Structural studies, homology modeling and molecular docking of novel non-competitive antagonists of GluK1/GluK2 receptors

Article abstract of DOI:10.1016/j.bmc.2013.12.013

Non-competitive ligands of kainate receptors have focused significant attention as medicinal compounds because they seem to be better tolerated than competitive antagonists and uncompetitive blocker of these receptors. Here we present structural studies (

Full text of DOI:10.1016/j.bmc.2013.12.013

Bioorganic & Medicinal Chemistry 22 (2014) 787–795
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Structural studies, homology modeling and molecular docking of
novel non-competitive antagonists of GluK1/GluK2 receptors
c
d
e
e
Agnieszka A. Kaczor a,b, , Zbigniew Karczmarzyk , Andrzej Fruzinski , Kalevi Pihlaja , Jari Sinkkonen ,
⇑
´
Kirsti Wiinämaki e, Christiane Kronbach f, Klaus Unverferth f, Antti Poso b, Dariusz Matosiuk a
a Department of Synthesis and Chemical Technology of Pharmaceutical Substances with Computer Modeling Lab, Faculty of Pharmacy with Division of Medical Analytics, 4A
´
Chodzki St., PL-20093 Lublin, Poland
b School of Pharmacy, University of Eastern Finland, Yliopistonranta 1, PO Box 1627, FI-70211 Kuopio, Finland
c Department of Chemistry, Siedlce University, 3 Maja 54 St., PL-08110 Siedlce, Poland
d
_
´
Institute of General and Ecological Chemistry, Technical University, Zeromskiego115 St.,PL-90924 Łódz, Poland
e Department of Chemistry, University of Turku, Vatselankatu 2, FI-20014 Turku, Finland
f Biotie Therapie GmbH, Meissner Str. 191, DE-01445 Radebul, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Non-competitive ligands of kainate receptors have focused significant attention as medicinal compounds
because they seem to be better tolerated than competitive antagonists and uncompetitive blocker of
these receptors. Here we present structural studies (X-ray structure determination, NMR and MS spectra)
of novel indole-derived non-competitive antagonists of GluK1/GluK2 receptors, homology models of
GluK1 and GluK2 receptors based on novel AMPA receptor template as well as molecular docking of
ligands to their molecular targets. We find that the allosteric site is in the receptor transduction domain,
in one receptor subunit, not between the two subunits as it was indicated by our earlier studies.
Ó 2013 Elsevier Ltd. All rights reserved.
Received 25 September 2013
Revised 2 December 2013
Accepted 5 December 2013
Available online 12 December 2013
Keywords:
Homology modeling
Indole derivatives
Kainate receptors
Molecular docking
X-ray structure determination
1. Introduction
potentiation mechanism. Thus, antagonists of kainate receptors are
potential anti-seizure and neuroprotective agents. Moreover, as
The glutamatergic system remains an attractive molecular
target for pharmacological intervention.1–3 Ligands acting on
non-competitive antagonists of AMPA receptors are well tolerated
in preclinical and clinical studies,6 it may be expected that it will
also concern such ligands of kainate receptors.
ionotropic glutamate receptors (iGluRs: NMDA, N-methyl-
D-aspar-
tate; AMPA, -amino-3-hydroxy-5-methyl-4-isooxazolepropionic
a
The research on non-competitive antagonists of kainate recep-
tors is hindered by the fact that only three series of such com-
pounds have been obtained up to now,7–9 (Fig. 1). Recently we
have reported 2,3,5-trisubstituted indoles 1–4 (intermediates)
and 1,2,3,5-tetrasubstituted indole derivatives 5–10 which belong
to most active non-competitive antagonists of GluK1 receptor and
are the first known such ligands of GluK2 receptor (Fig. 2).9 We
also proposed a pharmacophore model for these ligands.9 More-
over, we have suggested a binding site for them in the receptor
transduction domain10 thanks to construction of whole receptor
models.10,11 In order to build models of GluK1 and GluK2 receptors
we used separate templates for each domain (transmembrane
domain, transduction domain, ligand-binding domain and
N-terminal domain). Although the constructed models differed
significantly from the later obtained crystal structure of GluA2
receptor,12 they exhibited the correct twofold symmetry and
correct dimensions. Here we present structural studies of these
acid and kainate receptors) or metabotropic glutamate receptors
(mGluRs) are potential drug candidates for the treatment of neuro-
degenerative diseases (Alzheimer’s disease, Parkinson’s disease,
Huntington’s disease), epilepsy as well as schizophrenia, anxiety
and memory disorders.2,3 Although only a few glutamate receptors
ligands turned out to clinically useful (firstly, because of a crucial
role of glutamatergic system in many physiological processes and
secondly—due to the unfavorable psychotropic side effects,
traditionally linked with high-affinity NMDA receptor antagonists),
ligands of kainate receptor subfamily seem to be especially
promising.4,5 Kainate receptors are involved in epileptogenesis
and inducing synaptic plasticity, mainly via mossy fiber long-term
⇑
Corresponding author. Tel.: +48 815357365; fax: +48 815357355.
E-mail addresses: agnieszka.kaczor@umlub.pl (A.A. Kaczor), darek.matosiuk@
umlub.pl (D. Matosiuk).
0968-0896/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2013.12.013

788
A. A. Kaczor et al. / Bioorg. Med. Chem. 22 (2014) 787–795
Figure 1. Non-competitive antagonists of GluK1 receptors.
