Development of substituted N-[3-(3-methoxylphenyl)propyl] amides as MT 2-selective melatonin agonists: Improving metabolic stability

Article abstract of DOI:10.1016/j.bmc.2012.10.060

A series of novel and selective N-[3-(6-benzyloxy-3-methoxyphenyl)propyl] amides has recently been shown to possess sub-nanomolar range binding affinity to the type 2 melatonin receptor (MT₂). Pharmacokinetics studies suggested that these compounds were subject to vigorous CYP450-mediated metabolism, resulting in a series of metabolites with significantly decreased or diminished binding affinities toward MT₂ receptor. The ether bonds were found to be the major positions susceptible to metabolism. In this study, the benzyl ether bond was either removed or replaced with a carbon-carbon bond in an attempt to improve metabolic stability and enhance their resistance towards phase I oxidation. The synthesis, receptor binding affinity, intrinsic potency and metabolic stability of modified structures are reported in this article. By removal or replacement of metabolic labile ether linkerage with carbon linkers, a novel compound was identified with good potency and MT ₂ selectivity, and with increased metabolic stability.

Full text of DOI:10.1016/j.bmc.2012.10.060

Bioorganic & Medicinal Chemistry 21 (2013) 547–552
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Development of substituted N-[3-(3-methoxylphenyl)propyl] amides as
MT2-selective melatonin agonists: Improving metabolic stability
Yueqing Hu a, Jing Zhu a, King H. Chan a, Yung H. Wong a,b,
⇑
a Division of Life Science and the Biotechnology Research Institute, Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong
b State Key Laboratory of Molecular Neuroscience, and the Molecular Neuroscience Center, Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of novel and selective N-[3-(6-benzyloxy-3-methoxyphenyl)propyl] amides has recently been
shown to possess sub-nanomolar range binding affinity to the type 2 melatonin receptor (MT2). Pharma-
cokinetics studies suggested that these compounds were subject to vigorous CYP450-mediated metabo-
lism, resulting in a series of metabolites with significantly decreased or diminished binding affinities
toward MT2 receptor. The ether bonds were found to be the major positions susceptible to metabolism.
In this study, the benzyl ether bond was either removed or replaced with a carbon–carbon bond in an
attempt to improve metabolic stability and enhance their resistance towards phase I oxidation. The synthe-
sis, receptor binding affinity, intrinsic potency and metabolic stability of modified structures are reported
in this article. By removal or replacement of metabolic labile ether linkerage with carbon linkers, a novel
compound was identified with good potency and MT2 selectivity, and with increased metabolic stability.
Ó 2012 Elsevier Ltd. All rights reserved.
Received 9 August 2012
Revised 26 October 2012
Accepted 30 October 2012
Available online 20 November 2012
Keywords:
Melatonin
Agonist
Ligand
Phenylpropyl amide
1. Introduction
C6-debenzylation and C3-demethylation) and hydroxylation of
the phenyl ring.11
Melatonin is a multi-functional neurohormone primarily se-
creted by the pineal gland during the period of darkness. It regu-
lates the circadian rhythm and has been used to treat diseases
associated with the desynchronization of biological rhythms, such
as jet-lag, disturbed sleep–wake cycles and seasonal disorders.1,2
Besides its primary role of regulating circadian rhythm, melatonin
is involved in a number of additional physiological effects and has
a variety of therapeutic potentials for the treatment of depression,
cancer and neurodegenerative pathologies.3,4 Most of the biologi-
cal functions of melatonin are mediated through activation of
two subtypes of G protein-coupled receptors, MT1 and MT2, which
are widely expressed in different tissues.5 The exact role of MT1
and MT2 are not clearly defined due to the lack of receptor
subtype-selective ligands in clinical uses. Developing potent and
subtype-selective melatonin ligands have attracted huge interests
from medicinal chemists.6–9 We have previously characterized
In order to increase metabolic stability so as to ultimately im-
prove pharmacokinetic properties of this series of compounds,
additional N-phenylpropyl amides were designed and synthesized.
As the N-[3-(3-methoxylphenyl)propyl] amide was identified as an
optimal pharmacophore for binding towards melatonin receptor
and the hydrophobic phenyl substituent on C6 position of the
3-methoxylphenyl scaffold was proven to be a determinant of
subtype-selectivity, they were kept intact in the search for meta-
bolically more stable compounds. As shown in Figure 1, the posi-
tion for further structure modification was focused on the linker
connecting the two phenyl rings. The oxygen–carbon ether bond
of lead compound 1 was either removed or replaced with a car-
bon–carbon bond in an effort to block the degradation at this site.
2. Results and discussion
2.1. Chemistry
a
series of novel N-[3-(6-benzyloxy-3-methoxyphenyl)propyl]
amides exhibiting sub-nanomolar range binding affinity to MT2
selectively.10 Pharmacokinetic studies revealed that this series of
compounds generally exhibit less desirable pharmacokinetic prop-
erties due to high metabolic clearance, which may jeopardize their
in vivo pharmacological potency. Several sites susceptible to the
attack of phase I enzymes were identified. The major metabolic
pathways included carbon–oxygen ether bond cleavage (such as
The syntheses toward the novel N-phenylpropyl amides (7, 8, 9,
and 10) are shown in Scheme 1. All of the compounds have a 3-
methoxyphenyl group attached at C6 position on the scaffold 3-
methoxylphenyl ring. Between the two phenyl rings, compound
7 has a carbon–carbon triple bond, compound 8 has an ethylene
chain, compound 9 has one methylene carbon chain, and com-
pound 10 has no chain. All of these compounds were prepared
from a common triflite intermediate 6, through palladium medi-
ated carbon–carbon bond forming coupling reactions. The triflite
⇑
Corresponding author.
