Binding of methoxy-substituted N1-benzenesulfonylindole analogs at human 5-HT6 serotonin receptors

Article abstract of DOI:10.1016/j.bmcl.2006.04.094

Comparison of several amine-substituted and methoxy-substituted analogs of N₁-(4-aminobenzene)sulfonylindole suggests that these substituents might contribute to the 5-HT₆ serotonin receptor affinity of these agents via their electro

Full text of DOI:10.1016/j.bmcl.2006.04.094

Bioorganic & Medicinal Chemistry Letters 16 (2006) 3793–3796
Binding of methoxy-substituted N1-benzenesulfonylindole analogs
at human 5-HT6 serotonin receptors
Uma Siripurapu,a Renata Kolanos,a Małgorzata Dukat,a Bryan L. Rothb,c
and Richard A. Glennona,*
aDepartment of Medicinal Chemistry, School of Pharmacy, Virginia Commonwealth University, Richmond, VA 23298-0540, USA
bDepartment of Biochemistry, School of Medicine, Case Western Reserve University, Cleveland, OH 44106, USA
cDepartment of Psychiatry and Neurosciences, School of Medicine, Case Western Reserve University, Cleveland, OH 44106, USA
Received 23 March 2006; revised 13 April 2006; accepted 13 April 2006
Available online 18 May 2006
Abstract—Comparison of several amine-substituted and methoxy-substituted analogs of N1-(4-aminobenzene)sulfonylindole sug-
gests that these substituents might contribute to the 5-HT6 serotonin receptor affinity of these agents via their electronic effect
on the indolic nucleus. Their 1,2,3,4-tetrahydrocarbazole counterparts behave differently.
Ó 2006 Elsevier Ltd. All rights reserved.
N(CH3)2
5-HT6 receptors, one of seven families (5-HT1–5-HT7)
of serotonin receptors, are of interest because of their
possible involvement in certain neuropsychiatric and
neurological disorders.1–3 MS-245 (1; Ki = 2.1 nM) was
one of the first examples of a 5-HT6 receptor antago-
nist.4,5 1,2,3,4-Tetrahydrocarbazole 2 (Ki = 1.5 nM) rep-
resents an analog of MS-245 wherein the
conformationally flexible side chain has been somewhat
constrained.6 The 40-amino derivative of MS-245 (i.e., 3;
Ki = 0.8 nM) retains the affinity of its parent;7 likewise,
compound 4 (Ki = 2.0 nM) retains the affinity of 2.8
N(CH3)2
4
H3CO
H3CO
6
N
S
N
O
S
O
O
O
R
R
1 R = H
3 R = NH2
2 R = H
4 R = NH2
Both 3 and 4 can be abbreviated in structure (i.e., 5 and 6;
Ki = 10 and 29 nM, respectively) with only a small loss in
affinity, but with retention of 5-HT6 antagonist action.7–9
To this extent, it appears that the tetrahydrocarbazoles
behave much in the same manner as their simpler indole
counterparts with respect to binding at 5-HT6 receptors.
To account for these observations, we suggested two
possibilities: in some cases one of the methylamino
groups (rather than the primary amine) might interact
with the amine binding site of the receptor—presumably
the TM3 aspartate moiety—and/or (particularly when
the primary amine is present) the methylamino group(s)
might make an indirect (e.g., electronic) contribution to
binding.9
Amine-bearing analogs of 5, such as 7 (Ki = 21 nM) and
8
(Ki = 7.0 nM), retain affinity or, as with
9
(Ki = 1.9 nM), bind with enhanced affinity at 5-HT6
receptors.9 Interestingly, when an indolic methylamine
is present, the 40-position primary amine is not required
for binding (e.g., 10; Ki = 26 nM).9
N
S
N
S
O
O
O
O
Keywords: Serotonin receptors; Indole analogs; 5-HT6.
*
H2N
H2N
Corresponding author. Tel.: +1 804 828 8487; fax: +1 804 828
7404; e-mail: glennon@hsc.vcu.edu
5
6
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.04.094

