Novel and potent human and rat β3-adrenergic receptor agonists containing substituted 3-indolylalkylamines

Article abstract of DOI:10.1016/S0960-894X(03)00073-8

A novel series of 2-(3-indolyl)alkylamino-1-(3-chlorophenyl)ethanols was prepared and evaluated for in vitro ability to stimulate cAMP production in Chinese hamster ovary cells expressing cloned human β₃-AR. The optically active 30a was found to be the most potent and selective human β₃-AR agonist in this series with an EC₅₀ value of 0.062 nM. In addition, 30a selectivity for human β₃-AR was 210-fold and 103-fold that for human β₂-AR and β₁-AR, respectively. Furthermore, 30a showed potent agonistic activity at rat β₃-AR.

Full text of DOI:10.1016/S0960-894X(03)00073-8

Bioorganic & Medicinal Chemistry Letters 13 (2003) 1301–1305
Novel and Potent Human and Rat ꢀ3-Adrenergic Receptor
Agonists Containing Substituted 3-Indolylalkylamines
Hiroshi Harada,a,* Yoshimi Hirokawa,a Kenji Suzuki,a Yoichi Hiyama,a Mayumi Oue,b
Hitoshi Kawashima,b Naoyuki Yoshida,b Yasuji Furutanib and Shiro Katoa
aChemistry Research Laboratories, Dainippon Pharmaceutical Co., Ltd., Enoki 33-94, Suita 564-0053, Japan
bPharmacology & Microbiology Research Laboratories, Dainippon Pharmaceutical Co., Ltd., Enoki 33-94, Suita 564-0053, Japan
Received 5December 2002; accepted 16 January 2003
Abstract—A novel series of 2-(3-indolyl)alkylamino-1-(3-chlorophenyl)ethanols was prepared and evaluated for in vitro ability to
stimulate cAMP production in Chinese hamster ovary cells expressing cloned human b3-AR. The optically active 30a was found to
be the most potent and selective human b3-AR agonist in this series with an EC50 value of 0.062 nM. In addition, 30a selectivity for
human b3-AR was 210-fold and 103-fold that for human b2-AR and b1-AR, respectively. Furthermore, 30a showed potent ago-
nistic activity at rat b3-AR.
# 2003 Elsevier Science Ltd. All rights reserved.
Introduction
b-Adrenergic receptors (b-ARs) have been subclassified
as b1- and b2-ARs since 1967.1 A third b-AR, initially
referred to as ‘atypical’2 and later called b3-AR3,4 has
been found in a number of species,5ꢀ7 including man in
the early 1980s.4 The b3-AR is located on the cell sur-
face of both white and brown adipocytes and its stimu-
lation promotes both lipolysis and energy expenditure.8
Figure 1.
Since the discovery of b3-AR, a number of laboratories
have been engaged in developing potent and selective
b3-AR agonists for the treatment of obesity and non-
insulin dependent (Type-II) diabetes.9 Early b3-AR
agonists (the ‘first generation’ of potent and selective rat
b3-AR agonists) such as, BRL 37344,10 CL 316243,11
and SR 58611A,12 having a 3-chlorophenyl moiety in
the left-hand side and a carboxylic acid or an ester
functionality in the right-hand side as shown in Figure
1, were reported to be effective anti-obesity and anti-
diabetic agents in rodents.13
between human and rat b3-ARs, activity at the rat
b3-AR could not effectively predict that at the human
b3-AR.15 Thus, a new generation of human b3-AR ago-
nists with minimal side effects associated with activation
of human b1- and b2-ARs has long been needed.
