MT1 Ki
ꢀ10). The presence of an additional ether linkage influenced
intrinsic activity, converting a partial agonist (4a) into a full ag-
onist (4d). To investigate the preferred conformation of the
phenylbutyloxy chain, we replaced the flexible substituent
with either a meta- (4e) or para- (4 f) biphenylmethyloxy
group. Unfortunately, these bulkier aromatic ring systems re-
sulted in a great loss in binding affinity at both receptor sub-
types and to antagonist behavior for 4e. The decreased affinity
could be ascribed to the presence of a proximal phenyl ring. In
fact, the MT1 binding affinity of benzyloxy derivative 4g was
very modest, and the second phenyl ring in 4e and 4 f provid-
ed only a limited increase in binding affinity, irrespective of the
meta or para substitution. The 5-benzyloxy substituent showed
a different behavior on the melatonin scaffold, being tolerated
particularly at the MT2 receptor and leading to some selectivity
for this receptor subtype.[46,51] Rigidification of the arylalkyl side
chain in a more distal position in compound 4h, carrying a
b-naphthylethyloxy substituent, did not lead to any improve-
ment in binding affinity and subtype selectivity relative to the
biphenyl derivatives. Elongation of the alkyl chain of the phe-
nylalkyloxy substituent to six (4i) or eight (4j) methylene units
provided compounds with MT1 binding affinity lower than that
of tetramethylene derivative 4a, which is the most potent and
MT1-selective phenylalkyloxy derivative of the series. Removal
of the terminal phenyl ring in alkyloxy derivatives 4k–l mainly
influenced subtype selectivity. While affinity for MT1 receptors
was retained, the smaller alkyl groups were better tolerated at
the MT2 receptor. Compound 4l, with a hexyloxy substituent,
is a potent nonselective melatonin receptor agonist. A further
increase in alkyl chain length resulted in a considerable de-
crease in binding affinity at both receptors (4m versus 4l).
Finally, we investigated the effect of the 2-hydroxyethyloxy
Molecular modeling
Recently, a number of experimentally determined 3D structures
of agonist-bound GPCRs crystallized in their fully active state
have been reported, such as those of rhodopsin and the b2
adrenergic receptor. In addition to their importance in deci-
phering the receptor activation mechanism at a molecular
level, these agonist-bound receptors represent valuable tem-
plate structures for the building of novel homology models of
class A GPCRs. We therefore started from these active receptor
configurations to build a new MT1 homology model with the
aim of investigating the molecular basis of agonist recognition
and MT1 selectivity. Prediction of the 3D structure of melatonin
receptors can be considered a challenging task, given the
availability of structural templates showing only a limited per-
centage identity (<30% within the transmembrane (TM) do-
mains) with the target sequences. A further element which
makes melatonin receptor modeling even more challenging is
the lack of well-characterized ligand–receptor contacts. Muta-
genesis studies of melatonin receptors resulted in sparse infor-
mation and did not provide a definite binding scheme.[21] Diffi-
culty in the identification of the binding mode for melatonin
could be due to its lipophilic character and to the receptor
amino acid composition. Indeed, melatonin functional groups
cannot undertake strong polar interactions, and melatonin re-
ceptors are non-aminergic GPCRs that lack the complex pat-
tern of crucial amino acids that are known to anchor the polar
substituents of endogenous amines (e.g., the conserved as-
partic residue located on TM3). However, even if the modeling
of melatonin receptors must cope with the high degree of un-
certainty of available structural information, it could take ad-
vantage of the amount of ligand-based information available
thus far, such as structure–activity relationships and pharmaco-
phore models.
The crystal structure of the fully activated b2 adrenergic re-
ceptor (PDB code: 3P0G)[54] was used for homology modeling
of the MT1 receptor (see Experimental Section), and induced-fit
docking (IFD)[55] was applied to shape the putative binding site
around the structure of 2-phenylmelatonin (2PhMLT); 2PhMLT
was chosen because it is one of the most potent melatonin re-
ceptor agonists,[56] and its bulky structure was suitable to
define and model the MT1 binding site. 2PhMLT was rigidly
docked, with the conformation of the ethylamide side chain
and the methoxy group consistent with a previously devel-
oped pharmacophore model for melatoninergic agonists.[57]
Indeed, introduction of this ligand-based information allowed
us to focus only on those receptor–ligand complexes consis-
tent with known SARs. Resulting MT1 receptor–2PhMLT com-
plexes were evaluated according to both their docking scores
&4
&
www.chemmedchem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 12
ÝÝ
These are not the final page numbers!

MT1-Selective Melatonin Receptor Ligands
MED
and to agreement with the indications provided by mutagene-
sis. Site-directed mutagenesis performed on the MT1 receptor
suggested that His1955.46 (superscripts refer to Ballesteros-
Weinstein numbering[58]), located on TM5, is likely to be in-
volved in the stabilization of the 5-methoxy group of melato-
nin, as its mutation led to a three- to eightfold decrease in
binding affinity only for ligands carrying the 5-methoxy sub-
stituent.[59,60] A similar result was obtained by mutation of the
corresponding histidine (His2085.46) on the MT2 receptor.[61]
Therefore, the best ranked MT1 receptor–2PhMLT complex
having a 5-methoxy group in proximity to His1955.46 was se-
lected for further analysis. To evaluate the stability of the
ligand–receptor interactions in the resulting binding pose,
a molecular dynamics (MD) simulation was carried out in a sol-
vated lipid bilayer for 50 ns. As shown in Figure 2, the root
mean squared deviation (RMSD) calculated for TM alpha car-
bons reached a plateau after 15 ns of simulation, confirming
the stability of the receptor 3D structure.
Figure 3. a) Energy-minimized structures of the MT1 receptor in complex
with 2PhMLT (purple) or 4a (orange) after 50 ns MD simulation. Residues are
depicted as sticks, whereas ligand molecules are shown in ball-and-stick rep-
resentation. Hydrogen bonds are depicted as dashed blue lines. b) Close-up
view of the MT1 receptor–4a complex after MD simulation. The ligand mole-
cule is represented as a ball-and-sticks model with orange carbons. Residues
that differ between the MT1 and MT2 receptors are represented as orange
(MT1) and transparent green (MT2) spheres.
Figure 2. RMSD for the a-carbons of TM regions of the MT1 receptor in com-
plex with 2PhMLT (black) and 4a (gray) during a 50 ns MD simulation.
Visual inspection of trajectories revealed that ligand–recep-
tor interactions remained stable for the whole simulation, and
2PhMLT retained its initial conformation and orientation within
the ligand binding site. 2PhMLT was accommodated into
a binding site crevice between TM3, TM4, TM5, TM6, and TM7.
The agonist molecule formed two hydrogen bond interactions
between the amide oxygen and the hydroxy group of
Tyr2857.43 and between the methoxy oxygen and the hydroxy
group of Tyr1875.38 (Figure 3). No mutagenesis data on the MT1
receptor are available for these two tyrosines. However,
tagonist ZM241385,[63] resulted in an eightfold decrease in an-
tagonist binding.[64] Site-directed mutagenesis performed on
the V1a vasopressin receptor[65] and the a1B adrenergic recep-
tor[66] clearly highlighted the importance of the tyrosine resi-
due at position 5.38 in agonist binding. The indole ring of
2PhMLT was sandwiched between Gly1083.33 and His1955.46
and the 2-phenyl ring was accommodated within an additional
cavity formed by Trp2516.48, Leu2546.51, Asn2556.52, Tyr2817.39
,
,
Tyr2987.43 in the MT2 receptor, corresponding to MT1 Tyr2857.43
,
was found to be crucial for agonist stabilization, as mutation
of this residue to alanine completely abolished melatonin bind-
ing.[62] The involvement of position 5.38 in ligand binding has
been extensively investigated in different class A GPCRs
through mutagenesis studies, highlighting a key role of this
amino acid in ligand stabilization. For example, the non-ami-
nergic A2A adenosine receptor mutation of Met1775.38, which
directly interacts with the furan ring of the co-crystallized an-
and Ala2847.42; two edge-to-face interactions occurred be-
tween the side chains of Trp2516.48 and Tyr2817.39, and the 2-
phenyl ring further stabilized the ligand conformation within
the binding site (see Supporting Information figure S3 for a de-
tailed representation of the binding site). The amide function-
ality of 2PhMLT was within close proximity to Met1073.32, with
the methyl group interacting with the residue side chain. This
region could also accommodate longer amide substituents,
ChemMedChem 0000, 00, 1 – 12
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
&
5
&
ÞÞ
These are not the final page numbers!

