Novel 1-(1-benzyl-1H-indol-3-yl)-N,N,N-trimethylmethanaminium iodides are competitive antagonists for the human α4β2 and α7 nicotinic acetylcholine receptors

Article abstract of DOI:10.1039/c3md00042g

This work presents the synthesis and the pharmacological characterization of a series of novel 1-(1-benzyl-1H-indol-3-yl)-N,N,N-trimethylmethanaminium iodide derivatives at the human (h) α7 and α4β2 nicotinic acetylcholine receptors (nAChRs). The inhibito

Full text of DOI:10.1039/c3md00042g

MedChemComm
View Article Online
View Journal
CONCISE ARTICLE
Novel 1-(1-benzyl-1H-indol-3-yl)-N,N,N-
trimethylmethanaminium iodides are
competitive antagonists for the human
Cite this: DOI: 10.1039/c3md00042g
a4b2 and a7 nicotinic acetylcholine receptors†
a
a
b
a
*
´
´
Edwin G. Perez, Cristian Ocampo, Dominik Feuerbach, Jhon J. Lopez,
Guibeth L. Morelo,c Ricardo A. Tapiaa and Hugo R. Ariasd
This work presents the synthesis and the pharmacological characterization of a series of novel 1-(1-benzyl-
1H-indol-3-yl)-N,N,N-trimethylmethanaminium iodide derivatives at the human (h) a7 and a4b2 nicotinic
acetylcholine receptors (nAChRs). The inhibitory activity of the compounds was determined by Ca2+
influx assays on cells expressing either the ha7 or ha4b2 nAChR subtype. To determine whether the
observed inhibitory activity is mediated by a competitive or non-competitive mechanism, additional
radioligand binding assays were performed using [3H]methyllycaconitine, [3H]cytisine, and [3H]
imipramine. The results established that the compounds inhibit the nAChRs by
a competitive
mechanism and that the potencies are higher for the ha7 nAChR compared to that for the ha4b2
nAChR. Substitutions with oxygenated functional groups on the benzene ring increase the receptor
selectivity. In particular, the hydroxyl derivatives 4b and 4c present the highest selectivity for the ha7
nAChR subtype. Molecular docking results indicate that the hydroxyl group forms a hydrogen bond
with the carbonyl group at a7-Gln116, but not at b2-Phe115, supporting the observed receptor
selectivity at the molecular level.
Received 7th February 2013
Accepted 30th May 2013
DOI: 10.1039/c3md00042g
www.rsc.org/medchemcomm
interest in nAChR antagonists has increased because of their
possible therapeutic uses.6–8 In particular, a4b2 nAChR antag-
Introduction
Nicotinic acetylcholine receptors (nAChRs), members of the cys-
loop ligand-gated ion channel superfamily, are formed by ve
subunits surrounding a central pore permeable to Na+, K+, and
Ca2+ ions.1 Depending on whether nAChRs are formed by only
one or more than one subunit type, they can be divided into
homomeric (e.g., a7) and heteromeric (e.g., a4b2) nAChRs. a7
and a4b2 nAChRs are the most abundant receptor subtypes in
the brain. They are involved in important physiological func-
tions2 and in many relevant diseases, including neuropsychi-
atric, neurodegenerative, and pain-related diseases.3 In the last
decade, several laboratories have undertaken the synthesis of
selective agonists for these two nAChR subtypes, but unfortu-
nately, only a few are in clinical trials.4,5 More recently, the
onists could be benecial for the treatment of nicotine addic-
tion,7 whereas a7 nAChR antagonists could be benecial for the
treatment of cancer tumors6 and organophosphorus nerve
agent intoxication.8
Our laboratory has recently found that the indole ring might
be a useful scaffold for the development of novel potent and
selective nicotinic antagonists.9 The most potent derivatives are
those with the indole core attached to a short aminoalkyl
substituent and bearing a medium-length substituent on its
nitrogen (Fig. 1). Moreover, it has been shown that the
a
´
´
~
Facultad de Quımica, Ponticia Universidad Catolica de Chile, Av. Vicuna Mackenna
4860, Santiago 6094411, Chile. E-mail: eperezh@uc.cl; Fax: +56 0226864744; Tel:
+56 022 6864433
bNovartis Institutes for Biomedical Research, Basel, Switzerland
cFaculty of Chemical and Pharmaceutical Sciences, University of Chile, Santiago, Chile
dDepartment of Medical Education, California Northstate University College of
Medicine, Elk Grove, CA 95757, USA
† Electronic supplementary information (ESI) available: Full experimental details,
NMR spectra, calcium inux and radioligand binding procedures. See DOI:
10.1039/c3md00042g
Fig. 1 Plausible pharmacophore for nAChR antagonists.9 R ¼ alkyl group.