E-mail address: boyung@ust.hk (Y.H. Wong).
0968-0896/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2012.10.060

548
Y. Hu et al. / Bioorg. Med. Chem. 21 (2013) 547–552
Figure 1. Design of metabolically more stable compounds.
Scheme 1. Modification of the linker between the two phenyl rings. Reagents and conditions: (a) (1) BnBr, NaH, THF; (2) (EtO)2PO(CH2CN), NaH, THF, (b) LiAlH4, EtO2, (c)
EtCOCl, Et3N, (d) Pd/C, H2, MeOH, (e) Tf2O, Et3N, (f) m-MeOPhC„CH, Pd(PPh3)4, Cul, (g) m-MeOPhCH2ZnBr, Pd(PPh3)4, DMF, (h) m-MeOPhB(OH)2, Pd(PPh3)4, Na2CO3, toluene.
6 was prepared from commercially available starting materials in
six reaction steps. Starting with 2-hydroxy-5-methoxybenzalde-
hyde, alkylation with benzyl bromide under potassium carbonate
in DMF, followed by Horner–Emmons olefination with diethylcy-
anomethylphosphate under sodium hydride in THF provided the
2.2. Binding assay
Competitive binding characteristics of the compounds towards
human MT1 and MT2 melatonin receptors stably expressed in Chi-
nese hamster ovary (CHO) cells were determined by whole cell
binding assays using 1 nM [3H]melatonin as the probe. The Kd of
melatonin for MT1 and MT2 receptors was 0.296 and 0.429 nM,
respectively, as determined by saturation binding assays. The Ki
values of modified compound 7–10 for MT1 and MT2 as well as
their MT1/MT2 selectivity ratio are reported in Table 1. For compar-
ison, the binding data of the original lead compound 1 with an oxy-
gen–carbon linker was also listed. Compared with compound 1,
compound 7 with a carbon–carbon triple bond displayed even bet-
ter affinity (Ki 0.073 nM) and selectivity (MT1/MT2 527-folds) to-
ward MT2. Replacement of the oxygen–carbon linker with an
ethylene linker, compound 8, displayed slightly reduced binding
affinity toward MT2 (Ki 0.55 nM) but increased binding affinity to-
ward MT1 (Ki 22.2 nM), thus decreasing selectivity (40-fold) to-
ward MT2. With a shorter methylene linker, compound 9 showed
further reduction in binding affinity toward MT2 (Ki 3.39 nM).
a-b unsaturated nitriles 2 as a mixture of cis- and trans-isomers.
Reduction of the unsaturated nitriles 2 with lithium aluminum hy-
dride in refluxing diethyl ether generated the corresponding satu-
rated amine 3, which was reacted with propionyl chloride to give
the amide 4. Removing of the benzyl group of 4 by hydrogenation
provided free phenol 5, which was treated with trifluoroacetic
anhydride to generate common intermediate triflite 6. Coupling
of triflite 6 with m-methoxyphenyl–acetylene provided alkyne 7.
Reduction of the alkyne triple bond of 7 by hydrogenation yielded
compound 8 with an ethylene chain connecting the two phenyl
rings. Coupling of triflite 6 with 3-methoxybenzyl zinc bromide
in DMF gave compound 9 with a methylene chain connecting the
two phenyl rings. Reaction of triflite 6 with 3-methoxyphenyl
boronic acid bound the two phenyl rings directly to give biphenyl
compound 10.

Y. Hu et al. / Bioorg. Med. Chem. 21 (2013) 547–552
549
Table 1
Table 2
Binding affinity of compound 1 and 7–10 towards human MT1 and MT2 receptors
expressed in CHO cells
FLIPR dose–responses of compound
receptors expressed in CHO cells
1
and 7–10 towards human MT1 and MT2
O
O
HN
HN
O
O
L
O
L
O
a
–L–
Kia (nM)
MT1/MT2
Compound
–L–
EC50 (nM)
MT2
MT1/MT2
MT1
MT2
MT1
9.03
8.38
16.3
41.6
2.94
Melatonin
1
7
8
9
10
0.296
121
38.2
0.429
0.291
0.073
0.55
3.39
0.339
0.69
417
527
40
12
96
1
7
8
9
–OCH2–
–C„C–
–CH2CH2–
–CH2–
—
0.05
0.11
0.024
0.24
0.20
181
77
679
171
14
–OCH2–
–C„C–
–CH2CH2–
–CH2–
—
22.2
39.5
32.7
10
a
Test compound potency was expressed as EC50 (nM), while the ligand selec-
a
Test compound binding affinity was expressed as Ki (nM), while the ligand
tivity towards the two receptor subtypes was expressed as the MT1/MT2 EC50 ratio.
Data reported in the table were means of three or more experiments run at seven
different concentrations in triplicates.
selectivity towards the two receptor subtypes was expressed as the MT1/MT2 Ki
ratio. Data reported in the table were means of three trials done in duplicates. The
corresponding Ki values were calculated using the mean pIC50 values.
lower intrinsic efficacy of compound 7 is thus manifested as a
Interestingly, by removing the linker completely, compound 10 re-
gained the binding affinity toward MT2 (Ki 0.339 nM), providing
higher MT2 selectivity (96-fold). Overall, it is noteworthy that all
of the novel compounds retained good potency toward MT2 with
single or sub-digit nanomolar binding affinities and compound 7
with a carbon–carbon triple bond linker displayed the highest po-
tency and selectivity toward MT2.
higher EC50 value in FLIPR assays.