G. Spadoni et al.
MED
such as ethyl or propyl groups, consistent with SAR for differ-
ent series of melatonin receptor agonists.[67]
molecular basis of ligand binding and to propose structural
modifications aimed at improving ligand potency and selectivi-
ty.
To investigate the molecular basis of MT1 selectivity, com-
pound 4a was docked into the refined MT1 receptor model,
obtained from the MD simulation of the MT1–2PhMLT complex.
IFD was applied to facilitate the accommodation of the bulky
phenylbutyloxy chain of compound 4a. In this case, the stabili-
ty of the resulting MT1 receptor–4a complex was also assessed
through an MD simulation performed in a solvated lipid bilay-
er. The receptor structure was stable during the 50 ns MD sim-
ulation (Figure 2), as were the ligand–receptor interactions. As
can be seen in Figure 3a, the lipophilic chain of compound 4a
leaned on the tip of TM3 and TM4, forming hydrophobic inter-
Experimental Section
Chemistry
General methods: Melting points were determined on a Bꢁchi
B-540 capillary melting point apparatus and are uncorrected.
1H NMR spectra and 13C NMR spectra were recorded on a Bruker
Avance 200 spectrometer (1H: 200 MHz; 13C: 50 MHz) using CDCl3
as solvent unless stated otherwise. Chemical shifts (d scale) are re-
ported in parts per million (ppm) relative to the central peak of the
solvent. Coupling constants (J values) are given in Hz. EI-MS spec-
tra (70 eV) were taken on a Fisons Trio 1000 instrument. Only mo-
lecular ions [M]+ and base peaks are given. ESI-MS spectra were
taken on a Waters Micromass Zq instrument. Only molecular ions
[M+1] are given. Infrared spectra were obtained on a Nicolet
Avatar 360 FT-IR spectrometer; absorbances are reported in n˜
(cmÀ1). Elemental analyses for C, H, and N were performed on
a Carlo Erba analyzer, and results are within 0.4% of calculated
values. Column chromatography purifications were performed
under flash conditions using Merck 230–400 mesh silica gel. Analyt-
ical thin layer chromatography (TLC) was carried out on Merck
silica gel 60 F254 plates. The two radioligands, 2-[125I]iodomelatonin
(specific activity, 2000 CimmolÀ1) and [35S]GTPgS ([35S]guanosine-5’-
O-(3-thio-triphosphate); specific activity, 1000 CimmolÀ1) were pur-
chased from PerkinElmer. 2-(2-Bromoethyl)naphthalene was pre-
pared according to a published method.[68]
actions with several amino acids such as Gln1013.26, Gly1043.29
,
Phe1053.30, Val1594.57, Leu1634.61, and Leu168, Gln169, and
Tyr175 on extracellular loop 2 (ECL2). Interestingly, a sequence-
based comparison of MT1 and MT2 receptors revealed that the
MT2 subtype has some bulkier residues in positions surround-
ing the phenylbutyloxy chain of compound 4a (Figure 3b and
Supporting Information figure S1). Indeed, Gly1043.29
,
Val1594.57, and Leu1634.61 in the MT1 receptor are replaced by
Ala1173.29, Leu1724.57, and Phe1764.61, respectively, in the MT2
receptor. Based on this observation, it could be hypothesized
that the bulkier amino acids in the MT2 receptor hamper the
accommodation of the phenylbutyloxy substituent of 4a.
Therefore, the different amino acid composition may provide
a structural clue for the MT1 selectivity observed among chemi-
cally different melatoninergic ligands characterized by the
presence of bulky lipophilic chains in the position correspond-
ing to position 5 of melatonin.
General procedure for the synthesis of phenol derivatives 3a–c:
A solution of BBr3 (1m in CH2Cl2, 20 mL, 20 mmol) diluted with dry
CH2Cl2 (40 mL) was added dropwise to a solution of the suitable
methoxy derivative 2a–c (10 mmol) in dry CH2Cl2 (40 mL) at 08C,
and the resulting mixture was stirred at room temperature for
18 h. The reaction mixture was neutralized with a 2 n aqueous so-
lution of Na2CO3 and extracted with EtOAc. The organic phases
were combined, dried (Na2SO4), and concentrated under reduced
pressure to give a crude product that was purified by flash chro-
matography (silica gel; CH2Cl2/MeOH, 98:2).
Conclusions
A series of N-{[(3-O-substituted)anilino]alkyl}acetamides were
described as novel melatoninergic ligands, some of which
were endowed with high binding affinity and good MT1 sub-
type selectivity. Substituents in the C3 aniline position different-
ly affected binding affinity and intrinsic activity, depending on
their nature. Introduction of a C4- or C6-methylenoxy substitu-
ent was tolerated by both receptor subtypes (4k–l), whereas
an unfavorable effect was produced by the longer C8-alkyloxy
group (4m) or by conformationally constrained substituents
(4d–h). In contrast, the non-selective MT1/MT2 ligand 2a was
successfully converted into the high affinity MT1-selective
ligand 4a by replacing the ether methyl group with a phenyl-
butyl side chain. These results confirm the importance of re-
placing the methoxy group with a bulky substituent to achieve
MT1 selectivity and to help clarify the size and shape of the
two MT1 and MT2 binding pockets.
N-{2-[(3-Hydroxyphenyl)methylamino]ethyl}acetamide
(3a):
White solid (1.58 g, 76%): mp: 74–768C (Et2O/petroleum ether);
1H NMR (CDCl3): d=1.95 (s, 3H), 2.91 (s, 3H), 3.44 (m, 4H), 5.80
(brs, 1H), 6.29 (m, 3H), 6.71 (brs, 1H), 7.08 ppm (m, 1H); 13C NMR
(CDCl3): d=171.2, 157.4, 150.9, 130.2, 104.6, 104.3, 99.7, 51.6, 38.3,
37.4, 23.2 ppm; MS (EI, 70 eV): m/z 208 [M]+, 136 (100).[43]
N-{2-[(3-Hydroxyphenyl)methylamino]propyl}acetamide (3b): Oil
1
(1.51 g, 68%): H NMR ([D6]DMSO/CDCl3): d=1.51 (m, 2H), 1.65 (s,
3H), 2.90 (m, 2H), 3.06 ppm (m, 2H); ESI-MS (m/z): 223 [M+1];
13C NMR (CDCl3): d=171.2, 157.4, 150.9, 130.2, 104.7, 104.3, 99.7,
51.4, 38.8, 37.6, 26.5, 23.3 ppm.
N-{2-[(3-Hydroxyphenyl)phenylamino]ethyl}acetamide (3c): See
ref. [49] for details.
Structural information derived from the 3D model of the
MT1 receptor in its active state was combined with SAR, phar-
macophore, and mutagenesis data to propose a binding inter-
action scheme for agonists and a possible explanation for the
MT1 selectivity of compounds with a bulky lipophilic substitu-
ent on the ether oxygen. This receptor model, even with the
intrinsic limitations due to the building procedure and the lim-
ited structural information available, could help to clarify the
General procedure for the synthesis of N-[3-(substituted alkoxy)-
anilinoalkyl]acetamides 4a–g, 4i–m, and 4o: Sodium hydride
(80% in mineral oil, 0.017 g, 0.55 mmol) was added to a solution of
the suitable phenol derivative 3a–c (0.5 mmol) in dry DMF (2 mL)
at À108C under nitrogen atmosphere. After stirring for 5 min, the
suitable alkylating agent (0.8 mmol) was added to the reaction
mixture, and stirring was continued for 8–16 h, allowing the mix-
ture to rise to room temperature. The reaction mixture was poured
&6
&
www.chemmedchem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 12
ÝÝ
These are not the final page numbers!

MT1-Selective Melatonin Receptor Ligands
MED
into water and extracted 3ꢂ with EtOAc. The organic phases were
combined, washed once with brine, dried (Na2SO4), and concen-
trated to give the desired crude product, which was purified by
silica gel flash chromatography using EtOAc as eluent.
