Indoloxypropanolamine analogues as 5-HT1A receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2007.07.086

Analogues of pindolol, 1-(1H-indol-4-yloxy)-3-isopropylamino-propan-2-ol, were synthesized and evaluated as 5-HT_1A receptor antagonists. The structural features required for optimal binding to the 5-HT1A receptor are as follows: S-2-propanol linker, 4-indoloxy substituent, and a large lipophilic cyclic amine substituent.

Full text of DOI:10.1016/j.bmcl.2007.07.086

Bioorganic & Medicinal Chemistry Letters 17 (2007) 5600–5604
Indoloxypropanolamine analogues as 5-HT1A receptor antagonists
Joseph H. Krushinski, Jr.,* John M. Schaus, Dennis C. Thompson, David O. Calligaro,
David L. Nelson, Susan H. Luecke, David B. Wainscott and David T. Wong
Lilly Research Laboratories, Lilly Corporate Center, Indianapolis, IN 46285, USA
Received 7 June 2007; revised 26 July 2007; accepted 27 July 2007
Available online 22 August 2007
Abstract—Analogues of pindolol, 1-(1H-indol-4-yloxy)-3-isopropylamino-propan-2-ol, were synthesized and evaluated as 5-HT1A
receptor antagonists. The structural features required for optimal binding to the 5-HT1A receptor are as follows: S-2-propanol lin-
ker, 4-indoloxy substituent, and a large lipophilic cyclic amine substituent.
Ó 2007 Elsevier Ltd. All rights reserved.
Fluoxetine (ProzacÒ) has been a breakthrough therapy
in the fight against depression. In spite of this, depres-
sion continues to be a prevalent disease and is predicted
to be the second leading cause of illness-induced disabil-
ity in the world by 2020.1 The SSRIs exert their pharma-
cological action (elevation of synaptic serotonin levels)
in hours2 but require prolonged administration before
significant clinical improvement occurs.3 One proposal
for this delay is that these elevated serotonin levels cause
a negative feedback at the presynaptic 5-HT1A autore-
ceptors. This delayed onset of action could be overcome
by antagonizing the action of 5-HT at the presynaptic 5-
HT1A autoreceptors.4
the 5-HT1A receptor affinity and antagonist function.
Herein we describe the effect of the following five types
of structural change on activity: chain length, position
of attachment of the indole ring, role of the hydroxyl
group, alkylation of the C-2 position, and modification
of the alkyl amine. The compounds were studied as race-
mates and enantiomers of the most interesting com-
pounds were prepared to evaluate the importance of
stereochemistry. We report the synthesis, 5-HT1A recep-
tor binding affinities, and the intrinsic efficacy at the 5-
HT1A receptor. No attempt was made to modify the
b-adrenergic affinity of these compounds.
The regioisomeric 4- and 5-substituted 1-(1H-indolyl-
oxy)-3-(alkylamino)-2-propanols were prepared according
to Scheme 1. The 4-hydroxyindole 1 and 5-hydroxyin-
dole 2 were alkylated with epichlorohydrin9 to give the
intermediate epoxides, 3 and 4. The epoxide ring was
opened with the appropriate amine to give the indoloxy-
propanolamines (5a–e, 6a,b).
The indoloxypropanolamine b-adrenoceptor antagonist
pindolol is also a 5-HT1A receptor ligand. Pindolol
antagonizes several actions mediated by central 5-
HT1A receptors such as hypothermia5 and hormone
secretion.6 Pindolol has been reported to enhance and
accelerate the clinical effects of antidepressant drugs in
several placebo-controlled clinical trials.7 Pindolol has
also been reported to be a partial agonist at 5-HT1A
receptors. While other medicinal chemistry studies8 have
been performed on pindolol, our goal was to alter the
pindolol structure in such a way as to maximize both
Scheme 1. Reagents and conditions: (a) 5 N NaOH, epichlorohydrin,
60 °C, 1 h (3 (4-isomer), 68% and 4 (5-isomer), 83%); (b) RNH2,
MeOH, reflux (5a, R = cyclohexyl, 79%; 5b, R = 1-adamantyl, 63%;
5c, R = cyclopentyl, 34%; 5d, R = cycloheptyl, 37%; 5e, R = 3-pentyl,
77%; 6a, R = cyclohexyl, 82%; 6b, R = 1-adamantyl, 76%).
Keywords: 5-HT1A antagonist; Serotonin; Pindolol.