2.4. Metabolic stability
The metabolic stability of compound 7–10 in rat and human li-
ver microsomes was examined under the same experimental con-
ditions as reported previously.11 The in vitro half-life (t1/2) was
calculated according to the formula t1/2 = ꢁ0.693/k, where k is
the slope of the linear regression of the natural logarithm of the
parent remaining percentage versus incubation time. In vitro
intrinsic clearance (Clint) was calculated from the t1/2 value as re-
ported previously and was used to predict the hepatic clearance.
Compared with O-linkage structure 1, compound 7 exhibited con-
siderably improved stability with an in vitro half-life (t1/2) of 21.7
and 18.3 min in rat and human microsomes, respectively. The
in vitro intrinsic clearance (Clint) was 114.7 ml/min/kg in rat and
102.2 ml/min/kg in human for compound 7. On the contrary, com-
pounds 8, 9 and 10 showed no improvement in metabolic stability
(Table 3). To investigate the reactions triggering the degradation of
compounds 7–10, attempts were made to identify the metabolites.
Based on LC-UV/MS analysis, the principal biotransformation path-
ways of 7–10 were proposed to be demethylations of methoxyl
moieties, hydroxylations on aromatic rings and combined reac-
tions of demethylations and hydroxylations. No metabolite of
carbon–carbon bond linker cleavage was found in metabolism
medium, implying removal of carbon–oxygen ether linkage in ori-
ginal structure 1 successfully blocked the degradation in this site.
The major metabolites of compounds 7–10 were identified to
be C3- or C30-demethylated products and hydroxylated derivatives
of aromatic rings. Previously SAR studies indicated that the key
structural pharmacophores for both affinity and efficacy of melato-
nin receptor ligands are the presence and the relative spatial
position of C3-methoxy group and N-alkylamido side chain.7 Con-
sequently, demethylation on C3 site would compromise the affin-
ity of parent compound toward MT2 receptor and generate
metabolites with considerably reduced activity. In contrast, C30-
methoxy group plays a less important role for receptor binding.
Compound without a substituent at C30 also exhibited subnanom-
olar affinity to the MT2 receptor.10 Thus, metabolites derived from
C30-demethylation are expected to retain pharmacological activity.
Hydroxylation on the aromatic ring is another principal metabolic
2.3. Functional assay
The intrinsic potency of these compounds was evaluated using
Ca2+-based FLIPR assays and the results are shown in Table 2. Two
clonal cell lines expressing either hMT1 (CHO–hMT1) or hMT2
(CHO–hMT2) receptors with chimeric G
a protein 16z25 are effi-
ciently coupled to the phospholipase Cb/IP3/Ca2+ pathway and trig-
gered intracellular Ca2+ mobilization upon receptor activation.12
The prototypic agonist, melatonin, activated both MT1 and MT2
in a non-selective manner with EC50 values of 0.16 0.07 and
0.60 0.10 nM, respectively. All compounds were tested with the
assay system described above and appeared to be potent agonists
toward MT2 with MT1/MT2 selectivity ratio ranging from 14- to
679-folds. Comparing to compound 1 with an oxygen–carbon lin-
ker (EC50 0.05 nM, selectivity 181), compound 7 with a carbon–
carbon triple bond displayed similar potency toward MT1 (EC50
8.38 nM) and slightly reduced potency and selectivity toward
MT2 (EC50 0.11 nM, selectivity 77-fold); compound 8 with an eth-
ylene linker displayed better potency and selectivity toward MT2
(EC50 0.024 nM, selectivity 679-fold). Compound 9 with a methy-
lene linker and compound 10 without a linker displayed similar
MT2 potency (EC50 ꢀ0.2 nM), with compound 9 having higher
MT2 selectivity (171-fold). The most potent and selective com-
pound toward MT2 in this assay is compound 8 with an ethylene
linker. Although compound 7 has a higher binding affinity than
compound 8, it was less efficacious in FLIPR assays. This discrep-
ancy is probably due to differences in the compounds’ intrinsic
ability to induce conformational changes of the bound receptor.
Despite their structural similarity, differences in the linker be-
tween the two phenyl rings in compounds 7 and 8 may result in
distinct states of receptor conformation. The highly rigid alkyne-
linkered diphenyl structure in compound
7 may limit the
compound’s molecular flexibility, leading to reduced efficacy. The

550
Y. Hu et al. / Bioorg. Med. Chem. 21 (2013) 547–552
dehyde (4.79 g, 100%) was obtained as pale yellow solid. 1H NMR
(400 MHz, CDCl3): d 10.5 (s, 1H), 7.35–7.42 (m, 5H), 7.33 (d,
J = 3.2 Hz, 1H), 7.10 (dd, J = 8.8 Hz, 3.2 Hz, 1H), 6.99 (d, J = 8.8 Hz,
1H), 5.13 (s, 2H), 3.78 (s, 3H). 13C NMR (100 MHz, CDCl3): d
189.2, 155.6, 153.6, 136.1, 128.5 (2), 128.1, 127.2 (2), 125.3,
123.3, 114.9, 110.1, 71.2, 55.7.