N-(2-{[3-(Biphenyl-4-ylmethoxy)phenyl]methylamino}ethyl)aceta-
mide (4 f): This compound was prepared according to the general
procedure described above, starting from 3a and 4-phenylbenzyl
bromide to yield a white solid (60 mg, 32%): mp: 122–1238C
1
(Et2O); H NMR (CDCl3): d=1.94 (s, 3H), 2.95 (s, 3H), 3.46 (m, 4H),
N-(2-{Methyl-[3-(4-phenylbutoxy)phenyl]amino}ethyl)acetamide
(4a): This compound was prepared according to the general pro-
cedure described above, starting from 3a and 1-bromo-4-phenyl-
butane to yield a white solid (100 mg, 59%): mp: 578C (Et2O/petro-
5.11 (s, 2H), 5.57 (brs, 1H), 6.41–6.47 (m, 3H), 7.18 (m, 1H), 7.31–
7.65 ppm (m, 9H); 13C NMR (CDCl3): d=170.4, 160.0, 150.9, 140.9,
140.8, 136.3, 130.0, 128.8, 128.1, 127.4, 127.3, 127.1, 105.8, 102.5,
100.0, 69.7, 51.7, 38.5, 37.3, 23.3 ppm; IR (nujol): n˜ =3240,
1637 cmÀ1; MS (EI, 70 eV): m/z 374 [M]+, 167 (100); Anal. calcd for
C24H26N2O2: C 76.98, H 7.00, N 7.48, found: C 77.12, H 7.29, N 7.59.
1
leum ether); H NMR (CDCl3): d=1.82 (m, 4H), 1.94 (s, 3H), 2.69 (m,
2H), 2.93 (s, 3H), 3.46 (m, 4H), 3.97 (t, 2H, J=6.0), 5.60 (brs, 1H),
6.28–6.40 (m, 3H), 7.10–7.34 ppm (m, 6H); 13C NMR (CDCl3): d=
170.4, 160.4, 150.9, 142.3, 130.0, 128.4, 128.3, 125.8, 105.5, 102.2,
99.7, 67.6, 51.8, 38.4, 37.4, 35.6, 29.0, 27.9, 23.3 ppm; IR (nujol): n˜ =
3309, 1638 cmÀ1; MS (EI, 70 eV): m/z 340 [M]+, 268 (100); Anal.
calcd for C21H28N2O2: C 74.08, H 8.29, N 8.23, found: C 74.31, H
8.42, N 8.01.
N-{2-[(3-Benzyloxyphenyl)methylamino]ethyl}acetamide
(4g):
This compound was prepared according to the general procedure
described above, starting from 3a and benzyl bromide to yield
a white solid (56 mg, 38%): mp: 67–688C (Et2O/petroleum ether);
1H NMR (CDCl3): d=1.93 (s, 3H), 2.93 (s, 3H), 3.45 (m, 4H), 5.07 (s,
2H), 5.58 (brs, 1H), 6.34–6.43 (m, 3H), 7.15 (m, 1H), 7.32–7.47 ppm
(m, 5H); 13C NMR (CDCl3): d=170.4, 160.1, 150.8, 137.3, 130.0,
128.6, 127.9, 127.5, 105.8, 102.5, 100.0, 69.9, 51.7, 38.5, 37.3,
23.3 ppm; IR (nujol): n˜ =3323, 1649 cmÀ1; MS (EI, 70 eV): m/z 298
[M]+, 91 (100); Anal. calcd for C18H22N2O2: C 72.46, H 7.43, N 9.39,
found: C 72.49, H 7.38, N 9.09.
N-(3-{Methyl-[3-(4-phenylbutoxy)phenyl]amino}propyl)acetamide
(4b): This compound was prepared according to the general pro-
cedure described above, starting from 3b and 1-bromo-4-phenyl-
1
butane to yield an oil (97 mg, 55% yield): H NMR (CDCl3): d=1.80
(m, 6H), 1.94 (s, 3H), 2.69 (m, 2H), 2.89 (s, 3H), 3.31 (m, 4H), 3.95
(t, 2H, J=6.0), 5.53 (brs, 1H), 6.23–6.45 (m, 3H), 7.08–7.29 ppm (m,
6H); 13C NMR (CDCl3): d=170.1, 160.3, 150.6, 142.3, 130.0, 128.4,
128.3, 125.8, 105.6, 101.8, 99.8, 67.6, 50.5, 38.6, 37.8, 35.6, 29.0,
27.9, 27.0, 23.3 ppm; IR (neat): n˜ =3289, 1650 cmÀ1; ESI-MS (m/z):
355 [M+1]; Anal. calcd for C22H30N2O2: C 74.54, H 8.53, N 7.90,
found: C 74.41, H 8.70, N 7.95.
N-(2-{Methyl-[3-(6-phenylhexyloxy)phenyl]amino}ethyl)aceta-
mide (4i): This compound was prepared according to the general
procedure described above, starting from 3a and 1-bromo-6-phe-
nylhexane to yield a white solid (66 mg, 36%): mp: 58–598C (Et2O/
petroleum ether); 1H NMR (CDCl3): d=1.36–1.86 (m, 8H), 1.95 (s,
3H), 2.63 (m, 2H), 2.94 (s, 3H), 3.47 (m, 4H), 3.95 (t, 2H, J=6.5),
5.60 (brs, 1H), 6.24–6.41 (m, 3H), 7.10–7.33 ppm (m, 6H); 13C NMR
(CDCl3): d=170.4, 160.4, 150.9, 142.7, 130.0, 128.4, 128.2, 125.6,
105.5, 102.3, 99.7, 67.7, 51.8, 38.4, 37.4, 35.9, 31.4, 29.3, 29.0, 26.0,
23.3 ppm; IR (nujol): n˜ =3312, 1652 cmÀ1; MS (EI, 70 eV): m/z 368
[M]+, 91 (100); Anal. calcd for C23H32N2O2: C 74.96, H 8.75, N 7.60,
found: C 75.08, H 8.88, N 7.51.
N-(2-{Phenyl-[3-(4-phenylbutoxy)phenyl]amino}ethyl)acetamide
(4c): This compound was prepared according to the general proce-
dure described above, starting from 3c and 1-bromo-4-phenylbu-
tane to yield a white solid (90 mg, 45%): mp: 98–998C (Et2O/petro-
1
leum ether); H NMR (CDCl3): d=1.80 (m, 4H), 1.92 (s, 3H), 2.68 (m,
2H), 3.50 (m, 2H), 3.90 (m, 4H), 5.61 (brs, 1H), 6.48–6.61 (m, 3H),
7.00–7.34 pm (m, 11H); 13C NMR (CDCl3): d=170.4, 160.2, 149.1,
147.6, 142.2, 130.0, 129.5, 128.4, 128.3, 125.8, 122.2, 121.8, 112.7,
107.0, 106.9, 67.7, 51.0, 37.9, 35.6, 28.9, 27.9, 23.2 ppm; IR (nujol):
n˜ =3322, 1647 cmÀ1; ESI-MS (m/z): 403 [M+1]; Anal. calcd for
C26H30N2O2: C 77.58, H 7.51, N 6.96, found: C 77.42, H 7.26, N 6.98.
N-(2-{Methyl-[3-(8-phenyloctyloxy)phenyl]amino}ethyl)acetamide
(4j): This compound was prepared according to the general proce-
dure described above, starting from 3a and 1-bromo-8-phenyloc-
tane to yield a white solid (105 mg, 53%): mp: 65–668C (Et2O/pe-
troleum ether); 1H NMR (CDCl3): d=1.28–1.84 (m, 12H), 1.94 (s,
3H), 2.61 (m, 2H), 2.93 (s, 3H), 3.46 (m, 4H), 3.94 (t, 2H, J=6.5),
5.61 (brs, 1H), 6.28–6.40 (m, 3H), 7.10–7.32 ppm (m, 6H); 13C NMR
(CDCl3): d=170.4, 160.4, 150.9, 142.9, 130.0, 128.4, 128.2, 125.6,
105.4, 102.3, 99.7, 67.8, 51.8, 38.4, 37.4, 36.0, 31.5, 29.44, 29.38,
29.34, 29.26, 26.1, 23.3 ppm; IR (nujol): n˜ =3306, 1643 cmÀ1; MS (EI,
70 eV): m/z 396 [M]+, 91 (100); Anal. calcd for C25H36N2O2: C 75.72,
H 9.15, N 7.06, found: C 75.34, H 9.49, N 7.22.
N-(2-{[3-(2-Benzyloxyethoxy)phenyl]methylamino}ethyl)aceta-
mide (4d): This compound was prepared according to the general
procedure described above, starting from 3a and benzyl 2-bro-
1
moethyl ether to yield an oil (104 mg, 61%): H NMR (CDCl3): d=
1.93 (s, 3H), 2.92 (s, 3H), 3.44 (m, 4H), 3.83 (m, 2H), 4.15 (m, 2H),
4.64 (s, 2H), 5.58 (brs, 1H), 6.28–6.41 (m, 3H), 7.13 (m, 1H), 7.26–
7.38 ppm (m, 5H); 13C NMR (CDCl3): d=170.4, 160.1, 150.8, 138.1,
130.0, 128.4, 127.8, 127.7, 105.8, 102.3, 99.9, 73.4, 68.6, 67.3, 51.8,
38.5, 37.3, 23.3 ppm; IR (neat): n˜ =3297, 1655 cmÀ1; ESI-MS (m/z):
343 [M+1]; Anal. calcd for C20H26N2O3: C 70.15, H 7.65, N 8.18,
found: C 70.22, H 7.66, N 8.24.