*
Corresponding author. Tel.: +1 317 276 8261; fax: +1 317 276
7600; e-mail: krush@lilly.com
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.07.086

J. H. Krushinski et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5600–5604
5601
The 1-(1H-indol-4-yloxy)-3-(alkylamino) propanes were
prepared according to Scheme 2. The 4-hydroxyindole 1
was alkylated with 3-chloro-1-propanol under Mitsun-
obu conditions to give the intermediate chloride 7. The
chloride was displaced with the appropriate amine to
give the indoloxypropanamines 8a,b.
The (S)-1-(1H-indol-4-yloxy)-2-methyl-3-(alkylamino)-
2-propanols were prepared according to Scheme 5.
(2S)-(+)-methylglycidyl-3-nitrobenzene sulfonate was
prepared from (2R)-methylglycidol. The 4-hydroxyin-
dole 1 was alkylated with (2S)-(+)-methylglycidyl 3-
nitrobenzene sulfonate to give the intermediate epoxide
15.10 The epoxide ring was opened with the appropriate
amine to give the indoloxypropanolamines 16a,b.
The 1-(1H-indol-4-yloxy)-2-methoxy-3-(alkylamino) pro-
panes were prepared according to Scheme 3. Dimethyl
methoxymalonate was reduced with LAH to give
2-methoxy-1,3-propanediol in 54% yield and the hydro-
xyl groups were converted to the tosylates to give 2-
methoxy-1,3-diol bis(4-methylbenzenesulfonate) in 97%
yield. The hydroxyindole 1 was alkylated with 2-meth-
oxy-1,3-diol bis(4-methylbenzenesulfonate) to give the
intermediate tosylate 9. The tosylate was displaced with
the appropriate amine to give the indoloxymethoxypro-
panamines 10a,b.
The 4-(alkylamino)-1-(1H-indol-4-yloxy)-3-butanols were
prepared according to Scheme 6. The 4-hydroxyindole 1
was alkylated with 1,2-epoxy-4-butanol under Mitsun-
obu conditions to give the intermediate epoxide 17.
The epoxide ring was opened with the appropriate
amine to give the indoloxybutanolamines 18a–c.
The 1-(1H-indol-4-yloxy)-4-(alkylamino)-2-butanols were
prepared according to Scheme 7. The 4-hydroxyindole 1
was alkylated with 1,4-dichloro-2-butanol to give the
intermediate chloride 19. The chloride was displaced
with the appropriate amine to give the indoloxybutanol-
amines 20a,b.
The (S)- and (R)-1-(1H-indol-4-yloxy)-3-(alkylamino)-2-
propanols were prepared according to Scheme 4 (stereo-
chemistry in Scheme 4 is depicted for the S enantiomer).
The 4-hydroxyindole 1 was alkylated with (2S)-(+)-glyc-
idyl 3-nitrobenzene sulfonate or (2R)-(À)-glycidyl
3-nitrobenzene sulfonate to give their respective inter-
mediate epoxides, 11 and 12.10 The epoxide ring was
opened with the appropriate amine to give the indoloxy-
propanolamines (13a–c and 14a,b).
The 5-HT1A receptor binding assays and the [35S]GTPcS
binding assays were performed according to Rasmussen
et al.11
The importance of the amine substituent in modulation
of 5-HT1A receptor affinity can be readily seen in Table 1.
Increasing the size of the substituent from isopropyl
Scheme 2. Reagents and conditions: (a) diethyl azodicarboxylate,
PPh3, Cl(CH2)3OH, THF, 57%; (b) RNH2, Na2CO3, DMF, 100 °C,
(R = cyclohexyl, 43% and R = cyclopentyl, 28%).
Scheme 5. Reagents and conditions: (a) NaH, (2S)-(+)-methylglycidyl
3-nitrobenzene sulfonate, DMF, 45%; (b) RNH2, MeOH, reflux
(R = cyclohexyl, 85% and R = 1-adamantyl, 92%).
Scheme 3. Reagents and conditions: (a) NaH, 2-methoxypropane-1,3-
diol bis(4-methylbenzenesulfonate), DMF, 91%; (b) RNH2, DMF,
85 °C (R = cyclohexyl, 47%; R = 1-adamantyl, 54%).
Scheme 6. Reagents and conditions: (a) diethyl azodicarboxylate,
PPh3, 1,2-epoxy-4-butanol, THF, 27%; (b) RNH2, MeOH, reflux
(R = cyclohexyl, 77%; R = 1-adamantyl, 75%; R = isopropyl, 12%).
Scheme 4. Reagents and conditions: (a) NaH, (2S)-(+)-glycidyl 3-
nitrobenzene sulfonate or (2R)-(À)-glycidyl 3-nitrobenzene sulfonate,
DMF, S-isomer 88%; (b) RNH2, MeOH, reflux (13a, R = cyclohexyl,
63%; 13b, R = 1-adamantyl, 70%).
