Evaluation of isotryptamine derivatives at 5-HT2 serotonin receptors

Article abstract of DOI:10.1016/S0960-894X(01)00713-2

On the basis that meta-chlorophenylpiperazine (mCPP; 1) is a nonselective 5-HT_2C agonist, that benz-fused tryptamines (e.g., 5) display enhanced 5-HT₂ affinity, and that certain isotryptamines 3 reportedly bind with enhanced affinity and selectivity at 5-HT_2C receptors, we prepared and examined a series of isotryptamine-related analogues as potentially selective 5-HT_2C agonists. None of the compounds displayed selectivity for 5-HT_2C versus 5-HT_2A receptors. Detailed re-examination of a compound previously reported to display 100-fold 5-HT_2C selectivity [i.e., S(+)-5,6-difluoro-α-methylisotryptamine] revealed that its selectivity versus 5-HT_2A receptors was, at best, only 10-fold.

Full text of DOI:10.1016/S0960-894X(01)00713-2

Bioorganic & Medicinal Chemistry Letters 12 (2002) 155–158
Evaluation of Isotryptamine Derivatives at 5-HT2
Serotonin Receptors
a
a
a
b
Jean Chang-Fong, James Addo, Ma•gorzata Dukat, Carol Smith,
Nicholas A. Mitchell, Katharine Herrick-Davis, Milt Teitler
b
b
b
a,
and Richard A. Glennon *
aDepartment of Medicinal Chemistry, School of Pharmacy, Virginia Commonwealth University, Richmond, VA 23298, USA
bCenter for Neuropharmacology and Neuroscience, Albany Medical College, Albany, NY 12208, USA
Received 26 September 2001; accepted 15 October 2001
Abstract—On the basis that meta-chlorophenylpiperazine (mCPP; 1) is a nonselective 5-HT2C agonist, that benz-fused tryptamines
(
e.g., 5) display enhanced 5-HT affinity, and that certain isotryptamines 3 reportedly bind with enhanced affinity and selectivity at
2
5
-HT2C receptors, we prepared and examined a series of isotryptamine-related analogues as potentially selective 5-HT2C agonists.
None of the compounds displayed selectivity for 5-HT2C versus 5-HT2A receptors. Detailed re-examination of a compound pre-
viously reported to display 100-fold 5-HT2C selectivity [i.e., S(+)-5,6-difluoro-a-methylisotryptamine] revealed that its selectivity
versus 5-HT2A receptors was, at best, only 10-fold. # 2002 Elsevier Science Ltd. All rights reserved.
9
The identification of 5-HT -selective serotonergic
2
fied. More recently, Bo
¨
s et al. reported that certain
C
ligands—particularly 5-HT -selective agonists—has
C
been an elusive goal. 5-HT receptors have been impli-
2
isotryptamine derivatives (e.g., 3) display upwards of
100-fold selectivity for 5-HT2C versus 5-HT2A receptors.
Furthermore, certain pyrazino[1,2-a]indoles (e.g., 4),
which might be generally viewed as conformationally-
restricted analogues of mCPP with an embedded iso-
tryptamine moiety, have been identified as somewhat
2
cated in certain central disorders including depression,
schizophrenia, anxiety, and eating disorders.
1ꢀ3
How-
ever, agents with high affinity for 5-HT2C receptors
typically possess high affinity for 5-HT2A receptors and
rarely display >10-fold selectivity. Very few 5-HT -
6
selective 5-HT2C partial agonists.
2
C
selective agonists are known (e.g., refs 3 and 4). meta-
Chlorophenylpiperazine (mCPP; 1), although not
necessarily selective for 5-HT2 versus certain other
populations of 5-HT receptors, is an agent that displays
about 10-fold selectivity for 5-HT2C receptors. Because
mCPP behaves not as a 5-HT agonist but as a 5-HT2A
2
C
5
antagonist, it might represent an attractive lead for the
development of novel agents despite its low selectivity.
In fact, others have already reported this strategy.6
There exists some remote structural similarity between
7
We have previously demonstrated that benz-fusion of
tryptamines results in compounds with high affinity,
although not necessarily enhanced selectivity, for 5-HT2C
mCPP (1) and the tryptamines. Tryptamine derivatives,
versus 5-HT2A
C
though, lack selectivity for 5-HT
1
2
,2
receptors. We have previously shown that iso-
8
tryptamines (e.g., 2) bind at 5-HT receptors. Their
receptors (e.g., 5, 5-HT
10
K =9 nM; 5-HT
i
nM). Because it is claimed that certain isotryptamines 3
are 5-HT -selective, the present study employed the
K =3
2A i
2
C
2
subpopulation selectivity was not investigated at that
time because 5-HT2C receptors had not yet been identi-
2
C
strategy that if ring-fusion of tryptamines results in high-
affinity 5-HT ligands, and if the isotryptamine nucleus
imparts 5-HT2C selectivity, a ring-fused isotryptamine
2
(
i.e., 6) might display enhanced 5-HT2C selectivity. Also,
*Corresponding author. Tel.:+1-804-828-8487; fax:+1-804-828-7404;
e-mail: glennon@hsc.vcu.edu
given that pyrazino[1,2-a]indoles 4 possess an iso-
0
960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00713-2

156
J. Chang-Fong et al. / Bioorg. Med. Chem. Lett. 12 (2002) 155–158
tryptamine moiety and are structurally related to mCPP
1), we additionally explored several ring-substituted
5-chloro-3-methoxyindole-2-carboxylate was obtained,
the remainder of the synthesis followed the procedure
(
1
3
10
analogues of 4 incorporating substituents that might
mimic the lipophilic or electronic effects of a fused
ring. In particular, we prepared the 6-, 7-, and 8-chloro
and the 6-, 7-, and 8-methoxy derivatives of 4 (e.g.,
analogues 7).
shown in Scheme 1. Compounds 9 and 2a–2c were
prepared by literature methods. 5,6-Difluoroindole (mp
ꢁ
96–97 C from hexanes), required for the synthesis of 2d
and 10, was prepared by catalytic reduction of 3,4-
difluoro-6-nitrophenylacetonitrile. The indole was ela-
borated to 2d by treatment with NaH/HMPA and then
2-chloro-N,N-dimethylethylamine hydrochloride. Com-
pound 10 was obtained from 5,6-difluoroindole as pre-
viously described.
Chemistry
9
Ethyl benz[g]indole-2-carboxylate was prepared by the
1
1
method of Rydon and Siddappa except that hydro-
chloric acid was used in place of H SO to effect cycli-
zation; the ester was hydrolyzed to the acid, and the
2
4
Results and Discussion
ꢁ
resulting acid was decarboxylated by heating at 210 C
1
Compound 6 was found to bind both at 5-HT2A
(K =38ꢂ9 nM) and 5-HT (K =42ꢂ4 nM) receptors.
0,11
to afford benz[g]indole (11).
i
2C
i
Although 5-HT affinity was high, and higher than
A
2
9
Compound 11 was allowed to react with propylene
oxide to yield a 10:1 mixture of 12 and 13 which was
separated by column chromatography (10% EtOAc in
hexanes; silica gel) (Scheme 1). Reaction of 12 and 13
with diphenylphosphoryl azide (DPPA)/DBU gave the
expected azides as oils. Catalytic reduction (10% Pd/C)
nearly all of the isotryptamines already reported,
selectivity was lacking. Interestingly, however, com-
pound 8, a positional isomer of 6, was found to bind
with greatly reduced affinity at both populations of
receptors (5-HT2A K =4300ꢂ1130 nM; 5-HT2C
i
K >10,000 nM). The low affinity of 8 relative to 6 sug-
i
of the azides afforded the target amines 6 (oxalate salt,
ꢁ
gests a possible steric interaction involving the b-methyl
group that alters the conformation of the side chain.
The individual optical isomers of 6 and 8 should be
evaluated in future studies to determine the effect of
stereochemistry on binding.
mp 208–209 C from MeOH/Et O) and 8 (0.75 oxalate,
2
ꢁ
mp 123–124 C from MeOH/Et O), respectively. Com-
2
pound 14 was alternatively prepared in >90% yield by
conversion of 12 to its mesylate, followed by reaction
with sodium azide.
With few exceptions, the pyrazino[1,2-a]indoles 7 pos-
sessed low affinity at 5-HT2A and 5-HT2C receptors
(Table 1). At 5-HT2A receptors, only the 7-chloro deri-
vative 7c displayed appreciable affinity. Because 5-HT2A
agonists tend to show higher affinity for sites labeled by
The pyrazino[1,2-a]indoles 7 were prepared by the gen-
1
2
eral method of Rajur et al. (Scheme 2). Compound 7h
6
¨
has been reported by Bos et al.; certain of the other
1
2
derivatives had also been reported but their melting
ꢁ
3
points were 60–100 C lower than that found in the
present study (see Table 1).
an agonist radioligand (e.g., [ H]DOB) than by an
3
antagonist radioligand (e.g., [ H]ketanserin), 5-HT
2
A
sites were labeled with both radioligands (designated
herein as 5-HT2A(D) and 5-HT2A(K), respectively, in
Table 1) in order to obtain a more realistic perspective
of their affinity. Although somewhat higher affinity was
observed in each case (Table 1), the pyrazino[1,2-a]-
indoles still bind only with modest affinity. The pyr-
azino[1,2-a]indoles also displayed relatively low affinity
at 5-HT2C receptors, and showed no selectivity for 5-
The synthesis of compound 7h was initiated using the
6
method of Bo
¨
s et al.; however, once the necessary ethyl
HT2C relative to 5-HT2A receptors. Compound 7h has
9
been previously shown by Bo
¨
s et al. to be a 5-HT
2
C
partial agonist with about 8-fold selectivity for 5-HT2C
receptors. They have commented that the presence of
the 10-methoxy group contributes to affinity and 5-
HT2C selectivity; the present results (comparing 7h with
7
d) confirm their prior claim that the methoxy group
9
plays a role in binding. Data in Table 1 also show that
replacement of the 10-methoxy group by a methyl
ꢁ
Scheme 1. (a) Propylene oxide, THF, 0 C! room temperature, 16 h;
ꢁ
group (i.e., 7i) is tolerated, but that replacement of C10
(
(
b) DPPA/DBU, toluene, 0 C! reflux 16 h; (c) (i) H
2
, 10% Pd/C;
ii) oxalic acid.
with a nitrogen atom (i.e., 9) decreases affinity.
In an attempt to further determine why the pyr-
azino[1,2-a]indoles bind with low selectivity, we focused
on the structurally simpler isotryptamines 2. Although
compounds 2a and 2c bind with modest affinity, they
lack 5-HT2C selectivity (Table 1). There are several
possible explanations for the low affinity and selectivity
2 4 2
Scheme 2. (a) (i) NaH; (ii) ClCH CN, DMF; (b) (i) LiAlH /Et O; (ii)
oxalic acid or HCl.

J. Chang-Fong et al. / Bioorg. Med. Chem. Lett. 12 (2002) 155–158
157
Table 1. 5-HT2A and 5-HT2C binding data for compounds examined in this investigation
Mp ( C)a
ꢁ
K
i
, (nM) (ꢂSEM)
5-HT2A(D)b
5-HT2A
selectivityc
X
Y
5
-HT2A(K)b
5-HT2C
7
7
7
7
7
7
7
a
b
c
d
e
f
H
6-Cl
7-Cl
8-Cl
6-OMe
7-OMe
8-OMe
H
H
H
H
H
H
H
233d
223–225
221–222
230e
1040
630
58
1670
1000
5190
5830
(30)
(40)
(18)
(190)
(60)
635
330
80
320
300
1300
930
(75)
(40)
(10)
(50)
(20)
(120)
(20)
4780
520
560
1460
1990
5160
9930
(390)
(40)
7.5
1.6
7.0
4.6
6.6
4.0
10.7
(100)
(250)
(160)
(420)
(200)
225–227
204–206
(800)
(720)
f
g
238
7
7
9
h
i
8-Cl
8-Cl
—
OMe
Me
—
—
—
—
130
1250
>10,000
(10)
(250)
60
—
>10,000
(15)
330
450
>10,000
(50)
(20)
5.5
—
—
g
2
2
2
2
a
b
c
H
—
—
—
—
—
—
—
600
>10,000
1040
(80)
650
>10,000
—
230
15
(170)
720
>10,000
490
(80)
1.1
5-OMe
6-OMe
5-F,6-F
g
(100)
(90)
(100)
(20)
—
1.3
h
d
206–207
640
(60)
(2)
300
10
5-F,6-F
—
—
320
(50)
85
(15)
5.7
aAll new compounds (i.e., those for which melting points are provided), and 7a, 7d, and 7g were submitted for microanalysis and determined values
for C, H, and N were within 0.4% of theory. Analogues 7 were recrystallized from aqueous EtOH. All compounds were isolated as their oxalate salts
except for 7h (HCl), 9 (di HCl), and 10 (fumarate).
b
3
3
5
-HT2A(K) and 5-HT2A(D) are 5-HT2A sites labeled by [ H]ketanserin and [ H]DOB, respectively.
value/5-HT2A(D) value.