At the beginning of 1990 and on the basis of results
obtained from random screening for rat b3-AR agonists,
we found that a novel 2-[2-(3-indolyl)ethylamino]-1-(3-
chlorophenyl)ethanol (7) having the 3-chlorophenyl
moiety structure known to be required for b3-AR ago-
nistic activity, potently inhibited rat spontaneous colonic
contraction (b3-AR; EC50=22.9ꢁ3.1 nM) and only
slightly relaxed both rat uterus (b2-AR; EC50=
577.3ꢁ149.4 nM) and guinea-pig trachea (b1-AR;
EC50=>10,000 nM). In order to improve the selectivity
of lead compound 7, we focused on the introduction of
various substituents into the indole nucleus and the
However, human clinical trials with these drugs for the
treatment of metabolic disorders have been disappointing
due to a lack of selectivity and/or potency or poor
pharmacokinetics.14 Because of the structural differences
*Corresponding author. Tel.: +81-6-6337-5904; fax: +81-6-6338-
7656; e-mail: hiroshi-harada@dainippon-pharm.co.jp
0960-894X/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00073-8

1302
H. Harada et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1301–1305
side-chain at the 3-position of the indole ring, and per-
formed optical resolution of selected compounds.
Additionally, to develop potent and selective human
b3-AR agonists, we examined adenylyl cyclase activity
using Chinese hamster ovary (CHO) cell lines expressing
human b1-, b2-, and b3-ARs.
The 2-[3-(7-O-substituted 3-indolyl)-2-propylamino]-1-
(3-chlorophenyl)ethanols 24 and 30 as a mixture of dia-
stereomers and the selected optical isomers 23a,b and
30a,b listed in Tables 4 and 5 were synthesized as shown
in Scheme 4.
Structure–activity relationship (SAR) studies of various
2-(3-indolyl)alkylamino-1-(3-chlorophenyl)ethanols led
to the discovery of the optically active 30a, which is a
potent human and rat b3-AR agonist with low activity
for human b1- and b2-ARs.
In this paper, we describe the synthesis and SARs of
3-indolylethanolamine derivatives while keeping the
3-chlorophenyl moiety constant as an aryl group.
Scheme 4. (a) Boc2O; (b) Pd/C, H2, chlorobenzene; (c) ClCH2CO2Me,
K2CO3, KI; (d) aqueous HCl; (e) aqueous NaOH; (f) N,N0-carbonyl-
diimidazole; (g) separation by silica gel column chromatography;
(h) ClCH2CO2Me or MeI.
Chemistry
Protection of the secondary amine functionality of the
7-benzyloxytryptamine 28 with Boc group followed by
catalytic hydrogenation in the presence of chloro-
benzene to avoid removal of the 3-chlorine atom in the
benzene ring produced the 7-hydroxytryptamine 5.
Reaction of 5 with methyl chloroacetate, followed by
removal of the Boc protecting group under acidic con-
ditions furnished the 7-methoxycarbonylmethoxy-
tryptamine 29. Subsequent alkaline hydrolysis of 29 and
acid hydrolysis of 5 gave 30 and 24, respectively.
The requisite intermediate tryptamine derivatives were
prepared using procedures previously described in the lit-
eratures.16 In general, the 3-formylindoles obtained by
Vilsmeier reaction (POCl3, DMF) of the substituted
indoles were treated with nitroalkane in AcOH to produce
nitroolefins in good yield. Conjugated nitroolefins were
directly reduced by LiAlH4 to give saturated primary
amines although the yield was moderate to low (Scheme 1).
Most of the compounds (7–28 except 24) listed in Tables
1, 3, and 4 were prepared by coupling reaction of the
racemic 3-chlorostyrene oxide 1 or its optical isomer17
(R)-1 with various tryptamine derivatives in MeOH
(Scheme 2).
The optically active 23a,b and 30a,b having 7-methoxy-
and 7-carboxylmethoxy groups, respectively (Table 5),
were synthesized as follows. Protection of the amino-
ethanol moiety of 28 with a carbonyl group, followed by
silica gel column chromatography separation of the
resulting diastereomer 6 gave the optical isomers 6a,b
having R- and S-configuration in the a-methyl group,
respectively. A similar method to that described for the
The optically active 8a–d listed in Table 2 were synthesized
by treatment of the optically active 3-chloromandelic
acids18 (R)-2 and (S)-2 with the optically active a-methyl-
tryptamines19 (R)-3 and (S)-3 using BOP (benzotriazol-1-
yloxytris(dimethylamino)phosphonium hexafluorophos-
phate) as a coupling reagent followed by reduction of the
amide group of 4a–d with borane (Scheme 3).