3794
U. Siripurapu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3793–3796
It has been concluded that indole-containing ligands
need not bind at 5-HT6 receptors with superimposable
indolic nuclei, and that multiple amine groups, although
possibly having an affinity-enhancing effect, are not
required for binding.9
b
a
MeO
N
H
N
S
N
S
O
O
O
O
MeO
MeO
13
H3C
H2N
NH
12a,b
14
The purpose of the present investigation was 2-fold.
First, we wished to determine if introduction of a meth-
ylamino group would have an effect on the binding of
tetrahydrocarbazole 6 comparable to that seen upon
conversion of 5 to 7 or 8 (i.e., 11a and 11b). Second,
because the presence of an indolic methylamino group
might influence the binding of 7–10 directly, by inter-
action with the amine binding site, or indirectly, via its
electronic character, we proposed to examine several
derivatives of 5 bearing an electron-donating methoxy
group (i.e., 12 where R = –OMe) in place of the
methylamino group.
O
OMe
OMe
OMe
d
MeO
c
N
H
N
S
N
MeO
MeO
O
O
O
O
S
15
H2N
O2N
12c
16
Scheme1. Reagentsandconditions:(a)i—NaH(60%),DMF,80 °C/1 h;
ii—N-acetylsulfanilyl chloride, 0 °C/16 h; (b) 10% ethanolic HCl, D/3 h;
(c) i—50% NaOH, (n-Bu)4NHSO4, CH2Cl2, rt/20 min; ii—4-nitro-
benzenesulfonyl chloride, rt/16 h; (d) SnCl2Æ2H2O, absolute EtOH,
D/3 h.
R4
7
8
9
R4 = NHMe, R6 = H, R = NH2
R4 = H, R6 = NHMe, R = NH2
R4 = R6 = NHMe, R = NH2
R6
N
S
O
10 R4 = R6 = NHMe, R = H
O
O
NHCH3
R
EtO
NH
NH2
b
a
Compounds 12a and 12b were obtained (Scheme 1) by
reaction of N-acetylsulfanilyl chloride with the appro-
priately substituted methoxyindolyl anion. The protect-
ed amide intermediates 14 were hydrolyzed to the target
amines with dilute ethanolic HCl. An attempt to prepare
12c in the same manner was unsuccessful. Subsequently,
4,6-dimethoxyindole (15)10 was converted to its anion
and allowed to react with 4-nitrobenzenesulfonyl chlo-
ride, followed by reduction of the nitro intermediate
16 to the desired product (Scheme 1).
N
N
N
O
O
O
O
O
O
S
S
S
H2N
H3COCHN
H3COCHN
17
18
11a
Scheme 2. Reagents and conditions: (a) ClCOOEt, pyridine, DMF,
ꢀ10 °C, 45 min; (b) i—10% ethanolic HCl, reflux, 2 h; ii—LiAlH4,
THF, reflux, 5 h.
R5
All of the compounds examined displayed nanomolar
affinity for 5-HT6 receptors (Table 1). However, neither
methylaminotetrahydrocarbazole 11a (Ki = 165 nM)
nor 11b (Ki = 205 nM) showed an affinity comparable
to that of the unsubstituted tetrahydrocarbazole 6
(Ki = 29 nM). Unlike what was seen with the simpler
indole derivatives, introduction of the methylamino
group was not well tolerated and resulted in about
5- to 10-fold reduced affinity.
R
R7
N
S
N
S
O
O
O
O
H2N
H2N
11a R5 = NHMe, R7 = H
11b R5 = H, R7 = NHMe
11c R5 = R7 = OMe
12
Dimethoxycarbazole 11c was prepared in the same man-
ner employed for the synthesis of 12a and 12b beginning
with 5,7-dimethoxy-1,2,3,4-tetrahydrocarbazole.11 The
two methylamino carbazole analogs, 11a and 11b, were
prepared in a common manner as shown for 11a in
Scheme 2. The requisite nitro compound12 was reduced
(SnCl2Æ2H2O) to amine 17, and 17 was reacted with ethyl
chloroformate to afford intermediate 18. Careful hydro-
lysis of the amide followed by lithium aluminum hydride
reduction afforded the target compound 11a. Com-
pound 11b was prepared from the known 7-nitro coun-
terpart.12 Both products were isolated as their oxalate
salts, but it was not possible to remove all traces of
Et2O from the salts obtained.
In contrast, the affinity of 5 (Ki = 10 nM) was enhanced
upon introduction of a methoxy group at either the
4-position (12a; Ki = 3.3 nM) or 6-position (i.e., 12b;
Ki = 1.8 nM). As a consequence, the 4,6-dimethoxy-
substituted analog 12c (Ki = 0.8 nM) was prepared and
found to bind with >10 times higher affinity than 5.
As a further comparison between the indoles and tetra-
hydrocarbazoles, the corresponding dimethoxy tetra-
hydrocarbazole 11c was examined. Compound 11c
(Ki = 210 nM) displayed an affinity comparable to meth-
ylaminotetrahydrocarbazoles 11a and 11b, and nearly
10-fold lower than the parent tetrahydrocarbazole 6
(Ki = 29 nM).

U. Siripurapu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3793–3796
Table 1. Physicochemical and 5-HT6 receptor binding properties of target compounds
3795
5
4
5
R
R
N
S
N
S
7
6
O
O
7
O
O
H2N
H2N
6, 11
5, 12
R
Recrystallization solvent
Melting point (°C)
Empirical formulaa
Ki, nM ( SEM)13
6b
–H
—
—
—
C19H21N3O2S 1.25 C2H2O4
29
c
11a
11b
11c
5-NHMe
7-NHMe
5,7-Di OMe
CH2Cl2
CH2Cl2
MeOH
135–137
151–152
199–200
165 (24)
205 (20)
210 (30)
d
C19H21N3O2S 2.0 C2H2O4
C20H22N2O4S 0.25 H2O
5b
–H
4-OMe
—
—
151
—
10
12a
12b
12c
12db
EtOH
EtOH
MeOH
—
C
15H14N2O3S
C15H14N2O3S
C16H16N2O4S
3.3 (0.4)
1.8 (0.5)
0.8 (0.1)
6-OMe
4,6-Di OMe
4,6-Di NO2
143–145
200–201
—
—
980
1
a All compounds were homogeneous as determined using thin-layer chromatography, assigned structures are consistent with H NMR spectra, and
compounds analyzed within 0.4% of theory for C, H, and N. C2H2O4 = oxalate salt.
b Binding data for 5,9 6,8 and 12d9 were previously reported.
c Crystallized with 0.5 mol Et2O.
d Crystallized with 0.25 mol Et2O.
These new data show that replacement of the methyl-
amino group of 7 and 8 by methoxy (i.e., 12a and 12b,
respectively) was not only tolerated, but resulted in
slightly enhanced affinity. In addition, the dimethoxy
compound 12c displayed an affinity similar to that of
its dimethylamino counterpart 9, and >10 times the
affinity of its parent indole 5.
pounds possess a 40-amino group rather than an indolic
alkylamine as their sole amine substituent. Whereas evi-
dence suggests that analogs of 1 and 2 might bind in a
similar fashion,8 and although analogs of 11 and 12
may or may not bind similarly to one another, they
might not bind in a manner common to that of 1 and
2. Given this supposition, it would seem that the region
of the receptor into which 6 binds does not readily
accommodate indole substituents such as found in 11.
Because the electron-donating methoxy groups do not
likely contribute directly to an interaction with the
amine binding site, it would appear that they influence
binding indirectly either by their electronic character
or their ability to form a hydrogen bond with some
receptor-associated binding feature. These results are
also consistent with the earlier finding that the presence
of electron withdrawing groups decreases the affinity of
5. For example, 4,6-dinitro-N1-benzenesulfonylindole
(12d, Ki = 980 nM)9 binds at 5-HT6 receptors with
>500- to 1000-fold lower affinity than 8 and 12c. On this
basis, it would not seem that the secondary amine
groups of 7–9 interact directly with the amine binding
site. However, this conclusion may not apply to com-
pound 10 which lacks any other amine function for
binding.
Overall, then, it would appear that analogs of 5 and 6
might not always bind in a similar fashion at 5-HT6
receptors despite their structural similarity, and that
when an amino group is present at the benzenesulfonyl
ring 4-position, the indolic methylamino group(s) of 5-
derived compounds 7–9 can be replaced by a methoxy
group without unfavorable effect on affinity. Such infor-
mation should contribute to investigations aimed at
identification of pharmacophore models for 5-HT6
receptor binding.
Acknowledgments
This work was funded in part by MH 60599 (R.A.G.),
and U.S. was supported by a Werner Lowenthal
Neurological Sciences Award.
Despite their structural similarity, and the similar affin-
ity of 1 with 2, and 5 with 6, the results indicate that the
tetrahydrocarbazoles 11 do not necessarily behave in a
manner comparable to that of the structurally simpler
N1-benzenesulfonylindoles 12 upon introduction of a
methylamino group (viz. 11a and 11b) or a dimethoxy
group (viz. 11c). That is, in the presence of the partially
saturated benzenoid ring, introduction of these aromatic
substituents tends to decrease 5-HT6 receptor affinity.
In contrast, it has been demonstrated that there are a
number of binding similarities between analogs of 1
and 2.8 Differences are evident primarily when the com-
References and notes
1. Kroeze, W. K.; Kristiansen, K.; Roth, B. L. Curr. Top.
Med. Chem. 2002, 2, 507.
2. Glennon, R. A. J. Med. Chem. 2003, 46, 2795.
3. Woolley, M. L.; Marsden, C. A.; Fone, K. C. F. Curr.
Drug Top. 2004, 3, 59.
4. Glennon, R. A.; Lee, M.; Rangisetty, J. B.; Dukat, M.;
Roth, B. L.; Savage, J. E.; McBride, A.; Rauser, L.;

3796
U. Siripurapu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3793–3796
Hufesien, L.; Lee, D. K. H. J. Med. Chem. 2000, 43,
1011.
5. Tsai, Y.; Dukat, M.; Slassi, A.; MacLean, N.; Dem-
chyshyn, L.; Savage, J. E.; Roth, B. L.; Hufesein, S.;
Lee, M.; Glennon, R. A. Bioorg. Med. Chem. Lett.
2000, 10, 2295.
6. Lee, M.; Rangisetty, J. B.; Dukat, M.; Slassi, A.; Maclean,
N.; Lee, D. K. H.; Glennon, R. A. Med. Chem. Res. 2000,
10, 230.
7. Pullagurla, M.; Setola, V.; Roth, B. L.; Glennon, R. A.
Bioorg. Med. Chem. Lett. 2003, 13, 3355.
8. Chang-Fong, J.; Rangisetty, J. B.; Dukat, M.; Setola, V.;
Raffay, T.; Roth, B. L.; Glennon, R. A. Bioorg. Med.
Chem. Lett. 2004, 14, 1961.
9. Pullagurla, M.; Siripurapu, U.; Kolanos, R.; Bondarev,
M. L.; Dukat, M.; Setola, V.; Roth, B. L.; Glennon, R. A.
Bioorg. Med. Chem. Lett. 2005, 15, 5298.
10. Black, D. St. C.; Rothnie, N. E.; Wong, L. C. H. Aust.
J. Chem. 1983, 36, 2407.
11. Huffman, J. W.; Lu, J.; Dai, D.; Kitaygorodaskiy, A.;
Wiley, J. L.; Martin, B. R. Bioorg. Med. Chem. Lett. 2000,
8, 439.
13. The h5-HT6 radioligand binding assay was performed as
previously described.14 In brief, h5-HT6 cDNA was
transiently expressed in HEK-293 cells using Fugene6
according to the manufacturer’s recommendations. Twen-
ty-four hour after transfection, the medium was replaced;
24 h later, medium containing dialyzed serum (to remove
5-HT) was added. At 75 h after transfection, cells were
harvested by scraping and centrifugation. Cells were then
washed by centrifugation and re-suspension once in
phosphate-buffered saline (pH 7.40; PBS) and then frozen
as tight pellets at ꢀ80 °C until use. Binding assays were
performed at room temperature for 90 min in binding
buffer (50 mM Tris–Cl, 10 mM MgCl2, and 0.1 mM
EDTA, pH 7.40) with [3H]LSD (1 nM final concentration)
using 10 lM clozapine for non-specific binding. Various
concentrations of unlabeled test agent were used for Ki
determinations, with Ki values calculated using the
program LIGAND. Specific binding represented 80–90%
of total binding. Ki values are the result of triplicate
determinations.
14. Kohen, R.; Metcalf, M. A.; Khan, N.; Druck, T.;
Huebner, K.; Lachowicz, J. E.; Meltzer, H. Y.; Sibley,
D. R.; Roth, B. L.; Hamblin, M. W. J. Neurochem. 1996,
66, 47.
12. Moscalew, N.; Makosza, M. Tetrahedron Lett. 1999, 40,
5395.