Selectivity Optimization of Substituted 1,2,3-Triazoles as α7 Nicotinic Acetylcholine Receptor Agonists

Article abstract of DOI:10.1021/acschemneuro.5b00058

Three series of substituted anti-1,2,3-triazoles (IND, PPRD, and QND), synthesized by cycloaddition from azide and alkyne building blocks, were designed to enhance selectivity and potency profiles of a lead α7 nicotinic acetylcholine receptor (α7-nAChR) agonist, TTIn-1. Designed compounds were synthesized and screened for affinity by a radioligand binding assay. Their functional characterization as agonists and antagonists was performed by fluorescence resonance energy transfer assay using cell lines expressing transfected cDNAs, α7-nAChRs, α4β2-nAChRs, and 5HT_3A receptors, and a fluorescence cell reporter. In the IND series, a tropane ring of TTIn-1, substituted at N1, was replaced by mono- and bicyclic amines to vary length and conformational flexibility of a carbon linker between nitrogen atom and N1 of the triazole. Compounds with a two-carbon atom linker optimized binding with K_ds at the submicromolar level. Further modification at the hydrophobic indole of TTIn-1 was made in PPRD and QND series by fixing the amine center with the highest affinity building blocks in the IND series. Compounds from IND and PPRD series are selective as agonists for the α7-nAChRs over α4β2-nAChRs and 5HT_3A receptors. Lead compounds in the three series have EC₅₀s between 28 and 260 nM. Based on the EC₅₀, affinity, and selectivity determined from the binding and cellular responses, two of the leads have been advanced to behavioral studies described in the companion article (DOI: 10.1021/acschemneuro.5b00059).

Full text of DOI:10.1021/acschemneuro.5b00058

Research Article
pubs.acs.org/chemneuro
Selectivity Optimization of Substituted 1,2,3-Triazoles as α7 Nicotinic
Acetylcholine Receptor Agonists
Kuntarat Arunrungvichian,†,‡,§ Valery V. Fokin,§ Opa Vajragupta,*,† and Palmer Taylor‡
†Center of Excellence for Innovation in Drug Design and Discovery, Faculty of Pharmacy, Mahidol University, 447 Sri-Ayudhya Road,
Bangkok 10400, Thailand
‡Department of Pharmacology, Skaggs School of Pharmacy and Pharmaceutical Sciences, University of California, San Diego, 9500
Gilman Drive, La Jolla, California 92093-0650, United States
§Department of Chemistry, The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, California 92037, United States
S
* Supporting Information
ABSTRACT: Three series of substituted anti-1,2,3-triazoles
(IND, PPRD, and QND), synthesized by cycloaddition from
azide and alkyne building blocks, were designed to enhance
selectivity and potency profiles of a lead α7 nicotinic
acetylcholine receptor (α7-nAChR) agonist, TTIn-1. Designed
compounds were synthesized and screened for affinity by a
radioligand binding assay. Their functional characterization as
agonists and antagonists was performed by fluorescence
resonance energy transfer assay using cell lines expressing
transfected cDNAs, α7-nAChRs, α4β2-nAChRs, and 5HT3A
receptors, and a fluorescence cell reporter. In the IND series, a
tropane ring of TTIn-1, substituted at N1, was replaced by
mono- and bicyclic amines to vary length and conformational flexibility of a carbon linker between nitrogen atom and N1 of the
triazole. Compounds with a two-carbon atom linker optimized binding with Kd’s at the submicromolar level. Further
modification at the hydrophobic indole of TTIn-1 was made in PPRD and QND series by fixing the amine center with the
highest affinity building blocks in the IND series. Compounds from IND and PPRD series are selective as agonists for the α7-
nAChRs over α4β2-nAChRs and 5HT3A receptors. Lead compounds in the three series have EC50’s between 28 and 260 nM.
Based on the EC50, affinity, and selectivity determined from the binding and cellular responses, two of the leads have been
advanced to behavioral studies described in the companion article (DOI: 10.1021/acschemneuro.5b00059).
KEYWORDS: Neurologic disorders, α7-nAChRs, cycloaddition reactions, triazole, click-chemistry, nicotinic acetylcholine receptors,
pentameric ligand-gated ion channels
icotinic acetylcholine receptors (nAChRs) are widely
distributed in the central and peripheral nervous systems
Differences in subunit combinations and localization lead to
their functional and pharmacological variances.6,7 The orthos-
teric ligand binding pocket is located at the interface in the
extracellular domain between a principal (alpha) and a
complementary subunit.8 The α7-nAChR is unique in being
homomeric with identical binding sites formed between its
principal, C loop containing face and the opposing face of the
neighboring subunit.
Several α7-nAChR agonists have been developed for clinical
trials, namely, GTS-21,9,10 MEM3454,11 TC-5619,12 PNU-
282987,13 AR-R17779,14 ABT-107,15 and SEN34625/WYE-
10391416 (Figure 1). For α7-nAChR agonist design, the
primary pharmacophoric features of α7-nAChR agonists, a
cationic center, a hydrogen-bond acceptor, and a hydrophobic
moiety were modified.14,17,18 Despite comprehensive structural
N
(CNS and PNS), controlling peripheral voluntary motor,
autonomic, and central nervous system functions.1 Since they
are found in abundance in presynaptic locations in the CNS,
they can control release of other transmitters as well as
acetylcholine itself. Therefore, they have diverse modulatory
functions and are candidate targets for neurologic disorders of
development, schizophrenia and autism, and the aging process,
Parkinsonism and the Alzheimer dementias.2−4
Structurally, nAChRs are members of the pentameric Cys-
loop ligand-gated ion channel (LGIC) superfamily, which are
composed of five transmembrane spanning subunits, assembled
to surround a centrosymmetric ion pore. There are at least 12
distinct neuronal subunits (α2−α10 and β2−β4) characterized
in mammalian and avian systems.1,5 Structures of these
pentameric receptors are divided into three domains: N-
terminal extracellular, transmembrane, and intracellular.
Assembly of nAChRs is specific for certain subunit partner-
ships, and can be categorized as hetero- and homopentameric.
Received: February 11, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acschemneuro.5b00058
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
Research Article
Figure 1. α7-nAChR agonist prototypes.
Figure 2. α7-nAChR agonist pharmacophore. (A) lead compound TTIn-1, (B) acetylcholine (PDB ID: 3WIP), (C) nicotine (PDB ID: 1UW6), and
(D) epibatidine (PDB ID: 2BYQ, 3SQ6) with pharmacophore map.
studies, limitations in the development of α7-nAChR agonists,
encompassing selectivity, stability, and toxicity, remain.12,19
The availability of a soluble extracellular domain surrogate of
the homomeric α7-nAChR enables one to synthesize candidate
triazole ligands from azide and alkyne building blocks on a
homologous relative of the receptor target of drug action, the
acetylcholine binding protein (AChBP). These lead products
can be identified and determined structures of the complexes
with AChBP by mass spectrometry.20 Structural permutations
of these leads can be synthesized in small quantities and in the
absence of template by metal-catalyzed cycloadditions.21 The
copper-catalyzed azide−alkyne cycloaddition (CuAAC) has
several advantages: regiospecificity, requirements for only
stoichiometric amounts of starting materials for reaction
completion, reaction conditions paralleling biological temper-
atures, and ease of reaction and product purification.22,23
anti-1,2,3-Triazole-containing molecules from metal cata-
lyzed click chemistry have been reported to stimulate α7-
nAChRs as agonists24,25 and antagonize α4β2-nAChRs and
5HT3A receptors.21 The N2 and N3 of the triazole ring serve as
uncharged H-bond acceptors, and the triazole ring itself serves
as a stable, uncharged linker between the cationic center and
the hydrophobic ring.20,22
In this study, the cationic center and aromatic indole of the
triazole-based lead compound, TTIn-1 (Figure 2A), were
modified to improve the selectivity and potency profiles to α7-
nAChRs. Tertiary amines that can cross the blood-brain barrier
(BBB) and be protonated in the CNS were chosen to mimic
the cationic center in the α7-nAChR pharmacophore. Critical
interactions to stabilize the nicotinic ligand are either a
cation−π interaction for quaternary amine with an enveloping
cage of aromatic side chains or a H-bond interaction from a
protonated secondary or tertiary amine or an imine to carbonyl
backbone oxygen of a conserved tryptophan.26 The carbon
chain between the cationic center and the H-bond acceptor or
the triazole was varied to optimize connecting length to
maintain both H-bond interactions from the chosen basic
tertiary amine and the triazole with amino acid residues forming
the binding pocket of α7-nAChRs. The role of functional group
in the hydrophobic region, where the uncharged pyridine of
nicotine or epibatidine resides,27,28 was studied as well.