Table 1. Human b3-AR agonistic activity of 2-(3-indolyl)alkylamino-
1-(3-chlorophenyl)ethanols 7–11a
Compd
R
R1
Human b3-AR
Scheme 1. (a) MeNO2 or EtNO2, AcOH; (b) LiAlH4, THF.
EC50 (nM)
(IA%)b
cAMP accumulation
(% at 10ꢀ7 M)c
7d
8e
H
Me
Et
H
H
H
Me
69 (99)
12 (114)
9e
20
0
10d
Me
70 (118)
11e
Scheme 2. (a) MeOH.
ab3-AR agonistic activity was assessed by measuring cAMP accumulation in
CHO cells expressing human b3-AR (150,000 receptors/cell).
bThe maximal amount of cAMP obtained by (ꢀ)-isoproterenol and the amount
of cAMP in the absence of agonists were defined as 100 and 0%, respectively,
and the relative maximal response of each compound is expressed as intrinsic
activity (IA). EC50 value is a concentration of the test compound to be required
to achieve 50% of cAMP accumulation.
cActivity relative to (ꢀ)-isoproterenol.
eMixture of four diastereomers.
dRacemic mixture.
Scheme 3. (a) Benzotriazol-1-yloxytris(dimethylamino)phosphonium
.
hexafluorophosphate (BOP), DMF; (b) BH3 THF.

H. Harada et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1301–1305
1303
Table 2. b-AR agonistic activity of compound 8 and its individual
diastereomers at cloned human b1-, b2-, and b3-ARs and at the cloned
rat b3-ARa
Table 4. Human b3-AR agonistic activity of 7-O-substituted indole
derivatives 23–30a
Compdc
R3
Human b3-AR
Compd Configuration Configuration
of methyl
center
EC50
(nM) (IA%)b
EC50 (nM) (IA%)b
of hydroxy
center
23
24
25
26
27
28
29
30
Me
H
Et
Pr
0.67 (114)
1.7 (128)
0.96 (96)
14 (103)
2.8 (87)
11 (114)
0.92 (102)
0.11 (124)
Human Human Human
b3-AR
Rat
b3-AR
b2-AR b1-AR
8
Mixture
Mixture
12 (114) 23 (46)
NTd 0.97 (107)
iso-Pr
CH2Ph
CH2CO2Me
CH2CO2H
8a
8b
8c
8d
R
R
S
S
R
S
R
S
5.4 (110) 25 (50) 1.9 (65) 0.36 (98)
240 (97) 9.4 (50) 13 (96)
––c
220 (119) 330 (23) 47 (70) 11 (108)
3300 (62) 140 (47) 33 (108)
––c
aSee footnote a in Table 1.
bSee footnote b in Table 1.
ab-ARs agonistic activity were assessed by measuring cAMP accumulation in
CHO cells expressing various b-ARs. Expression levels21 of b-ARs were 150,000
receptors/cell, 30,000 receptors/cell, 12,000 receptors/cell and 880,000 receptors/
cell for human b3-, b2-, b1-, and rat b3-ARs.
cMixture of R,R and R,S diastereomers.
bSee footnote b in Table 1.
c—, could not be calculated because of low IA.
dNT, not tested.
