Synthesis and structure-activity relationship of novel benzoxazole derivatives as melatonin receptor agonists

Article abstract of DOI:10.1016/j.bmcl.2004.04.082

A series of benzoxazole derivatives was synthesized and evaluated as melatoninergic ligands. The binding affinity of these compounds for human MT₁ and MT₂ receptors was determined using 2-[ ¹²⁵I]-iodomelatonin as the radio

Full text of DOI:10.1016/j.bmcl.2004.04.082

Bioorganic & Medicinal Chemistry Letters 14 (2004) 3799–3802
Synthesis and structure–activity relationship of novel
benzoxazole derivatives as melatonin receptor agonists
Li-Qiang Sun,* Jie Chen, Marc Bruce, Jeffrey A. Deskus,
James R. Epperson, Katherine Takaki, Graham Johnson,à Lawrence Iben,
Cathy D. Mahle,§ Elaine Ryan and Cen Xu
Bristol-Myers Squibb Pharmaceutical Research Institute, 5 Research Parkway, Wallingford, CT 06492, USA
Received 23 March 2004; revised 26 April 2004; accepted 27 April 2004
Available online
Abstract—A series of benzoxazole derivatives was synthesized and evaluated as melatoninergic ligands. The binding affinity of these
compounds for human MT1 and MT2 receptors was determined using 2-[125I]-iodomelatonin as the radioligand. From this series of
benzoxazole derivatives, compounds 14 and 17 were identified as melatonin receptor agonists.
Ó 2004 Elsevier Ltd. All rights reserved.
The hormone melatonin (N-acetyl-5-methoxy-trypt-
amine) (Fig. 1) is synthesized and released primarily by
the pineal gland in a circadian manner that closely fol-
lows the daily light/dark cycle.1;2 It plays a major role in
the regulation of circadian rhythms and the control of
seasonal cycles.3;4 Melatonin alleviates jet lag, regulates
delayed sleep phase syndrome,5 and induces sleep.6
It has been demonstrated that many of the effects of
melatonin are mediated through G-protein-coupled
receptors expressed primarily in the brain, retina, pitu-
itary, and blood vessels.7 The recent cloning of several
G-protein-coupled melatonin receptor genes has
revealed at least three melatonin receptor subtypes, two
of which are defined as MT1 and MT2 and have been
found in mammals.8 The human MT1 and MT2 receptor
subtypes show 55% overall identity at the amino acid
level and about 70% identity within the transmembrane
domains where ligand binding is thought to occur. Both
MT1 and MT2 receptors are negatively coupled to
adenylate cyclase and exhibit subnanomolar binding
affinity for melatonin.9;10 While it is known that the MT1
and MT2 receptors are expressed both centrally and
peripherally, the physiological roles of these receptors
are not as well defined. A few MT2 selective ligands have
been described to date.11–18 Several physiological
responses mediated through the activation of MT2
receptor have been identified using some of these com-
pounds.19 This receptor appears to play the major role in
mammalian circadian entrainment since in mice lacking
the MT1 receptor, melatonin retains phase shifting
activity that can be blocked by a MT2-selective antag-
onist.20 However, there are only two examples of MT1
selective ligands provided in the literature until last
year,17;21 the lack of which has made it difficult to discern
the physiological role of the subtype. Therefore, a need
exists for ligands that demonstrate selectivity at
the receptor level and particularly for MT1. Such
ligands would be valuable tools to help define the subtle
O
R1
O
HN
HN
O
O
N
R
N
H
1
Melatonin
Figure 1.
Keywords: Melatonin; Melatonin receptors; Agonists; Benzoxazole.
* Corresponding author. Tel.: +1-203-677-7460; fax: +1-203-677-7702;
e-mail: sunl@bms.com
Present address: Arena Pharmaceuticals, Inc., 6166 Nancy Ridge
Drive, San Diego, CA 92121, USA.
Present address: Rib-X Pharmaceuticals, Inc., 300 George Street,
à
New Haven, CT 06511, USA.
Present address: Bayer Pharmaceutical Research Center, 400 Morgan
Lane, West Haven, CT 06516, USA.