Receptor binding and functional assays of synthesized
compounds were carried out by radioligand binding and
B
DOI: 10.1021/acschemneuro.5b00058
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
Research Article
Figure 3. Optimization of the lead compound, TTIn-1.
fluorescence resonance energy transfer (FRET) assays,
respectively, on intact cells in culture expressing α7-nAChRs,
α4β2-nAChRs, and 5HT3A receptors.
steric hindrance on the binding affinity and potency. Projected
distances between the basic nitrogen atom and the triazole N1
of the designed compounds were in the range of two to four
C−C bonds (3.7−6.4 Å) based on structures of acetylcholine
(2 C atoms, 4.7 Å, Figure 2B) and SEN34625/WYE-10391416
(4 C atoms, 7.9 Å between the piperidine N and urea NH,
Figure 1). Simple monocyclic amines (IND1−IND6) with
lower pKa values of 6.93−9.66 (Table 1) were chosen to
replace the tropane ring of TTIn-1 (pKa 9.67) for
pharmacokinetic advantages. Once the unprotonated amine
crosses the blood-brain barrier, the pharmacologically active
protonated state forms that is required for hydrogen bond
donation to the backbone carbonyl of a conserved Trp on the
principal face of the nAChR.26 Therefore, the pKa should
accommodate a balance of blood-brain penetration (pharma-
cokinetics) and receptor binding of the protonated species
(pharmacodynamics). Other alicyclic rings, bicyclic analogs and
ring size reduction and expansion were employed. The number
of carbon atoms between the basic amine N and the triazole
ring was fixed at two or three C atoms.
RESULTS AND DISCUSSION
■
1. Design Strategies. The lead tertiary amine TTIn-1 was
optimized based on projected components of an α7-nAChR
pharmacophore (Figure 2) which encompasses (i) the basic
center with its linker (R1 of IND series) and (ii) hydrophobic
aromatic moieties (R2 of PPRD and QND series) as shown in
Figure 3.
In the first set of 14 compounds (Figure 3A), the basic center
was the prime focus. The tropane of TTIn-1 was replaced with
different amine fragments: (i) simple alicyclic rings (IND1−
IND6, IND9), that is, piperidine, morpholine, and piperazine;
(ii) fused alicyclic ring (IND7, IND8); (iii) semirigid analogues
of two-carbon linker (IND10, IND11); (iv) fully flexible
aliphatic amine (IND12); and (v) reduced and enlarged
alicyclic rings (IND13, IND14). Modifications were made to
investigate distances spanning the basic nitrogen and the
triazole ring, the flexibility or spatial configuration, and the
C
DOI: 10.1021/acschemneuro.5b00058
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
Research Article
a
Table 1. Physicochemical Properties and Distance between Basic Amine and N1 and N2 of Triazole Ring
3.8 Å
3.7 Å
pKa
9.17
compd
pKa
9.67
log P
distance to N1 (Å)
distance to N2 (Å)
compd
pKa
9.13
log P
compd
log P
TTIn-1
IND1
IND2
IND3
IND4
IND5
IND6
IND7
IND8
IND9
IND10
IND11
IND12
IND13
IND14
2.7
4.1
3.8
5.1
6.4
3.8
5.1
6.4
3.8
3.7
3.8
3.8
3.8
3.8
3.8
3.8
4.6
4.8
6.1
7.3
4.5
6.1
7.3
4.8
4.1
4.8
4.5
4.5
4.7
4.8
4.5
PPRD1
PPRD2
PPRD3
PPRD4
PPRD5
PPRD6
PPRD7
PPRD8
PPRD9
PPRD10
PPRD11
PPRD12
PPRD13
PPRD14
PPRD15
2.55
2.09
2.76
3.44
2.73
3.75
4.57
2.62
3.24
0.41
3.02
3.02
3.53
3.27
2.75
QND1
QND2
QND3
QND4
QND5
QND6
QND7
QND8
QND9
QND10
QND11
QND12
QND13
2.05
1.6
9.13
9.66
9.65
6.93
7.46
7.45
9.54
9.17
9.28, 4.65
9.67
8.85
8.63
9.26
9.59
3.02
3.08
3.6
9.14, 3.18
9.13
9.17, 3.12
9.17
2.27
2.94
2.24
3.26
4.08
2.13
2.75
−0.08
2.53
2.53
3.03
9.13
9.17
1.95
2.01
2.53
3.15
2.53
1.63
2.25
2.64
2.17
2.58
3.47
9.13
9.17
9.13
9.17
9.13
9.17
8.9
8.93
9.13
9.17
8.70, 5.11
9.11
9.09, 5.11
9.17
9.12
9.17
9.13
9.17
9.4
9.14, 1.07
QND15
9.17, 1.06
2.25
a
Distance between basic amine and N1 of triazole ring of 3D structures cleaned by MarvinSketch was measured with PyMOL. The physicochemical
properties were predicted by MarvinSketch.
a
Scheme 1. Synthesis of the Azide Building Blocks
a
Reagents and conditions: (i) Et3N, EtOH, reflux, overnight; (ii) H2N−NH2, EtOH, reflux 45 min; (iii) freshly prepared TfN3 in toluene, K2CO3,
CuSO4·5H2O, H2O:CH3OH, RT, overnight; (iv) CH3SO2Cl, Et3N, CH2Cl2, 0 °C to RT, 4 h; (v) NaN3, MeCN, reflux, 6 h; (vi) NaN3, H2O, reflux,
overnight; (vii) NaOH, Et2O, 0 °C, 30 min.
D
DOI: 10.1021/acschemneuro.5b00058
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
Research Article
a
Scheme 2. Synthesis of Terminal Alkynes
a
Reagents and conditions: (i) TMSCCH, CuI, PdCl2(PPh3)2, DMF, RT, overnight; (ii) TBAF in THF or K2CO3 in CH3OH, RT, 1 h; (iii)
K2CO3, DMF, RT, overnight.
Scheme 3. Synthesis of anti-1,2,3-Triazole Containing Molecules
In the second set of 29 compounds in the R2 series (Figure
3B), R1 residues were fixed with leads from the first compound
set that showed higher binding affinity to α7-nAChR (Kd < 0.34
μM), R1 of the PPRD series is the 2-(piperidin-1-yl)ethyl
moiety of IND1 and that for the QND series is the quinuclidine
of IND8. The aromatic R2 of both series was separated into two
general categories, (i) mono- and disubstituted benzenes and
(ii) bicyclic aromatic systems: indole and substituted indole,
naphthalene and adenine rings. For the first category (Figure 3-
B1), the p-position of the benzene ring was substituted with H-
bond donors (−NH2 and −OH), H-bond acceptor (−OCH3),
or groups of different electronic influence on the ring system
(−CH3, −Cl, −Ph) to verify the role of functional group at p-
position of benzene ring. The m-substitution of a carboxy
methyl ester was included in the structures of PPRD14,
PPRD15, and QND15 to study the effect from the additional
heteroatom substitution. For the second category (Figure 3-
B2), the indole ring was modified by ring substitutions and
linkage to the triazole.
Predicted physicochemical properties (pKa, log P) and
distance between basic nitrogen atom and N1 of triazole ring
of compounds in IND, PPRD, and QND series are shown in
Table 1.
2. Synthesis. The general synthesis of the designed
compounds started with the synthesis of azide and alkyne
building blocks followed by the CuAAC reaction,20 yielding the
final compounds as shown in Schemes 1−3. Azide building
blocks were prepared by two different methods: the diazo
transfer reaction of amine29 for azide building blocks of IND1−
IND3, IND5−IND10, IND14, PPRD1−PPRD15, QND1−
QND13 and QND15, and the nucleophilic substitution of a
leaving group with sodium azide30,31 for azide building blocks
of IND4, IND11−IND13. Then, the azide building blocks were
reacted with the terminal alkynes that are commercially
available or were prepared in house by the Sonogashira cross-
coupling reaction32 or nucleophilic substitution to yield 1,2,3-
triazoles with 20−95% yield.
3. Biological Evaluation. The binding affinities (1/Kd) of
all synthesized compounds were evaluated in cell based
neurotransmitter fluorescent engineered reporter (CNiFERs)
expressing α7-nAChRs, α4β2-nAChRs, and 5HT3A receptors.
Ligand binding was assayed by competitive radioligand
displacement using [3H]-( )-epibatidine for α7- and α4β2-
nAChRs and [3H]-granisetron for 5HT3A receptors. The α7/
5HT3A chimeric receptor, having an α7-nAChR extracellular
domain and 5HT3A transmembrane and cytoplasmic domains,
was used instead of α7-nAChRs because the signal-to-noise
ratio of the α7/5HT3A chimeric receptors was significantly
higher than the α7-nAChRs. Also, the chimeric receptors
produced more robust and reproducible results (data not
shown). Compounds at 10 μM concentration that compet-
itively dissociated [3H]-( )-epibatidine or [3H]-granisetron
binding to LGIC-CNiFERs more than 50% were further
analyzed over a range of concentrations for Kd values.