Table 1
Pharmacological activity of novel non-competitive antagonists of GluK1/GluK2
receptors, nd–not determined8
R1
R2
R3
R4
IC50, lM
GluK1
GluK2
1
2
3
H
H
H
Ph
4-OMePh
Ph
H
CH3
CH3
OCH3
H
OCH3
nd
nd
nd
nd
nd
nd
4
H
OCH3
nd
nd
5
6
7
8
9
C2H5
4-ClBn
C2H5
C2H5
C2H5
4-OMePh
4-OMePh
Ph
4-OMePh
Ph
CH3
CH3
H
CH3
CH3
OCH3
OCH3
OCH3
H
4.0
0% (100
12.0
32% (100
32% (100
0.7
20% (100
6.7
10.0
6.0
l
M)
lM)
l
l
M)
M)
OCH3
10
C2H5
OCH3
0% (100
lM)
1% (100 lM)
Figure 2. Synthesis of indole derivatives.
0.20 ꢀ 0.17 ꢀ 0.2 of 6,
x scans. The multi-scan absorption correc-
tion was applied (SADABS13). The structures were solved by direct
methods using SIR9214 and refined by full-matrix least-squares
with SHELXL97.15 All hydrogen atoms in 5 were located in a differ-
ence Fourier map and their coordinates were refined isotropically
[C–H = 0.94(3)–1.11(4) Å and Uiso(H) = 1.5Ueq(C)]. In the case of 7
the H atoms were treated as riding on their parent C atoms with
C–H distances of 0.93 (aromatic), 0.97 (CH2) and 0.96 (CH3) and
Uiso(H) = 1.5Ueq(C). The assumed absolute crystal structure of 5
was confirmed by refinement of the Flack parameter.16 The crystal
and experimental data are listed in Table 2. Molecular graphics
were prepared using ORTEP3 for Windows.17 PARST18 and PLA-
TON19 were used for geometrical calculations. All calculations were
performed using WINGX v. 1.64.05 package.20 CCDC-930477 for 5
and CCDC-930478 for 6 contain the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge at http://www.ccdc.cam.ac.uk/conts/retrieving.html [or
from the Cambridge Crystallographic Data Centre (CCDC), 12 Union
Road, Cambridge CB2 1EZ, UK; fax: +44(0) 1223 336 033; email:
deposit@ccdc.cam.ac.uk].
compounds, homology models of GluK1 and GluK2 receptors built
using a novel AMPA GluA2 receptor template12 and molecular
docking to the transduction domain of the receptors.
2. Materials and methods
2.1. Chemistry
The intermediates 1–4 were synthesized by Fischer or Bischler
indolization method and were alkylated with ethyl bromide or
4-chlorobenzyl chloride in the presence of sodium hydride in
anhydrous DMF to give final compounds 5–10 as reported
previously (Fig. 2).9
2.2. Pharmacology
The final products, 5–10 were tested for their affinity to GluK1
and GluK2 receptors as reported previously.9 In particular, the
investigations with the 3H-kainate binding assay showed no
inhibition, which makes it possible to conclude about the
non-competitive type of antagonism for compounds 5–10.
Pharmacological data are presented in Table 1.
2.4. NMR studies
NMR spectra were acquired using Bruker Avance 500 spectrom-
eter (equipped with BBO-5 mm-Zgrad probe) operating at
500.13 MHz for 1H and 125.77 MHz for 13C, respectively. Spectra
were recorded at 25 °C using CDCl3 as a solvent with a non-
spinning sample in 5 mm NMR-tubes. Spectra were processed by
a PC with Windows XP operating system and XWin-NMR software.
Proton and carbon spectra were referenced internally to TMS signal
using value 0.00 ppm.
2.3. X-ray structure analysis
Prismatic colorless crystals of 5 and 6 suitable for X-ray
diffraction analysis were grown by slow evaporation of an ethanol
solution. X-ray data were collected on the Bruker SMART APEX CCD
diffractometer; crystal sizes: 0.25 ꢀ 0.22 ꢀ 0.05 of
5
and

A. A. Kaczor et al. / Bioorg. Med. Chem. 22 (2014) 787–795
789
Table 2
Crystal data and structure refinement for 5and 6
5
6
Empirical formula
Formula weight
Temperature (K)
Crystal system
Space group
C
19H21NO2
C24H22NO2Cl
391.88
293(2)
295.37
293(2)
Orthorhombic
P212121
Triclinic
ꢀ
P1
Unit cell parameters
a (Å)
b (Å)
c (Å)
8.720(2)
9.553(2)
20.024(4)
9.088(2)
12.477(3)
18.704(4)
87.93(3)
a
(°)
b (°)
87.82(3)
c
(°)
80.37(3)
2088.5(7)
4
1.246
824
V (Å3)
1668.2(6)
4
1.176
Z
Dcalc (g cmꢁ3
F(000)
)
632
k (Cu K
a) (Å)
1.54178
1.54178
Cell parameters from
h range for lattice parameters (°)
Absorption coefficient
870 reflections
38.35–68.90
0.600
104 reflections
2.36–70.14
1.761
l )
(mmꢁ1
Tmin/Tmax
h Range for data collection (°)
Index ranges h, k, l
0.864/0.971
4.42–70.05
ꢁ10/10, ꢁ11/11, ꢁ24/24
19196
0.720/0.966
9.30–39.80
ꢁ11/11, ꢁ15/15, ꢁ22/22
24160
No. of measured reflections
No. of independent reflections
No. of observed reflections
Refinement method
3166(Rint = 0.0192)
7612 (Rint = 0.0187)
3054 with I > 2
r(I)
6148 with I > 2r(I)
Full-matrix least-squares on F2
Full-matrix least-squares on F2
Final R indices: R, wR(F2)
Goodness-of-fit on F2, S
Data/parameters
0.0337, 0.1019
1.034
3166/264
0.0070(7)
0.1(2)
0.0502, 0.1476
1.137
7612/506
0.0011(3)
0.1(2)
Extinction coefficient
Flack parameter
+0.134 and ꢁ0.131
+0.309 and ꢁ0.364
Largest diff. peak and hole (e Åꢁ3
)
0.000
0.001
(D/r)
max
1H NMR spectra and 13C NMR proton-decoupled spectra were
physiological pH Epik module was used.26 The molecular struc-
tures of the investigated compounds in the ground state (in vacuo)
were further optimized with the B3LYP DFT (the variant of DFT
method using Becke’s three parameter hybrid functional (B3)27
with correlation functional such as the one proposed by Lee, Yang,
and Parr (LYP)28) using 6-31G(d,p) as included in GAUSSIAN 09.29
Molecular docking was performed with Glide from the Schrodinger
Suite. Ligand-receptor complexes were inserted into POPC lipid
bilayer and water with Schrodinger suite of programs and sodium
and potassium ions were added to balance protein charges and
then till concentration of 0.15 M. The stability of ligand-receptor
complexes was assessed by molecular dynamics simulations with
Desmond v. 3.0.3.1.30 The ligands-receptor complexes in lipid
bilayer were minimized and subjected to MD first in the NVT
ensemble for 1 ns and then in NPT ensemble for 20 ns.