§
0960-894X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.04.082

3800
L.-Q. Sun et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3799–3802
contributions of individual melatonin receptor subtypes
to the multiple functions of melatonin, while also
elucidating the pharmacological profile of subtype
selective melatoninergic agents.
OH
OH
H
N
O2N
CN
R
CN
a
O
3
10
R
R
O
We have previously reported the discovery and binding
characteristics toward human MT1 and MT2 melatonin
receptors of a class of N-(2-benzoxazol-7-yl-cyclopro-
pylmethyl)-amides such as 1 in Figure 1.22 These mel-
atoninergic ligands possess high binding affinity toward
human MT1 and MT2 receptor subtypes, comparable to
or higher than the affinity of melatonin itself. Of par-
ticular interest, these ligands bind to MT1 receptor at the
subnanomolar level and exhibit up to a 14-fold selec-
tivity ratio over the MT2 subtype. These results high-
lighted the utility of introducing conformational
restraint in the alkyl tether that links the benzoxazole
core with a terminal amide functionality as a means of
modulating receptor affinity and selectivity. The earlier
work represented part of a broader initiative to under-
stand the SAR associated with benzoxazole-based mel-
atoninergic agents in the context of MT1 and MT2
binding properties. In this article, we present a detailed
description of the SAR associated with the two termini
of this parent acylic benzoxazole chemotype. More
specifically, the synthesis and biological activity of a new
series of benzoxazole derivatives as novel and potent
melatoninergic ligands are discussed. These studies
incorporate a propyl amide motif that was shown to be
optimal in a series of phenylalkyl amide derivatives.23
O
b
c
N
N
CN
NH2
12
11
R
O
O
N
d
N
R1
13-47
H
Scheme 2. Reagents and conditions: (a) (RCO)2O, 10% Pd/C, H2,
THF; (b) PPTS, xylene, reflux; (c) Raney Ni, H2, MeOH/NH4OH;
(d) RCOCl, Et3N, CH2Cl2 or RNCO, benzene.
zation. In Scheme 2, the order of these synthetic pro-
cesses is reversed.
As outlined in Scheme 1, the synthesis of the designed
compounds 7–9 started from commercially available
3-nitrosalicylaldehyde 2, which was homologated with
diethylcyanomethylphosphonate using
a
Horner–
Wadsworth–Emmons reaction to afford nitrile 3 in
excellent (95%) yield. Cinnamonitrile 3 was converted to
carbamate 4 in a one-pot procedure wherein 3 was
hydrogenated over 10% Pd/C in the presence of
(Boc)2O. Further reduction of nitrile 4 with Raney
Nickel in the presence of the appropriate anhydride
yielded a series of amides 5. Removal of the Boc group,
using hydrochloric acid, produced aniline 6, which was
closed to the benzoxazole heterocycle using triethyl
ortho-formate in the presence of a catalytic amount of
PPTS under reflux to provide target compounds 7–9.
The synthetic routes utilized to procure the compounds
of interest are shown in Schemes 1 and 2. These routes
differ in that in Scheme 1 the oxazole ring is introduced
after derivatization of the acyclic amine to its final form.
This order for the synthetic transformations was found
to be essential because the unsubstituted-oxazole ring
system present in compounds 7–9 is unstable toward the
conditions required for amine formation and derivati-
The general route to the 2-substituted benzoxazole
derivatives 13–47 is described in Scheme 2. In brief, the
nitrile 3 was hydrogenated over 10% Pd/C in the pres-
ence of an anhydride followed by quenching of this
reaction with aqueous sodium hydroxide to afford amide
10. Heating 10 with a catalytic amount of PPTS effected
dehydration to give the benzoxazole 11. Hydrogenation
of the nitrile moiety of 11 with Raney Nickel afforded
the primary amine 12, which was derivatized with an
acid chloride or an isocyanate to give the desired amide
and urea products, respectively. These compounds, 13–
47, are compiled in Table 1.