Scheme 7. Reagents and conditions: (a) KOH, 1,4-dichloro-2-butanol,
H2O, 65 °C, 38%; (b) RNH2, IPA, reflux (R = cyclohexyl, 90% and
R = 1-adamantyl, 49%).

5602
J. H. Krushinski et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5600–5604
(pindolol = 22.4 nM) to the 3-pentyl (5e = 28.6 nM)
maintained 5-HT1A receptor affinity. But amine substi-
tution with a carbocyclic ring system (5a = 3.58 nM,
5b = 9.29 nM, 5c = 10.6 nM, and 5d = 6.82 nM) gave a
marked increase in binding affinity for the 5-HT1A
receptor. With this information the cyclohexyl and 1-
adamantyl amine substituents became the focus of the
subsequent SAR.
acceptor properties of the hydroxyl group are impor-
tant, although the steric bulk of the methoxy group
may attenuate the 5-HT1A affinity. These modifications
also decrease 5-HT1A receptor antagonist activity.
The role of the stereochemical configuration of the hy-
droxyl group was explored. The affinity of the S isomers
(Table 2, 13a = 1.33 nM and 13b = 3.8 nM) was superior
to the
R isomers (Table 2, 14a = 26.1 nM and
The hydroxyl group is important for potent 5-HT1A
receptor affinity. Removal of the hydroxyl group
(Table 1, 8a = 78 nM and 8b = 94 nM) diminished
5-HT1A receptor affinity 8 to 21-fold and alkylation
of the hydroxyl group (Table 1, 10a = 40.6 nM and
10b = 32 nM) resulted in a 3 to 10-fold loss in affinity.
This suggests that both the hydrogen bond donor and
14b = 18.3 nM) and to the racemates (Table 1, 5a =
3.58 nM and 5b = 9.29 nM). The S isomers were substi-
tuted with a methyl group at the 2 position creating a
quaternary center (16a and 16b). This steric bulk led
to a significant loss in receptor affinity. All these com-
pounds showed a similar low degree of partial agonist
activity.
Table 1. 5-HT1A binding affinities and functional data of hydroxyl group modifications
Compound
A
R
5-HT1A Kia (nM)
Inhibition of 5-HT-stimulated
GTPcS binding
c
Emin (%)
Kib (nM)
Pindolol
5a
5b
OH
OH
OH
OH
OH
OH
H
Isopropyl
22.4 3.8
3.58 0.25
9.29 1.53
10.6 2.9
6.82 0.59
28.6 6.1
78.4 8.7
94.3 15.7
40.6 9.6
32.0 5.8
81.1 9.88
9.83 1.83
7.84 0.76
35.5 3.77
11.3 0.63
49.8 3.19
ndd
9.76 1.37
9.88 0.68
6.56 0.83
11.8 1.08
8.90 1.41
4.74 0.72
31.3 1.5e
24.8 0.22
36.2 2.1e
28.9 1.1e
Cyclohexyl
1-Adamantyl
Cyclopentyl
Cycloheptyl
3-Pentyl
5c
5d
5e
8a
Cyclohexyl
Cyclopentyl
Cyclohexyl
1-Adamantyl
8b
H
233 28.7
ndd
ndd
10a
10b
OCH3
OCH3
a Affinities were determined in vitro by radioligand binding using cell lines expressing the human 5-HT1A receptor. Each value is means SEM of at
least three determinations.
b Ki calculated from the inhibition of 300 nM 5-HT mediated stimulation of [35S]GTPcS binding in mouse LM(tkÀ) cells expressing the human
5-HT1A receptor.
c Degree of maximal inhibition of 300 nM 5-HT mediated stimulation of [35S]GTPcS binding expressed as % of the stimulation produced by 10 lM
5-HT.
d Ki not determined.
e Degree of inhibition of 300 nM 5-HT mediated stimulation of [35S]GTPcS binding at 10 lM, the maximal concentration of compound that could be
accurately tested. Data expressed as % of the stimulation produced by 10 lM 5-HT.