This journal is ª The Royal Society of Chemistry 2013
Med. Chem. Commun.

View Article Online
MedChemComm
Concise Article
trimethylammonium group increases ligand affinity to
nAChRs.9–14 In this paper, we describe the synthesis of a series
of novel 1-(1-benzyl-1H-indol-3-yl)-N,N,N-trimethylmethan-
aminium iodide derivatives by using a one-pot microwave
heating procedure. Subsequently, the pharmacological activity
of these compounds at human (h) a7 and a4b2 nAChRs was
determined by Ca2+ inux assays. In addition, radioligand
competition binding experiments were performed to determine
the inhibitory mechanism of these compounds. Specically,
[3H]methyllycaconitine ([3H]MLA) and [3H]cytisine were used as
selective ligands for the a7 and a4b2 nAChR agonist sites,
respectively, and [3H]imipramine was used to test the interac-
tion of these compounds with the nAChR ion channel.
Results and discussion
Chemical synthesis
The classical synthesis of N1-benzylindole derivatives involves
the deprotonation of the N–H group present in the indole ring
and the subsequent coupling with a benzyl halide.9 This
strategy is efficient when it is not necessary to protect the
functional groups. However, when acidic functional groups are
present (e.g., phenol group), the synthesis needs two additional
steps: protection and deprotection of the functional group. To
avoid these additional steps, we used a one-step microwave
heating strategy that permits the coupling of indoline with
different benzaldehydes producing N1-benzylindole derivatives
without the need for protecting protocols.15,16
Fig. 2 Chemical structures of the novel 1-(1-benzyl-1H-indol-3-yl)-N,N,N-trime-
thylmethanaminium iodide compounds, 4a–j.
Scheme 1 illustrates the synthesis of compounds 4a–j in
three steps. First, treatment of indoline (1) with substituted
benzaldehydes (2a–j) in toluene under microwave heating
conditions15,16 gave N1-benzylindoles (3a–j). The subsequent
Mannich reaction with dimethylamine and formaldehyde fol-
lowed by quaternization using CH3I produced the desired
compounds 4a–j (see Fig. 2 for their structures).
Ca2+ inux assays
The functional and binding properties of compounds 4a–j were
subsequently determined on the ha7 and ha4b2 nAChRs using
procedures published elsewhere (see ESI†).17–20 The inhibitory
potencies (IC50) of our compounds were determined in GH3-
ha7 (Fig. 3A) and HEK293-ha4b2 (Fig. 3B) cells by Ca2+ inux
assays. The results indicate that the compounds inhibit the ha7
nAChR with potencies 2–15 times higher than that for the
ha4b2 nAChR (Table 1). Interestingly, the compound 4j showed Fig. 3 Inhibitory effect of compounds shown in Table 1 on (ꢀ)-epibatidine-
induced Ca2+ influx in HEK293-ha4b2 (A) and GH3-ha7 (B) cells. The EC50 values
for (ꢀ)-epibatidine (D) are (A) 30 ꢀ 5 nM (n ¼ 21) and (B) 52 ꢀ 4 nM (n ¼ 17),
respectively. The symbols are: 4a ( ), 4b ( ), 4c ( ), 4d ( ), 4e ( ), 4f ( ), 4g ( ), 4h
(
), 4i ( ), and 4j ( ). The plots are representative of three determinations, where
the error bars correspond to the standard deviation (SD). The calculated IC50 and
H values are summarized in Table 1.
n
no activity on the ha4b2 subtype at concentrations as high as
100 mM (Fig. 3B), suggesting that this compound as well as
compounds 4b and 4c are more selective antagonists for the
ha7 nAChR.
Scheme 1 Reagents and conditions: (a) toluene, microwave heating, 30 min; (b)
aqueous (CH3)2NH, aqueous HCHO, acetic acid, rt, 16 h; (c) CH3I, acetone, 16 h.
Med. Chem. Commun.