acquired with single-pulse excitation and 30° flip angle. 1 Hz expo-
nential weighting was applied prior to Fourier transformation in
carbon spectra. Gradient selected DQF-COSY spectra were acquired
with cosygpmfqf pulse program (pulse programs refer to original
ones installed by Bruker). Gradient selected NOESY spectra were
acquired with noesygpph pulse program. Gradient selected
1H–13C HSQC spectra were acquired with hsqcetgpsisp.2 pulse pro-
gram (using shaped pulses). Gradient selected 1H–13C HMBC spec-
tra were acquired with hmbcgplpndqf pulse program.
2.5. MS studies
The electron ionization (EI) mass spectra were recorded on a VG
ZABSpec mass spectrometer (VG Analytical, Division of Fisons,
Manchester, UK), that was equipped with Opus V3.3X program
package (Fisons Instruments, Manchester, UK).
The following software were also used for visualization of
results: Chimera v.1.5.3,31 Mercury v.2.4,32 VegaZZ v.2.4.0.25,33
Yasara Structure v.11.9.18,34 PyMol v.0.99,35 Discovery Studio v.
3.136 and ArgusLab.37
2.6. Molecular modeling
Multiple alignment was performed with Muscle20 and MOE
Molecular Operating Environment.21 Crystal structure of AMPA
GluA2 receptor (PDB ID: 3KG212) was selected as the main tem-
plate. Additional templates were used for N-terminal domain
(crystal structure of the GluK2/GluK5 NTD tetramer assembly,
PDB ID: 3QLV22) and ligand binding domain (crystal structure of
GluK1 ligand-binding domain (S1S2) in complex with an antago-
nist, PDB ID: 4DLD23). Homology modeling was carried out with
Modeller v. 9.11.24 Input conformations of the investigated com-
pounds were prepared using the LigPrep protocol from the Schro-
dinger Suite.25 To sample different protonation states of ligands in
3. Results and discussion
3.1. X-ray investigation
For X-ray evaluation we chose 2-methoxyphenyl-3-methyl-5-
methoxy-indole derivatives 5 and 6 because they differ consider-
ably in their GluK1/GluK2 kainate receptor affinity. The complete
crystal structure analysis of these compounds was expected to
yield information concerning the conformation and especially
mutual orientation of pharmacophoric aromatic substituents. The
X-ray investigation revealed that the compound 5 crystallizes in

790
A. A. Kaczor et al. / Bioorg. Med. Chem. 22 (2014) 787–795
Figure 3. A view of the molecules 5 (A) and 6 (B) with the atomic labelling. Non-H atoms are represented by displacement ellipsoids of 50% probability.
Table 3
Table 4
Selected torsion angles [°] for 5 and 6 in their crystals
Intermolecular interaction geometry (Å, °) for 6
Torsion angles
5
6A
6B
D–Hꢂ ꢂ ꢂA
D-H
Hꢂ ꢂ ꢂA
Dꢂ ꢂ ꢂA
D-Hꢂ ꢂ ꢂA
141.6(17)
145.8(19)
154(2)
137.7(18)
N1–C2–C21–C22
C4–C5–O51–C52
C23–C24–O27–C28
C2–N1–C10–C11
N1–C10–C11–C12
ꢁ75.85(18)
ꢁ4.6(2)
90.5(3)
ꢁ121.6(2)
ꢁ13.7(3)
ꢁ166.8(2)
ꢁ111.2(2)
22.1(3)
C22–H221ꢂ ꢂ ꢂCg1
C23–H231ꢂ ꢂ ꢂCg2(i)
C25–H251ꢂ ꢂ ꢂCg(ii)
C26–H261ꢂ ꢂ ꢂCg1(ii)
0.99(2)
1.00(2)
0.96(2)
0.97(2)
2.836(19)
2.90(3)
2.62(3)
2.79(3)
3.6592(16)
3.7637(16)
3.5078(17)
3.5695(16)
ꢁ179.0(2)
ꢁ176.5(2)
ꢁ90.9(2)
6.3(5)
ꢁ101.8(2)
23.0(3)
Symmetry codes: (i) 1ꢁx, ꢁ1/2 + y, 1/2ꢁz; (ii) 1ꢁx, 1/2 + y, 1/2ꢁz.