N-{2-[(3-Butoxyphenyl)methylamino]ethyl}acetamide (4k): This
compound was prepared according to the general procedure de-
scribed above, starting from 3a and 1-bromobutane to yield
a white solid (79 mg, 60%): mp: 688C (Et2O/petroleum ether);
1H NMR (CDCl3): d=0.98 (t, 3H, J=7.0), 1.50 (m, 2H), 1.77 (m, 2H),
1.94 (s, 3H), 2.94 (s, 3H), 3.46 (m, 4H), 3.96 (t, 2H, J=6.5), 5.61
(brs, 1H), 6.29–6.41 (m, 3H), 7.14 ppm (m, 1H); 13C NMR (CDCl3):
d=170.4, 160.4, 150.9, 130.0, 105.5, 102.3, 99.7, 67.5, 51.8, 38.4,
37.4, 31.4, 23.3, 19.3, 13.9 ppm; IR (nujol): n˜ =3231, 1637 cmÀ1; MS
(EI, 70 eV): m/z 264 [M]+, 192 (100); Anal. calcd for C15H24N2O2: C
68.15, H 9.15, N 10.60, found: C 68.23, H 9.30, N 10.69.
N-(2-{[3-(Biphenyl-3-ylmethoxy)phenyl]methylamino}ethyl)aceta-
mide (4e): This compound was prepared according to the general
procedure described above, starting from 3a and 3-phenylbenzyl
bromide to yield a white solid (133 mg, 71%): mp: 80–818C (Et2O/
petroleum ether); 1H NMR (CDCl3): d=1.92 (s, 3H), 2.94 (s, 3H),
3.45 (m, 4H), 5.13 (s, 2H), 5.58 (brs, 1H), 6.39–6.43 (m, 3H), 7.16
(m, 1H), 7.30–7.59 ppm (m, 9H); 13C NMR (CDCl3): d=170.4, 160.1,
150.8, 141.6, 140.9, 137.8, 130.1, 129.0, 128.8, 127.4, 127.2, 126.8,
126.5, 126.4, 105.8, 102.5, 100.0, 70.0, 51.7, 38.5, 37.3, 23.2 ppm; IR
(nujol): n˜ =3280, 1635 cmÀ1; MS (EI, 70 eV): m/z 374 [M]+, 167
(100); Anal. calcd for C24H26N2O2: C 76.98, H 7.00, N 7.48, found: C
76.73, H 6.74, N 7.35.
N-{2-[(3-Hexyloxyphenyl)methylamino]ethyl}acetamide (4l): This
compound was prepared according to the general procedure de-
scribed above, starting from 3a and 1-bromohexane to yield
a white solid (98 mg, 67%): mp: 568C (Et2O/petroleum ether);
ChemMedChem 0000, 00, 1 – 12
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
&
7
&
ÞÞ
These are not the final page numbers!

G. Spadoni et al.
MED
1H NMR (CDCl3): d=0.91 (t, 3H, J=6.5), 1.27–1.62 (m, 6H), 1.77 (m,
2H), 1.94 (s, 3H), 2.94 (s, 3H), 3.46 (m, 4H), 3.96 (m, 2H), 5.64 (brs,
1H), 6.28–6.41 (m, 3H), 7.14 ppm (m, 1H); 13C NMR (CDCl3): d=
170.4, 160.4, 150.9, 130.0, 105.4, 102.3, 99.7, 67.8, 51.8, 38.4, 37.4,
31.6, 29.4, 25.8, 23.2, 22.6, 14.05 ppm; IR (nujol): n˜ =3230,
1630 cmÀ1; MS (EI, 70 eV): m/z 292 [M]+, 220 (100); Anal. calcd for
C17H28N2O2: C 69.83, H 9.65, N 9.58, found: C 69.51, H 9.57, N 9.23.
170.6, 160.0, 150.8, 130.1, 105.7, 102.5, 99.5, 69.0, 61.4, 51.7, 38.5,
37.3, 23.3 ppm; IR (CDCl3): n˜ =3456, 1668 cmÀ1; ESI MS (m/z): 253
[M+1]; Anal. calcd for C13H20N2O3: C 61.88, H 7.99, N 11.10, found:
C 61.98, H 8.11, N 11.36.
Pharmacology
Binding affinities were determined using 2-[125I]iodomelatonin as
the labeled ligand in competition experiments with cloned human
MT1 and MT2 receptors expressed in NIH3T3 rat fibroblast cells. The
characterization of NIH3T3 MT1 and MT2 cells was already described
in detail.[69,70] Membranes were incubated for 90 min at 378C in
binding buffer (50 mm Tris/HCl, pH 7.4). The final membrane con-
centration was 5–10 mg of protein per tube. The membrane pro-
tein level was determined in accordance with a previously reported
method.[71] 2-[125I]Iodomelatonin (100 pm) and different concentra-
tions of the new compounds were incubated with the receptor
preparation for 90 min at 378C. Nonspecific binding was assessed
with 10 mm melatonin; IC50 values were determined by nonlinear
fitting strategies with Prism (GraphPad SoftWare Inc., San Diego,
CA). The pKi values were calculated from the IC50 values in accord-
ance with the Cheng–Prusoff equation.[53] The pKi values are the
mean of at least three independent determinations performed in
duplicate.
N-{2-[Methyl-(3-octyloxyphenyl)amino]ethyl}acetamide
This compound was prepared according to the general procedure
(4m):
above described starting from 3a and 1-bromooctane. White solid
1
(93 mg, 58%); mp 63–48C (Et2O/petroleum ether); H NMR (CDCl3):
d=0.89 (m, 3H), 1.25–1.49 (m, 10H), 1.77 (m, 2H), 1.94 (s, 3H),
2.94 (s, 3H), 3.46 (m, 4H), 3.94 (t, 2H, J=6.5), 5.63 (brs, 1H), 6.28–
6.40 (m, 3H), 7.13 ppm (m, 1H); 13C NMR (CDCl3): d=170.4, 160.4,
150.9, 129.9, 105.4, 102.3, 99.7, 67.8, 51.8, 38.4, 37.4, 31.8, 29.4,
29.4, 29.3, 26.1, 23.3, 22.7, 14.1 ppm; IR (nujol): n˜ =3219,
1610 cmÀ1; MS (EI, 70 eV): m/z 320 [M]+, 248 (100); Anal. calcd for
C19H32N2O2: C 71.21, H 10.06, N 8.74, found: C 71.39, H 10.23, N
8.58.
{3-[(2-Acetylaminoethyl)methylamino]phenoxy}acetic
acid
methyl ester (4o): This compound was prepared according to the
general procedure described above, starting from 3a and methyl
chloroacetate. The crude residue was purified by flash chromatog-
raphy over silica gel, using EtOAc as an eluent to yield an amor-
1
phous solid (116 mg, 83%): H NMR (CDCl3): d=1.94 (s, 3H), 2.94 (s,
To determine the functional activity of the new compounds at MT1
and MT2 receptor subtypes,[35S]GTPgS binding assays in NIH3T3
cells expressing human-cloned MT1 or MT2 receptors were per-
formed. The amount of bound [35S]GTPgS is proportional to the
level of the analogue-induced G protein activation and is related to
the intrinsic activity of the compound under study. The detailed
description and validation of this method were reported else-
where.[69,72] Membranes (15–25 mg of protein, final incubation
volume 100 mL) were incubated at 308C for 30 min in the presence
and absence of melatonin analogues, in assay buffer consisting of
[35S]GTPgS (0.3–0.5 nm), GDP (50 mm), NaCl (100 mm), and MgCl2
(3 mm). Nonspecific binding was defined using [35S]GTPgS (10 mm).
In cell lines expressing human MT1 or MT2 receptors, melatonin
produced concentration-dependent stimulation of basal [35S]GTPgS
binding with a maximal stimulation above basal levels of 370%
and 250% in MT1 and MT2 receptors, respectively. Basal stimulation
is the amount of [35S]GTPgS specifically bound in the absence of
compounds and was taken as 100%. The maximal G protein activa-
tion was measured in each experiment using melatonin (100 nm).