OH
OH
O2N
a
c
O2N
CN
O
b
2
3
Boc OH
N
Boc OH
N
O
d
CN
H
H
N
R1
O
H
The Ki values of compounds 7–9 and 13–47 for human
MT1 and MT2 melatonin receptor subtypes were
determined in assays using 2-[125I]-iodomelatonin as the
ligand according to described methods.24 The chemical
structures, Ki values and MT2/MT1 selectivity ratios of
these compounds are reported in Table 1. Compared to
melatonin, the benzoxazoles 7–9, in which there is a
hydrogen at the 2-position of the benzoxazole ring
(R ¼ H), demonstrated high affinity for both MT1 and
MT2 receptors. On the other hand, these ligands showed
a slight change in binding affinity ongoing from
acetamide 7 to butyramide 9 with the exception that
5
4
O
OH
O
N
H2N
e
N
R1
N
R1
H
H
6
7-9
Scheme 1. Reagents and conditions: (a) (EtO)2POCH2CN, NaH,
THF, 0 °C, 95%; (b) Boc2O, 10% Pd/C, H2, THF, 94%; (c) (RCO)2O,
Raney Ni, H2, THF; (d) 4 N HCl, EtOAc, 40 °C; (e) CH(OEt)3, PPTS,
xylene, reflux.

L.-Q. Sun et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3799–3802
3801
Table 1. Binding affinity of compounds 7–9 and 13–47 for human MT1
and MT2 melatonin receptors stably expressed in NIH3T3 cells. Values
represent mean from experiments performed in duplicate
Homologation of the methyl group of the amide side
chain to ethyl (14), propyl (15), or iso-propyl (16) re-
sulted in a higher increase in MT1 affinity compared to
MT2 receptor binding, leading to a reduction in receptor
selectivity. Interestingly, pursuing a similar substitution
pattern with the cyclopropyl derivative 17 resulted in the
achievement of subnanomolar MT1 affinity that was
accompanied by a twofold increase in MT2 binding.
Moreover, 17 showed little selectivity toward the MT1
receptor subtype (ratio ¼ 3.8). Replacing the terminal
amide of 17 with a simple ethyl urea provided a com-
pound, 18, which demonstrated a marked reduction in
binding affinity at both receptors. Compounds 19–24, in
which R is an ethyl moiety instead of a methyl group,
showed a trend toward increased MT2 affinity and
reduced MT1 binding. Of this series, only the amide 21
and urea 24 exhibited modest selectivity for the MT2
receptor subtype. However, the introduction of an
n-propyl, iso-propyl, or a phenyl group at the 2-position
of the benzoxazole ring, compounds 25–42, resulted in a
significant decrease in binding affinity toward both
human MT1 and MT2 receptors while retaining the same
limited level of subtype selectivity as melatonin itself.
Increased selectivity for the MT1 subtype was observed
when a 4-phenylbutyl group was introduced at the
benzoxazole 2-position, as exemplified by compounds
43–47. This series demonstrates excellent affinity for the
MT1 subtype and relatively modest binding affinity for
the MT2 receptor. Consequently these analogues show
MT2/MT1 selectivity ratios ranging from 9 to 35 with
compound 47 the most selective ligand identified in this
study.