Table 2. 5-HT1A binding affinities and functional data of hydroxyl group stereochemistry and C2 substitution
Compound
A
B
R
5-HT1A Kia (nM)
Inhibition of 5-HT-stimulated
GTPcS binding
c
Emin (%)
Kib (nM)
13a
13b
13c
14a
14b
16a
16b
OH
OH
OH
H
H
Cyclohexyl
1.33 0.14
3.89 0.58
11.2 0.2
26.1 1.6
18.3 4.9
1450 300
868 75
6.68 1.05
4.65 0.57
24.0 3.45
50.8 4.94
29.9 5.91
nt
7.52 0.36
6.38 0.64
8.75 1.75
8.87 1.48
3.15 0.68
nt
H
1-Adamantyl
2-Adamantyl
Cyclohexyl
H
OH
OH
CH3
CH3
H
1-Adamantyl
Cyclohexyl
1-Adamantyl
OH
OH
nt
nt
See Table 1 legend (nt, not tested).

J. H. Krushinski et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5600–5604
Table 3. 5-HT1A binding affinities and functional data of chain length modification
5603
f
5-HT1A EC50 (nM)
g
5-HT1A Emax (%)
Compound
m
n
R
5-HT1A Kia (nM)
18a
18b
18c
20a
20b
2
2
2
1
1
1
1
1
2
2
Isopropyl
1772 396
872
nt
nt
Cyclohexyl
1-Adamantyl
Cyclohexyl
1-Adamantyl
6
nt
nt
551 97
125 14
76.1 8.0
nt
nt
394 16
854 110
45.6 1.3
51.8 1.1
See Table 1 legend (nt, not tested).
f EC50 for stimulation of [35S] GTPcS binding in mouse LM(tkÀ) cells expressing the human 5-HT1A receptor.
g Maximum stimulation of [35S] GTPcS binding expressed relative to the maximal effect of 5-HT.
nol derivatives (Table 3, 18a = 1772 nM, 18b = 872 nM,
and 18c = 551 nM) and 2-butanol derivatives (Table 3,
20a = 124.7 nM and 20b = 76.1 nM) suffered a signifi-
cant loss in affinity for the 5-HT1A receptor. An increase
in agonist efficacy was observed in the 2-butanol
derivatives.
Table 4. 5-HT1A binding affinities of the positional isomers of indole
Compound
Ar
R
5-HT1A Kia (nM)
5a
5b
6a
6b
4-Indole
4-Indole
5-Indole
5-Indole
Cyclohexyl
1-Adamantyl
Cyclohexyl
1-Adamantyl
3.58 0.25
9.29 1.53
>3000
The 4-indole substituent was replaced with the 5-indole
substituent. The 5-indole derivatives (Table 4,
6a = >5000 nM and 6b = 833 nM) were poor ligands at
the 5-HT1A receptor.
833 30
See Table 1 legend.
Table 5. Blockade of 8-OH-DPAT-induced increase in rat serum
corticosterone concentrations by compound 13a
We have shown that the optimal structural features for
potent 5-HT1A receptor affinity are the 4-indole substi-
tuent, the 2-propanol linker, the S configuration of the
hydroxyl group, and larger, lipophilic substituent on
the amine. Compounds 13a and 13b represent these
requirements. Compounds 13a and 13b show a low de-
gree of partial agonist efficacy, Emin = 7.52% and
6.38%, respectively.
Treatment
Corticosterone,
ng/ml
Percent
antagonism
Vehicle
33.6 2.4
398.4 13.4
—
—
8-OH-DPAT alone
13a + 8-OH-DPAT
0.03 mg/kg, sc
0.1 mg/kg, sc
0.3 mg/kg, sc
1.0 mg/kg, sc
350.4 15.4
304.8 46.1
206.0 60.6
62.4 9.6
13
26*
53*
92*
Compound 13a, upon further in vivo evaluation,
blocked the 8-OH DPAT-induced increase in rat serum
corticosterone concentrations with an ED50 = 2.5 mg/
kg, sc (Table 5).12
Male Sprague–Dawley rats were administered compound 13a subcu-
taneously (sc) 15 min prior to 0.3 mg/kg, sc 8-OH-DPAT, and rats
were sacrificed 1 h after 8-OH-DPAT. Serum corticosterone levels were
measured using H-corticosterone kits (ICN Biomedicals, Costa Mesa,
CA) according to kit instructions. Means and standard errors for five
rats per group are shown.
Profiling of compounds 13a and 13b at the human b1
and b2 receptors13 showed comparable receptor affinity
to pindolol (Table 6).
* P 6 0.05.
In conclusion, we have developed compound 13a, which
is a potent antagonist at the 5-HT1A receptor. This com-
pound is not a candidate for further development be-
cause of its affinity for the b1 and b2 receptors. This
platform is an appropriate starting place to develop a
compound which is devoid of b affinity but retains the
5HT1A affinity and antagonism.