Table 3
Rat and human microsomal metabolic stability of compound 1 and 7–10
RLMa HLMb
t1/2c (min) Clintd (ml/min/kg) t1/2c (min) Clintd (ml/min/kg)
Compound
Melatonin 75.3
1
7
8
9
10
33.1
262.3
114.7
300.6
290.1
332.6
73.2
5.5
18.3
5.1
6.0
5.3
25.5
339.8
102.2
366.7
311.7
352.8
9.5
21.7
8.3
8.6
7.5
To a dried flask charged with sodium hydride (60% in mineral
oil, 594 mg) in 60 mL THF at 0 °C under nitrogen, was added
diethyl cyanomethyl phosphonate (2.1 mL, 13.62 mmol) dropwise.
After warmed up to room temperature for 30 min, 2-benzyloxy-5-
methoxy-benzaldehyde (3.0 g, 12.38 mmol) in 5 mL THF was
added via cannulation. The reaction was stirred at room tempera-
ture for 1 h and quenched with water. Aqueous workup and puri-
fication by column chromatography on silica gel provided the
desired products (3.14 g, 95% yield) as mixtures of cis- and trans
isomers. ESI-MS: 266.14 (M+1) cis-isomer: 1H NMR (400 MHz,
CDCl3): d 7.74 (d, J = 2.4 Hz, 1H), 7.61 (d, J = 12.4 Hz, 1H), 7.33–
7.39 (m, 5H), 6.90–6.96 (m, 2H), 5.39 (d, J = 12.0 Hz, 1H), 5.04 (s,
2H), 3.82 (s, 3H). 13C NMR (100 MHz, CDCl3): d 153.5, 151.0,
143.2, 136.4, 128.5 (2), 128.0, 127.3 (2), 123.5, 118.8, 117.6,
114.0, 111.9, 94.7, 71.3, 55.8. trans-Isomer: 1H NMR (400 MHz,
CDCl3): d 7.60 (d, J = 16.8 Hz, 1H), 7.37 (m, 5H), 6.89 (m, 3H),
5.96 (d, J = 16.8 Hz, 1H), 5.03 (s,1H), 3.74 (s, 3H).
a
b
c
RLM, rat liver microsomes.
HLM, human liver microsomes.
t1/2, calculated in vitro elimination half-life.
Clint, in vitro intrinsic clearance.
d
pathway. The metabolites that retained structural characteristics
necessary for binding to the MT2 receptor might have a capacity
to stimulate the MT2 receptor.
3. Conclusion
In this study, we have described the successful synthesis and
evaluations of novel N-[3-(3-methoxylphenyl)propyl] amides
derivatives by modification of the metabolic liable benzyl ether
linkage. Radioligand binding assay demonstrated that all of the
new derivatives possessed high affinity toward MT2 and moderate
affinity toward MT1. A carbon–carbon triple bond was found to be
the best linkage in terms of binding affinity and selectivity toward
MT2. Calcium mobilization functional assay revealed that all of
derivatives were potent agonists toward MT2 with sub-digit nano-
molar EC50 values. Compound 8 with an ethylene linker behaved
as the most potent and selective agonist. Metabolic stability assays
demonstrated that the initial benzyl ether cleavage was successfully
blocked, and the major pathways involved combination of deme-
thylations and hydroxylations of the phenyl rings for the new deriv-
atives. Among them, compound 7 with a triple bond linker was
found to have improved half-life compared to the original lead com-
pound with an ether linker. These findings suggest that the modifi-
cation of the ether linkage was successful in terms of retaining good
potency and selectivity while improving metabolic stability.
Although neither compound 7 or 8 is an ideal drug candidate yet,
both compounds can serve as new lead compounds for further
structure modifications. Strategies involving protection against
demethylation and hydroxylation of the phenyl rings will be em-
ployed, and the results will be described in subsequent publications.
4.1.2. 3-(2-Benzyloxy-5-methoxy-phenyl)-propylamine (3)
To a solution of 3-(2-benzyloxy-5-methoxy-phenyl)-acrylonitrile
(3.87 g, 15.57 mmol) in 140 mL diethyl ether under nitrogen, was
added lithium borohydride (1.66 g, 43.73 mmol). The resulted mix-
ture was heated to reflux overnight. The reaction was cooled to room
temperature and quenched with 2.1 mL water dropwise. Another
3.2 mL aqueous sodium hydroxide (10%) and 5.5 mL water were
added while stirring. After the gray solid turned white, the mixture
was filtered through a plug of sodium sulfate and rinsed with a mix-
ture of methanol/dichloromethane. The filtrate was concentrated
and purified by column chromatography on silica gel to give the de-
sired product (2.58 g, 65%) as pale yellow oil. ESI-MS: 272.08 (M+1).
1H NMR (400 MHz, CDCl3): d 7.31–7.43 (m, 5H), 6.82 (d, J = 8.8 Hz,
1H), 6.75 (d, J = 2.8 Hz, 1H), 6.67 (dd, J = 8.4, 2.8 Hz, 1H), 5.02 (s,
2H), 3.75 (s, 3H), 2.66–2.71 (m, 4H), 1.71–1.79 (m, 4H). 13C NMR
(100 MHz, CDCl3): d 153.5, 150.6, 137.4, 132.1, 128.4 (2), 127.6,
127.0 (2), 116.1, 112.8, 110.8, 70.7, 55.6, 41.8, 34.0, 27.6.
4.1.3. N-[3-(2-Benzyloxy-5-methoxy-phenyl)-propyl]-
propionamide (4)
To a solution of 3-(2-benzyloxy-5-methoxy-phenyl)-propyl-
amine (0.70 g, 2.58 mmol) in 13 mL dichloromethane, was added
triethylamine (0.65 mL, 4.64 mmol) followed by propionyl chloride
(0.34 mL, 3.87 mmol). The reaction was stirred at room tempera-
ture for 1 h and quenched with water. After aqueous workup and
purification by column chromatography on silica gel, eluting with
50% ethyl acetate in petroleum ether, the desired product (0.79 g,
94%) was obtained as a white solid.