R
O
O
N
N
H
R1
7-9 and 13-47
Compd
R
R1
MT1 Ki
(nM)
MT2 Ki
(nM)
Ratio
MT2/MT1
Mel
7
––
H
––
0.4
30
0.3
4.3
1.8
3.7
5.3
2.8
2.2
2.3
2.7
26
0.75
0.14
0.27
0.37
0.22
1.0
Me
Et
8
H
6.6
9
H
n-Pr
Me
9.9
24
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
Me
Me
Me
Me
Me
Me
Et
Et
2.8
2.3
3.0
0.7
n-Pr
i-Pr
c-Pr
NHEt
Me
0.96
0.8
3.8
50
0.52
0.25
0.28
0.12
0.34
1.0
9.0
3.5
8.1
13.2
4.5
75
2.3
1.0
1.0
4.5
4.7
12
Et
Et
Et
n-Pr
i-Pr
c-Pr
NHEt
Me
Et
Et
Et
0.16
0.19
0.53
0.13
0.48
0.17
0.07
0.30
0.11
0.13
2.2
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
i-Pr
i-Pr
i-Pr
i-Pr
i-Pr
i-Pr
Ph
183
35
27
Et
51
166
60
n-Pr
i-Pr
c-Pr
NHEt
Me
22
29
44
7.6
36
492
116
142
119
40
35
15
Et
The two most attractive compounds emerging from this
series, 14 and 17, were further tested for their functional
activity in NIH3T3 cells expressing melatonin MT1 or
MT2 receptor.25 The agonist activity of these com-
pounds was assessed by comparing their ability to in-
hibit forskolin-stimulated cAMP accumulation with
that of melatonin. Full agonist activity was confirmed
for both compounds at both the human melatonin MT1
and MT2 receptor subtypes. At the MT1 receptor the
EC50s for compounds 14 and 17 were measured as 0.26
and 0.17 nM, respectively, while the relative intrinsic
activity of these two analogues was 1.1 and 1.0,
respectively. At MT2 receptor, the EC50 values obtained
for benzoxazoles 14 and 17 were 0.32 and 0.74 nM,
respectively, with intrinsic activities of 0.98 and 0.96
compared to that of melatonin, respectively.
n-Pr
i-Pr
c-Pr
NHEt
Me
16
88
34
114
52
3.3
194
40
0.27
2.2
89
10
Ph
Ph
Et
13.1
0.77
0.28
0.66
1.6
n-Pr
i-Pr
c-Pr
NHEt
60
87
17
57
Ph
Ph
23
340
38
61
Ph
0.18
12
18
Ph(CH2)4 Me
Ph(CH2)4 Et
Ph(CH2)4 n-Pr
Ph(CH2)4 i-Pr
Ph(CH2)4 c-Pr
1.3
16
14
0.76
3.3
29
18
9
14
1.3
0.63
22
35
Data reported in the text are means of 1–2 experiments run at five
different concentrations in duplicates. Standard errors were typically
within 10% of mean value. Melatonin was run as standard reference in
every assay with reproducible Ki.
In conclusion, the constrained cyclopropane moiety
of N-(2-benzoxazol-7-yl-cyclopropylmethyl)amide was
successfully replaced by a conformationally flexible
alkyl side chain. This structural modification led to the
discovery of a series of benzoxazole derivatives26 as
novel melatoninergic ligands and the subsequent iden-
tification of benzoxazole derivative 14 and 17 as mela-
tonin receptor agonists and compound 47 as a highly
potent human MT1 ligand with moderate receptor
selectivity. Structure–activity relationships have been
defined around both the terminal amide moiety and
the C2 position of the benzoxazole ring. Further-
more, a 4-phenybutyl moiety at the 2-position of the
acetamide 7 exhibited relatively weaker MT1 binding.
This series demonstrated poor to modest selectivity for
the MT2 receptor with a ratio of MT1/MT2 ranging
from 2 to 7.
The introduction of a methyl group at the 2-position of
the benzoxazole heterocycle provided a series of ligands
13–18 from which the parent compound 13 is a potent
MT2 ligand that exhibits fourfold weaker MT1 affinity,
providing a modest index of receptor selectivity.

3802
L.-Q. Sun et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3799–3802
17. Audinot, V.; Mailliet, F.; Lahaye-Brasseur, C.; Bonnaud,
A.; Gall, A. L.; Amosse, C.; Dromaint, S.; Rodriguez, M.;
Nagel, N.; Galizzi, J.-P.; Malpaux, B.; Guillaumet, G.;
Lesieur, D.; Lefoulon, F.; Renard, P.; Delagrange, P.;
Boutin, J. A. Naunyn-Schmiedeberg’s Arch. Pharmacol.
2003, 367, 553.
benzoxazole heterocycle provides compounds with
enhanced MT1 receptor activity and selectivity.
Acknowledgements
18. Epperson, J. R.; Deskus, J. A.; Gentile, A. J.; Iben, L. G.;
Ryan, E.; Sarbin, N. S. Bioorg. Med. Chem. Lett. 2004, 14,
1023.