Compounds were added at three different concentrations (one
concentration equivalent to 100 nm melatonin, a second one 10-
fold smaller, and a third one 10-fold larger), and the percent stimu-
lation above basal was determined. The equivalent concentration
was estimated on the basis of the ratio of the affinity of the test
compound to that of melatonin. It was assumed that, at the equiv-
alent concentration, the test compound occupies the same
number of receptors as 100 nm melatonin. All of the measure-
ments were performed in triplicate. The relative intrinsic activity
(IAr) values were obtained by dividing the maximum ligand-
induced stimulation of [35S]GTPgS binding by that of melatonin, as
measured in the same experiment.
3H), 3.46 (m, 4H), 3.82 (s, 3H), 4.64 (s, 2H), 5.79 (brs, 1H), 6.22 (dd,
1H, J=2.0, 8.0), 6.37 (d, 1H, J=2.0), 6.42 (dd, 1H, J=2.0, 8.0),
7.14 ppm (t, 1H, J=8.0); 13C NMR (CDCl3): d=170.5, 169.7, 159.0,
150.8, 130.1, 106.5, 101.5, 99.9, 65.2, 52.2, 51.6, 38.5, 37.2,
23.2 ppm; ESI MS (m/z): 281 [M+1].
N-(2-{Methyl-[3-(2-naphthalen-2-ylethoxy)phenyl]amino}ethyl)-
acetamide (4h): K2CO3 (0.033 g, 0.24 mmol) was added to a solu-
tion of 3a (0.05 g, 0.24 mmol) in CH3CN (0.8 mL), and the resulting
mixture was stirred and heated at 608C under nitrogen atmos-
phere for 20 min. 2-(2-Bromoethyl)naphthalene[68] was added to
the reaction mixture, then heating and stirring was continued for
16 h. The reaction mixture was poured into water and extracted 3x
with EtOAc. The organic phases were combined, washed once with
brine, dried (Na2SO4), and concentrated to give the crude desired
product, which was purified by silica gel flash chromatography
with EtOAc as eluent to yield an oil (7 mg, 8%): 1H NMR (CDCl3):
d=1.93 (s, 3H), 2.94 (s, 3H), 3.27 (t, 2H, J=7.0), 4.28 ppm (t, 2H,
J=7.0); 13C NMR (CDCl3): d=170.5, 160.4, 150.9, 136.3 133.5, 132.4,
130.0, 128.3, 127.7, 127.6, 127.3, 126.9, 126.2, 125.7, 105.4, 102.2,
99.7, 67.7, 51.8, 38.4, 37.4, 35.6, 23.3 ppm; IR (neat): n˜ =3292,
1658 cmÀ1; MS (EI, 70 eV): m/z 362 [M]+, 155 (100); Anal. calcd for
C23H26N2O2: C 76.21, H 7.23, N, 7.73, found: C 76.02, H 7.15, N 7.69.
N-(2-{[3-(2-Hydroxyethoxy)phenyl]methylamino}ethyl)acetamide
(4n): LiAlH4 (0.085 g, 2.24 mmol) was added portionwise to a stir-
ring solution of 4o (0.330 g, 1.2 mmol) in dry THF (1 mL) under ni-
trogen atmosphere at 08C, and the resulting mixture was stirred at
08C for 1 h. Water was added dropwise to destroy the excess hy-
dride, the resulting mixture was filtered over Celite, and the filtrate
was concentrated in vacuo and partitioned between EtOAc and
water. The combined organic phases were washed with brine,
dried (Na2SO4), and concentrated to afford a crude residue, which
was purified by flash chromatography over silica gel (EtOAc/MeOH,
95:5 as eluent) to yield a white solid (270 mg, 64%): mp: 86–878C
(Et2O/petroleum ether); 1H NMR (CDCl3): d=1.94 (s, 3H), 2.24 (t,
1H), 2.94 (s, 3H), 3.44 (m, 4H), 3.96 (m, 2H), 4.10 (m, 2H), 5.62 (brs,
1H), 6.28–6.40 (m, 3H), 7.14 ppm (m, 1H); 13C NMR (CDCl3): d=
Molecular modeling
A number of X-ray structures have been obtained for GPCRs crys-
tallized in their agonist-bound form, including rhodopsin, b1 and b2
adrenergic receptors, and A2A adenosine receptor. However, de-
spite the presence of co-crystallized agonists, only a limited
&8
&
www.chemmedchem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 12
ÝÝ
These are not the final page numbers!

MT1-Selective Melatonin Receptor Ligands
MED
number of these GPCR structures shows the predicted pattern of
structural rearrangements associated with the fully activated
state.[73,74] In particular, the outward movement of the intracellular
end of TM6, required to accommodate the C-terminus of the cog-
nate G protein, was observed only in some agonist-bound forms of
rhodopsin and the b2 adrenergic receptor. The b2 adrenergic recep-
tor was selected as the template structure for homology modeling
of the MT1 receptor, sharing higher sequence identity with the TM
regions (24%) than rhodopsin (22%). The amino acid sequence of
the human MT1 receptor and the X-ray crystal structure of the b2
adrenergic receptor were retrieved from the Universal Protein Re-
source[75] (UniProt ID: P48039) and the Protein Data Bank (PDB
code: 3P0G),[53] respectively. An initial sequence alignment was cre-
ated with ClustalW2[76] and was subsequently refined, taking into
account conserved sequence motifs among class A GPCRs.[58,77] In
particular, highly conserved residues located on TM domains, as
well as the conserved disulfide bridge connecting ECL2 and TM3,
were considered during alignment refinement (Supporting Infor-
mation figure S1).
Met1073.32 and His1955.46, setting the enclosing and bounding
boxes to default dimensions. 2PhMLT was rigidly docked into the
MT1 receptor binding site to retain the postulated bioactive confor-
mation. Ligand docking was performed in the standard precision
mode, collecting fifty poses for subsequent analysis. Resulting
ligand–receptor complexes were then submitted to a protein struc-
ture refinement stage; once amino acid side chains that had previ-
ously been removed were re-introduced, residues within a shell of
6 ꢃ around any ligand pose were refined by a side chain conforma-
tional search, followed by energy minimization of the residues and
the ligand molecule. In the final docking stage, 2PhMLT was rigidly
re-docked into each receptor structure produced in the previous
refinement step, applying default Glide settings. The IFD score, ac-
counting for protein–ligand interaction energy and the total
energy of the system, was used to rank the final ligand–receptor
complexes. Mutagenesis studies of the MT1 receptor[59,60] suggested
a role of His1955.46 in the stabilization of the 5-methoxy group of
melatonin (see main text). Therefore, the best ranked ligand–recep-
tor complex with the 5-methoxy substituent within proximity of
His1955.46 was selected for further analysis. The MT1 receptor–
2PhMLT complex was minimized using the Amber force field[84] im-
plemented in MacroModel 9.7[85] and by applying the Polak–Ribiere
Comparative modeling was carried out with Modeller 9.7,[78] and an
ensemble of 30 MT1 receptor models was initially generated. The
3D coordinates of residues 23 to 227 and 266 to 344 of the b2 re-
ceptor were used as the template. For crystallization purposes, in-
tracellular loop 3 of the b2 receptor (residues 228–265) was re-
placed with the T4 lysozyme molecule, thus Modeller was used to
rebuild the corresponding sequence of the MT1 receptor (residues
216–230). The MT1 receptor models begin at residue 17 and end at
residue 314. Stereochemical quality assessment of the receptor
models was performed with Procheck[79] and with the Protein
Report tool implemented in Maestro 9.0.[80] Selection of the best
model was based on the quality of geometrical parameters and on
the Modeller objective function. The selected MT1 receptor model
was then processed through the Protein Preparation Wizard work-
flow[81] available in Maestro 9.0. Hydrogen atoms were added to
the structure, and protonation states for ionizable side chains were
chosen to be consistent with physiological pH. The overall hydro-
gen bonding network was optimized by adjusting the tautomeriza-
tion states of histidine residues and by sampling the orientation of
hydroxy and thiol groups, together with the side chain amides of
asparagine and glutamine residues. Protein C- and N-termini were
capped with neutral groups (acetyl and methylamino, respectively),
and the all-hydrogen receptor model was then subjected to re-
strained minimization using the OPLS2001[82] force field to an
RMSD of 0.5 ꢃ. The Ramachandran plot for the final refined struc-
ture is reported in Supporting Information figure S2.
conjugate gradient method to
a convergence threshold of
0.05 kJmolÀ1 ꢃÀ1; the ligand and residues within 8 ꢃ proximity of
the ligand were free to move, while all other atoms were fixed.