4. Experimentals
4.1. Chemistry
All reagents and starting materials were purchased from com-
mercial sources and used without further purifications. The reac-
tion condition and the yield are reported as it was and not
optimized. 1H NMR were recorded using a Varian 300 MHz spec-
trometer. Chemical shifts were measured in parts per million rela-
tive to tetramethylsilane as the internal standard. Coupling
constants were measured in hertz.
Mp: 66–68 °C. HR-TOF-MS: Calcd for
C20H26NO3 (M+1)
328.1913. Found: 328.1915. 1H NMR (400 MHz, CDCl3): d 7.35–
7.43 (m, 5H), 6.87 (d, J = 8.8 Hz, 1H), 6.70 (s, 1H), 6.71 (d,
J = 8.4 Hz, 1H), 5.60 (s, 1H), 5.00 (s, 2H), 3.76 (s, 3H), 3.18 (m,
2H), 2.67 (t, J = 7.0 Hz, 2H), 1.75–1.86 (m, 4H), 1.00 (t, J = 7.6 Hz,
3H). 13C NMR (100 MHz, CDCl3): d 173.5, 153.8, 150.6, 137.0,
131.4, 128.6(2), 128.0, 127.6(2), 116.2, 112.9, 111.3, 71.1, 55.6,
38.2, 30.1, 29.6, 27.2, 10.1.
4.1.1. 3-(2-Benzyloxy-5-methoxy-phenyl)-acrylonitrile (2)
To a flask charged with 2-hydroxy-5-methoxy-benzaldehye
(3.0 g, 19.72 mmol) and benzyl bromide (2.8 mL, 23.66 mmol) in
80 mL THF under nitrogen, was added sodium hydride (60% in
mineral oil, 1.26 g). The reaction was heated to 60 °C overnight.
After aqueous workup and purification by column chromatography
on silica gel, the desired product 2-benzyloxy-5-methoxy-benzal-
4.1.4. N-[3-(2-Hydroxy-5-methoxy-phenyl)-propyl]-
propionamide (5)
To a solution of N-[3-(2-benzyloxy-5-methoxy-phenyl)-propyl]-
propionamide (0.47 g, 1.44 mmol) in 10 mL methanol, was added 5%
palladium on carbon (0.12 g). The resulted mixture was stirred

Y. Hu et al. / Bioorg. Med. Chem. 21 (2013) 547–552
551
under hydrogen atmosphere (balloon pressure) for 1 day. After fil-
tered through a plug of celite, the filtrate was concentrated. Purifica-
tion by column chromatography on silica gel provided the desired
product (0.32 g, 92%) as a white solid. Mp: 83–84 °C. ESI-MS
224.07 (M+1). 1H NMR (400 MHz, CDCl3): d 8.15 (s, 1H), 6.77 (d,
J = 8.8 Hz, 1H), 6.56–6.61 (m, 3H), 3.69 (m, 3H), 3.20 (q, J = 6.2Hz,
2H), 2.61 (t, J = 6.8 Hz, 2H), 2.16 (q, J = 7.6 Hz, 2H), 1.76 (m, 2H),
1.09 (t, J = 7.6 Hz, 3H). 13C NMR (100 MHz, CDCl3): d 174.6, 152.8,
148.3, 128.6, 115.9, 115.6, 111.8, 55.6, 38.6, 29.8 (2), 27.0, 10.0.
in petroleum ether, provided the desired product (12 mg,
0.034 mmol, 91%). ESI-MS: 356.38 (M+1). TOF-HRMS Calcd for
C
(300 MHz, CDCl3): d 7.20 (t, J = 7.7 Hz, 1H), 7.08 (d, J = 9.3Hz, 1H),
6.70–6.78 (m, 5H), 5.46 (br s, 1H), 3.78 (s, 6H), 3.29 (q, J = 6.6 Hz,
2H), 2.82 (s, 4H), 2.58 (t, J = 7.8 Hz, 2H), 2.15 (q, J = 7.6 Hz, 2H),
1.74 (m, 2H), 1.12 (t, J = 7.6 Hz, 3H). 13C NMR (75 MHz, CDCl3): d
173.7, 159.6, 157.9, 143.5, 140.5, 131.5, 130.2, 129.3, 120.8, 114.7,
114.2, 111.4, 111.1, 55.2, 55.1, 39.3, 37.8, 33.7, 30.8, 30.2, 29.7, 9.9.
22H30NO3 (M+H)+ 356.2220. Found: 356.2216. 1H NMR
4.1.5. N-[3-(2-Trifluoromethylsulfonyl-5-methoxy-phenyl)-
propyl]-propionamide (6)
4.1.8. N-{3-[5-Methoxy-2-(3-methoxy-benzyl)-phenyl]-propyl}-
propionamide (9)
To a solution of N-[3-(2-hydroxy-5-methoxy-phenyl)-propyl]-
propionamide (279 mg, 1.18 mmol) in 6 mLdry dichloromethane
under N2 at 0 °C, was added triethylamine (0.43 mL, 3.06 mmol),
followed by trifluoroacetic anhydride (0.26 mL, 1.65 mmol) drop-
wise. The reaction was allowed to warm up to room temperature
and stir for 3 h. Water was added and the mixture was extracted
with dichloromethane (ꢂ3). The combined organic extract was
dried, filtered and concentrated. Purification by column chroma-
tography on silica gel recovered starting material 56 mg (20%)
and provided the desired product (186 mg, 0.50 mmol, 43%). ESI-
MS: 370.14 (M+1).