Otherwise their Kd values are listed as >10 μM.
Compounds were functionally screened as agonists and
antagonists, for selectivity and potency using FRET assay of
LGIC-CNiFERs transfected cells.33 All LGIC-CNiFERs for
E
DOI: 10.1021/acschemneuro.5b00058
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
Research Article
Table 2. Dissociation Constants for Ligand Binding, Kd, Activation Parameters, EC50, and Functional Antagonism, KA, Values of
Indole Series (IND1−IND14), the Lead Compound TTIn-1, and the Reference α7-nAChR Agonist PNU-282987 from the
a
Intact Cell Binding Assay (Kd) and FRET Assay Using LGIC-CNiFERs
a
Data were analyzed from at least three independent experiments with each experiment containing two or more samples. Values are reported as
means standard deviation (SD). Kd > 10 μM, EC50 > 10 μM, and KA > 10 μM were indicated for the compounds that did not pass the initial
#
screening for binding affinity and functional characterization. *partial agonist; Ccompetitive antagonist; NCnoncompetitive antagonist; p < 0.05
§
δ
compared with TTIn-1; p < 0.05 compared with PNU-282987 from subanalysis; p < 0.05 compared with TTIn-1 from subanalysis. Values are
calculated as described in Methods.
functional characterization are the same as those used for the
competitive radioligand binding with the exception that α7-
nAChRs were used instead of the aforementioned chimeric
receptor. Initially, the agonist properties of tested compounds
(final concentration = 13.3 μM) were assessed directly. Then,
antagonism was subsequently evaluated (final concentration =
10 μM) by adding the standard agonists, ( )-epibatidine for
α7- and α4β2-nAChRs and 5-hydroxytryptamine (5-HT) for
5HT3A receptors. All α7-nAChR functional assays were
performed in the presence of PNU-120596, an α7-nAChR
positive allosteric modulator (PAM), to increase signal and
delay formation of the desensitized state of α7-nAChRs. This
F
DOI: 10.1021/acschemneuro.5b00058
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
Research Article
addition enables measurement of more prolonged FRET
signals.33 Compounds having agonist responses of ≥0.20 of
the maximal response were determined for EC50, and
compounds having inhibition fraction of ≥0.50 were
determined for KA.21 Compounds with EC50 and KA values
>10 μM were not taken for further quantitation of receptor
occupation and responses.
4. Effect of Variation in Basic Amine Ring System (IND
Series). 4.1. Effect on Binding Affinity (1/Kd) or Dissociation
Constant (Kd). Kd values with α7-nAChRs as shown in Table 2
indicate that two compounds (IND1, IND8) exhibited
statistically enhanced affinity or lower Kd for the α7-nAChRs
(Kd = 0.34 and 0.12 μM, respectively) relative to the lead
compound, TTIn-135 (Kd = 4.3 μM). When compared the
binding affinity between these three compounds (TTIn-1,
IND1, IND8) and PNU-282987, a reference α7-nAChR
agonist, the Kd values reveal that the affinity of TTIn-1 is
lower (p < 0.05), but the of IND1 and IND8 are not
significantly different from PNU-282987 (Kd = 0.27 μM).
A comparison of Kd values indicated that extending the
distance between the basic nitrogen and the triazole ring
increased the Kd for α7-nAChRs: IND2 (length 5.1 Å, Kd = 7.2
μM) and IND3 (length 6.4 Å, Kd = 13.8 μM) compared with
IND1 (3.8 Å linker length, Kd = 0.34 μM). Besides linker
elongation, the additional heteroatom (IND4−IND6, IND9) in
the simple alicyclic ring dramatically increases Kd for α7-
nAChRs. This is probably due to (i) the lower pKa of IND4−
IND6 leading to a smaller fraction of the protonated forms
(25−54%) forming H-bond to the backbone carbonyl of the
conserved tryptophan on the principal face of the alpha subunit,
or (ii) protonation of the secondary amine of IND9 at pH 7.4
resulting in an excessive distance (6.6 Å). Both of the fused
alicyclic rings in this study (IND7, IND8) interact with α7-
nAChRs, but only IND8 has the lower dissociation constant.
The lower Kd of IND8 might come from the optimal distance
to maintain both hydrogen bond interactions of IND8. Other
modifications, that is, semirigid analogue (IND10, IND11),
freely flexible aliphatic amine (IND12), and ring size alteration
(IND13, IND14) did not appreciably affect the Kd for α7-
nAChRs.
Most compounds that are able to interact with α7-nAChRs
also bind to 5HT3A receptors as may be expected from the
sequence homology (Supporting Information Figure S1),11,34
but there is a far smaller crossover to the α4β2-nAChRs where
only IND8 showed a low Kd.
4.2. Effect of Basic Amine Center on Agonist and
Antagonist (Pharmacologic Activity). The α7-nAChR agonist
potencies are in general agreement with the binding affinities.
EC50 values of agonists and KA values of antagonists are shown
in Table 2. α7-nAChR dose−response curves of IND1−IND14
are shown in Figure 4.
Only IND1 and IND8 are more potent than TTIn-1, with
EC50 values of 0.17 and 0.028 μM versus 0.57 μM, respectively,
and comparable to PNU-282987 (EC50 0.11 μM). The higher
potency profile of IND1 and IND8 over other alicyclic amines
may arise from the stronger base of piperidine (pKa 9.13) as
well as quinuclidine (pKa 9.17) and optimizing distance
between basic amine and N1 of triazole (3.7−3.8 Å) to
facilitate H-bond formation in the binding pocket of α7-
nAChRs. However, compounds IND10−IND14, which have
the same 3.8 Å distance, did not show enhanced α7-nAChR
potency. This indicates that other factors such as flexibility and
steric constraints govern the potency of α7-nAChR agonist.
Figure 4. α7-nAChR agonist dose−response curves of (A) IND1−
IND3, IND6−IND8 and (B) IND9−IND14. The α7-nAChR
functional assays were conducted in the presence of 10 μM PNU-
120596. FRET ratios were normalized to the maximum response by
100 nM ( )-epibatidine. The concentration−response curves show
Hill coefficients of ∼2, but because of the limited concentration points
and rapid responses, we have not estimated Hill coefficients for the 13
agonists.
For α4β2-nAChRs, IND7 showed noncompetitive antago-
nism with low potency profile (KA = 22.2NC μM) (Supporting
Information Figure S2). Surprisingly, IND8 showed high
affinity binding to α4β2-nAChRs, but did not elicit agonist or
antagonist properties. It is possible that the interacting residues
are not the key residues of α4β2-nAChRs responsible for gating
the cation channel.
The 5HT3A functional responses for 8 of 14 compounds did
not correlate well with the binding affinity measurements. Some
compounds, that is, IND1, IND2, and IND7 apparently bound
to 5HT3A receptors, but did not yield detectable agonist or
antagonist responses. Interestingly, one compound (IND8)
turned out to be 5HT3A partial agonist (EC50 = 0.21 μM).
5HT3A receptor agonist and antagonist dose−response curves
are shown in Supporting Information Figures S3 and S4.
The significant improvement of selectivity and potency
profiles of IND1 and IND8 over the lead compound indicates
that 3.7−3.8 Å is an optimal distance between the basic
nitrogen atom and N1 of the triazole ring for α7-nAChR
agonist activity. IND1, a six-membered monocyclic ring with a
2 methylene linker, is the most potent α7-nAChR agonist
(EC50 = 0.17 μM) and highly selective for α7 over the α4β2-
nAChR, and the 5HT3A receptor. IND8, a bicyclic amine, is the
most potent compound in the series (EC50 = 0.028 μM), but
with a diminished degree of receptor selectivity. Therefore, the
piperidine and quinuclidine rings from IND1 and IND8
representing mono- and bicyclic rings were selected as the
cation center (R1) for further modification of the hydrophobic
R2 in the second compound set, PPRD and QND series,
respectively.
5. Variation in Aromatic Indole Substitution of the
anti-Triazole (PPRD and QND Series). 5.1. Effect on
G
DOI: 10.1021/acschemneuro.5b00058
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
Research Article
a
Table 3. Kd Values of Compounds in PPRD and QND Series from Intact Cell Binding Assay Using LGIC-CNiFERs
a
Kd > 10 μM was indicated in the compounds that did not pass the initial screen. Kd of TTIn-1 is 4.3 μM for α7-nAChRs, >10 μM for α4β2-
#
§
nAChRs, and 6.0 μM for 5HT3A receptors. p < 0.05 compared with IND1 for PPRD series; p < 0.05 compared with IND8 for QND series.