The resulting structure was then prepared for MD simulation using
Desmond version 2.2.6.2.3:[86,87] accordingly, the energy-minimized
ligand–receptor complex was embedded in a POPC lipid bilayer by
aligning the receptor to the 3P0G crystal structure deposited into
the Orientations of Protein in Membranes (OPM) database,[88] with
at least 20 ꢃ between the protein and its closest periodic image.
The protein membrane system was solvated by approximately
10700 T3P water molecules in a simulation box of 77 ꢃꢂ74 ꢃꢂ
99 ꢃ. The Amber99SB[89] force field was used to model the protein,
while ligand and lipids were parameterized using GAFF.[90] Partial
atomic charges of 2PhMLT were computed by the Antechamber
module[91] of AmberTools 10 at the AM1-BCC level. The system was
relaxed using a modified version of a membrane relaxation proto-
col implemented in the Desmond package. The equilibration
phase was followed by a 50 ns long MD simulation performed in
the NPT ensemble at 310 K and 1 atm using the Langevin coupling
scheme.[92] All bond lengths to hydrogen atoms were constrained
using M-SHAKE.[93] Short-range electrostatic interactions were cut
off at 9 ꢃ, whereas long-range electrostatic interactions were com-
puted using the Smooth Particle Mesh Ewald method.[94] A RESPA
integrator[95] was used with a time-step of 2 fs, and long-range
electrostatics were computed every 6 fs.
The structure of 2PhMLT was built with Maestro 9.0 and optimized
to fit a pharmacophore model previously developed for nonselec-
tive melatoninergic agonists.[57] The acetylaminoethyl side chain of
2PhMLT showing the best fit is perpendicular to the indole ring in
its extended conformation (torsion angles: C3a-C3-Cb-Ca ꢀÀ908, C3-
Cb-Ca-Nꢀ1808, Cb-Ca-N-COꢀ1808, Figure 1). The 5-methoxy methyl
group lies in the plane of the indole ring, directed toward posi-
tion 4. The final ligand structure was minimized using the
Compound 4a was optimized with the OPLS2005 force field to
a convergence threshold of 0.05 kJmolÀ1 ꢃÀ1. An IFD protocol was
then applied to dock 4a into the MT1 structure obtained from the
energy-minimized final snapshot of the MD simulation performed
on the MT1–2PhMLT complex. Glide scoring grids for subsequent
docking calculations were centered on the 2PhMLT pose obtained
at the end of the MD simulation. Compound 4a was docked flexi-
bly, imposing the formation of a hydrogen bond between the
Tyr1875.38 hydroxy group and the phenolic oxygen of the ligand,
and between the Tyr2857.43 hydroxy group and the amide oxygen,
to produce the same pattern of polar interactions as 2-PhMLT. To
favor the accommodation of the bulky phenylbutyloxy chain of
compound 4a, Leu1634.61 and Gln181, located on ECL2, were tem-
porarily mutated to alanines. The MT1–4a complex, having the
best IFD score and with the amide side chain conformation of 4a
consistent with the pharmacophore model for nonselective mela-
OPLS2005 force field[83] to
a
convergence threshold of
0.05 kJmolÀ1 ꢃÀ1. An IFD protocol was applied to account for both
ligand and receptor flexibility during ligand docking.[55] An initial
softened-potential docking run was performed, applying van der
Waals radii scaling of 0.7 and 0.5 on protein and ligand non-polar
atoms, respectively. Amino acids hampering accommodation of
2PhMLT into the binding crevice (Val1113.36
, , and
Trp2516.48
Tyr2817.39) were temporarily mutated to alanines. Energy grids gen-
erated for the initial softened-potential docking were centered on
ChemMedChem 0000, 00, 1 – 12
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
&
9
&
ÞÞ
These are not the final page numbers!

G. Spadoni et al.
MED
[24] O. Nosjean, M. Ferro, F. Cogꢆ, P. Beauverger, J. M. Henlin, F. Lefoulon,
J. L. Fauchere, P. Delagrange, E. Canet, J. A. Boutin, J. Biol. Chem. 2000,
275, 31311–31317.
[25] C. Ekmekcioglu, Biomed. Pharmacother. 2006, 60, 97–108.
[26] M. L. Dubocovich, M. Markowska, Endocrine 2005, 27, 101–110.
[27] X. Jin, C. von Gall, R. L. Pieschl, V. K. Gribkoff, J. H. Stehle, S. M. Reppert,
D. R. Weaver, Mol. Cell. Biol. 2003, 23, 1054–1060.
[28] C. von Gall, M. L. Garabette, C. A. Kell, S. Frenzel, F. Dehghani, P. M.
Schumm-Draeger, D. R. Weaver, H. W. Korf, M. H. Hastings, J. H. Stehle,
Nat. Neurosci. 2002, 5, 234–238.
tonin receptor agonists, was chosen. Ligand atoms and residues
within a sphere of 8 ꢃ around the ligand were submitted to an
energy minimization procedure, applying the Amber force field, to
a convergence threshold of 0.05 kJmolÀ1 ꢃÀ1. The final MT1 recep-
tor–4a complex was re-introduced in the pre-equilibrated lipid bi-
layer obtained from the MD simulation performed on the MT1–
2PhMLT complex. The resulting system was then submitted to
a 50 ns MD simulation applying the protocol previously described.
[29] K. Baba, N. Pozdeyev, F. Mazzoni, S. Contreras-Alcantara, C. Liu, M. Kasa-
matsu, T. Martinez-Merlos, E. Strettoi, P. M. Iuvone, G. Tosini, Proc. Natl.
Acad. Sci. USA 2009, 106, 15043–15048.
[30] C. Ersahin, M. I. Masana, M. L. Dubocovich, Eur. J. Pharmacol. 2002, 439,
71–172.
[31] M. I. Masana, S. Doolen, C. Ersahin, W. M. Al-Ghoul, S. P. Duckles, M. L.
Dubocovich, D. N. Krause, J. Pharmacol. Exp. Ther. 2002, 302, 1295–
1302.
[32] D. L. Drazen, R. J. Nelson, Neuroendocrinology 2001, 74, 178–184.
[33] M. L. Dubocovich, M. I. Masana, S. Iacob, D. M. Sauri, Naunyn Schmiede-
bergs Arch. Pharmacol. 1997, 355, 365–375.
Acknowledgements
The authors thank Dr. Jeff Wereszczynski (University of California,
San Diego, USA) for providing the GAFF templates for Desmond.
This work was supported by Universities of Parma, Milano and
Urbino “Carlo Bo”.
[34] M. L. Dubocovich, K. Yun, W. M. Al-Ghoul, S. Benloucif, M. I. Masana,
FASEB J. 1998, 12, 1211–1220.
[35] R. Ochoa-Sanchez, S. Comai, B. Lacoste, F. R. Bambico, S. Dominguez-
Lopez, G. Spadoni, S. Rivara, A. Bedini, D. Angeloni, F. Fraschini, M. Mor,
G. Tarzia, L. Descarries, G. Gobbi, J. Neurosci. 2011, 31, 18439–18452.
[36] D. P. Zlotos, Arch. Pharm. 2005, 338, 229–247.
[37] M. Mor, S. Rivara, D. Pala, A. Bedini, G. Spadoni, G. Tarzia, Expert Opin.
Ther. Pat. 2010, 20, 1059–1077.
[38] D. P. Zlotos, Curr. Med. Chem. 2012, 19, 3532–3549.
[39] C. Descamps-Francois, S. Yous, P. Chavatte, V. Audinot, A. Bonnaud, J. A.
Boutin, P. Delagrange, C. Bennejean, P. Renard, D. J. Lesieur, J. Med.
Chem. 2003, 46, 1127–1129.
[40] C. Mꢆsangeau, B. Pꢆrꢅs, C. Descamps-Francois, P. Chavatte, V. Audinot,
S. Coumailleau, J. A. Boutin, P. Delagrange, C. Bennejean, P. Renard,
D.-H. Caignard, P. Berthelot, S. Yous, Bioorg. Med. Chem. 2010, 18, 3426–
3436.
[41] B. Di Giacomo, A. Bedini, G. Spadoni, G. Tarzia, F. Fraschini, M. Pannacci,
V. Lucini, Bioorg. Med. Chem. 2007, 15, 4643–4650.
[42] C. Larraya, J. Guillard, P. Renard, V. Audinot, J. A. Boutin, P. Delagrange,
C. Bennejean, M.-C. Viaud-Massuard, Eur. J. Med. Chem. 2004, 39, 515–
526.