Table 6. b1 and b2 receptor binding affinities of selective compounds
Compound
b1, Kia (nM)
b2, Kia (nM)
Pindolol
13a
13b
0.52
2.37
7.92
0.40
3.75
6.63
a Affinities were determined in vitro by radioligand binding using cell
lines expressing the human b receptors. Each value is the mean of two
determinations with a 95% confidence interval.
References and notes
1. Murray, C. J. L.; Lopez, A. D. Lancet 1997, 349, 1498.
2. (a) Perry, K. W.; Fuller, R. W. Life Sci. 1992, 50, 1683; (b)
Bymaster, F. P.; Zhang, W.; Carter, P. A.; Shaw, J.;
Chernet, E.; Phebus, L.; Wong, D. T.; Perry, K. W.
Psychopharmacology 2002, 160, 353.
The effect of the chain length on receptor affinity was
investigated. A methylene unit was inserted on either
the indoloxy side of the 2-propanol chain to give the
3-butanol analogues or on the amine side of the 2-pro-
panol chain to give the 2-butanol analogues. The 3-buta-
3. (a) Pinder, R. M. Pharma News 1996, 3, 11; (b) Blier, P.;
de Montigny, C. Trends Pharmacol. Sci. 1994, 15, 220; (c)

5604
J. H. Krushinski et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5600–5604
Asberg, M.; Eriksson, B.; Martensson, B.; Traskman-
Bendz, L.; Wagner, A. J. Clin. Psychiatry 1986, 47(suppl.),
23.
8. (a) van Niel, M.; Beer, M. S.; Castro, J. L.; Cheng, S. K.
F.; Evans, D. C.; Heald, A.; Hitzel, L.; Hunt, P.;
Mortishire-Smith, R.; O’Conner, D. O.; Watt, A. P.;
MacLeod, A. M. Bioorg. Med. Chem. Lett. 1999, 9, 3243;
(b) Shvedov, V. I.; Safonova, O. A.; Bulanova, T. I.; Il’ina,
A. S.; Shvarts, G. Ya.; Ozhakov, S. D.; Mashkovskii, M.
D. Khim.-Farm.Zh. 1980, 14, 67.
9. Farina, J. S.; Jackson, S. A.; Cummings, C. L. Org. Prep.
Proced. Int. 1989, 21, 173.
10. Klunder, J. M.; Onami, T.; Sharpless, K. B. J. Org. Chem.
1989, 54, 1295.
11. Rasmussen, K.; Calligaro, D. O.; Czachura, J. F.;
Dreshfield-Ahmad, L. J.; Evans, D. C.; Hemrick-Luecke,
S. K.; Kallman, M. J.; Kendrick, W. T.; Leander, J. D.;
Nelson, D. L.; Overshiner, C. D.; Wainscott, D. B.;
Wolff, M. C.; Wong, D. T.; Branchek, T. A.; Zgombick,
J. M.; Xu, Y.-C. J. Pharmacol. Exp. Ther. 2000, 294, 688.
12. Hemrick-Luecke, S. K.; Evans, D. C. Neuropharmacology
2002, 42, 162.
4. Artigas, F. Trends Pharmacol. Sci. 1993, 14, 262.
5. (a) Hadrava, V.; Blier, P.; Dennis, T.; Ortemann, C.; de
Montigny, C. Neuropharmacology 1995, 34, 1311; (b)
Scott, P. A.; Chou, J. M.; Tang, H.; Frazer, A. J.
Pharmacol. Exp. Ther. 1994, 270, 198; (c) O’Connell, M.
T.; Saran, G. S.; Curzon, G. Br. J. Pharmacol. 1992, 106,
603.
6. (a) Koenig, J. I.; Gudelsky, G. A.; Meltzer, H. Y. Eur.
J. Pharm. 1987, 137, 1; (b) Gilbert, F.; Brazell, C.;
Tricklebank, M. D.; Stahl, S. M. Eur. J. Pharm. 1988, 147,
431; (c) Lesch, K. P.; Sohnle, K.; Poten, B.; Schoellnham-
mer, G.; Rupprecht, R.; Schulte, H. M. J. Clin. Endocri-
nol. Metab. 1990, 70, 670.
7. (a) Artigas, F.; Celada, P.; Laruelle, M.; Adell, A. Trends
Pharmacol. Sci. 2001, 22, 224; (b) Artigas, F.; Romero, L.;
de Montigny, C.; Blier, P. Trends Neurosci. 1996, 19, 378;
(c) Artigas, F.; Perez, V.; Alvarez, E. Arch. Gen. Psychi-
atry 1994, 51, 248.
13. Bylund, D. B.; Snyder, S. H. Mol. Pharmacol. 1976, 12,
568.