Binding Affinity (1/Kd) or Dissociation Constant (Kd).
Modifications at R2 showed that the anti-triazoles in the
QND series have higher affinity than the corresponding
compounds in PPRD series for both α7-nAChRs and 5HT3A
receptors. Kd values of all compounds that passed the initial
screening with LGIC receptors are shown in Table 3.
5.1.1. Mono- and Disubstituted Benzenes. Binding affinities
to α7-nAChRs for compounds where one substituent on
benzene is a H-bond donor (PPRD2 and PPRD8) appeared to
be greater than those bearing H-bond acceptor (PPRD3) or
heteroatom (PPRD13). The −NH2 of PPRD2 and −OH of
PPRD8 resemble bioisosteres of the −NH of indole; the Kd
values of PPRD2 and PPRD8 are not significantly different
from that of IND1. Since PPRD7 does not bind to all tested
LGIC receptors, the larger biphenyl moiety cannot be
accommodated in the interface binding pocket. An increase
in Kd from steric hindrance is also observed in compounds
bearing benzene ring with disubstitution, PPRD14 (Kd > 10
H
DOI: 10.1021/acschemneuro.5b00058
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
Research Article
a
Table 4. EC50 and KA Values of PPRD Series
agonist EC50 SD (μM)
antagonist KA SD (μM)
α4β2
compd
IND1
α7
5HT3A
>10
α7
5HT3A
0.17 0.05
>10
>10
mono- and disubstituted benzenes
PPRD2
PPRD3
PPRD4
PPRD7
PPRD8
PPRD13
PPRD14
PPRD15
0.38 0.08
15.0 0.6*,#
3.1 0.8
>10
>10
>10
>10
>10
−
−
−
−
−
−
−
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
2.9 0.7
17.8 4.6*,#
>10
39.0 2.7C
>10
>10
−
−
1.4 0.1C
6.0 1.5C
3.1 0.2
−
bicyclic ring systems
PPRD1
PPRD5
PPRD6
PPRD9
PPRD10
PPRD11
PPRD12
7.4 0.7#
>10
>10
>10
−
>10
>10
−
−
−
−
−
>10
>10
1.1 0.2
>10
>10
>10
12.9 4.6*,#
>10
>10
>10
10.5 3.1C, 25.5 12.8NC
15.1 4.0C, 56.2 31.0NC
11.6 3.5C, 18.9 4.8NC
6.6 2.2C, 6.6 1.7NC
4.5 1.5C, 11.3 4.5NC
>10
6.5 1.7C
>10
−
−
>10
2.2 0.3C
>10
a
Data were averaged from at least three independent experiments. All values are reported as means SD. EC50 and KA > 10 μM were indicated for
the compounds that did not pass the initial screen. *Partial agonist; Ccompetitive antagonist; NCnoncompetitive antagonist; #p < 0.05 compared with
IND1. Values are calculated as described in Methods.
a
Table 5. EC50 and KA Values of QND Series
agonist EC50 SD (μM)
antagonist KA SD (μM)
α4β2
compd
IND8
α7
5HT3A
α7
5HT3A
0.028 0.010
0.21 0.08*
−
>10
−
mono- and disubstituted benzenes
QND2
QND3
QND4
QND7
QND8
QND13
QND15
0.052 0.018
0.097 0.025
0.20 0.08
−
0.037 0.009
1.0 0.3§
1.3 0.0§
1.6 0.3*
5.7 1.7*,§
−
−
−
−
>10
−
−
>10
>10
0.50 0.27C, 0.53 0.18NC
>10
6.7 2.4C, 26.2 9.8NC
>10
6.2 1.7C
11.2 2.6C, 22.7 6.4NC
11.6 1.0C
>10
0.90 0.09*
4.9 1.2§
0.26 0.03*
−
−
−
−
−
−
bicyclic ring systems
QND1
QND5
QND6
QND9
QND10
QND11
QND12
0.12 0.04
0.30 0.04*
0.16 0.04*
>10
−
−
−
−
−
−
−
>10
−
−
>10
0.26 0.00
22.1 6.2*,§
1.0 0.0
29.5 7.5*,§
1.6 0.4
>10
7.0 4.5C, 6.2 0.4NC
5.2 1.2C, 6.5 0.2NC
>10
1.4 0.5
>10
−
>10
−
9.3 4.3C, 36.8 21.8NC
0.081 0.009NC
0.45 0.14
0.34 0.13*
>10
−
a
Data were averaged from at least three independent experiments. All values are reported as means SD. EC50 and KA > 10 μM were indicated for
the compounds that did not pass the initial screening. *partial agonist; competitive antagonist; NCnoncompetitive antagonist; §p < 0.05 compared
with IND8. Values are calculated as described in Methods.
C
μM), but not for PPRD15 (Kd = 4.4 μM) when compared to
corresponding compounds with monosubstitutions (PPRD8
and PPRD2). The modified compounds in the PPRD series
were generally selective for α7-nAChRs.
QND15 increases the binding affinity to 5HT3A receptor (Kd =
0.043 μM) that is lower than that for QND2, its corresponding
structure (Kd = 1.2 μM) and equivalent to IND8 (Kd = 0.052
μM).
In the QND series, Kd’s for α7-nAChRs of six compounds
where indole was replaced by mono- and disubstituted benzene
are not significantly different from IND8, except for QND15.
The Kd value of QND8 where the phenyl hydroxyl group acts
as H-bond donor, similar to the −NH of indole, was found to
be the lowest (0.081 μM). Most compounds in QND series
that bind to α7-nAChRs show binding to 5HT3A receptors with
Kd’s in the range of 0.04−2.3 μM. The ester extension of
5.1.2. Bicyclic Ring Systems. The Kd values of all tested
LGIC receptors of compounds in PPRD series that contain
aromatic bicyclic systems increased when compared with IND1.
Methylation of indole ring in PPRD9 and addition of carbonyl
group in the indole ring in PPRD5 were found to increase α7-
nAChR Kd values (Kd = 12.1 and 3.3 μM, respectively).
Changing the −NH indole orientation in the structures of
I
DOI: 10.1021/acschemneuro.5b00058
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
Research Article
PPRD11 and PPRD12 resulted in a reduction of binding
affinity to α7-nAChRs.
PPRD14, and PPRD15 (KA = 39.0, 1.4, and 6.0 μM,
respectively) (Supporting Information Figure S5).
As shown in Figure 5B, the potencies of QND2 (EC50 =
For the QND series, the Kd’s for α7-nAChRs of all
compounds are higher than that for IND8 except for QND1
(Kd = 0.16 μM) and QND12 (Kd = 0.92 μM). Methylation of
the indole ring led to a 15-fold increase in Kd of QND9 (Kd =
1.9 μM) when compared to IND8. Steric hindrance caused by
acyl substitution on the indole ring (QND5, QND11) or
methoxy substitution on the naphthalene ring (QND6)
increased the α7-nAChR Kd from 0.1 μM to >2 μM. However,
changing the orientation of the indole in QND12 did not affect
the α7-nAChR Kd. For α4β2-nAChRs, only two compounds,
QND6 and QND9, showed similar Kd’s to IND8. The Kd’s of
all modified compounds in the QND series for 5HT3A
receptors are same as IND8 except for QND11.
5.2. Effect of the Aromatic Moiety on Agonist and
Antagonist Activity. The compounds in the QND series have
higher α7 agonist potencies than PPRD series. The EC50 and
KA values of PPRD and QND series are shown in Tables 4 and
5, respectively.
5.2.1. Mono- and Disubstituted Benzenes. The potencies of
compounds in PPRD series are in accord with their binding
affinities (Table 4); the α7-nAChR dose−response curves are
shown in Figure 5A. Compounds containing H-bond donor as
0.052 μM), QND3 (EC50 = 0.097 μM), QND4 (EC50 = 0.20
μM), and QND8 (EC50 = 0.037 μM) are not significantly
different from that of IND8 (EC50 = 0.028 μM). The
heteroatom in QND13 and steric hindrance from additional
methyl ester substituent in QND15 decreased α7-nAChR
agonist potency (EC50 = 1.0 and 1.3 μM, respectively), while
the more bulky biphenyl group turned QND7 into a weak α7-
nAChR antagonist (Supporting Information Figure S6). Three
compounds (QND8, QND13, QND15) showed moderate to
low antagonist potencies for α4β2-nAChRs (Supporting
Information Figure S7). Interestingly, all compounds having
heteroatom substitutions at p-position (QND2, QND3, QND8,
QND13, QND15) have 5HT3A receptor agonist properties in
contrast to the 5HT3A antagonist properties of TTIn-1 and
QND4 (Supporting Information Figures S7 and S8).