Lit. mp 130 C; also reported as HCl salt.
c
Selectivity=5-HT2C
K
i
K
i
d
12
ꢁ
ꢁ
7
e
12
Lit. mp 162 C.
f
; lit.12 mp 180 C for oxalate salt.
ꢁ
Present compound is C11H11ClN .0.75C H O
2 2 2 4
g
Binding data not obtained.
Compound recrystallized from MeOH.
h
3
observed with these derivatives. One of the highest affi-
nity and most 5-HT -selective agents reported by Bo
tors and [ H]5-HT to label 5-HT receptors, whereas,
2C
3
s
although [ H]DOB was used herein to label 5-HT
2A
2
C
¨
9
3
et al. is the 5,6-difluoro analogue of 3; perhaps the
fluoro groups are important contributors to selectivity.
We examined the 5,6-difluoro derivative of 2a (i.e., 2d),
but 2d failed to display substantially improved affinity
or selectivity over 2a. Another explanation for low
affinity and lack of selectivity is that compound 3 is
an a-methylisotryptamine whereas compounds 2 are
N,N-dimethylisotryptamines. Hence, we prepared com-
pound 10. Compound 10 has been reported to bind at 5-
receptors, [ H]mesulergine was employed to label 5-
HT2C receptors. Compound 10 was re-examined using
various assay conditions (Table 2). The results show
that compound 10 binds with similar affinity at rat and
human 5-HT2A receptors when an agonist radioligand is
employed, and that there is 20-fold increase in affinity
3
3
when [ H]DOB is used in place of [ H]ketanserin.
Compound 10 binds at human 5-HT2C receptors with
about 20-fold higher affinity when agonist radioligand is
3
HT2A sites with reduced affinity (K ca. 100 nM) relative
i
employed in place of [ H]mesulergine. Nevertheless, the
selectivity of compound 10 for h5-HT2C receptors over
h5-HT2A receptors is, at best, only about 10-fold using
to 5-HT2C sites (K ca. 1 nM) resulting in a compound
i
9
with 100-fold selectivity. We found that compound 10
binds with somewhat higher affinity at 5-HT2A sites
(
K =15 nM), and with lower affinity at 5-HT sites
i
2C
Table 2. Binding of compound 10 at 5-HT2A and 5-HT2C receptors
using different assay conditions
(K =85 nM) than previously reported.
i
Not only did the compounds in the present study lack
selectivity for 5-HT2C receptors, compound 10, which
was previously reported to be selective, also failed to
exhibit the expected selectivity. Hence, the results can-
not be solely attributed to structure; a remaining expla-
nation for the apparent disparity might be the assay
conditions. The earlier study used human 5-HT2A and
, nM (SEM)a
K
i
Receptor
5-HT2A
Species
Radioligand
3
9
Rat
Rat
[ H]Ketanserin
3
320 (ꢂ50)
15 (ꢂ2)
[ H]DOB
3
Human
Human
Human
Human
Human
[ H]DOB
3
39 (ꢂ4)
[ H]5-HT
[ H]Mesulergine
12 (ꢂ1)
3
5-HT2C
85 (ꢂ15)
3.9 (ꢂ0.8)
3.8 (ꢂ0.5)
3
[ H]DOB
3
[ H]5-HT
5
rat 5-HT2A receptors and human 5-HT receptors. The
-HT2C receptors whereas the present investigation used
a
2C
i
K values, obtained at least in triplicate, are followed by SEM. Details
3
14
earlier study also used [ H]DOB to label 5-HT recep-
2
of the assay procedures are provided.
A

158
J. Chang-Fong et al. / Bioorg. Med. Chem. Lett. 12 (2002) 155–158
assay conditions that match the literature conditions as
closely as possible. The h5-HT2A binding results were
8. Glennon, R. A.; Jacyno, J. M.; Young, R.; McKenney,
J. D.; Nelson, D. J. Med. Chem. 1984, 27, 41.
9. Bos, M.; Jenck, F.; Martin, J. R.; Moreau, J. L.; Sleight,
¨
A. J.; Wichmann, J.; Widmer, U. J. Med. Chem. 1997, 40, 2762.
3
confirmed using a second agonist radioligand ([ H]5-
3
HT), as were the h5-HT2C results [using ([ H]DOB). In
fact, under the latter conditions, compound 10 dis-
played only 3-fold selectivity for 5-HT2C receptors.
1
1
1
1
0. Glennon, R. A.; Bartyzel, P.; Teitler, M. Med. Chem. Res.
991, 1, 201.
1. Rydon, H. N.; Siddappa, S. J. Chem. Soc. 1951, 2462.
2. Rajur, S. B.; Merwade, A. Y.; Hendi, S. B.; Basanagou-
In conclusion, none of the examined pyrazinoindoles
nor isotryptamines displayed substantial selectivity for
dar, L. D. Indian J. Chem. 1989, 28B, 1065.
3. von Schmutz, J.; Kunzle, F. Helv. Chim. Acta 1956, 39, 1144.
14. Cell culture and transfection (for human 5HT2A assays):
1
5
-HT versus 5-HT2A receptors. As reported by Bo
2
¨
s et
C
6
al., the presence of a 10-methoxy substituent on the
pyrazinoindole nucleus might be necessary for enhanced
COS-7 cells were grown in DMEM (Life Technologies) with
10% fetal bovine serum (Life Technologies) in 5% CO at 37
2
C and subcultured 1:12 twice a week. Twenty-four hours
before transfection, cells were seeded at 80% confluence in
5
-HT2C affinity. In contrast, we were unable to confirm
the selective nature of compound 10. As a consequence,
caution is advised when interpreting results of pharma-
cological studies using 10 as a selective 5-HT2C agonist.
1
00-mm dishes. Cells were transfected with human 5-HT2A
DNA by Lipofectamine (Life Technologies). This was accom-
plished by combining 20 mL Lipofectamine with 5 mg of plas-
mid per dish. Transfections were performed in serum-free
ꢁ
DMEM for 4.5 h at 37 C. Radioligand binding: Thirty-six
hours after transfection, membranes were prepared from COS-
Acknowledgements
7
0.5 mM EDTA, 5 mM MgCl , pH 7.4 (tissue buffer) and cen-
cells (for human 5-HT2A) by scraping in 50 mM Tris–HCl,
This work was supported in part by DA 01642.
2
trifugation at 10,000g for 30 min. Membranes were re-sus-
pended in tissue buffer, homogenized and centrifuged again.
For human 5-HT2C, NIH3T3 cells stably transfected with
human 5-HT2C INI receptors were grown to confluence and
prepared in the same manner as the COS-7 cells. After re-sus-
pension in assay buffer (tissue buffer+0.1% ascorbate), 0.5-
mL membrane aliquots were added to each assay tube con-
References and Notes
1
. Glennon, R. A.; Dukat, M.; Westkaemper, R. B. Serotonin
Receptor Subtypes and Ligands. In Psychopharmacology, a
Generation of Progress. CD ROM Version, 1999.
. Serotonin Receptors and their Ligands, Olivier, B., van
Wijngaarden, I., Soudin, W., Eds. Elsevier: Amsterdam, 1997.
. Issac, M. Drugs Future 2001, 26, 383.
. Isaac, M.; Slassi, A.; O’Brien, A.; Edwards, L.; MacLean,
N.; Bueschkens, D.; Lee, D. K. H.; McCallum, K.; De Lan-
noy, I.; Demchyshyn, L.; Kamboj, R. Bioorg. Med. Chem.
Lett. 2000, 10, 919.
. Kennett, G. Curr. Opin. Invest. Drugs 1993, 2, 317.
. Bos, M.; Jenck, F.; Martin, J. R.; Moreau, J. L.; Mutel, V.;
Sleight, A. J.; Widmer, U. Eur. J. Med. Chem. 1997, 32, 253.
. Mokrosz, J.; Boksa, J.; Bojarski, A. J.; Charakchieva-
Minol, S. Med. Chem. Res. 1993, 3, 240.
3
3
taining either 2 nM [ H]DOB or 0.4 nM [ H]ketanserin for 5-
3
3
2
HT2A or 2 nM [ H]5-HT or 1 nM [ H]mesulergine for 5-HT
2C
and varying concentrations of competing drug in a final
volume of 1 mL mianserin (10 mM) was used to define non-
specific binding. Samples were incubated at room temperature
3
4
ꢁ
for 30 min (5-HT2A) or at 37 C for 30 min (5-HT2C), filtered
through glass fiber filters (presoaked in 0.3% poly-
ethyleneimine) on a Brandel cell harvester, and counted in
Ecoscint cocktail in a liquid scintillation counter (Beckman,
Berkeley, CA, USA) at 40% efficiency. Data analyses: Data
analyses were performed using Prism software (GraphPad,
San Diego, CA, USA). The Cheng-Prusoff equation was used
to calculate K values from the IC values.
5
6
¨
7
i
50