Cg1–centroid of the pyrrole ring; Cg2–centroid of the benzene ring C4ꢂ ꢂ ꢂC9.
chiral P212121 space group, while the compound 6 crystallizes in
ꢀ
P1 space group with two molecules A and B in the asymmetric part
methoxy-indole part is common for both analyzed structures
and the geometry of this molecular fragment in 5 and 6 is very
similar. Its conformation is described by three torsion angles
of the unit cell. The view of the molecules with numbering of the
atoms are shown in Figure 3. The 2-methoxyphenyl-3-methyl-5-
Figure 4. Overlay of molecules 5 and 6A (A), 5 and 6B (B) and 6A and 6B (C) by least-squares fitting of the indole rings.

A. A. Kaczor et al. / Bioorg. Med. Chem. 22 (2014) 787–795
791
N1–C2–C21–C22, C4–C5–O51–C52 and C23–C24–O27–C28 (Ta-
ble 3). The first one with ꢁ75.85(18)° for 5, 90.5(3)° for 6A and
ꢁ121.6(2)° for 6B indicates similar spatial orientation of the 2-
methoxyphenyl substituent in relation to the fused bicyclic system
in 5 and 6B, and slightly different in 6A. The other two torsion an-
gles show the same, respectively, cis and trans conformation of the
both methoxy groups in 5 and 6B and opposite trans and cis confor-
mation of these substituents in 6A. The similarities and differences
in the conformation of the investigated molecules 5 and 6 (A and B)
are shown in Figure 4. The chlorobenzyl substituent in 6 adopts
gauche-trans conformation with respect to the indole ring with tor-
sion angles C2–N1–C10–C11 of ꢁ101.8(2), ꢁ111.2(2)° and N1–
C10–C11–C12 of 23.0(3) and 22.1(3)° for A and B molecules,
respectively. This conformation is stabilized by the respective
C12–H16ꢂꢂꢂN1 intramolecular hydrogen bonds in molecule A and
B (Table 4).
The planar indole ring system is slightly folded along the junc-
tion C8–C9; the dihedral angle between the six- and five-mem-
bered rings being 1.16(5)° for 5, 0.83(6)° for 6A and 1.71(6)° for
6B. The classic hydrogen bonds are not present in the crystal struc-
ture of 5. Phenyl and indole rings in the crystal of 5 located perpen-
dicularly to each other belonging to the molecules related by 21
axis parallel to [010] direction interact via C–Hꢂ ꢂ ꢂ
p interactions
(Table 3). Aromatic C22–H221 and C23–H231 from one side and
C25–H251 and C26–H261 groups from the other side of the indole
ring are involved in these type of interaction (Fig. 5). Similar inter-
actions were found in the crystal structure of 6. They are accompa-
nied by C–Hꢂ ꢂ ꢂO intermolecular hydrogen bonds (Table 5) linking
into pairs inversion-related molecules A and B (Fig. 6).
Figure 5. Unit-cell packing in crystal of 5. Dashed lines indicate C–Hꢂꢂꢂ
p intermo-
lecular interactions.
Table 5
3.2. NMR spectra
Hydrogen-bond geometry (Å, °) for 6
D–Hꢂ ꢂ ꢂA
D–H
Hꢂ ꢂ ꢂA
2.58
2.58
2.49
Dꢂ ꢂ ꢂA
2.902(3)
2.899(3)
3.409(3)
D–Hꢂ ꢂ ꢂA
101
101
168
All studied compounds were fully characterized by NMR spec-
troscopy. The assignment of 1H and 13C NMR chemical shifts were
done by the aid of techniques DQF-COSY, HSQC, HMBC and NOESY.
1H NMR spectrum consisted of signals from aromatic indole core
and different substituents. In a case of 3-methylindole, the proton
C12A–H12Aꢂ ꢂ ꢂN1A
C12B–H12Bꢂ ꢂ ꢂN1B
C25B–H25Bꢂ ꢂ ꢂO51Ai
0.93
0.93
0.93
Symmetry codes: (i) 1ꢁx, ꢁy, 1ꢁz.
Figure 6. Unit-cell packing in crystal of 6. Dashed lines indicate C–HꢂꢂꢂO intermolecular hydrogen bonds.

792
A. A. Kaczor et al. / Bioorg. Med. Chem. 22 (2014) 787–795
correlations are presented for compounds 9 and 10 in Figure 7. This
figure shows also the numbering of the structures which is adopted
in Tables 6–9.