[43] G. Spadoni, A. Bedini, P. Orlando, S. Lucarini, G. Tarzia, M. Mor, S. Rivara,
V. Lucini, M. Pannacci, F. Scaglione, Bioorg. Med. Chem. 2011, 19, 4910–
4916.
Keywords: melatonin · molecular modeling · MT1 selectivity ·
structure–activity relationships
[1] J. Barrenetxe, P. Delagrange, J. A. Martinez, J. Physiol. Biochem. 2004, 60,
61–72.
[2] S. R. Pandi-Perumal, I. Trakht, V. Srinivasan, D. W. Spence, G. J. Maestroni,
N. Zisapel, D. P. Cardinali, Prog. Neurobiol. 2008, 85, 335–353.
[3] J. Arendt, J. Biol. Rhythms 2005, 20, 291–303.
[4] V. Srinivasan, M. Smits, W. Spence, A. D. Lowe, L. Kayumov, S. R. Pandi-
Perumal, B. Parry, D. P. Cardinali, World J. Biol. Psychiatry 2006, 7, 138–
151.
[5] M. D. Maldonado, R. J. Reiter, M. A. Perez-San-Gregorio, Hum. Psycho-
pharmacol. 2009, 24, 391–400.
[6] G. J. Maestroni, Expert Opin. Invest. Drugs 2001, 10, 467–476.
[7] F. Simko, O. Pechanova, J. Pineal Res. 2009, 47, 127–133.
[8] M. Ambriz-Tututi, H. I. Rocha-Gonzꢄlez, S. L. Cruz, V. Granados-Soto, Life
Sci. 2009, 84, 489–498.
[9] J. C. Mayo, R. M. Sainz, D. X. Tan, R. Hardeland, J. Leon, C. Rodriguez,
R. J. Reiter, J. Neuroimmunol. 2005, 165, 139–149.
[10] R. Hardeland, Endocrine 2005, 27, 119–130.
[11] R. Hardeland, Molecules 2009, 14, 5054–5102.
[12] D. E. Blask, L. A. Sauer, R. T. Dauchy, Curr. Top. Med. Chem. 2002, 2, 113–
132.
[44] a) L. Q. Sun, J. Chen, M. Bruce, J. A. Deskus, J. R. Epperson, K. Takaki, G.
Johnson, L. Iben, C. D. Mahle, E. Ryan, C. Xu, Bioorg. Med. Chem. Lett.
2004, 14, 3799–3802; b) L. Q. Sun, J. Chen, K. Takaki, G. Johnson, L.
Iben, C. D. Mahle, E. Ryan, C. Xu, Bioorg. Med. Chem. Lett. 2004, 14,
1197–1200; c) L. Q. Sun, K. Takaki, J. Chen, L. Iben, J. O. Knipe, L. Pajor,
C. D. Mahle, E. Ryan, C. Xu, Bioorg. Med. Chem. Lett. 2004, 14, 5157–
5160.
[45] V. Audinot, F. Mailliet, C. Lahaye-Brasseur, A. Bonnaud, A. Le Gall, C.
Amossꢆ, S. Dromaint, M. Rodriguez, N. Nagel, J. P. Galizzi, B. Malpaux, G.
Guillaumet, D. Lesieur, F. Lefoulon, P. Renard, P. Delagrange, J. A. Boutin,
Naunyn Schmiedebergs Arch. Pharmacol. 2003, 367, 553–561.
[46] C. Markl, W. P. Clafshenkel, M. I. Attia, S. Sethi, P. A. Witt-Enderby, D. P.
Zlotos, Arch. Pharm. 2011, 344, 666–674.
[47] F. Lezoualc’h, R. Jockers, I. Berque-Bestel, Curr. Pharm. Des. 2009, 15,
719–729.
[48] S. Rivara, A. Lodola, M. Mor, A. Bedini, G. Spadoni, V. Lucini, M. Pannacci,
F. Fraschini, F. Scaglione, R. O. Sanchez, G. Gobbi, G. Tarzia, J. Med.
Chem. 2007, 50, 6618–6626.
[49] S. Rivara, F. Vacondio, A. Fioni, C. Silva, C. Carmi, M. Mor, V. Lucini, M.
Pannacci, A. Caronno, F. Scaglione, G. Gobbi, G. Spadoni, A. Bedini, P.
Orlando, S. Lucarini, G. Tarzia, ChemMedChem 2009, 4, 1746–1755.
[50] S. Rivara, S. Lorenzi, M. Mor, P. V. Plazzi, G. Spadoni, A. Bedini, G. Tarzia,
J. Med. Chem. 2005, 48, 4049–4060.
[13] E. Mills, P. Wu, D. Seely, G. Guyatt, J. Pineal Res. 2005, 39, 360–366.
[14] N. Bouatia-Naji, A. Bonnefond, C. Cavalcanti-ProenÅa, T. Sparsø, J.
Holmkvist, M. Marchand, J. Delplanque, S. Lobbens, G. Rocheleau, E.
Durand, F. De Graeve, J. C. Chꢅvre, K. Borch-Johnsen, A. L. Hartikainen,
A. Ruokonen, J. Tichet, M. Marre, J. Weill, B. Heude, M. Tauber et al., Nat.
Genet. 2009, 41, 89–94.
[15] V. Lyssenko, C. L. Nagorny, M. R. Erdos, N. Wierup, A. Jonsson, P. Spꢆgel,
M. Bugliani, R. Saxena, M. Fex, N. Pulizzi, B. Isomaa, T. Tuomi, P. Nilsson,
J. Kuusisto, J. Tuomilehto, B. Boehnke, D. Altshuler, F. Sundler, J. G. Eriks-
son, A. U. Jackson et al., Nat. Genet. 2009, 41, 82–88.
[16] A. G. Wade, I. Ford, G. Crawford, A. D. McMahon, T. Nir, M. Laudon, N. Zi-
sapel, Curr. Med. Res. Opin. 2007, 23, 2597–2605.
[17] M. Miyamoto, CNS Neurosci. Ther. 2009, 15, 32–51.
[18] C. de Bodinat, B. Guardiola-Lemaitre, E. Mocaꢇr, P. Renard, C. MuÇoz,
M. J. Millan, Nat. Rev. Drug Discovery 2010, 9, 628–642.
[19] D. P. Cardinali, V. Srinivasan, A. Brzezinski, G. M. Brown, J. Pineal Res.
2012, 52, 365–375.
[20] G. Spadoni, A. Bedini, S. Rivara, M. Mor, CNS Neurosci. Ther. 2011, 17,
733–741.
[21] M. L. Dubocovich, P. Delagrange, D. N. Krause, D. Sugden, D. P. Cardinali,
J. Olcese, Pharmacol. Rev. 2010, 62, 343–380.
[22] R. Jockers, P. Maurice, J. A. Boutin, P. Delagrange, Br. J. Pharmacol. 2008,
154, 1182–1195.
[51] D. Sugden, H. Pickering, T. Mui-Teck, P. J. Garratt, Biol. Cell 1997, 89,
531–537.
[23] S. M. Reppart, D. R. Weaver, C. Godson, Trends Pharmacol. Sci. 1996, 17,
100–102.
&10
&
www.chemmedchem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 12
ÝÝ
These are not the final page numbers!

MT1-Selective Melatonin Receptor Ligands
MED
[52] R. Nonno, V. Lucini, G. Spadoni, M. Pannacci, A. Croce, D. Esposti, C. Bal-
samini, G. Tarzia, F. Fraschini, B. M. Stankov, J. Pineal Res. 2000, 29, 234–
240.
[75] The UniProt Consortium, Nucleic Acids Res. 2011, 40, D71–D75.
[76] J. D. Thompson, D. G. Higgins, T. J. Gibson, Nucleic Acids Res. 1994, 22,
4673–4680.
[53] C. Yung-Chi, W. H. Prusoff, Biochem. Pharmacol. 1973, 22, 3099–3108.
[54] S. G. F. Rasmussen, B. T. DeVree, Y. Zou, A. C. Kruse, K. Y. Chung, T. S. Ko-
bilka, F. S. Thian, P. S. Chae, E. Pardon, D. Calinski, J. M. Mathiesen,
S. T. A. Shah, J. A. Lyons, M. Caffrey, S. H. Gellman, J. Steyaert, G. Skinio-
tis, W. I. Weis, R. K. Sunahara, B. K. Kobilka, Nature 2011, 477, 549–555.