5.2.2. Bicyclic Ring Systems. The functional responses of
PPRD series that contain bicyclic amines are in agreement with
data from the binding affinity analysis. Removal of H-bond
donor, the −NH indole, markedly reduced the agonist EC50 of
PPRD1 and PPRD9 compared to IND1 (Figure 6A). PPRD9
Figure 6. α7-nAChR agonist dose−response curves for indole
modification, other bicyclic rings, and nitrogen-rich ring for (A)
PPRD and (B) QND series. Assays were performed in the presence of
10 μM PNU-120596. FRET ratios were normalized to the maximum
response by 100 nM ( )-epibatidine.
Figure 5. α7-nAChR agonist dose−response curves of mono- and
disubstituted benzenes for (A) PPRD and (B) QND series. Assays
were performed in the presence of 10 μM PNU-120596. FRET ratios
were normalized to the maximum response with 100 nM
( )-epibatidine.
also acts α4β2-nAChR and 5HT3A receptor antagonists
(Supporting Information Figure S10 and S5, respectively).
Change in position of the NH-indole, the H-bond donor,
turned PPRD11 and PPRD12 into weak and nonselective α7-
nAChR antagonists (Supporting Information Figure S6).
The α7 agonist properties of compounds in QND series
showed a common pattern with the PPRD series. The potency
and selectivity profiles for α7-nAChRs were not enhanced
when compared with IND8. Compounds having heteroatom at
p-position (QND1, QND5, QND9) and QND12 were 5HT3A
a substituent (PPRD2 and PPRD8) have higher potency (0.38
and 2.9 μM) than those with H-bond acceptor (PPRD3) or
heteroatom (PPRD13) (15 and 17.8 μM). Additional m-
substitution decreased the potencies of PPRD14 (EC50 > 10
μM) and PPRD15 (EC50 = 3.1 μM) compared to the
corresponding monosubstituted compounds, PPRD8 (EC50
=
2.9 μM) and PPRD2 (EC50 = 0.38 μM), respectively. Three
compounds from this modified series also show competitive
antagonism behavior to 5HT3A receptors, that is, PPRD8,
J
DOI: 10.1021/acschemneuro.5b00058
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
Research Article
agonists (Supporting Information Figure S8), whereas QND11
is a 5HT3A antagonist (Supporting InformationFigure S9).
Replacement of the indole with an electron-rich adenine ring
altered PPRD10 into α7-nAChR antagonist with moderate
potency (KA = 11.6C, 18.9NC μM) (Supporting Information
Figure S6), whereas QND10 showed modest α7-agonist
activity (EC50 = 29.5 μM).
hydrophobic interactions in the binding pocket of α7-nAChR
for α7-agonist action.
IND1, which has piperidine as the basic amine with a 2-
methylene linker to triazole, is a selective and potent α7-
nAChR agonist in the series. Its agonist potency is 3.5-fold
higher than that of the lead compound TTIn-1 and is highly
selective for α7 over α4β2-nAChRs and the 5HT3A receptors.
IND8 with quinuclidine as the basic center is the most potent
α7 agonist with a moderate selectivity profile: its α7 agonistic
potency increases 20-fold from that of TTIn-1. Modification of
the aromatic motifs in PPRD series did not further enhance
affinity and potency profiles, whereas six compounds of QND
series showed higher α7-nAChR agonistic potency than TTIn-
1. However, alterations of the indole substitution reduced the
selectivity profile with antagonism on α4β2-nAChRs and
antagonism/agonism on 5HT3A receptors.
6. General Structure−Activity Considerations. All
QND compounds have higher affinity and potency than the
corresponding compounds in PPRD series. The activity profiles
might arise from the rigidity of the quinuclidine ring that
restricts the distance between the cationic center and the
hydrogen bond acceptor in the triazole to the optimal 4.1 Å
distance to N2 or 5.3 Å distance to N3, conferring key
interactions in the binding pocket to achieve agonist properties.
Also the quinuclidine compounds present a greater angle of
access to form a hydrogen bond with the backbone carbonyl
oxygen of Trp149 (α7 numbering). The hydrophobic R2 is also
an important motif affecting to the selectivity and potency
profiles. The proper hydrophobic motifs for potent α7-nAChR
agonists for 1,2,3-triazole based molecules appeared to be 5-
substituted indole, benzodioxole, and their isosteric p-
substituted benzene of small group, that is −NH2, −OCH3,
−CH3, and −OH. Information obtained from optimizing
triazole derivatives via 43 compounds in three series (IND,
PPRD, and QND series) revealed that the basic center with
3.7−3.8 Å length of linker, the triazole ring as H-bond acceptor,
and the designated motif are the key pharmacophoric features
of the triazole derivatives to become α7-nAChR agonists
(Figure 7). This investigation also provides an analysis of
METHODS
■
1. General Experimental Detail. Chemical reagents are from
Sigma-Aldrich, Oakwood, Alfa Aesar, Fluka, and Acros, whereas all
solvents are from Fisher. All starting materials with at least 95% purity
were used without further purification. The reactions were monitored
by TLC and LC-MS (HP 1100 Series LC/MSD). Melting points of
the final compounds were measured (Thomas-Hoover Unimelt), and
the structures were elucidated by using FTIR (Thermo Scientific
Nicolet iS5), NMR (Variance Mercury-300, Bruker AMX-400, Bruker
DRX-500), and HRMS (Agilent ESI-TOF). Purity of final compounds
was determined by chromatographic integration of the diode array UV
acquired on an HP 1100 series LC/MSD combining liquid
chromatography with electron spray ionization (ESI) mass spectrom-
etry. Analyses were conducted using 30−100% MeCN/water
containing 0.1% trifluoroacetic acid (TFA); all compounds were
determined to be >95% purity. The physicochemical properties were
predicted by MarvinSketch.
1.1. Azide Preparation (Scheme 1). When starting with amines, the
azide building blocks were prepared by diazo transfer reaction29
(IND1−IND3, IND5−IND10, IND14, PPRD1−PPRD15, QND1−
QND13, QND15) whereas nucleophilic substitution was used for the
molecules bearing good leaving groups,30,31 that is, alicyclic alcohols
(IND4 and IND11) and alkyl chlorides (IND12, IND13) (Scheme 1).
All azido compounds were confirmed by LC-MS and used in CuAAC
without further purification (more than 90% purity).
For diazo transfer reaction, the trifluoromethanesulfonyl azide
(TfN3) was prepared first by the reaction between trifluoromethane-
sulfonic anhydride and 5 equiv of sodium azide in the mixture of water
and toluene. The reaction mixture was stirred vigorously at room
temperature for 2 h and then extracted with toluene to obtain TfN3.
Then the primary amine was reacted with 1.8 equiv of freshly prepared
TfN3 in the mixture of H2O/CH3OH/toluene in the ratio of 3:6:5 at
room temperature overnight to give the azido compound. The
methylene linker of some azide building blocks (IND3, IND5, IND14)
was extended before azidation. Starting amine 1.3 equiv was reacted
with N-(2-bromoethyl) phthalimide, N-(3-bromopropyl)phthalimide,
or N-(4-bromobutyl)phthalimide to vary the length of linker38 using 2
equiv of triethylamine (Et3N) as catalyst. The reaction was refluxed
overnight in EtOH. Then, the phthalimide group was deprotected by
refluxing with hydrazine monohydrate for 45 min to give the primary
amine.
For nucleophilic substitution of alicyclic alcohols, the starting
alcohol was mesylated first by 1.5 equiv of methanesulfonyl chloride
(MsCl) using 1.6 equiv of Et3N as a basic catalyst. The reaction
mixture was run in CH2Cl2 from 0 °C to room temperature for 4 h to
get mesylated alcohol. After that, this mesylated alcohol was refluxed
with 1.2 equiv of sodium azide in MeCN for 6 h to obtain the azido
compound.
Figure 7. Pharmacophore model of α7-nAChR agonist obtained from
this study.
specificity for α7-nAChR agonist activity in relation to the
prevalent α4β2-nAChR and the LGIC homologous 5HT3A
receptor. With the triazole series there is minimal overlap in
potency with α4β2-nAChR, and when observed, it is usually an
α7 agonist versus an α4β2 antagonist. A great parallelism is
seen with α7 agonists and 5HT3 antagonists/agonists.
However, it remains to be determined what complement of
these two activities would be therapeutically advantageous.36
Through high resolution crystallographic studies, it should be
possible to investigate the binding poses, side chain positions
and conformational differences torsional angles in the ligand
bound complexes with AChBP in greater depth.37
The α7-nAChR agonist properties of the compounds in the
three series of 1,4-disubstituted (anti) 1,2,3-triazoles verify that
the triazole ring plays a role not only as the H-bond acceptor
but as a linker for optimizing distance and orientation for key
interactions: H-bond from basic amine with carbonyl backbone
of conserved Trp, the H-bond with the triazole, and
For alkyl chlorides, the starting material was refluxed overnight with
3 equiv of sodium azide in water. After that, diethyl ether was added in
the reaction mixture followed by 4 equiv of NaOH. The reaction
K
DOI: 10.1021/acschemneuro.5b00058
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
Research Article
mixture was stirred in an ice bath for 30 min. Then, the reaction
mixture was extracted with diethyl ether, dried over MgSO4
anhydrous, and concentrated in vacuo to obtain the azido compound.