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Concise Article
MedChemComm
Table 1 Inhibitory potencies for the studied compounds on the ha4b2 and ha7
nAChRs determined by Ca2+ influx experimentsa
to ha7 and ha4b2 nAChR agonist sites, respectively, in the
0.6–2.9 mM concentration range, whereas compound 4b inhibits
[3H]imipramine binding to AChR ion channels (Fig. 4C) only at
concentrations higher than 100 mM (Table 2).
ha4b2
ha7
ha4b2/
ha7
IC50
IC50
The Hill coefficient (nH) values for the ha4b2 nAChR are
close to unity, suggesting that the ligand-induced [3H]cytisine
binding inhibition is mediated by a non-cooperative mecha-
nism (Table 2). In addition, the nH values for the ha7 nAChR are
closer to two, suggesting that the ligand-induced [3H]MLA
binding inhibition is mediated by a cooperative mechanism
and thus that the compounds bind to more than one [3H]MLA
binding site. This result is in agreement with the fact that the
ha7 nAChR has ve putative binding sites and it needs three of
them to be occupied for full activation.21 In contrast, the nH
value for the binding of compound 4b to the [3H]imipramine
site at the ha7 nAChR is close to 0.5, suggesting a negative
cooperative mechanism, and thus, supporting an allosteric
mode of inhibition.22 These results support the idea that these
compounds inhibit ha7 and ha4b2 nAChRs by a competitive
mechanism and rule out the possibility of a noncompetitive
mechanism. Our results indicate that compounds with a tri-
methylammonium group attached to the indole ring by a
methylenic linker (see Fig. 1) behave as nAChR competitive
antagonists, which is in agreement with trends reported in the
literature for other quaternary ammonium compounds.9–14
The presence of a variety of functional groups on the benzene
ring was evaluated to establish both the optimal functional
group and its position on this aromatic ring (Table 1). The initial
results indicate that 4a, a compound without substitution on the
benzene ring, has inhibitory potency in the micromolar range
but with only negligible ha7/ha4b2 selectivity. Subsequent result
analyses of compounds with substitutions on the benzene ring
indicated that they are more selective. For example, compounds
4b and 4c, containing only one hydroxyl group on the benzene
ring, are the most selective antagonists for the ha7 subtype.
Interestingly, the methylation of the hydroxy group present in 4b
to furnish 4d maintains similar potency but decreases the
observed receptor selectivity.
Compound (mM)
nH
(mM)
nH
ratio
4a
4b
4c
4d
4e
4f
4g
4h
4i
4.2 ꢀ 1.5 1.02 ꢀ 0.28 2.3 ꢀ 0.9 2.15 ꢀ 0.20 1.9
14.7 ꢀ 1.2 1.32 ꢀ 0.35 1.0 ꢀ 0.2 1.81 ꢀ 0.02 14.7
24.9 ꢀ 8.0 1.05 ꢀ 0.01 2.2 ꢀ 0.4 2.24 ꢀ 0.39 11.3
4.4 ꢀ 2.5 1.02 ꢀ 0.13 2.0 ꢀ 0.6 2.25 ꢀ 0.40 2.2
9.5 ꢀ 4.3 0.94 ꢀ 0.05 3.4 ꢀ 0.2 2.24 ꢀ 0.37 2.8
9.1 ꢀ 5.2 0.93 ꢀ 0.25 2.2 ꢀ 0.1 2.46 ꢀ 0.29 4.1
14.8 ꢀ 5.0 1.13 ꢀ 0.31 2.2 ꢀ 0.8 2.03 ꢀ 0.22 6.7
3.9 ꢀ 1.0 1.23 ꢀ 0.07 1.3 ꢀ 0.2 2.00 ꢀ 0.10 3.0
20.0 ꢀ 0.4 1.45 ꢀ 0.22 2.7 ꢀ 0.8 2.03 ꢀ 0.23 7.4
4j
>100
—
41.3 ꢀ 15.5 2.16 ꢀ 0.01
—
a
The IC50 and nH values were obtained from Fig. 3A and B, respectively.
Radioligand binding assays
Our radioligand binding results clearly indicate that the
compounds are competitive antagonists for the ha7 (Fig. 4A)
and ha4b2 (Fig. 4B) nAChR subtypes. More specically,
compounds 4b and 4c inhibit [3H]MLA and [3H]cytisine binding
Additional substitutions (e.g., hydroxy, methoxy, or methyl-
endioxy) on the benzene ring were next examined. The results
indicate that the antagonist activity of these compounds (i.e.,
4e, 4f, 4g, and 4h) at ha4b2 nAChRs improved, diminishing
consequently the ha7 nAChR selectivity. Finally, we centered
our attention to dicationic compounds because it has been
Fig. 4 Ligand interaction with different binding sites at the ha7 and ha4b2
nAChRs. Inhibition of (A) [3H]MLA binding to ha7 nAChR agonist sites and (B) [3H]
cytisine binding to ha4b2 nAChR agonist sites mediated by compounds 4b (,)
and 4c (-), respectively. (C) Inhibition of [3H]imipramine binding to ha7 (B) and
ha4b2 (C) nAChR ion channels mediated by compound 4b.