Table 5. b-AR agonistic activity of optically active 7-O-substituted
indole derivatives 23a,b, 30a,b, and reference compounds at cloned
human b1-, b2-, and b3-ARs and at the cloned rat b3-ARa
Table 3. Human b3-AR agonistic activity of substituted indole deri-
vatives 12–22a
Compd
Configuration
of methyl
center
R3
EC50
(nM) (IA%)b
Compd
R
R2
Human b3-AR
Human Human Human
b3-AR b2-AR b1-AR
Rat
b3-AR
EC50 (nM)
(IA%)b
cAMP accumulation
(% at 10ꢀ7 M)c
23a
23b
R
S
R
S
Me
Me
0.36 (89) 5.2 (46) 0.13 (118) 0.15 (147)
c
12d
13d
14d
15e
16d
17e
18d
19e
20e
21e
22e
H
H
H
Me
H
Me
H
Me
Me
Me
Me
1-Me
2-Me
4-Me
4-Me
5
6-Me
7-Me
6-MeO
7-MeO
6-Cl
6
5
20
120 (51)
—
130 (83) 6.5(121)
30a (AJ-9677)
30b
BRL 37344
CH2CO2H 0.062 (116) 13 (26) 6.4 (26) 0.016 (110)
c
CH2CO2H 10 (84)
—
14 (107) 1.2 (106)
21 (95) 290 (31) 1700 (17) 0.095 (109)
96 (165)
-Me
7
aSee footnote a in Table 2.
bSee footnote b in Table 2.
cSee footnote c in Table 2.
12 (95)
0
22 (102)
1.7 (113)
28 (131)
38 (98)
6-Br
aSee footnote a in Table 1.
bSee footnote b in Table 1.
cSee footnote c in Table 1.
dSee footnote d in Table 1.
eSee footnote e in Table 1.
preparation of 30 from 28 was applied to the preparation
of the desired optical isomers 23a,b and 30a,b. The abso-
lute configuration of 6b (S-configuration) and 30a
(R-configuration) was confirmed by X-ray crystal-
lography, and the ORTEP diagram of 30a is shown in
Figure 2.20
Figure 2. The ORTEP drawing of 30a with thermal ellipsoids at 50%
probabilities.
Introduction of a methyl group (yielding 8) into the
tryptamine side-chain of 7 resulted in good improve-
ment in b3-AR agonistic activity (EC50=12 nM, intrin-
sic activity (IA) value of 114%). However, introduction
of an ethyl group (yielding 9) resulted in low activity at
the human b3-AR. The a,a-dimethyltryptamine 10 had
an activity nearly equipotent to that of the parent com-
pound (7), and the tetrahydrocarbazole 11 showed sig-
nificantly poor activity.
Results and Discussion
Activation of b-ARs by the various compounds pre-
pared in this study was assessed by measuring cAMP
accumulation in CHO cells expressing cloned human b1-,
b2-, and b3-ARs and rat b3-AR. As shown in Table 1,
the lead compound 7 exhibited a relatively modest ago-
nistic activity at the human b3-AR (EC50=69 nM).
From the study on the b-ARs agonistic activity of the four
optical isomers of BRL 37344, the (R,R)-configuration

1304
H. Harada et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1301–1305
had proved to be important for enhancing rat b3-AR
agonistic activity.22 Thus, the four optical isomers 8a–d
of the selected compound 8 were prepared and their
agonistic activity at the human b1-, b2-, and b3-ARs and
rat b3-AR was evaluated. As expected, the optical iso-
mer 8a, having R-configuration in both the hydroxy and
a-methyl centers, exhibited a potent agonistic activity at
the human and rat b3-ARs compared with other stereo-
isomers (8b–d). Although 8a had an agonistic activity at
human and rat b3-ARs 2–3 times more potent than that
of the original compound 8, it was completely non-
selective. Other enantiomers (8b–d) showed low activity
at all b-ARs (Table 2).
activity at human and rat b3-ARs with selectivity for
human b3-AR over 100-fold that for the b1-AR and
200-fold that for the b2-AR. Introduction of
a
carboxylmethoxy group into the indole ring of 30a led
to a much more potent agonistic activity at the human
b3-AR and ca. 6-fold increase in activity at the rat
b3-AR as compared to BRL 37344. The presence of the
7-carboxylmethoxy group and the R-configuration for
the a-methyl group were therefore found to be necessary
for potent agonistic activity and selectivity.