[55] a) Schrçdinger Suite 2011 Induced Fit Docking protocol; b) Glide ver-
sion 5.7, Schrçdinger, LLC, New York, NY, 2011; c) Prime version 3.0,
Schrçdinger, LLC, New York, NY, 2011.
[77] T. Mirzadegan, G. Benko, S. Filipek, K. Palczewski, Biochemistry 2003, 42,
4310–4310.
[78] A. Sali, T. L. Blundell, J. Mol. Biol. 1993, 234, 779–815.
[79] R. A. Laskowski, M. W. MacArthur, D. S. Moss, J. M. Thornton, J. Appl.
Crystallogr. 1993, 26, 283–291.
[80] Maestro, version 9.0, Schrçdinger, LLC, New York, NY, 2009.
[81] a) Schrçdinger Suite 2009 Protein Preparation Wizard; b) Epik version
2.0, Schrçdinger LLC, New York, NY, 2009; c) Impact version 5.5, Schrç-
dinger LLC, New York, NY, 2009; d) Prime version 2.1, Schrçdinger, LLC,
New York, NY, 2009.
[56] G. Spadoni, B. Stankov, A. Duranti, G. Biella, V. Lucini, A. Salvatori, F. Fra-
schini, J. Med. Chem. 1993, 36, 4069–4074.
[57] S. Rivara, G. Diamantini, B. Di Giacomo, D. Lamba, G. Gatti, V. Lucini, M.
Pannacci, M. Mor, G. Spadoni, G. Tarzia, Bioorg. Med. Chem. 2006, 14,
3383–3391.
[58] J. A. Ballesteros, H. Weinstein in Methods in Neurosciences, (Eds.: S. C.
Sealfon, P. M. Conn), Elsevier, 1995, pp. 366–428.
[59] S. Conway, S. J. Canning, P. Barrett, B. Guardiola-Lemaitre, P. Delagrange,
P. J. Morgan, Biochem. Biophys. Res. Commun. 1997, 239, 418–423.
[60] T. Kokkola, M. A. Watson, J. White, S. Dowell, S. M. Foord, J. T. Laitinen,
Biochem. Biophys. Res. Commun. 1998, 249, 531–536.
[61] M. J. Gerdin, F. Mseeh, M. L. Dubocovich, Biochem. Pharmacol. 2003, 66,
315–320.
[62] P. Mazna, V. Obsilova, I. Jelinkova, A. Balik, K. Berka, Z. Sovova, R. Ettrich,
P. Svoboda, T. Obsil, J. Teisinger, J. Neurochem. 2004, 91, 836–842.
[63] V.-P. Jaakola, M. T. Griffith, M. A. Hanson, V. Cherezov, E. Y. T. Chien, J. R.
Lane, A. P. IJzerman, R. C. Stevens, Science 2008, 322, 1211–1217.
[64] V.-P. Jaakola, J. R. Lane, J. Y. Lin, V. Katritch, A. P. IJzerman, R. C. Stevens,
J. Biol. Chem. 2010, 285, 13032–13044.
[65] M. Conner, S. R. Hawtin, J. Simms, D. Wootten, Z. Lawson, A. C. Conner,
R. A. Parslow, M. Wheatley, J. Biol. Chem. 2007, 282, 17405–17412.
[66] A. Cavalli, F. Fanelli, C. Taddei, P. G. De Benedetti, S. Cotecchia, FEBS Lett.
1996, 399, 9–13.
[82] W. L. Jorgensen, D. S. Maxwell, J. Tirado-Rives, J. Am. Chem. Soc. 1996,
118, 11225–11236.
[83] G. A. Kaminski, R. A. Friesner, J. Tirado-Rives, W. L. Jorgensen, J. Phys.
Chem. B 2001, 105, 6474–6487.
[84] S. J. Weiner, P. A. Kollman, D. A. Case, U. C. Singh, C. Ghio, G. Alagona, S.
Profeta, P. Weiner, J. Am. Chem. Soc. 1984, 106, 765–784.
[85] MacroModel, version 9.7, Schrçdinger, LLC, New York, NY, 2009.
[86] a) Desmond Molecular Dynamics System, version 2.2, D. E. Shaw Re-
search, New York, NY, 2009; b) Maestro-Desmond Interoperability Tools,
version 2.2, Schrçdinger, New York, NY, 2009.
[87] E. C. K. J. Bowers, H. Xu, R. O. Dror, M. P. Eastwood, B. A. Gregersen, J. L.
Klepeis, I. Kolossvary, M. A. Moraes, F. D. Scaerdoti, J. K. Salmon, Y. Shan,
D. E. Shaw, Proceedings of the 2006 ACM/IEEE Conference on Super-
computing (SC06), 2006.
[88] M. A. Lomize, A. L. Lomize, I. D. Pogozheva, H. I. Mosberg, Bioinformatics
2006, 22, 623–625.
[89] V. Hornak, R. Abel, A. Okur, B. Strockbine, A. Roitberg, C. Simmerling,
Proteins Struct. Funct. Bioinf. 2006, 65, 712–725.
[90] J. Wang, R. M. Wolf, J. W. Caldwell, P. A. Kollman, D. A. Case, J. Comput.
Chem. 2004, 25, 1157–1174.
[91] J. Wang, W. Wang, P. A. Kollman, D. A. Case, J. Mol. Graphics Modell.
2006, 25, 247–260.
[67] S. Rivara, M. Mor, A. Bedini, G. Spadoni, G. Tarzia, Curr. Top. Med. Chem.
2008, 8, 954–968.
[68] G. J. Ellames, J. M. Herbert, J. Labelled Compd. Radiopharm. 2010, 53,
801–803.
[69] R. Nonno, V. Lucini, M. Pannacci, C. Mazzucchelli, D. Angeloni, F. Fraschi-
ni, B. M. Stankov, Br. J. Pharmacol. 1998, 124, 485–492.
[70] R. Nonno, M. Pannacci, V. Lucini, D. Angeloni, F. Fraschini, B. M. Stankov,
Br. J. Pharmacol. 1999, 127, 1288–1294.
[92] S. E. Feller, Y. Zhang, R. W. Pastor, B. R. Brooks, J. Chem. Phys. 1995, 103,
4613–4621.
[93] R. A. Lippert, K. J. Bowers, R. O. Dror, M. P. Eastwood, B. A. Gregersen,
J. L. Klepeis, I. Kolossvary, D. E. Shaw, J. Chem. Phys. 2007, 126, 046101.
[94] U. Essmann, L. Perera, M. L. Berkowitz, T. Darden, H. Lee, L. G. Pedersen,
J. Chem. Phys. 1995, 103, 8577–8593.
[95] M. Tuckerman, B. J. Berne, G. J. Martyna, J. Chem. Phys. 1992, 97, 1990–
2001.
[71] M. M. Bradford, Anal. Biochem. 1976, 72, 248–254.
[72] G. Spadoni, C. Balsamini, A. Bedini, G. Diamantini, B. Di Giacomo, A. Ton-
tini, G. Tarzia, M. Mor, P. V. Plazzi, S. Rivara, R. Nonno, M. Pannacci, V.
Lucini, F. Fraschini, B. M. Stankov, J. Med. Chem. 1998, 41, 3624–3634.
[73] X. Deupi, J. Standfuss, Curr. Opin. Struct. Biol. 2011, 21, 541–551.
[74] G. Lebon, T. Warne, C. G. Tate, Curr. Opin. Struct. Biol. 2012, DOI:
10.1016/j.sbi.2012.03.007.
Received: June 19, 2012
Revised: July 27, 2012
Published online on && &&, 0000
ChemMedChem 0000, 00, 1 – 12
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
&
11
&
ÞÞ
These are not the final page numbers!

FULL PAPERS
S. Rivara, D. Pala, A. Lodola, M. Mor,
V. Lucini, S. Dugnani, F. Scaglione,
A. Bedini, S. Lucarini, G. Tarzia,
G. Spadoni*
Focusing on selectivity: A new series
of MT1-selective agonists was synthe-
sized by modulating the versatile N-ani-
linoethylamide scaffold through intro-
duction of lipophilic (aryl)alkyl substitu-
ents on the ether oxygen atom. A com-
bination of molecular modeling studies
and ligand-based information provided
hypotheses for ligand–receptor inter-
actions and for MT1 subtype selectivity.
&& – &&
MT1-Selective Melatonin Receptor
Ligands: Synthesis, Pharmacological
Evaluation, and Molecular Dynamics
Investigation of N-{[(3-O-
Substituted)anilino]alkyl}amides
&12
&
www.chemmedchem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 12
ÝÝ
These are not the final page numbers!