1.2. Alkyne Preparation. The halogen containing molecule was
reacted with 3 equiv of ethynyltrimethylsilane having PdCl2(PPh3)2
(0.03−0.05 equiv), CuI (0.03−0.05 equiv) as catalyst, and Et3N as
base in DMF at room temperature under N2 atmosphere for 3 h to
overnight via the Sonogashira reaction. After extraction with diethyl
ether or EtOAc and purification by column chromatography, the silyl
group was removed by adding 1 M of TBAF in THF or K2CO3 in
CH3OH. The crude product was purified by column chromatography
to give terminal alkyne products; for example, 5-ethynyl-1H-indole was
prepared via method described above using 5-iodoindole as starting
reagent and purified by column chromatography (10% EtOAc in
initially screened at 10 μM final concentration. Nonspecific binding
was measured by using 100 μM MLA (Tocris Bioscience, Bristol,
U.K.) for α7/5HT3A chimeric receptors, 10 μM varenicline (Tocris
Bioscience, Bristol, U.K.) for α4β2-nAChR, and 300 nM tropisetron
(Tocris Bioscience, Bristol, U.K.) for 5HT3A receptors. After that, the
radioactive ligands were added at 20 nM final concentration of [3H]-
( )-epibatidine for α7/5HT3A chimeric receptors, 5 nM [3H]-
( )-epibatidine for α4β2-nAChR, and 10 nM [3H]-granisetron for
5HT3A receptors. The mixtures were measured on the Wallac 1450
MicroBeta Trilux with 1 h intervals for a total of 15 measurements.
Compounds that displaced over 50% of the bound radioligand from
the LGIC-CNiFERs were analyzed in separate experiments for Kd
values by the Cheng-Prusoff equation using GraphPad Prism. Mean Kd
values and standard deviations were calculated from at least three
independent experiments. The difference of Kd values were statistically
analyzed by one-way ANOVA followed by Tukey’s multiple
comparison test using GraphPad Prism.
1
hexane) to yield intermediate compound (90.39%). H NMR (300
MHz, CDCl3) δ 8.19 (s, 1H), 7.82 (s, 1H), 7.35−7.27 (m, 2H), 7.22−
7.17 (m, 1H), 6.55−6.50 (m, 1H), 0.28 (s, 9H). 13C NMR (76 MHz,
CDCl3) δ 135.6, 127.7, 126.1, 125.3, 125.2, 114.3, 111.1, 107.1, 103.0,
91.3, 0.3. After desilylation and purification (10% EtOAc in hexane), 5-
2.2. Functional Assays for Agonists and Antagonists.33 Agonist
and antagonist elicited responses were characterized using the above
CNiFER cell expressing LGIC receptors. A fluorometric imaging plate
reader system (FlexStation 3; Molecular Devices, Sunnyvale, CA) was
used to detect FRET responses from HEK cells expressed α7-nAChRs,
α4β2-nAChRs, and 5HT3A receptors. A 96-well black, clear-bottom
microplate (Greiner bio-one, Germany) was coated with poly-D-lysine
(Sigma-Aldrich, St. Louis, MO) at 50 μL/well for 30 min and then
washed with phosphate-buffered saline (PBS) (Corning, Cellgro,
Manassas, VA) before plating the cells. After plating and incubation at
37 °C for 1 day, the media was replaced with 100 μL of artificial
cerebrospinal fluid (aCSF: 121 mM NaCl, 5 mM KCl, 26 mM
NaHCO3, 1.2 mM NaH2PO4·H2O, 10 mM glucose, 2.4 mM CaCl2,
1.3 mM MgSO4, 5 mM HEPES, pH 7.4) for α4β2 and 5HT3A
receptors, whereas 10 μM PNU-120596 (Tocris Bioscience, Bristol,
U.K.) in aCSF with 30 min incubation at 37 °C was used for α7-
nAChR. The tested compounds were prepared in aCSF for all
receptors except α7-nAChR, where 10 μM PNU-120596, a positive
allosteric modulator (PAM), was used. The prepared compounds were
added in a separate 96-well polypropylene plate (Costar, Corning,
NY). Experiments were conducted at 37 °C using 436 nm excitation.
Emitted light was collected at 485 and 528 nm. Basal fluorescence was
recorded for 30 s, followed by addition of 50 μL of ligand (first
addition). Measurements were made at 3.84 s intervals for 2 min to
measure the agonist response. After that, the agonist compound, which
was 100 nM final concentration of ( )-epibatidine (Tocris Bioscience,
Bristol, U.K.) for α7- and α4β2-nAChRs, and 1 μM of 5-
hydroxytryptamine (5-HT) (Tocris Bioscience, Bristol, U.K.) for the
5HT3A receptor were added to evaluate the antagonist response of the
test compounds. Agonist and antagonist properties were sequentially
screened at the final concentration of 13.3 and 10 μM, respectively,
which differed from the screening concentration from intact cell
binding assay (10 μM). Therefore, some compounds that do not bind
to LGIC receptors at screening concentrations still exhibit agonist and
antagonist responses in functional screening characterizations.
Compounds whose fraction of the maximal response (Δ/Δmax) was
higher than 0.20 were further evaluated to determine their EC50,
whereas compounds that inhibit Δ/Δmax more than 0.50 were further
characterized to determine the type of antagonism and also calculate
the antagonist dissociation constant (KA). The KA for competitive
antagonists and noncompetitive antagonists were calculated from the
Schild eqs 1 and 2.
1
ethynyl-1H-indole as yellow solid was obtained (79.19%). H NMR
(500 MHz, CDCl3) δ 8.20 (s, 1H), 7.85 (s, 1H), 7.38−7.29 (m, 2H),
7.24−7.19 (m, 1H), 6.57−6.52 (m, 1H), 3.02 (s, 1H). 13C NMR (126
MHz, CDCl3) δ 135.8, 127.8, 126.1, 125.5, 125.3, 113.3, 111.2, 103.0,
85.4, 74.8.
1.3. Copper-Catalyzed Azide−Alkyne Cycloaddition (CuAAC).
The azide building blocks were reacted with an alkyne building block
(1:1) in a mixture of t-BuOH and water (1:1) using 5 mol % of
CuSO4·5H2O and 20 mol % of sodium ascorbate as catalyst. The
reaction mixture was run at room temperature for 2.5−24 h and
purified by column chromatography; for example, 1-(2-azidoethyl)-
piperidine (0.0641 g, 0.42 mmol) was reacted with 5-ethynyl-1H-
indole (0.0499 g, 0.35 mmol) in a 1:1 ratio of t-BuOH and water with
CuSO4·5H2O (0.0050 g, 0.02 mmol) and sodium ascorbate (0.0158 g,
0.08 mmol) as catalyst. The reaction was stirred at room temperature
for 5 h. After that, the reaction mixture was extracted with CH2Cl2.
The crude product was purified by column chromatography (EtOAc)
to give a pale yellow solid compound of 5-(1-(2-(piperidin-1-yl)ethyl)-
1H-1,2,3-triazol-4-yl)-1H-indole (IND1) (68.86%). FTIR (ATR)
(cm−1) 3318, 3148, 3083, 2924, 1615, 1547, 1452, 1438, 1346,
1225, 1121, 1054, 892, 795, 771, 745. 1H NMR (500 MHz, DMSO-d6)
δ 8.41 (s, 1H), 8.00 (dd, J = 1.6, 0.8 Hz, 1H), 7.58 (dd, J = 8.4, 1.5 Hz,
1H), 7.44 (d, J = 8.4 Hz, 1H), 7.36 (dd, J = 2.9, 2.2 Hz, 1H), 6.50−
6.44 (m, 1H), 4.48 (t, J = 6.5 Hz, 2H), 2.76 (t, J = 6.5 Hz, 2H), 2.47−
2.34 (m, 4H), 1.51−1.44 (m, 4H), 1.42−1.33 (m, 2H). 13C NMR
(126 MHz, DMSO-d6) δ 147.7, 135.6, 127.8, 126.0, 122.0, 120.3,
119.0, 116.7, 111.7, 101.4, 57.8, 53.8, 47.0, 25.5, 23.9; mp = 148−149
°C. HRMS calculated (C17H21N5, MH+) 296.1871, found 296.1871.