Table 2 Ligand binding affinities for the agonist and ion channel binding sites from ha7 and ha4b2 nAChRsa
ha7
ha4b2
Radioligand
[3H]MLA
nAChR sites
Agonist
Ligand
Ki (mM)
nH
Ki (mM)
nH
4b
4c
4b
4c
4b
0.6 ꢀ 0.1
2.9 ꢀ 0.2
—
1.18 ꢀ 0.11
1.26 ꢀ 0.10
—
—
—
—
—
[3H]Cytisine
Agonist
1.6 ꢀ 0.2
0.6 ꢀ 0.1
112 ꢀ 22
0.72 ꢀ 0.06
0.86 ꢀ 0.04
0.75 ꢀ 0.12
—
—
[3H]Imipramine
Ion channel
145 ꢀ 36
0.54 ꢀ 0.08
a
The Ki and nH values were obtained from Fig. 4A–C, respectively.
This journal is ª The Royal Society of Chemistry 2013
Med. Chem. Commun.

View Article Online
MedChemComm
Concise Article
Fig. 5 Molecular interactions of compounds 4b and 4c with the (A and B) a7 and (C and D) a4b2 nAChR orthosteric sites. Residues from the principal and
˚
complementary components are colored green and yellow, respectively, whereas the ligands are colored cyan. Distances in dashed lines are in Angstrom (A).
reported that these types of compounds are potent antagonists N1-benzyl-3-trimethylmethanammonium iodides improve ha7
for different nAChR subtypes.23,24 The results show that the nAChR selectivity by forming a hydrogen bond with the polar
p-trimethylammonium derivative (i.e., 4i) presents a 7-fold group only present in ha7-Gln116. This increased selectivity
higher selectivity for the ha7 nAChR compared to the ha4b2 could be exploited for the treatment of disorders where ha7
nAChR, whilst the m-methylpyridinium derivative (i.e., 4j) is nAChRs are implicated such as cancer tumors6 and organo-
inactive at the ha4b2 nAChR and shows low activity at the ha7 phosphorus nerve agent intoxication.8
nAChR. Although the ha7 nAChR inhibition is low, the lack of
activity at the ha4b2 nAChR opens the door for the development
of potentially selective ha7 nAChR antagonists structurally
related to this m-methylpyridinium derivative.
Acknowledgements
This work was supported by FONDECYT grant 11090120, ICM
grant P10-035-F (E.G.P) and ICM Grant P10-003-F (R.A.T).
Molecular docking25 of compounds 4b and 4c to the nAChR
orthosteric sites explained the observed receptor selectivity
(Fig. 5A–D). The results indicate that the trimethylammonium
group forms cation–p interactions with the aromatic box at
both nAChR subtypes. Nevertheless, the hydroxyl group (meta
and para positions at the benzyl moiety of 4b and 4c, respec-
tively) forms a hydrogen bond with the carbonyl group from the
ha7-Gln116 side chain at the complementary component
(Fig. 5A and B). Interestingly, this hydrogen bond is absent in
the ha4b2 nAChR because this residue is replaced by b2-Phe115
which does not have the polar group to interact with the
ligand hydroxyl moiety (Fig. 5C and D). Our results coincide
with other publications indicating that ha7-Gln116 is a key
residue to dene the selectivity between a7 and a4b2 nAChR
subtypes.17,26,27
Notes and references
1 D. Paterson and A. Nordberg, Prog. Neurobiol., 2000, 61, 75.
2 A. Taly, P.-J. Corringer, D. Guedin, P. Lestage and
J.-P. Changeuxand, Nat. Rev. Drug Discovery, 2009, 8, 733.
3 M. N. Romanelli, P. Gratteri, L. Guandalini, E. Martini,
C. Bonaccini and F. Gualtieri, ChemMedChem, 2007, 2, 746.
4 A. A. Mazurov, J. D. Speake and D. Yohannes, J. Med. Chem.,
2011, 54, 7943.
5 G. C. Dumbar, R. V. Kuchibhatla and G. Lee, J. Psycho-
pharmacol., 2011, 25, 1020.