To a dry flask under N2, was added zinc powder (131 mg,
2.0 mmol), DMF (0.4 mL) and 1,2-dibromoethane (0.014 mL,
0.08 mmol). The mixture was heated to 70 °C for 10 min, and cooled
to room temperature. TMSCl (0.015 mL, 0.06 mmol) was added and
the reaction was stirred for 30 min. After the mixture was cooled to
0 °C, a solution of 3-methoxybenzyl bromide (0.28 mL, 2.0 mmol) in
4 mL DMF was added dropwise in 2 h. After stirring for another 2 h,
the zinc reagent was ready for use. To a flask, was added N-[3-(2-tri-
fluoromethylsulfonyl-5-methoxy-phenyl) -propyl]-propionamide
(36 mg, 0.097 mmol), Pd(PPh3)4 (8 mg, 0.0069 mmol). The flask
was degassed three times and filled with N2. A solution of the
freshly prepared zinc reagent (1.5 mL) was added and the mixture
was heated to 70 °C for 24 h. The reaction was stopped and the sol-
vent was removed under reduced pressure. A solution of 10% HCl
(ꢀ10 mL) was added and the mixture was extracted with dichloro-
methane (ꢂ3). The combined extract was dried, filtered and con-
centrated. The resulting residue was purified by column
chromatography on silica gel, eluting with 10% acetone in petro-
leum ether to give the desired product (14 mg, 0.041 mmol, 42%).
ESI-MS: 342.21 (M+1). TOF-HRMS Calcd for C21H28NO3 (M+H)+
342.2064. Found: 342.2045. 1H NMR (300 MHz, CDCl3): d 7.06 (t,
J = 7.0 Hz, 1H), 6.71 (d, J = 8.8 Hz, 1H), 6.64–6.68 (m, 5H), 5.20 (br
s, 1H), 3.92 (s, 2H), 3.79 (s, 3H), 3.75 (s, 3H), 3.22 (q, J = 6.5 Hz,
2H), 2.56 (t, J = 7.8 Hz, 2H), 2.12 (q, J = 7.5 Hz, 2H), 1.69 (m, 2H),
1.11 (t, J = 7.6 Hz, 3H). 13C NMR (75 MHz, CDCl3): d 173.7, 159.7,
158.4, 143.0, 141.2, 131.7, 130.2, 129.3, 120.9, 115.0, 114.5, 111.2,
111.0, 55.2, 55.1, 39.2, 38.3, 31.9, 30.3, 29.4, 9.9.
1H NMR (300 MHz, CDCl3): d 7.13 (d, J = 9.0 Hz, 1H), 6.83 (d,
J = 2.8 Hz, 1H), 6.77 (dd, J = 8.9 Hz, 2.8 Hz, 1H), 5.70 (br s, 1H), 3.81
(s, 3H), 3.32 (q, J = 6.4 Hz, 2H), 2.70 (t, J = 7.6 Hz, 2H), 2.20 (q,
J = 7.6 Hz, 2H), 1.85 (m, 2H), 1.15 (t, J = 7.6 Hz, 3H). 13C NMR
(75 MHz, CDCl3): d 174.1, 159.0, 141.3, 135.6, 122.2, 115.9, 112.8,
55.6, 38.7, 29.7, 29.6, 27,6, 9.7 and 124.8, 120.6, 116.4, 112.1 (quar-
tet for CF3).
4.1.6. N-{3-[5-Methoxy-2-(3-methoxy-phenylethynyl)-phenyl]-
propyl}-propionamide (7)
To a two-necked flask attached with a condenser, was added N-
[3-(2-trifluoromethylsulfonyl-5-methoxy-phenyl)-propyl]-propi-
onamide (47 mg, 0.127 mmol), Pd(PPh3)4 (14 mg, 0.012 mmol), CuI
(7 mg, 0.037 mmol) and nBu4NI (66 mg, 0.18 mmol). And the flask
was degassed and filled with N2 three times. A pre-degassed mix-
ture of triethylamine and DMF (1:5, 1.5 mL) was added, followed
by 1-ethynyl-3-methoxy-benzene (0.064 mL, 0.51 mmol). The mix-
ture was heated to 70 °C for 24 h. The reaction was stopped and the
solvent was evaporated under reduced pressure. Water was added
and the mixture was extracted with dichloromethane (ꢂ3). The
combined extract was dried, filtered and concentrated. Purification
by column chromatography on silica gel, eluting with 20–25% ace-
tone in petroleum ether, provided the desired product (24 mg,
0.068 mmol, 53%) together with starting material (22 mg, 46%
recovery). ESI-MS: 352.29 (M+1). TOF-HRMS Calcd for C22H26NO3
(M+H)+ 352.1907. Found: 352.1907. 1H NMR (300 MHz, CDCl3): d
7.45 (d, J = 8.2Hz, 1H), 7.26 (t, J = 8.4 Hz, 1H), 7.11 (d, J = 7.8 Hz,
1H), 7.04 (s, 1H), 6.89 (d, J = 8.4 Hz, 1H), 6.73–6.77 (m, 2H), 5.52
(br s, 1H), 3.83 (s, 3H), 3.82 (s, 3H), 3.30 (m, 2H), 2.88 (t, J = 7.3Hz,
2H), 2.03 (q, J = 7.6 Hz, 2H), 1.92 (m, 2H), 1.06 (t, J = 7.6 Hz, 3H).