In conclusion, the synthesis and SAR studies of sub-
stituted tryptamine derivatives based on human b3-AR
agonistic activity have been discussed. Introduction of a
carboxylmethoxy group at the 7-position of the indole
ring resulted in the identification of a potent human
(EC50=0.062 nM) and rat (EC50=0.016 nM) b3-ARs
full agonist 30a (AJ-9677). Additionally, this compound
(30a) showed good selectivity for human b3-AR as
compared to that for human b1- and b2-ARs.
Next, we examined the agonistic activity of various
substituted tryptamine derivatives. Introduction of a
methyl group at 1-, 2-, 4-, 5-, and 7-positions of the
indole ring of 7 and 8 resulted in low agonistic
activity at the human b3-AR (compounds 12–16 and
18, Table 3). The 6-methyl group was well tolerated
and 17 displayed comparable agonistic activity to
that of its parent 8. The 6-methoxytryptamine 19 and
the tryptamine derivatives 21 and 22 with chlorine
and bromine at the 6-position, respectively, were weak
human b3-AR agonists compared with 8. Fortunately,
the 7-methoxyindole counterpart 20 showed much
more potent agonistic activity than 8. From the above
SAR studies, the 7-methoxytryptamine 20 was found to
exhibit the most preferred agonistic activity at human
b3-AR.
References and Notes
1. Lands, A. M.; Arnold, A.; McAuliff, J. P.; Luduena, F. P.;
Brown, T. G., Jr. Nature 1967, 214, 597.
2. Arch, J. R. S. Proc. Nutrition Soc. 1989, 48, 215.
3. Emorine, L. J.; Marullo, S.; Briend-Sutren, M.-M.; Patey,
G.; Tate, K.; Delavier-Klutchko, C.; Strosberg, A. D. Science
1989, 245, 1118.
4. Tan, S.; Curtis-Prior, P. B. Int. J. Obesity 1983, 7, 409.
5. Granneman, J. G.; Lahners, K. N.; Chaudhry, A. Mol.
Pharmacol. 1991, 40, 895.
6. Muzzin, P.; Revelli, J.-P.; Kuhne, F.; Gocayne, J. D.;
McCombie, W. R.; Venter, J. C.; Giacobino, J.-P.; Fraser,
C. M. J. Biol. Chem. 1991, 266, 24053.
Because the (R)-hydroxy isomers of the hydroxy center
were more potent than the corresponding (S)-hydroxy
derivatives, 7-O-substituted indole analogues with an
(R)-hydroxy group were prepared and tested for their
agonistic activity at the human b3-AR. As shown in
Table 4, the 7-methoxyindole derivative 23 showed an
activity ca. 2.5-fold more potent than that of 20.
Removal of the methyl group on the methoxy substitu-
tion gave derivative 24, which exhibited a decreased
agonistic activity at the human b3-AR. When the meth-
oxy group of 23 was replaced with an ethoxy and
methoxycarbonylmethoxy groups, the resultants 25 and
29 exhibited approximately equal agonistic activity at
human b3-AR. A significant loss in activity was
observed with the 7-propoxy (26), 7-isopropoxy (27),
and 7-benzyloxy (28) derivatives. Replacement of the
methoxy group of 23 by a carboxylmethoxy group
(yielding 30) led to a 6-fold improvement in human
b3-AR agonistic activity.
7. Nahmias, C.; Blin, N.; Elalouf, J.-M.; Mattei, M. G.;
Strosberg, A. D. EMBO. J. 1991, 10, 3721.
8. Lafonate, M.; Berlan, M. J. Lipid Res. 1993, 34, 1057.
9. (a) Weyer, C.; Gautier, J. F.; Danforth, E., Jr. Diabetes
Metab. 1999, 25, 11. (b) Souza, J. C.; Burkey, B. F. Curr.
Pharm. Des. 2001, 7, 1433. (c) Weber, A. E. Annu. Rep. Med.
Chem. 1998, 33, 193.
10. Cantello, B. C. C.; Smith, S. A. Drugs Future 1991, 16,
797.