6 L. Paleari, A. Cesario, M. Fini and P. Russo, Drug Discovery
Today, 2009, 14, 822.
7 K. M. Tobey, D. M. Walentiny, J. L. Wiley, F. I. Carroll,
M. I. Damaj, M. R. Azar, G. F. Koob, O. George, L. S. Harris
and R. E. Vann, Psychopharmacology, 2012, 223, 159.
8 Y. Peng, Q. Zhang, G. L. Snyder, H. Zhu, W. Yao, J. Tomesch,
R. L. Papke, J. P. O'Callaghan, W. J. Welsh and
L. P. Wennogle, Bioorg. Med. Chem. Lett., 2010, 20, 4825.
Conclusions
Taken together, the results conrm the importance of the
indole ring as a novel moiety for the development of competitive
antagonists for ha7 and ha4b2 nAChR subtypes. More impor-
tantly, hydroxyl substitutions on the benzene ring from
´
¨
9 E. G. Perez, B. K. Cassels, C. Eibl and D. Gundisch, Bioorg.
Med. Chem., 2012, 20, 3719.
Med. Chem. Commun.
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Concise Article
MedChemComm
10 L. P. Dwoskin, B. M. Joyce, G. Zheng, N. M. Neugebauer, 19 H. R. Arias, A. Rosenberg, K. M. Targowska-Duda,
V. K. Manda, P. Lockman, R. L. Papke, M. T. Bardo and
D. Feuerbach, K. Jozwiak, R. Moaddel and I. W. Weiner,
P. A. Crooks, Biochem. Pharmacol., 2007, 74, 1271.
Int. J. Biochem. Cell Biol., 2010, 42, 1007.
´
´
´
11 P. Iturriaga-Vasquez, E. G. Perez, E. Y. Slater, I. Bermudez 20 H. R. Arias, R.-X. Gu, D. Feuerbach, B.-B. Gu, Y. Ye and
and B. K. Cassels, Bioorg. Med. Chem., 2007, 15, 3368. D.-Q. Wei, Biochemistry, 2011, 50, 5263.
12 C. Gotti, B. Balestra, M. Moretti, G. E. Rovati, L. Maggi, 21 D. Rayes, M. J. De Rosa, S. M. Sine and C. Bouzat, J. Neurosci.,
G. Rossoni, F. Berti, L. Villa, M. Pallavicini and
F. Clementi, Br. J. Pharmacol., 1998, 124, 1197.
13 R. Simsek, J. Chang-Fong, M. Lee, M. Dukat, M. I. Damaj,
2009, 29, 6022.
22 H. R. Arias, D. Feuerbach, K. M. Targowska-Duda and
K. Jozwiak, Nerochem. Int., 2010, 36, 595.
B. R. Martin and R. A. Glennon, Bioorg. Med. Chem. Lett., 23 G. Zheng, Z. Zhang, C. Dowell, E. Wala, L. P. Dwoskin,
2003, 13, 2917.
J. R. Holtman, J. M. McIntosh and P. A. Crooks, Bioorg.
Med. Chem. Lett., 2011, 21, 2476.
24 Z. Zhang, G. Zheng, M. Pivavarchyk, A. G. Deaciuc,
L. P. Dwoskin and P. A. Crooks, Bioorg. Med. Chem. Lett.,
2011, 21, 88.
14 H. M. Ragab, J. S. Kim, M. Dukat, H. Navarro and
R. A. Glennon, Bioorg. Med. Chem. Lett., 2006, 16, 4283.
15 H. Mao, R. Xu, J. Wan, Z. Jiang, C. Sun and Y. Pan, Chem.–
Eur. J., 2010, 16, 13352.
16 I. Deb, D. Das and D. Seidel, Org. Lett., 2011, 13, 812.
25 See ESI† for further details.
17 H. R. Arias, H. Gu, D. Feuerbach and D.-Q. Wei, Biochemistry, 26 X. Huang, F. Zheng, C. Xi, P. A. Crooks, L. P. Dwoskin and
2010, 49, 4169. C.-G. Zhan, J. Med. Chem., 2006, 49, 7661.
18 H. R. Arias, D. Feuerbach, K. M. Targowska-Duda, 27 Y. Xiao, P. S. Hammond, A. A. Mazurov and D. Yohannes,
M. M. Russell and K. Jozwiak, Biochemistry, 2010, 49, 5734. J. Chem. Inf. Model., 2012, 52, 3064.
This journal is ª The Royal Society of Chemistry 2013
Med. Chem. Commun.