19. Masana, M. I.; Dubocovich, M. L. Sci. STKE 2001, 6,
107.
We thank Ms. Yi-Xin Li for NMR data support, Mr.
Robert Kane for MS data support, and Dr. Nicholas A.
Meanwell for valuable discussion.
20. Dubocovich, M. L.; Yun, K.; Al-Ghoul, W. M.; Benlou-
cif, S.; Masana, M. I. FASEB J. 1998, 12, 1211.
21. Descamps-Francois, C.; Yous, S.; Chavatte, P.; Audinot,
V.; Bonnaud, A.; Boutin, J. A.; Delagrange, P.; Bennejean,
C.; Renard, P.; Lesieur, D. J. Med. Chem. 2003, 46, 1127.
22. Sun, L.-Q.; Chen, J.; Takaki, K.; Johnson, G.; Iben, L.;
Mahle, C. D.; Ryan, E.; Xu, C. Bioorg. Med. Chem. Lett.
2004, 14, 1197.
References and notes
1. Yu, H.-S.; Reiter, R. J. Melatonin. Biosynthesis, Physio-
logical Effects, and Clinic Applications; CRC Press: Boca
Raton, 1993.
2. Arendt, J. Melatonin and the Mammalian Pineal Gland;
Chapman and Hall: London, 1995.
3. Cassone, V. M. Trends Neurosci. 1990, 13, 457.
4. Bartness, T. J.; Goldman, B. D. Experientia 1989, 45, 939.
5. Arendt, J.; Deacon, S.; English, J.; Hampton, S.; Morgan,
L. J. Sleep Res. 1995, 4, 74.
23. Garratt, P. J.; Travard, S.; Vonhoff, S. J. Med. Chem.
1996, 39, 1797.
24. Sun, L.-Q.; Chen, J.; Mattson, R. J.; Epperson, J. R.;
Deskus, J. A.; Li, W.-S.; Takaki, K.; Hodges, D. B.; Iben,
L.; Mahle, C. D.; Ortiz, A.; Molstad, D.; Ryan, E.;
Yeleswaram, K.; Xu, C.; Luo, G. Bioorg. Med. Chem.
Lett. 2003, 13, 4381.
6. Wirz-Justice, A.; Amstrong, S. M. J. Sleep Res. 1996, 5,
137.
7. Morgan, P. J.; Barrett, P.; Howell, H. E.; Helliwell, R.
Neurochem. Int. 1994, 24, 101.
25. Mattson, R. J.; Catt, J. D.; Keavy, D.; Sloan, C. P.;
Epperson, J.; Gao, Q.; Hodges, D. B.; Iben, L.; Mahle, C.
D.; Ryan, E.; Yocca, F. D. Bioorg. Med. Chem. Lett. 2003,
13, 1199.
8. Reppert, S. M.; Weaver, D. R.; Godson, C. Trends
Pharmacol. Sci. 1996, 17, 100.
9. Reppert, S. M.; Weaver, D. R.; Ebisawa, T. Neuron 1994,
13, 1177.
10. Reppert, S. M.; Godson, C.; Mahle, C. D.; Weaver, D. R.;
Slaugenhaupt, S. A.; Gusella, J. F. Proc. Natl. Acad. Sci.
1995, 92, 8734.
11. Teh, M.-T.; Sugden, D. Naunyn-Schmiedeberg’s Arch.
Pharmacol. 1998, 358, 522.
12. Faust, R.; Garratt, P. J.; Jones, R.; Yeh, L.-K.; Tsotinis,
A.; Panoussopoulou, M.; Calogeropoulou, T.; Teh, M.-T.;
Sugden, D. J. Med. Chem. 2000, 43, 1050.
13. Spadoni, G.; Balsamini, C.; Diamantini, G.; Tontini, A.;
Tarzia, G.; Mor, M.; Rivara, S.; Plazzi, P. V.; Nonno, R.;
Lucini, V.; Pannacci, M.; Fraschini, F.; Stankov, B. M.
J. Med. Chem. 2001, 44, 2900.
14. Wallez, V.; Durieux-Poissonnier, S.; Chavatte, P.; Boutin,
J. A.; Audinot, V.; Nicolas, J.-P.; Bennejean, C.; Delag-
range, P.; Renard, P.; Lesieur, D. J. Med. Chem. 2002, 45,
2788.
15. Mattson, R. J.; Catt, J. D.; Keavy, D.; Sloan, C. P.;
Epperson, J.; Gao, Q.; Hodges, D. B.; Iben, L.; Mahle, C.
D.; Ryan, E.; Yocca, F. D. Bioorg. Med. Chem. Lett. 2003,
13, 1199.
26. Selected spectral data. Compound 3: 1H NMR (300 MHz,
CDCl3) d 8.20 (dd, J ¼ 8:4, 1.6 Hz, 1H), 7.73–7.70 (m,
1H), 7.64 (d, J ¼ 16:8 Hz, 1H), 7.08 (t, J ¼ 7:7 Hz, 1H),
6.26 (d, J ¼ 16:8 Hz, 1H); 13C NMR (75 MHz, CDCl3) d
153.8, 143.7, 136.4, 134.3, 127.3, 125.0, 120.1, 118.0, 100.6;
MS (ESI) 189 (M)H)þ. Compound 10: (N-[3-((1E)-2-
cyanoethyl)-2-hydroxyphenyl]-5-phenylpentanamide): 1H
NMR (300 MHz, CDCl3) d 7.61 (s, 1H), 7.54–7.06 (m,
6H), 6.91–6.72 (m, 2H), 3.03 (t, J ¼ 7:3 Hz, 2H), 2.80–2.65
(m, 4H), 2.48 (t, J ¼ 5:5 Hz, 2H), 1.84–1.67 (m, 4H); 13C
NMR (75 MHz, CDCl3) d 174.0, 147.7, 142.0, 129.3,
128.6, 128.5, 126.1, 126.0, 125.8, 121.8, 120.3, 120.0, 36.8,
35.7, 30.9, 27.7, 25.5, 17.5; MS (ESI) 321 (M)H)þ.
Compound 11: (3-[2-(4-phenylbutyl)benzoxazol-7-yl]pro-
panenitrile): 1H NMR (300 MHz, CDCl3) d 7.62 (d,
J ¼ 7:8 Hz, 1H), 7.31–7.06 (m, 7H), 3.26 (t, J ¼ 7:5 Hz,
2H), 3.02 (t, J ¼ 7:3 Hz, 2H), 2.94–2.78 (m, 4H), 1.96–1.74
(m, 4H); 13C NMR (75 MHz, CDCl3) d 167.2, 149.2, 142.0,
141.3, 128.6, 128.5, 126.0, 124.9, 124.8, 121.1, 119.0, 118.8,
35.6, 31.0, 28.6, 26.5, 26.4, 17.9; MS (ESI) 305 (M+H)þ.
1
Compound 20: H NMR (300 MHz, CDCl3) d 7.47 (dd,
J ¼ 7:8, 0.9 Hz, 1H), 7.18 (t, J ¼ 7:6 Hz, 1H), 7.05 (d,
J ¼ 7:3 Hz, 1H), 5.56 (br s, 1H), 3.41 (q, J ¼ 6:9 Hz, 2H),
2.95–2.81 (m, 4H), 2.15 (q, J ¼ 7:6 Hz, 2H), 1.94–1.80 (m,
16. Yous, S.; Durieux-Poissonnier, S.; Lipka-Belloli, E.;
Guelzim, H.; Bochu, C.; Audinot, V.; Boutin, J. A.;
Delagrange, P.; Bennejean, C.; Renard, P.; Lesieur, D.
Bioorg. Med. Chem. 2003, 11, 753.
2H), 1.41 (t, J ¼ 7:5 Hz, 3H), 1.04 (t, J ¼ 7:6 Hz, 3H); 13
C
NMR (75 MHz, CDCl3) d 173.8, 168.0, 149.5, 140.7,
124.7, 124.6, 124.4, 117.3, 39.1, 29.8, 29.7, 27.4, 22.2, 11.0,
10.0; MS (ESI) 259 (M)1)þ.