3.3. MS spectra
All of the compounds 1–10 do not show many fragments. In
all cases the molecular ion forms the base peak of the spectrum
and double-charged molecular ion is always present (4–10%)
although for 6 it is very weak. [MꢁH]+ ion is also abundant when
R1 = H (2–4) or in the case of compound 10 which also explains
why this ion is the most abundant for 4. Somehow R3 = H (1, 6)
seems to limit the formation of the [MꢁH]+ ion. In the other
compounds this ion is not very abundant since in all of them R1
is other than H and R3 equal to methyl. Compound 8 for which
R3 = CH3 and R4 = H (like for compound 2) gives 12% of [MꢁH]+
ion. All compounds also exhibit the ion [MꢁCH3]+ which is fairly
abundant for compounds 1–3 but very weak for four-membered
compound 4. The [MꢁCH3]+ ion is missing from the spectrum of
compound 6 due to the dominant loss of the R1 = CH2C6H4CL-p
corresponding to m/z 266 (54%). The loss of methyl can occur from
Figure 7. Numbering system and observed NOESY correlations (in arrows) for
compounds 9 and 10.
signals can be easily identified based on coupling constants. When
3-methyl was lacking, DQF-COSY information was needed to assign
indole protons. All substituents had their own characteristic proton
signals (arom-CH3: 2.1–2.5 ppm; OCH3: 3.8–4.0 ppm; CH2CH3:
1.1–1.4 and 4.0–4.2 ppm; aromatic signals 6.9–7.7 ppm). When
the proton signals were assigned, HSQC provided directly corre-
sponding proton bearing carbons. The quaternary carbons were
thereafter assigned based on the long range 1H–13C correlations
from HMBC spectrum. Finally, the assignments were proved by
observed NOESY cross peaks. As an example, decisive NOESY
Table 6
1H NMR data (chemical shifts, multiplicity and coupling constants) for compounds 1–3 and 5–9 in CDCl3 and at 25 °C (TMS = 0.00 ppm)
Compound
H-3/3-Me
H-4
H-5/5-OMe
H-6
H-7
H-ortho*
H-meta*
H-para*
Others
a
1
6.76
br. s
2.43
s
2.44
s
2.20
s
2.26
s
6.45
br. s
2.23
s
2.21
s
7.09
d 2.5
7.58
d 7.8
7.35
d 2.4
7.03
d 2.4
7.05
d 2.4
7.11
d 2.4
7.59
d 7.8
7.04
d 2.4
3.87
s
7.13
dd 7.2, 7.8
3.90
s
3.89
s
3.89
s
3.87
s
7.14
dd 7.2, 7.8
3.89
s
6.86
dd 2.5, 8.8
7.19
dd 7.2, 8.0
6.87
dd 2.4, 8.7
6.89
dd 2.4, 8.8
6.81
dd 2.4, 8.8
6.90
dd 2.4, 8.8
7.23
dd 7.2, 8.0
6.90
7.28
d 8.8
7.35
d 8.0
7.26
d 8.7
7.24
d 8.8
7.02
d 8.8
7.29
d 8.8
7.35
d 8.0
7.25
d 8.8
7.64
d 7.5
7.51
d 8.8
7.57
d 7.5
7.31
d 8.7
7.21
d 8.7
7.49
d 7.4
7.32
d 8.8
7.39
d 7.6
7.43
t 7.5
7.01
d 8.8
7.47
t 7.5
7.02
d 8.7
6.93
d 8.7
7.46
t 7.4
7.02
d 8.8
7.48
t 7.6
7.31
t 7.5
—
b
c
d
e
f
2
3
5
6
7
8
9
7.35
t 7.5
—
—
7.40
t 7.4
—
g
h
7.41
t 7.6
dd 2.4, 8.8
*
Only vicinal couplings are presented for substituted aryl group protons.
a
8.22 (br s, NH).
3.87 (s p-OMe), 7.94 (br s, NH).
7.90 (br s, NH).
1.19 (t 7.2, CH3), 3.88 (s p-OMe), 4.02 (q 7.2, CH2).
3.84 (s p-OMe), 5.14 (s CH2), 6.85 (d 8.5, H-20), 7.18 (d 8.5, H-30).
1.31 (t 7.2, CH3), 4.17 (q 7.2, CH2).
1.21 (t 7.2, CH3), 3.89 (s p-OMe), 4.06 (q 7.2, CH2).
b
c
d
e
f
g
h
1.18 (t 7.1, CH3), 4.03 (q 7.1, CH2).
Table 7
13C NMR chemical shifts for compounds 1–3 and 5–9 in CDCl3 and at 25 °C (TMS = 0.00 ppm)
Compound
C-2
C-3
99.85
C-3a
C-4
C-5
C-6
C-7
C-7a
C-3-Me
C-5-OMe
C-ipso
C-ortho
C-meta
C-para
Others
1
2
3
5
6
7
8
9
138.61
133.99
134.97
137.76
138.24
141.67
136.96
137.92
129.73
130.06
130.42
128.81
129.20
128.60
128.65
128.35
102.25
118.74
100.83
100.80
100.85
102.27
118.76
100.86
154.51
119.44
154.14
153.87
154.15
154.27
118.92
153.91
112.64
121.96
112.44
111.42
111.75
111.79
121.35
111.67
111.63
110.50
111.44
110.16
110.66
110.62
109.37
110.27
132.00
135.64
131.00
130.89
131.78
132.44
135.79
131.33
—
55.84
—
132.44
125.91
133.40
124.76
124.11
133.22
124.67
132.53
125.06
129.00
127.66
131.64
131.59
129.27
131.71
130.47
129.02
114.28
128.80
113.82
113.88
128.49
113.82
128.35
127.65
158.98
127.28
159.24
159.35
127.88
159.25
127.76
—
a
107.75
108.50
108.11
108.70
101.73
108.47
108.41
9.59
9.76
9.32
9.49
—
55.97
56.08
55.96
55.94
—
—
b
c
d
e
f
9.24
9.32
56.05
a
b
c
d
e
f
55.37 (p-OMe).
15.45 (CH3), 38.65 (CH2), 55.31 (p-OMe).
46.99 (CH2), 55.29 (p-OMe), 127.39 (C-20), 128.72 (C-30), 132.71 (C-40), 137.19 (C-10).
15.47 (CH3), 38.88 (CH2).
55.31 (p-OMe).
15.42 (CH3), 38.74 (CH2).

A. A. Kaczor et al. / Bioorg. Med. Chem. 22 (2014) 787–795
793
Table 8
1H NMR data (chemical shifts, multiplicity and coupling constants) for compounds 4 and 10 in CDCl3 and at 25 °C (TMS = 0.00 ppm)
Compound
H-1
H-2
H-3 / 3-OMe
H-4
H-5
H-6
H-7
H-8-OMe
H-9
H-10
Others
a
4
7.28
dd 1.4, 7.6
7.52
7.23
t 7.6
7.30
dd 7.4, 7.8
7.15
dt 1.4, 7.6
7.17
7.26
d 7.6
7.32
d 7.4
3.05
br t 7.2
2.96
m
2.94
m
2.87
m
6.99
d 2.4
7.01
d 2.4
3.87
s
3.88
s
6.84
dd 2.4, 8.8
6.88
7.24
d 8.8
7.26
d 8.8
b
10
dd 1.2, 7.8
dt 1.4, 7.4
dd 2.4, 8.8
a
8.07 (br s NH).
1.51 (t 7.1, CH3), 4.41 (q 7.1, CH2).
b
Table 9
13C NMR chemical shifts for compounds 4 and 10 in CDCl3 and at 25 °C (TMS = 0.00 ppm)
Compound C-1 C-2 C-3 C-4 C-4a C-5 C-6 C-6a C-6b
C-7
C-8
C-9
C-10
C-10a
C-11a
C-11b
Others
a
4
10
119.75 126.66 126.62 128.47 136.47 29.55 19.72 112.48 128.94 100.59 154.30 112.40 111.82 132.11 133.87 127.77
121.97 126.73 126.18 128.71 138.18 30.78 20.17 113.66 126.37 100.43 154.12 112.25 110.28 133.70 134.74 129.66
b
a
55.89 (8-OMe).
15.74 (CH3), 39.83 (CH2), 55.93 (8-OMe).
b
Table 10
Electron impact mass spectra of 4 2,3,5-trisubstituted 1–4 and 6 1,2,3,5-tetrasubstituted indoles 5–10
Compound
M+Å
M2+
[MꢁH]+
[MꢁCH3]+
[MꢁC2H3O]+
[MꢁR1ꢁC2H3O]+
[MꢁR3ꢁCH3O]+Å
[MꢁHꢁR1ꢁC2H3O]+
[MꢁR2]+
[MꢁHꢁR1]+
1
2
3
4
5
6
7
8
9
223(100)
237(100)
237(100)
249(100)
295(100)
391(100)
251(100)
265(100)
265(100)
277(100)
111.6(8)
118.5(7)
118.6(8)
124.3(10)
147.5(9)
Very weak
125.7(8)
132.6(4)
132.5(6)
138.7(8)
222(4)
236(34)
236(22)
248(61)
294(6)
390(2)
250(2)
164(12)
264(7)
276(28)
208(24)
222(25)
222(27)
234(4)
280(22)
—
236(42)
250(28)
250(34)
262(18)
180(47)
194(5)
—
206(12)
252(8)
—
208(14)
—
222(9)
234(6)
179(5)
193(9)
193(9)
205(13)
—
223(9)
179(6)
193(4)
193(6)
205(8)
191(6)
191(7)
191(7)
217(9)a
249(5)
—
—
—
130(5)
160(5)
—
—
—
—
—
—
—
—
192(7)
—
204(14)
—
247(6)d
—
—
—
—
193(6)b
219(8)b
—
204(14)
192(5)
192(5)
204(11)
235(8)
—
247(5)
—
248(6)
10
217(5)c
Other fragments for 1: [MꢁHꢁR1ꢁC4H5O]+ 152(8), Ph+(6); 2: [MꢁHꢁC4H5O]+ 167(7); 4: [MꢁHꢁCH3]+Å 233(6), [MꢁC2H3OꢁC8H8]+Å 152(8); 6: 267(11), [MꢁR1]+ 266(54),
[MꢁR1ꢁR3]+Å 251(6), [MꢁHꢁR1 ꢁR4ꢁC2H3O]+Å 191(6), [R1]+ 125(17); 7: [MꢁCH3ꢁC2H3O]+Å 193(6), [MꢁH2ꢁC4H5O]+Å 180(5); 8: [MꢁHꢁCH3ꢁR1]+ 220(6), [MꢁHꢁR1ꢁCH3O]+Å
204(6), [MꢁH2ꢁR1ꢁC2H3O]+ 191(7). [MꢁH2ꢁC4H5O]+Å 180(5); 10: [MꢁH2]+Å 275(6), [MꢁR1]+ 248(6), [MꢁR1ꢁCH3O]+Å217(5).
a
[MꢁR1ꢁCH3O]+Å
.
.
b
c
[ꢁCH3ꢁCH3O]+Å
[MꢁC2H5ꢁCH3O]+Å
.
d
[MꢁH2]+Å
.
the methoxy groups (1, 7, 8, 9 and 10) and to some extent also from
the loss of R3 (2, 3, 5, 8 and 9) or from R1 = C2H5 (5, 7, 8, 9 and 10).
Compounds 1, 2, 4, 5, 7, 9 and 10 exhibit the ion [MꢁC2H3O]+
which is strongest for 2 (47%) and all compounds except 5 give also
relatively weak ion [MꢁR1ꢁC2H3O]+Å. Compounds 1–5 and 8 exhi-
bit also some amounts of the ion [MꢁR3ꢁCH3O]+Å. It is interesting
that only 2 and 3 show small amount of the ion [MꢁR2]+ (Table 10).
Compounds 2, 4, 8–10 exhibit also the ion [MꢁHꢁR1ꢁC2H3O]+. It is
interesting that both four-membered derivatives 4 and 10 exhibit
the ion [MꢁH2]+Å which proves that the origin of this ion must be
other than [MꢁHꢁR1]+Å as it is for 8 and 10. A few more ions can
be observed for most of the compounds as shown below Table 10.
separate templates for each domain.10,11 However, they are also
characterized by dimer of dimers topology. They consist of
N-terminal domain, ligand-binding domain, transduction domain
and transmembrane domain. The final model of GluK2 receptor,
a very similar to GluK1 receptor model, is presented in Figure 8.
3.3.1. Homology modeling
Homology models of GluK1 and GluK2 receptors were built on
novel AMPA GluA2 receptor template with application of addi-
tional kainate receptor templates for ligand binding domain and
N-terminal domain (see Experimental section for details).
Sequences identity between the GluA2 template and GluK1 and
GluK2 receptors was 39.73% and 40.58% respectively. To our
knowledge, these are first kainate receptor models built on novel
AMPA receptor template.
Homology models of GluK1 and GluK2 receptors have similar
architecture to the used GluA2 receptor template and differ
significantly from previously constructed models based on
Figure 8. Homology model of GluK2 receptor, A-front view; B-side view.

794
A. A. Kaczor et al. / Bioorg. Med. Chem. 22 (2014) 787–795
Figure 9. Compound 5 in the binding pocket of GluK1 (A, B) and GluK2 (C, D) receptor.
Figure 10. Compound 7 in the binding pocket of GluK1 (A, B) and GluK2 (C, D) receptor.
3.3.2. Ligand-receptor interactions
on the basis of sequence differences between GluK1 and GluK2
receptors in the transduction domain. There are no differences in
S1-M1 linker and in S2-M4 linker Asp823 and Asn824 in GluK1
corresponds to Glu808 and Ser809 in GluK2. Indeed, these residues
were found in the binding pocket (Figs. 9 and 10). In case of GluK1
receptor it was determined that the pharmacophoric9 methoxy
group at position 5 interacts via the hydrogen bond with the side
The binding site for non-competitive GluK1/GluK2 receptors
was identified in the receptor transduction domain, that is, in the
domain which connects ligand-binding domain and transmem-
brane domain. This assumption was made on the basis of studies
by Balannik et al.38 for AMPA receptors as well as on our earlier
molecular modelling studies.10 The exact binding site was found

A. A. Kaczor et al. / Bioorg. Med. Chem. 22 (2014) 787–795
3. Stawski, P.; Janovjak, H.; Trauner, D. Bioorg. Med. Chem. 2010, 18, 7759.
795
chain of Thr761. This was true for all the compounds having meth-
oxy group at this position and it is in contrast with earlier studies.
In case of GluK2 receptor an analogous interaction was not identi-
fied but some residues turned out to have potential for such con-
tacts (like the main chain nitrogen atom of Ala812). Instead, the
derivatives having methoxy group at position 2 formed hydrogen
bond with the main of Glu756. Further derivatives which complete
the binding pockets for both receptors can be found in Figures 9
and 10 drawn for derivatives 5 and 7. It was also determined that
inactive derivatives 6 (with chlorobenzyl derivative in position 1)
and 10 (carbazol derivatives) are too big to adopt well to the bind-
ing pocket. In contrast to our previous molecular docking studies,11
it was found that the binding pocket for non-competitive antago-
nists is situated within one receptor subunit, not on the border
of two subunits. The stability of ligand-receptor complexes was
confirmed in short molecular dynamics simulations.
4. Jane, D. E.; Lodge, D.; Collingridge, G. L. Neuropharmacology 2009, 56, 90.
5. Matute, C. CNS Neurosci. Ther. 2011, 17, 661.
6. Szénási, G.; Vegh, M.; Szabo, G.; Kertesz, S.; Kapus, G.; Albert, M.; Greff, Z.; Ling,
I.; Barkoczy, J.; Simig, G.; Spedding, M.; Harsing, L. G., Jr. Neurochem. Int. 2008,
52, 166.
7. Valgeirsson, J.; Nielsen, E. Ø.; Peters, D.; Varming, T.; Mathiesen, C.; Kristensen,
A. S.; Madsen, U. J. Med. Chem. 2003, 46, 5834.
8. Valgeirsson, J.; Nielsen, E. O.; Peters, D.; Mathiesen, C.; Kristensen, A. S.;
Madsen, U. J. Med. Chem. 2004, 47, 6948.
9. Kaczor, A. A.; Kronbach, C.; Unverferth, K.; Pihlaja, K.; Wiinamaki, K.;
Sinkkonen, J.; Kijkowska-Murak, U.; Wróbel, T.; Stachal, T.; Matosiuk, D. Lett.
Drug Des. Discov. 2012, 9, 891.
10. Kaczor, A. A.; Kijkowska-Murak, U. A.; Kronbach, C.; Unverferth, K.; Matosiuk,
D. J. Chem. Inf. Model. 2009, 49, 1094.
11. Kaczor, A. A.; Kijkowska-Murak, U. A.; Matosiuk, D. J. Med. Chem. 2008, 51,
3765.
12. Sobolevsky, A. I.; Rosconi, M. P.; Gouaux, E. Nature 2009, 462, 745.
13. Sheldrick, G.M. SADABS, Ver. 2.06. University of Göttingen, Germany 2002.
14. Altomare, A.; Cascarano, G.; Giacovazzo, C.; Guagliardi, A. J. Appl. Cryst. 1993,
26, 343.
15. Sheldrick, G. M. Acta Cryst. 2007, 64, 112.
16. Flack, H. D. Acta Cryst. 1983, 39, 876.
17. Farrugia, L. J. J. Appl. Cryst. 1997, 30, 565.
4. Conclusions
18. Nardelli, M. Comp. Chem. 1983, 7, 95.
19. Spek, A. L. J. Appl. Cryst. 2003, 36, 7.
20. Edgar, R. C. Nucleic Acids Res. 2004, 32, 1792.
21. MOE Molecular Operating Environment 2009.10, Chemical Computing Group.
22. Kumar, J.; Schuck, P.; Mayer, M. L. Neuron 2011, 71, 319.
In this Letter we performed extensive structural studies of novel
indole-derived non-competitive antagonist of GluK1 and GluK2
receptors in view of their pharmacological activity. We also built
the first homology model of kainate receptors based on the novel
AMPA receptor template. Furthermore, we identified the site of
allosteric modification of kainate receptors in the transduction
domain. The obtained results are of crucial relevance for elabora-
tion of more potent non-competitive antagonists of GluK1 and
GluK2 receptors as the constructed models may be used in future
structure-based drug design studies. The presented compounds
may find application as pharmacological tools to study kainate
receptors and as lead compounds for further modifications.
23. Venskutonyte, R.; Frydenvang, K.; Valadés, E. A.; Szyman´ ska, E.; Johansen, T. N.;
˙
Kastrup, J. S.; Pickering, D. S. ChemMedChem 2012, 7, 1793.
24. Eswar, N.; Marti-Renom, M. A.; Webb, B.; Madhusudhan, M. S.; Eramian, D.;
Shen, M.; Pieper, U.; Sali, A. Current Protocols in Bioinformatics Suppl 15; John
Wiley & Sons Inc, 2006; pp 5.6.1–5.6.30.
25. LigPrep, Version 2.4: Schrödinger, LLC, New York, 2010.
26. Epik, Version 2.1: Schrödinger, LLC, New York, 2010.
27. Becke, A. J. Chem. Phys. 1993, 98, 5648.
28. Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B Condens. Matter 1988, 37, 785.
29. GAUSSIAN 09, Revision A.1 Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G.
E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.;
Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A.
F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.;
Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai,
H.; Vreven, T.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.;
Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand,
J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.;
Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.;
Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.;
Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.;
Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas,
Ö.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J; Gaussian Inc:
Wallingford CT, 2009.
30. Bowers, K. J.; Chow, E.; Xu, H.; Dror, R. O.; Eastwood, M. P.; Gregersen, B. A.;
Klepeis, J. I.; Kolossváry, I.; Moraes, M. A.; Sacerdoti, F. A.; Salmon, J. K.; Shan,
Y.; Shaw, D. E. Scalable Algorithms for Molecular Dynamics Simulations on
Commodity Clusters, Proceedings of the ACM/IEEE Conference on
Supercomputing (SC06), Tampa, Florida, November 11–17, 2006.
31. Pettersen, E. F.; Goddard, T. D.; Huang, C. C.; Couch, G. S.; Greenblatt, D. M.;
Meng, E. C.; Ferrin, T. E. J. Comput. Chem. 2004, 25, 1605.
Acknowledgments
This work was funded by the Polish Ministry of Science and
Higher Education, grant number 405 021 31/1121. The paper was
developed using the equipment purchased within the project
‘The equipment of innovative laboratories doing research on new
medicines used in the therapy of civilization and neoplastic
diseases’ within the Operational Program Development of Eastern
Poland 2007–2013, Priority Axis I Modern Economy, operations
I.3 Innovation promotion. The research was partially prepared
during the postdoctoral fellowship of Agnieszka A. Kaczor at
University of Eastern Finland, Kuopio, Finland under Marie Curie
fellowship. Calculations were partially performed under a compu-
tational grant by Interdisciplinary Center for Mathematical and
Computational Modeling (ICM), Warsaw, Poland, grant number
G30-18. Desmond calculations were performed using resources
of CSC, Finland.
32. Macrae, C. F.; Bruno, I. J.; Chisholm, J. A.; Edgington, P. R.; McCabe, P.; Pidcock,
E.; Rodriguez-Monge, L.; Taylor, R.; Van De Streek, J.; Wood, P. A. J. Appl. Cryst.
2008, 41, 466.
33. Pedretti, A.; Villa, L.; Vistoli, G. J. Comput. Aided Mol. Des. 2004, 18, 167.
34. Krieger, E.; Vriend, G. Bioinformatics 2002, 18, 315.
35. The PyMOL Molecular Graphics System, Version 0.99, Schrödinger LLC.
36. Discovery Studio 3.1, Accelrys.
References and notes
37. http://www.arguslab.com/arguslab.com/ArgusLab.html.
38. Balannik, V.; Menniti, F. S.; Paternain, A. V.; Lerma, J.; Stern-Bach, Y. Neuron
2005, 48, 279.
1. Kaczor, A. A.; Matosiuk, D. Curr. Med. Chem. 2010, 17, 2608.
2. Kew, J. N. C.; Kemp, J. A. Psychopharmacology (Berl.) 2005, 179, 4.