13C NMR (75 MHz, CDCl3): d 173.8, 159.9, 159.4, 145.4, 133.6,
129.5, 124.5, 123.9, 116.2, 114.6, 114.5 (2), 111.8, 91.5, 88.1, 55.3
(2), 38.8, 32.0, 30.4, 29.7, 9.9.
4.1.9. N-[3-(4,30-Dimethoxy-biphenyl-2-yl)-propyl]-
propionamide (10)
To a flask, was charged with N-[3-(2-trifluoromethylsulfonyl-5-
methoxy-phenyl)-propyl]-propionamide (30 mg, 0.081 mmol) and
3-methoxyphenylboronic acid (18.5 mg, 0.122 mmol). After the
flask was degassed and filled with N2 three times, toluene (0.6 mL)
and aqueous sodium carbonate solution (2 M, 0.2 mL) were added,
followed by Pd(PPh3)4 (9 mg, 0.0081 mmol). The mixture was de-
gassed and filled with nitrogen again, and heated to 80 °C for over-
night. After the reaction was stopped, the mixture was extracted
with ethyl acetate (ꢂ3). The combined extract was dried, filtered
and concentrated. Purification by column chromatography on silica
gel, eluting with 20–25% acetone/petroleum ether, provided the de-
sired product (10 mg, 0.03 mmol, 38%). ESI-MS: 328.31 (M+1). TOF-
HRMS Calcd for C20H26NO3 (M+H)+ 328.1907. Found: 328.1900. 1H
NMR (300 MHz, CDCl3): d 7.32 (t, J = 7.9 Hz, 1H), 7.14 (d, J = 8.2Hz,
1H), 6.78–6.91 (m, 5H), 5.00 (br s, 1H), 3.84 (s, 3H), 3.93 (s, 3H),
3.09 (q, J = 6.3Hz, 2H), 2.64 (t, J = 7.5 Hz, 2H), 2.02 (q, J = 7.6 Hz,
2H), 1.61–1.70 (m, 2H), 1.06 (t, J = 7.6 Hz, 3H). 13C NMR (75 MHz,
CDCl3): h173.6, 159.3, 159.0, 143.0, 140.3, 134.3, 131.0, 129.2,
122.0, 115.3, 114.7, 112.1, 111.3, 55.3, 55.2, 38.6, 31.0, 30.2, 29.6, 9.8.
4.1.7. N-(3-{5-Methoxy-2-[2-(3-methoxy-phenyl)-ethyl]-
phenyl}-propyl)-propionamide (8)
To a flask containing N-{3-[5-methoxy-2-(3-methoxy-phenyl-
ethynyl)-phenyl]-propyl}-propionamide (13 mg, 0.037 mmol),
was added methanol (1 mL) and 10% palladium on carbon (5 mg).
The mixture was degassed three times and stirred under hydrogen
(1 atm) at room temperature overnight. The reaction was stopped
and the mixture was filtered through a plug of celite. Purification
by column chromatography on silica gel, eluting with 20% acetone
4.2. Radioligand binding assay
CHO cells stably expressing human MT1 or MT2 receptor have
been described and characterized previously.13 Competitive

552
Y. Hu et al. / Bioorg. Med. Chem. 21 (2013) 547–552
binding assays were performed as described14,15 with modifica-
tions for intact cells. Briefly, 1.5 ꢂ 105 cells were suspended in
binding buffer (50 mM Tris, 2 mM MgCl2, 1 mM EGTA, pH 7.4) con-
taining 1 nM [3H]melatonin and increasing concentrations of a test
compound. Assays were carried out at 4 °C for 60 min with occa-
sional agitation and then terminated by rapid filtration through
GF/C filters pre-soaked in 10 mM Tris, pH 7.4. Bound radioactivity
was counted in Wallac 1450 Microbeta Jet scintillation counter.
Competitive curves were fitted using a one-site competition non-
linear regression (GraphPad Prism 3.03). Data were means of 2–3
independent experiments performed in duplicates. Standard errors
were typically within 10% of the mean value. Melatonin was em-
ployed as standard reference in every assay with reproducible Ki.
Ki values were calculated using the Cheng–Prusoff equation.
fire C18 column (4.6 ꢂ 150 mm, 5
lM) was used as an analytical
column. The UV detection wavelength was 290 nM for compound
7 and 220 nM for other compounds. The mobile phase consisted
of 55% acetonitrile and 45% water. The flow-rate was 1 ml/min.
The quantification methods were validated by determining linear-
ity, accuracy and inter-/intra-day precise.
The metabolite identification was performed using a HPLC sys-
tem coupled with an online Thermo-Finnigan LCQ Classic ion trap
mass spectrometer equipped with electrospray ionization source.
The separation of metabolites and parents were carried out by a
Waters Sunfire C18 column (4.6 ꢂ 150 mm, 5
lM). Mobile phase
consisted of acetonitrile and water and was delivered in a gradient
program at the flow-rate 0.4 ml/min: 0 min, 40% ACN; 19 min, 40%
ACN; 30 min, 60% ACN. For MS detection, positive mode was em-
ployed for the identification of metabolites. The major working
parameters for mass spectrometer were as follows: sheath gas N2
flow at 60 arbitrary units, auxiliary gas N2 flow at 40 arbitrary
units, ion-spray voltage 4.5 kV, capillary temperature at 200 °C.
The metabolites were identified by MS full scan mode (m/z 100–
1000) and CID-MS spectra were produced by collision-induced dis-
sociation of each molecular ion of interest, using normalized colli-
sion energy of 20–50%.
4.3. Functional assay
Fluorometric assay was performed using two clonal cell lines
with stable expression of both melatonin receptor and chimeric
G
a
protein 16z25 to measure intracellular Ca2+ mobilization.16
Cells in normal growth medium were seeded into clear-bottomed
black-walled 96-well plates a day before assay. Cells in each well
were labeled with 2 lM Fluo-4 (Invitrogen) in 200 ll of Hank’s bal-
anced salt solution (pH 7.5) containing 2.5 mM probenecid for 1 h
at 37 °C prior to the addition of test compounds or melatonin.
Changes in fluorescence upon the addition of test compounds or
melatonin were detected in the fluorometric imaging plate reader
Acknowledgment
This study was supported in part by Grants from the Hong Kong
Jockey Club, the Innovation and Technology Fund (ITS/113/03), Re-
search Grants Council (660108 and TBRS/T13-607/12R) and Uni-
versity Grants Council (AoE/B-15/01) of Hong Kong.
FLIPRTETRA
™ (Molecular Devices/MDS Analytical Technologies)
with an excitation wavelength of 488 nM. The real-time fluores-
cent signal was monitored for 3 min. Results were expressed as
changes in relative fluorescence units (RFU). Concentration-re-
sponse curves were generated by determining the maximal change
in RFU of each data set. Numerical analysis of the statistics and
EC50 (median effective concentration) values were performed on
GraphPad Prism version 3.03.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmc.2012.10.060.
References and notes
4.4. Metabolic stability evaluation
1. Arendt, J. J. Biol. Rhythm 2005, 20, 291.
2. Samel, A. Eur. J. Med. Res. 1999, 4, 385.
In vitro metabolism was conducted in a system consisted of 50 ll
of NADPH generating system (1 mM NADP, 1 mM NADH, 1 mM glu-
cose-6-phosphate, 2 unit/ml glucose-6-phosphate dehydrogenase
3. Delagrange, P.; Boutin, J. Chronobiol. Int. 2006, 23, 413.
4. Olakowska, E.; Marcol, W.; Kotulska, K.; Lewin-Kowalik, J. Bratisl. Lek. Listy
2005, 106, 171.
5. Ekmekcioglu, C. Biomed. Pharmacother. 2006, 60, 97.
6. Mor, M.; Rivara, S.; Pala, D.; Bedini, A.; Spadoni, G.; Tarzia, G. Expert Opin. Ther.
Pat. 2010, 20, 1059.
7. Spadoni, G.; Bedini, A. Melatonin: From Molecules to Therapy; Pandi-Perumal, S.
R., Cardinali, D. P., Eds.; Nova Science: New York, 2007; p 33.
8. Mesangeau, C.; Fraise, M.; Delagrange, P.; Caignard, D. H.; Boutin, J. A.;
Berthelot, P.; Yous, S. Eur. J. Med. Chem. 2011, 46, 1835.
9. Koike, T.; Hoashi, Y.; Takai, T.; Nakayama, M.; Yukuhiro, N.; Ishikawa, T.; Hirai,
K.; Uchikawa, O. J. Med. Chem. 2011, 54, 3436.
10. Hu, Y. Q.; Ho, M. K. C.; Chan, K. H.; New, D. C.; Wong, Y. H. Bioorg. Med. Chem.
Lett. 2010, 20, 2582.
11. Zhu, J.; Hu, Y. Q.; Ho, M. K. C.; Wong, Y. H. Xenobiotica 2011, 41, 35.
12. Liu, A. M. F.; Ho, M. K. C.; Wong, C. S. S.; Chan, J. H. P.; Pau, A. H. M.; Wong, Y. H.
J. Biomol. Screen. 2003, 8, 39.
13. New, D. C.; Wong, Y. H. Assay Drug Dev. Technol. 2004, 2, 269.
14. Ho, M. K. C.; New, D. C.; Wong, Y. H. Neurosignals 2002, 11, 115.
15. Tian, Y.; New, D. C.; Yung, L. Y.; Allen, R. A.; Slocombe, P. M.; Twomey, B. M.;
Lee, M. M. K.; Wong, Y. H. Eur. J. Immunol. 2004, 34, 785.
16. Liu, A.; Ho, M.; Wong, C.; Chan, J.; Pau, A.; Wong, Y. J. Biomol. Screen. 2003, 8, 39.
and 4 mM MgCl2), 5
buffer (50 mM, pH 7.4). The mixture was pre-incubated at 37 °C
for 30 min. Reactions were initiated by adding 12.5 l rat/human li-
ll of 1 mM test compound and 425
ll TrisꢃCl
l
ver microsomal suspensions (20 mg protein/ml) and shaken at 37 °C
for 30 min with air exposure, and were subsequently terminated by
adding 2 ml ice-cold dichloromethane containing internal stan-
dards. After extraction by shaking the sample tube and centrifuga-
tion at 3000 rpm for 5 min, the organic phase was transferred to a
new tube and evaporated under N2. The residue was reconstituted
in 100 ll methanol, and analyzed by HPLC–DAD/MS system for
evaluating stability and identifying metabolites.
Sample analysis was carried out using a HPLC system equipped
with a Waters 600 controller quaternary pump, Waters 717 plus
autosampler injector and Waters 2996 photodiode array detector
(Waters Products, Milford, MA, USA). A reverse phase Water Sun-