2. Activity Testing. Cell lines expressing specific subtypes of
receptors and a genetically encoded fluorescence resonance energy
transfer (FRET)-based calcium sensor (TN-XXL) were prepared from
sequential transfections and clone selection.33 Human α7-nAChRs and
α4β2-nAChRs were expressed in HEKtsA201 cells, whereas hα7/
m5HT3A chimeric receptors and mouse 5HT3A receptors were
expressed in HEK293 cells. Cells were cultured in DMEM (Corning,
Cellgro, Manassas, VA) supplemented with 10% FBS (Gibco, Life
Technologies, Grand Island, NY) and 1% glutamine (Gibco, Life
Technologies, Grand Island, NY) and incubated at 37 °C with 10%
CO2.
2.1. Intact Cell Binding Assay. Cell based neurotransmitter
fluorescent engineered reporters (CNiFERs) expressing Ca2+ per-
meable LGIC receptors were used to determine the Kd values of the
compounds. Cells were harvested by centrifugation (Beckman coulter,
Brea, CA) at 500 rpm for 5 min, and then mixed and incubated on ice
for 30 min, after counting, with Wheat Germ Agglutinin SPA beads
(20 mg/mL) (PerkinElmer, Waltham, MA) in Hank’s balanced salt
solution (HBSS) (Gibco, Life Technologies, Grand Island, NY) at the
specified concentrations: (1) hα7/m5HT3A chimera-TN-XXL,
200 000 cells/well, 1 mg/mL SPA beads; (2) hα4β2-TN-XXL,
100 000 cells/well, 0.5 mg/mL SPA beads; and (3) m5HT3A-TN-
XXL, 100 000 cells/well, 1 mg/mL SPA beads. The compounds were
KA = [A]/[DR − 1]
(1)
KA = [A]/[(Δmax/Δ) − 1]
(2)
where [A] is the concentration of compound, DR (dose ratio) is the
EC50 ratio of tested compound over the control compound, which is
( )-epibatidine for α7- and α4β2-nAChR, and 5HT for 5HT3A, and
Δ/Δmax is the fraction of the maximal response. Mean values and
standard deviations were calculated from at least three independent
experiments. The difference of EC50 and KA values were statistically
L
DOI: 10.1021/acschemneuro.5b00058
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
Research Article
analyzed by one-way ANOVA followed by Tukey’s multiple
comparison test using GraphPad Prism.
(4) Schaaf, C. P. (2014) Nicotinic acetylcholine receptors in human
genetic disease. Genet. Med. 16 (9), 649−656.
(5) Jensen, A. A., Frolund, B., Liljefors, T., and Krogsgaard-Larsen, P.
(2005) Neuronal nicotinic acetylcholine receptors: structural reve-
lations, target identifications, and therapeutic inspirations. J. Med.
Chem. 48 (15), 4705−4745.
ASSOCIATED CONTENT
* Supporting Information
■
S
Synthesis and characterization of alkyne and 1,2,3-triazoles.
Sequence alignment of human α7-nAChR with human and
mouse 5HT3A receptors (Figure S1). IND, PPRD, and QND
series dose−response curve of α4β2-nAChRs and 5HT3A
receptors (Figures S2−S10). The Supporting Information is
available free of charge on the ACS Publications website at
DOI: 10.1021/acschemneuro.5b00058.
(6) Gaimarri, A., Moretti, M., Riganti, L., Zanardi, A., Clementi, F.,
and Gotti, C. (2007) Regulation of neuronal nicotinic receptor traffic
and expression. Brain Res. Rev. 55 (1), 134−143.
(7) Lippiello, P., Bencherif, M., Hauser, T., Jordan, K., Letchworth, S.,
and Mazurov, A. (2007) Nicotinic receptors as targets for therapeutic
discovery. Expert Opin. Drug Discovery 2 (9), 1185−1203.
(8) Taly, A., Corringer, P. J., Guedin, D., Lestage, P., and Changeux,
J. P. (2009) Nicotinic receptors: allosteric transitions and therapeutic
targets in the nervous system. Nat. Rev. Drug Discovery 8 (9), 733−750.
(9) Briggs, C. A., Anderson, D. J., Brioni, J. D., Buccafusco, J. J.,
Buckley, M. J., Campbell, J. E., Decker, M. W., Donnelly-Roberts, D.,
Elliott, R. L., Gopalakrishnan, M., Holladay, M. W., Hui, Y. H.,
Jackson, W. J., Kim, D. J., Marsh, K. C., O’Neill, A., Prendergast, M. A.,
Ryther, K. B., Sullivan, J. P., and Arneric, S. P. (1997) Functional
characterization of the novel neuronal nicotinic acetylcholine receptor
ligand GTS-21 in vitro and in vivo. Pharmacol., Biochem. Behav. 57 (1−
2), 231−241.
AUTHOR INFORMATION
Corresponding Author
*Phone: 662 6448695. E-mail: opa.vaj@mahidol.ac.th.
Author Contributions
Study conception and design: K.A., V.V.F., O.V., and P.T..
Acquisition of data: K.A. Analysis and interpretation of data:
K.A. Drafting of manuscript: K.A. Critical revision: K.A., V.V.F.,
O.V., and P.T.
■
(10) Hibbs, R. E., Sulzenbacher, G., Shi, J., Talley, T. T., Conrod, S.,
Kem, W. R., Taylor, P., Marchot, P., and Bourne, Y. (2009) Structural
determinants for interaction of partial agonists with acetylcholine
binding protein and neuronal alpha7 nicotinic acetylcholine receptor.
EMBO J. 28 (19), 3040−3051.
(11) Wallace, T. L., Callahan, P. M., Tehim, A., Bertrand, D.,
Tombaugh, G., Wang, S., Xie, W., Rowe, W. B., Ong, V., Graham, E.,
Terry, A. V., Jr., Rodefer, J. S., Herbert, B., Murray, M., Porter, R.,
Santarelli, L., and Lowe, D. A. (2011) RG3487, a novel nicotinic
alpha7 receptor partial agonist, improves cognition and sensorimotor
gating in rodents. J. Pharmacol. Exp. Ther. 336 (1), 242−253.
(12) Toyohara, J., and Hashimoto, K. (2010) alpha7 Nicotinic
Receptor Agonists: Potential Therapeutic Drugs for Treatment of
Cognitive Impairments in Schizophrenia and Alzheimer’s Disease.
Open Med. Chem. J. 4, 37−56.
(13) Vicens, P., Ribes, D., Torrente, M., and Domingo, J. L. (2011)
Behavioral effects of PNU-282987, an alpha7 nicotinic receptor
agonist, in mice. Behav Brain Res. 216 (1), 341−348.
(14) Mazurov, A., Hauser, T., and Miller, C. H. (2006) Selective
alpha7 nicotinic acetylcholine receptor ligands. Curr. Med. Chem. 13
(13), 1567−1584.
(15) Radek, R. J., Robb, H. M., Stevens, K. E., Gopalakrishnan, M.,
and Bitner, R. S. (2012) Effects of the novel alpha7 nicotinic
acetylcholine receptor agonist ABT-107 on sensory gating in DBA/2
mice: pharmacodynamic characterization. J. Pharmacol. Exp. Ther. 343
(3), 736−745.
(16) Ghiron, C., Haydar, S. N., Aschmies, S., Bothmann, H.,
Castaldo, C., Cocconcelli, G., Comery, T. A., Di, L., Dunlop, J., Lock,
T., Kramer, A., Kowal, D., Jow, F., Grauer, S., Harrison, B., La Rosa, S.,
Maccari, L., Marquis, K. L., Micco, I., Nencini, A., Quinn, J.,
Robichaud, A. J., Roncarati, R., Scali, C., Terstappen, G. C., Turlizzi,
E., Valacchi, M., Varrone, M., Zanaletti, R., and Zanelli, U. (2010)
Novel alpha-7 nicotinic acetylcholine receptor agonists containing a
urea moiety: identification and characterization of the potent, selective,
and orally efficacious agonist 1-[6-(4-fluorophenyl)pyridin-3-yl]-3-(4-
piperidin-1-ylbutyl) urea (SEN34625/WYE-103914). J. Med. Chem. 53
(11), 4379−4389.
(17) Horenstein, N. A., Leonik, F. M., and Papke, R. L. (2008)
Multiple pharmacophores for the selective activation of nicotinic
alpha7-type acetylcholine receptors. Mol. Pharmacol. 74 (6), 1496−
1511.
(18) Bunnelle, W. H., Dart, M. J., and Schrimpf, M. R. (2004) Design
of ligands for the nicotinic acetylcholine receptors: the quest for
selectivity. Curr. Top Med. Chem. 4 (3), 299−334.
Funding
This work was supported by Thailand Research Fund (TRF)
through the Royal Golden Jubilee Ph.D. Program (Grant No.
