Tetrahydroisoquinoline derivatives as melatonin MT2 receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2004.09.024

A series of tetrahydroisoquinolines has yielded potent MT₂ receptor antagonists, which are selective versus the MT₁ receptor. A series of tetrahydroisoquinolines has yielded potent MT₂ receptor antagonists, which are selective versus the MT₁ receptor.

Full text of DOI:10.1016/j.bmcl.2004.09.024

Bioorganic & Medicinal Chemistry Letters 14 (2004) 5881–5884
Tetrahydroisoquinoline derivatives as melatonin MT2
receptor antagonists
George N. Karageorge,* Stephen Bertenshaw, Lawrence Iben, Cen Xu, Nathan Sarbin,
Anthony Gentile and Gene M. Dubowchik
Bristol-Myers Squibb Pharmaceutical Research Institute, Wallingford, CT 06492-7660, USA
Received 14 July 2004; revised 9 September 2004; accepted 10 September 2004
Available online 30 September 2004
Abstract—A series of tetrahydroisoquinolines has yielded potent MT2 receptor antagonists, which are selective versus the MT1
receptor.
Ó 2004 Elsevier Ltd. All rights reserved.
Melatonin (N-acetyl-5-methoxytryptamine) has been
shown to exert its biological effects through binding to
specific G-protein coupled receptors in the brain.1 Clon-
ing of several melatonin receptor genes has revealed at
least three melatonin receptor proteins. Two of these
receptors are G-protein coupled receptors and have been
designated as MT1 and MT2.2,3 Specific 2-[125I]iodome-
latonin binding within the hypothalamus in the brain
is completely localized to the suprachiasmatic nucleus,
the area in the brain thought to regulate the bodyÕs
internal clock. Further evidence suggests that the mela-
tonin subtype primarily responsible for this action is the
MT2 receptor.4 This hypothesis has been further
strengthened by recent studies showing that MT2 recep-
tor agonists advance circadian rhythms5 while MT2
antagonists block melatonin-mediated phase advances
of circadian rhythms.6
Limited previous research has led to the identification of
a small number of selective MT1 or MT2 melatoninergic
ligands. Furthermore, much of this earlier work focused
on MT1 receptor ligands.12 Few reports have appeared
on selective MT2 receptor agonists.13–15 and MT2 recep-
tor antagonists.16–18 In this paper we present a novel ser-
ies of MT2 receptor antagonists. This series of
compounds was discovered via directed high-through-
put screening, and the central feature around which we
focused our initial SAR efforts was the tetrahydroiso-
quinoline scaffold as shown in compound 4 (Scheme 1).
2
3
R2
R3
R4
NH2
a
Ar
OHC
+
R1
Accordingly, this body of evidence suggests that selec-
tive MT2 receptor agonists should be particularly useful
for the treatment of sleep and chronobiotic disorders,
including jet lag and work shift syndrome.7,8 By focus-
ing on selective MT2 receptor antagonism, other effects
of endogenously secreted melatonin should not be inter-
rupted. This is important as melatonin has also been
thought to be involved in seasonal affective disorder9,10
immune disorders, premenstrual syndrome, and repro-
ductive disorders.11
R5
R2
R2
1
R3
R4
R3
R4
b
NH
R5 CH2
Ar R1
N
R6
R5 CH2
Ar R1
O
4
Scheme 1. Reagents and conditions: (a) formic acid 95°C, ꢀ50% yield;
(b) R6(C@O)Cl, Et3N, CH2Cl2 or R6(C@O)Cl, poly(4-vinylpyridine),
dichloroethane, Wa21J (Supelco) polyamine scavenging resin, rt (42–
88% yield).
Keywords: Melatonin; Antagonist.
*
Corresponding author. Tel.: +1 203 677 6351; fax: +1 203 677
7702; e-mail: karageog@bms.com
0960-894X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.09.024

5882
G. N. Karageorge et al. / Bioorg. Med. Chem. Lett. 14 (2004) 5881–5884
Table 2. Melatonin receptor binding of compounds 1p–u
The general synthetic route to these compounds is
described in Scheme 1. Pictect–Spengler19 reaction of
amine 2 with aldehyde 3 in formic acid afforded interme-
diate 4 in approximately 50% yield. Acylation of com-
pound 4 using an appropriate activated acyl derivative
was accomplished either singly or in parallel fashion
using automated high-speed synthesis techniques to
yield amide 1.
MeO
N
R6
MeO
CH2
O
H
Example
R6
MT2
IC50 (nM)
MT1
IC50 (nM)
MT1/MT2
Using the methodology described above with 3,3-diphe-
nylproprionaldehyde (i.e., R1 and Ar = Ph in 4) and
3,4-dimethoxyphenethylamine, a diverse set of racemic
compounds (1a–o) were prepared and evaluated for
their melatonergic properties (Table 1) in a human
MT1 and MT2 receptor binding assay.20,21 Acetamide
1a had the most potent MT2 receptor binding affinity
with a greater than 100-fold selectivity for the MT2
receptor. Urea 1g, formamide 1b, and carbamate 1i were
also potent MT2 receptor ligands. However, the latter
two possessed less selectivity versus the MT1 receptor,
when compared to 1a. Larger substituents at R6 (i.e.,
pentyl 1l, hexyl 1o, and c-hexyl 1k) had significantly
reduced MT2 receptor affinity. These results indicated
that MT2 receptor affinity could be achieved with small
substituents at R6, but selectivity versus MT1 receptor
binding was optimized with a specific substitution
pattern found in 1a.
1p
1q
1r
1s
1t
Me
76
24
>1000
—
—
>13
—
c-Bu
c-Hex
4-CF3Ph
H
220
1000
83
—
—
—
>1000
>1000
>12
>6
1u
NH2
160
(1q), there was an increase in MT2 affinity. For example,
c-Bu derivative 1j had an IC50 = 310nM( Table 1),
whereas, des-phenyl analog 1q had an IC50 of 24nM
(Table 2). These results suggest that the lipophilic bind-
ing pocket occupied by the second phenyl ring in 1a is
adjacent to the area where R6 binds, and that a larger
R6 may be able to partially fill that pocket when the
phenyl ring is removed.
Finally, substitution on the aromatic phenyl ring of the
tetrahydroisoquinoline was evaluated. This was accom-
plished with a variety of phenethylamines 2 in the Pic-
tect–Spengler reaction (Scheme 1). The results of these
efforts are shown in Table 3.
Removal of one of the side-chain phenyl rings was inves-
tigated. Using the methodology described in Scheme 1
with 3-phenylproprionaldehyde (i.e., R1 = H and
Ar = Ph in 4), afforded compounds 1p–u (Table 2).
When R6 = M e (1p), the removal of the phenyl group
(i.e., R1 = H) led to a decrease in MT2 receptor affinity.
For example, compound 1a has an MT2 affinity of
9.7nM( Table 1), while des-phenyl analog 1p had an
IC50 of 76nM( Table 2). However, when R6 was larger
Proper substitution of the phenyl ring of the tetrahydro-
isoquinoline was necessary for potent affinity at the MT2
receptor. The absence of substitution on the phenyl ring
led to no affinity for the MT1 and MT2 receptor (i.e.,
1w). Substitution at R2 was uniformly deleterious to
MT2 and MT1 receptor affinity (i.e., 1x,ab, and 1ad).
In contrast, substitution at R4 led to potent MT2 recep-
tor ligands (i.e., 1a,z, 1ae). In fact, 1z with only a single
methoxy substituent at R4 is the most potent compound
encountered in this study, and it was reasonably selec-
tive versus the MT1 receptor.
Table 1. Melatonin receptor binding of compounds 1a–o
MeO
N
R6
MeO
CH2
O
Additional substitution at R3 had significant effects on
selectivity. Addition of a second methoxy substituent
at R3 (1a) improved selectivity, while addition of bro-
mine (1ae) diminished selectivity. This diminution of
selectivity for 1ae was a result of both increased MT1
receptor affinity and a loss of MT2 receptor affinity.
Example
R6
MT2
IC50 (nM)
MT1
IC50 (nM)
MT1/MT2
1a
1b
1c
1d
1e
1f
Me
9.7
9.9
62
>1000
621
>103
62
H
Et
>1000
>1000
>1000
>1000
>1000
>1000
673
>16
>12
>8
>9
>33
>17
30
Pr
Ph
79
Compounds 1a,b,e, and 1f were selected for evaluation
of their functional activity at the MT2 receptor. Agon-
ism of MT2 receptors leads to inhibition of adenylate cy-
clase, leading to a decrease in intracellular cAMP. Thus,
inhibition of forskolin-stimulated cAMP accumulation
in NIH-3T3 cells stably expressing the MT2 receptor
was used as our functional assay.5 Compounds 1a, 1b,
1e, and 1f demonstrated full antagonism in this assay
(intrinsic activity = 0), indicating these compounds, as
representatives of the series, are full MT2 receptor
antagonists.
120
100
30
CF3
1g
1h
1i
NH2
NHMe
OMe
c-Bu
c-Hex
Pentyl
c-Bu
4-CF3Ph
Hexyl
60
22
1j
310
370
280
240
810
400
—
—
—
—
1k
1l
—
—
—
—
1m
1n
1o
—
—
—
—

G. N. Karageorge et al. / Bioorg. Med. Chem. Lett. 14 (2004) 5881–5884
Table 3. Melatonin receptor binding of amides 1v–ae
5883
R2
R3
R4
N
Me
R5 CH2
O
Example
R2
R3
R4
R5
MT2 IC50 (nM)
MT1 IC50 (nM)
MT1/MT2
1a
H
OMe
H
OMe
H
H
9.7
130
>1000
>1000
>1000
>1000
512
>103
>7
—
1w
1x
H
H
OMe
H
H
H
H
>1000
270
5.5
1y
1z
OMe
H
H
H
>3
93
H
OMe
H
H
1aa
1ab
1ac
1ad
1ae
H
OMe
OMe
F
OMe
H
45
>1000
140
>1000
>1000
>1000
>1000
230
>22
—
OMe
H
H
H
H
>7
—
CH2CH2O
H
OMe
H
>1000
24
H
Br
H
9
14. Conway, S.; Canning, S. J.; Howell, H. E.; Mowat, E. S.;
Barrett, P.; Drew, J. E.; Delagrange, P.; Lesieur, D.;
Morgan, P. J. Eur. J. Pharmacol. 2000, 15, 390.
15. 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.
16. Dubocovich, M. L.; Masana, M. I.; Iacob, S.; Sauri, D.
M. N–S Arch. Pharmacol. 1997, 355, 365–375.
17. Spandoni, G.; Balsamini, C.; Diamantin, 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, 2900–2912.
In summary, we have identified a novel series of tetra-
hydroisoquinolines,22 which are potent and selective
MT2 receptor antagonists. In particular, compound 1a
demonstrated single-digit nanomolar binding affinity,
a > 100-fold selectivity for the MT2 receptor versus the
MT1 receptor, and functional antagonism for the MT2
receptor. This compound represents a useful pharmaco-
logical tool to further investigate the chronobiotic func-
tion of the MT2 receptor. Behavioral aspects will be
disclosed in further publications.
18. 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, 753–759.
19. McElvain, S. M.; Bolliger, K. M. Org. Synth. Coll. 1932, 1,
473.
References and notes
1. Li, P.-K.; Witt-Enderby, P. A. Drugs Future 2000, 25, 945.
2. Reppert, S. M.; Weaver, D. R.; Godson, C. Trends
Pharmacol. Sci. 1996, 17, 100.
3. IUPHAR reclassified the Mel 1a and Me1 1b receptors to
MT1 and MT2, The IUPHAR Compendium of Receptor
Characterization and Classification, IUPHAR media:
London, 2000; pp 271–277.
20. (a) Reppert, S. M.; Weaver, D. R.; Ebisawa, T. Neuron
1994, 13, 1177; (b) Reppert, S. M.; Godson, C.; Mahle, C.
D.; Weaver, D. R.; Slaugenhaupt, S. A.; Gusella, J. F.
Proc. Natl. Acad. Sci. U.S.A. 1995, 92, 8734.
4. Dubocovich, D. L. et al. Adv. Exp. Med. Biol. 1999, 460,
181–190.
5. 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, 1199–1202.
6. Dubocovich, M. L. et al. FASEB J. 1998, 12, 1211–1219.
7. Arendt, J.; Deacon, S.; English, J.; Hampton, S.; Morgan,
L. J. Sleep Res. 1995, 4, 74.
8. Wirz-Justice, A.; Armstrong, S. M. J. Sleep Res. 1996, 5,
137.
9. Cassone, V. M. Trends Neurosci. 1990, 13, 457.
10. Bartness, T. J.; Goldman, B. D. Experientia 1989, 45, 939.
11. Nelson, R. J.; Drazen, D. L. Reprod. Nutr. Dev. 1999, 39,
383.
12. Descamps-Francois, C.; Yous, S.; Chavatte, P.; Audinot,
V.; Bonnaud, A.; Boutin, J. A.; Delagrange, P.; Benne-
jean, C.; Renard, P.; Lesieur, D. J. Med. Chem. 2003, 46,
1127.
21. Takaki, K. S.; Sun, L.-Q.; Johnson, G.; Epperson, J. R.;
Bertenshaw, S. B. U.S. Patent 6,569,894, 2003.
22. Selected spectral data. Compound 1a: H NMR (CDCl3)
1
(approx. 1.8:1 ratio of rotomers, A:B, respectively): d 7.32
(rotomers A and B, m, 10H), 6.62 (rotomer A, s, 1H), 6.60
(rotomer B, s, 1H), 6.53 (rotomer A, s, 1H), 6.38 (rotomer
B, s 1H), 5.68 (rotomer A, m, 1H), 4.60 (rotomer B, m,
1H), 4.12 (rotomer B, t, 1H), 4.03 (rotomer A, m, 1H),
3.92 (rotomer A, s, 3H), 3.88 (rotomers A and B, s, 3H),
3.77 (rotomer B, s, 3H), 3.43 (rotomers A and B, m, 1H),
3.21 (rotomer A, m, 1H), 2.94 (rotomer A, m, 1H), 2.80
(rotomers A and B, m, 1H), 2.67 (rotomer A, m, 1H), 2.54
(rotomer A, m, 1H), 2.35 (rotomer B, m, 1H), 1.96
(rotomer B, s, 3H), 1.65 (rotomer A, s, 3H). Mass spec.
416 (MH)+. Anal. Calcd for C27H29NO3–2/3H2O: C,
75.85; H, 7.15; N, 3.28. Found: C, 75.89; H, 7.03; N, 3.24.
1
Compound 1b: H NMR (CDCl3) (approx. 1.8:1 mixture
of rotomers A:B, respectively): d 8.09 (rotomer B, s, 1H),
7.89 (rotomer A, s, 1H), 7.30 (rotomers A and B, m, 10H),
6.61 (rotomer A, s, 1H), 6.55 (rotomer B, s, 1H), 6.47
(rotomer A, s, 1H), 6.30 (rotomer B, s, 1H), 5.40 (rotomer
B, q, 1H), 4.52 (rotomer A, q, 1H), 4.22 (rotomer A, q,
13. 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.

5884
G. N. Karageorge et al. / Bioorg. Med. Chem. Lett. 14 (2004) 5881–5884
1H), 4.16 (rotomer B, t, 1H), 4.01 (rotomer A, q, 1H), 3.89
(CDCl3) (approx. 2:1 ratio of rotomers): d 7.32 (rotomers
A and B, m, 10H), 6.58 (rotomers A and B, s, 1H), 6.51
(rotomer A, s, 1H), 6.30 (rotomer B, s, 1H), 5.67 (rotomer
B, t, 1H), 4.66 (rotomer A, s, 1H), 4.15 (rotomer B, t, 1H),
4.00 (rotomer A, q, 1H), 3.90 (rotomer A, s, 3H), 3.85
(rotomers A and B, s, 3H), 3.68 (rotomer B, s, 3H),
3.43 (rotomer B, m, 2H), 2.80 (rotomers A and B, m,
4H), 2.31 (rotomer A, m, 2H), 2.15 (rotomer B, m, 2H),
1.78 (rotomer A, m, 1H), 1.18 (rotomers A and B, m, 2H),
1.05 (rotomer B, t, 3H), 0.87 (rotomer A, t, 3H). Mass
spec. 442 (MH)+. Anal. Calcd for C29H33NO3–1/3H2O: C,
77.55; H, 7.55; N, 3.12. Found: C, 77.38; H, 7.50; N, 3.19.
Compound 1f: 1H NMR (CDCl3): d 7.30 (m, 10H), 6.59 (s,
1H), 6.29 (s, 1H), 5.50 (q, 1H), 4.07 (t, 1H), 3.83 (s, 3H),
3.77 (s, 3H), 3.62 (m, 1H), 2.83 (m, 4H), 2.45 (m, 1H).
Mass spec. 470 (MH)+. Anal. Calcd for C27H26NO3F3: C,
69.07; H, 5.58; N, 2.98. Found: C, 68.90; H, 5.58; N,
2.91.
(rotomer A, s, 3H), 3.86 (rotomer A, s, 3H), 3.83 (rotomer
B, s, 3H), 3.74 (rotomer B, s, 3H), 3.48 (rotomer A, m,
1H), 2.80 (rotomers A and B, m, 4H). Mass spec. 402
(MH)+. Anal. Calcd for C26H27NO3: C, 77.78; H, 6.78; N,
3.49. Found: C, 77.74, H, 6.93; N, 3.30. Compound 1c: 1H
NMR (CDCl3) (approx. 2:1 ratio of A:B rotomers,
respectively): d 7.34 (rotomers A and B, m, 10H), 6.60
(rotomers A and B, s, 1H), 6.52 (rotomer A, s, 1H), 6.30
(rotomer B, s, 1H), 5.69 (rotomer B, t, 1H), 4.64 (rotomer
A, m, 1H), 4.10 (rotomer B, t, 1H), 4.01 (rotomer A, q,
1H), 3.90 (rotomer A, s, 3H), 3.87 (s, 3H), 3.71 (rotomer
B, s, 2H), 3.44 (rotomer B, m, 2H), 2.80 (rotomers A and
B, m, 4H), 2.33 (rotomer A, m, 2H), 2.17 (rotomer B, m,
2H), 1.80 (rotomer A, m, 4H), 1.14 (rotomer B, t, 3H),
0.91 (rotomer A, t, 3H). Mass spec. 430 (MH)+. Anal.
Calcd for C28H31NO3: C, 78.29; H, 7.27; N, 3.26. Found:
C, 78.12; H, 7.23; N, 3.15. Compound 1d: 1H NM R