11. Bloom, J. D.; Claus, T. H. Drugs Future 1994, 19, 23.
12. Cecchi, R.; Croci, T.; Boigegrain, R.; Boveri, S.; Baroni,
M.; Boccardi, G.; Guimbard, J. P.; Guzzi, U. Eur. J. Med.
Chem. 1994, 29, 259.
13. (a) Arch, J. R. S.; Wilson, S. Int. J. Obesity 1996, 20, 191.
(b) Howe, R. Drugs Future 1993, 18, 529.
14. (a) Lipworth, B. J. J. Clin. Pharmacol. 1996, 42, 291. (b)
Strosberg, A. A. Annu. Rev. Pharmacol. Toxicol. 1997, 37, 421.
15. Liggett, S. Mol. Pharm. 1992, 42, 634.
16. (a) 3-(2-Aminopropyl)indole, 3-(2-aminobutyl)indole:
Heinzelman, R. V.; Anthony, W. C.; Lyttle, D. A.; Szmusz-
kovicz, Z. J. Org. Chem. 1960, 25, 1548. (b) Bergman, J.;
Finally, the 7-methoxy and 7-carboxylmethoxyindole
derivatives (23 and 30, respectively) with potent agonis-
tic activity at human b3-AR were selected for diaster-
eomers separation and agonistic activity examination at
human b1-, b2-, and b3-ARs and rat b3-AR. As shown
in Table 5, the optical isomers 23a and 30a with an
(R)-a-methyl group were more potent than the corres-
ponding diastereomers 23b and 30b. However, 23a
exhibited poor selectivity for human b3-AR as it
potently stimulated both human b1- and b2-ARs. On the
other hand, the selectivity of 30a for human b3-AR was
high. The optically active 30a showed a potent agonistic
Backvall, J.-E.; Lindstrom, J.-O. Tetrahedron 1973, 29, 971. 3-
¨
¨
(2-Amino-2-methylpropyl)indole: Snyder, H. R.; Katz, L.
J. Am. Chem. Soc. 1947, 69, 3140. 3-Amino-1,2,3,4-tetra-
hydrocarbazole: Mooradian, A.; Dupont, P. E.; Hlavac, A. G.;
Aceto, M. D.; Pearl, J. J. Med. Chem. 1977, 20, 487.
17. Bloom, J. D.; Dutia, M. D.; Johnson, B. D.; Wissner, A.;
Burns, M. G.; Largis, E. E.; Dolan, J. A.; Claus, T. H. J. Med.
Chem. 1992, 35, 3081.
18. Collet, A.; Jacques, J. Bull. Soc. Chim. Fr. 1973, 3330.

H. Harada et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1301–1305
1305
19. Repk, D. B.; Ferguson, W. J. J. Heterocyclic Chem. 1976,
13, 775.
J=7.8 Hz), 6.94 (1H, d, J=2.0 Hz), 6.99 (1H, d, J=7.8 Hz),
7.27–7.33 (3H, m), 7.46 (1H, s), 11.02 (1H, s).
29
20. Mp 225–226 ꢂC (decomp.) (aqueous NH3/MeOH). ½aꢃD
21. Takeda, Y.; Chou, K. B.; Takeda, J.; Sachais, B. S.;
Krause, J. E. Biochem. Biophys. Res. Comm. 1991, 179, 1232.
22. Oriowo, M. A.; Chapman, H.; Kirkham, D. M.; Sennitt,
M. V.; Ruffolo, R. R.; Cawthorne, M. A. J. Pharmacol. Exp.
Ther. 1996, 277, 22.
ꢀ23.3ꢂ (c 1.0, 1 mol NaOH). IR (KBr) nmax; 3134, 1609, 1406,
1250, 1057, 783 cmꢀ1 1H NMR (dimethylsulfoxide-d6); 0.93
.
(3H, d, J=6.0 Hz), 2.5–3.3 (8H, m), 4.55 (2H, s), 4.85 (1H, dd,
J=9.8, 2.6 Hz), 6.42 (1H, d, J=7.8 Hz), 6.78 (1H, t,