PHD/0113/2551) to K.A. and O.V., the Office of the High
Education Commission and Mahidol University under the
National Research Universities to O.V., National Institutes of
Health GM18360 to P.T., and GM087620 to V.V.F.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We acknowledge Dr. Ida Drier, Dr. Rakesh Sit, and Dr. James
Oakdale for providing some synthesis materials and advice, and
Dr. Akos Nemecz for suggestions on the pharmacological
evaluation.
ABBREVIATIONS
■
5HT, 5-hydroxytryptamine; AChBP, acetylcholine binding
protein; aCSF, artificial cerebrospinal fluid; AD, Alzheimer’s
disease; BBB, blood-brain barrier; CNiFERs, cell based
neurotransmitter fluorescent engineered reporters; CuAAC,
copper-catalyzed azide−alkyne cycloaddition; DMEM, Dulbec-
co’s modified Eagle’s medium; DR, dose−response; ESI,
electrospray ionization; FRET, fluorescence resonance energy
transfer; FTIR, Fourier transform infrared spectroscopy; HBSS,
Hank’s balanced salt solution; HRMS, high resolution mass
spectrometry; LC-MS, liquid chromatography mass spectrom-
etry; LGIC, ligand-gated ion channel; Ls, Lymnaea stagnalis;
MLA, methyllycaconitine; nAChR, nicotinic acetylcholine
receptor; NMR, nuclear magnetic resonance; RBA, radioligand
binding assay; SPA, signal proximity assay
REFERENCES
■
(1) Gotti, C., and Clementi, F. (2004) Neuronal nicotinic receptors:
from structure to pathology. Prog. Neurobiol. 74 (6), 363−396.
(2) Dani, J. A., and Bertrand, D. (2007) Nicotinic acetylcholine
receptors and nicotinic cholinergic mechanisms of the central nervous
system. Annu. Rev. Pharmacol. Toxicol. 47, 699−729.
(3) Gotti, C., Zoli, M., and Clementi, F. (2006) Brain nicotinic
acetylcholine receptors: native subtypes and their relevance. Trends
Pharmacol. Sci. 27 (9), 482−491.
M
DOI: 10.1021/acschemneuro.5b00058
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
Research Article
(19) Mazurov, A. A., Speake, J. D., and Yohannes, D. (2011)
Discovery and development of alpha7 nicotinic acetylcholine receptor
modulators. J. Med. Chem. 54 (23), 7943−7961.
(20) Grimster, N. P., Stump, B., Fotsing, J. R., Weide, T., Talley, T.
T., Yamauchi, J. G., Nemecz, A., Kim, C., Ho, K. Y., Sharpless, K. B.,
Taylor, P., and Fokin, V. V. (2012) Generation of candidate ligands for
nicotinic acetylcholine receptors via in situ click chemistry with a
soluble acetylcholine binding protein template. J. Am. Chem. Soc. 134
(15), 6732−6740.
(21) Yamauchi, J. G., Gomez, K., Grimster, N., Dufouil, M., Nemecz,
A., Fotsing, J. R., Ho, K. Y., Talley, T. T., Sharpless, K. B., Fokin, V. V.,
and Taylor, P. (2012) Synthesis of selective agonists for the alpha7
nicotinic acetylcholine receptor with in situ click-chemistry on
acetylcholine-binding protein templates. Mol. Pharmacol. 82 (4),
687−699.
(22) Hein, C. D., Liu, X. M., and Wang, D. (2008) Click chemistry, a
powerful tool for pharmaceutical sciences. Pharm. Res. 25 (10), 2216−
2230.
nicotinic receptor agonist, is also a potent agonist of the 5-HT3
receptor. Eur. J. Pharmacol. 452 (2), 137−144.
(35) Yamauchi, J. G. (2012) Approaches in developing selective
pharmacological lead compounds for the alpha7 nicotinic acetylcholine
receptor with click-chemistry. Ph.D. Dissertation, University of
California, San Diego, CA.
(36) Hashimoto, K. (2015) Tropisetron and its targets in Alzheimer’s
disease. Expert Opin. Ther Targets 19 (1), 1−5.
(37) Kaczanowska, K., Harel, M., Radic, Z., Changeux, J. P., Finn, M.
G., and Taylor, P. (2014) Structural basis for cooperative interactions
of substituted 2-aminopyrimidines with the acetylcholine binding
protein. Proc. Natl. Acad. Sci. U. S. A. 111 (29), 10749−10754.
(38) Coppo, F. T., Maskell, E. S. L., Redshaw, S., Skidmore, J., Ward,
R. W., and Wilson, D. M. (2007) 2-Phenyl-5-amino-1,3,4-oxadiazoles
and their use as nicotinic acetylcholine receptor ligands. European
Patent WO2007138033.
(23) Rostovtsev, V. V., Green, L. G., Fokin, V. V., and Sharpless, K. B.
(2002) A stepwise huisgen cycloaddition process: copper(I)-catalyzed
regioselective “ligation” of azides and terminal alkynes. Angew. Chem.,
Int. Ed. Engl. 41 (14), 2596−2599.
(24) Pin, F., Vercouillie, J., Ouach, A., Mavel, S., Gulhan, Z., Chicheri,
G., Jarry, C., Massip, S., Deloye, J. B., Guilloteau, D., Suzenet, F.,
Chalon, S., and Routier, S. (2014) Design of alpha7 nicotinic
acetylcholine receptor ligands in quinuclidine, tropane and quinazoline
series. Chemistry, molecular modeling, radiochemistry, in vitro and in
rats evaluations of a [(18)F] quinuclidine derivative. Eur. J. Med. Chem.
82, 214−224.
(25) Routier, S., Suzenet, F., Pin, F., Chalon, S., Vercouillie, J., and
Guilloteau, D. (2014) 1,4-disubstituted 1,2,3-triazoles, methods for
preparing same, and diagnostic and therapeutic uses thereof. U.S.
Patent US 20140030191 A1.
(26) Rucktooa, P., Smit, A. B., and Sixma, T. K. (2009) Insight in
nAChR subtype selectivity from AChBP crystal structures. Biochem.
Pharmacol. 78 (7), 777−787.
(27) Celie, P. H., van Rossum-Fikkert, S. E., van Dijk, W. J., Brejc, K.,
Smit, A. B., and Sixma, T. K. (2004) Nicotine and carbamylcholine
binding to nicotinic acetylcholine receptors as studied in AChBP
crystal structures. Neuron 41 (6), 907−914.
(28) Hansen, S. B., Sulzenbacher, G., Huxford, T., Marchot, P.,
Taylor, P., and Bourne, Y. (2005) Structures of Aplysia AChBP
complexes with nicotinic agonists and antagonists reveal distinctive
binding interfaces and conformations. EMBO J. 24 (20), 3635−3646.
(29) Beckmann, H. S., and Wittmann, V. (2007) One-pot procedure
for diazo transfer and azide-alkyne cycloaddition: triazole linkages
from amines. Org. Lett. 9 (1), 1−4.
(30) Suzuki, T., Ota, Y., Ri, M., Bando, M., Gotoh, A., Itoh, Y.,
Tsumoto, H., Tatum, P. R., Mizukami, T., Nakagawa, H., Iida, S.,
Ueda, R., Shirahige, K., and Miyata, N. (2012) Rapid discovery of
highly potent and selective inhibitors of histone deacetylase 8 using
click chemistry to generate candidate libraries. J. Med. Chem. 55 (22),
9562−9575.
(31) Dimitrov, I., Jankova, K., and Hvilsted, S. (2010) Synthesis of
polystyrene-based random copolymers with balanced number of basic
or acidic functional groups. J. Polym. Sci., Part A: Polym. Chem. 48 (9),
2044−2052.
(32) Md Tohid, S. F., Ziedan, N. I., Stefanelli, F., Fogli, S., and
Westwell, A. D. (2012) Synthesis and evaluation of indole-containing
3,5-diarylisoxazoles as potential pro-apoptotic antitumour agents. Eur.
J. Med. Chem. 56, 263−270.
(33) Yamauchi, J. G., Nemecz, A., Nguyen, Q. T., Muller, A.,
Schroeder, L. F., Talley, T. T., Lindstrom, J., Kleinfeld, D., and Taylor,
P. (2011) Characterizing ligand-gated ion channel receptors with
genetically encoded Ca2++ sensors. PLoS One 6 (1), e16519.
(34) Broad, L. M., Felthouse, C., Zwart, R., McPhie, G. I., Pearson, K.
H., Craig, P. J., Wallace, L., Broadmore, R. J., Boot, J. R., Keenan, M.,
Baker, S. R., and Sher, E. (2002) PSAB-OFP, a selective alpha 7
N
DOI: 10.1021/acschemneuro.5b00058
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX