Oxime Ethers of (E)-11-Isonitrosostrychnine as Highly Potent Glycine Receptor Antagonists

Article abstract of DOI:10.1021/acs.jnatprod.6b00479

A series of (E)-11-isonitrosostrychnine oxime ethers, 2-aminostrychnine, (strychnine-2-yl)propionamide, 18-oxostrychnine, and N-propylstrychnine bromide were synthesized and evaluated pharmacologically at human α1 and α1β glycine receptors in a functional

Full text of DOI:10.1021/acs.jnatprod.6b00479

Article
pubs.acs.org/jnp
Oxime Ethers of (E)‑11-Isonitrosostrychnine as Highly Potent Glycine
Receptor Antagonists
Amal M. Y. Mohsen,† Yasmine M. Mandour,† Edita Sarukhanyan,‡ Ulrike Breitinger,† Carmen Villmann,⊥
Maha M. Banoub,† Hans-Georg Breitinger,† Thomas Dandekar,‡ Ulrike Holzgrabe,§
Christoph Sotriffer,§ Anders A. Jensen,∥ and Darius P. Zlotos*,†
†Faculty of Pharmacy and Biotechnology, The German University in Cairo, New Cairo City, 11835 Cairo, Egypt
‡Biocenter, Department of Bioinformatics and §Institute of Pharmacy and Food Chemistry, University of Wurzburg, 97074 Wurzburg,
̈
̈
Germany
⊥Institute of Clinical Neurobiology, University of Wurzburg, 97078 Wurzburg, Germany
̈
̈
∥Department of Drug Design and Pharmacology, Faculty of Health and Medical Sciences, University of Copenhagen, DK-2100
Copenhagen, Denmark
S
* Supporting Information
ABSTRACT: A series of (E)-11-isonitrosostrychnine oxime
ethers, 2-aminostrychnine, (strychnine-2-yl)propionamide, 18-
oxostrychnine, and N-propylstrychnine bromide were synthe-
sized and evaluated pharmacologically at human α1 and α1β
glycine receptors in a functional fluorescence-based and a
whole-cell patch-clamp assay and in [3H]strychnine binding
studies. 2-Aminostrychnine and the methyl, allyl, and
propargyl oxime ethers were the most potent α1 and α1β
antagonists in the series, displaying IC50 values similar to those
of strychnine at the two receptors. Docking experiments to the
strychnine binding site of the crystal structure of the α3 glycine
receptor indicated the same orientation of the strychnine core for all analogues. For the most potent oxime ethers, the ether
substituent was accommodated in a lipophilic receptor binding pocket. The findings identify the oxime hydroxy group as a
suitable attachment point for linking two strychnine pharmacophores by a polymethylene spacer and are, therefore, important for
the design of bivalent ligands targeting glycine receptors.
For a long time, understanding of the molecular mechanisms
of GlyR function has been hampered by a lack of high-
resolution structures of the receptors, and only recently have
homology models based on structures of other Cys-loop
receptors become available.12 Recently, however, electro-cryo-
microscopy structures of the zebra fish α1 GlyR in complex
with strychnine and glycine13 and a crystal structure of human
α3 GlyR in complex with strychnine14 have been published,
providing a detailed insight into the molecular recognition of
agonists and antagonists and mechanisms of GlyR activation
and inactivation.
The natural product strychnine has been optimized for
strong interactions with biological macromolecules through
evolutionary selection. Indeed, it displays nanomolar binding
affinity and inhibitory potency at recombinant and native GlyRs
in binding studies and functional assays.2,3,15 In previous
structure−activity relationship (SAR) studies any structural
change to strychnine, including N-quaternization, C-2-nitration,
lactam reduction, hydrogenation, oxidation, or relocation of the
trychnine, the major alkaloid from the plant Strychnos nux
Svomica, exhibits pharmacological activity at several neuro-
transmitter receptors, including a number of ligand-gated ion
channels, such as muscle-type and neuronal nicotinic acetylcho-
line receptors and glycine receptors (GlyRs). The major
pharmacological action of strychnine is a strong antagonistic
activity at the latter receptors, which are often referred to as
strychnine-sensitive GlyRs.1−3 The GlyRs are anion-selective
channels that mediate inhibitory synaptic transmission in the
spinal cord and brainstem, and they are believed to play crucial
roles in motor coordination and pain signal transmission.4,5
Disruption of the normal GlyR function caused by heritable
mutations is associated with neuromotor disorders such as
hyperekplexia or some forms of spasticity.5−8 The GlyRs are
pentameric subunit complexes belonging to the superfamily of
Cys-loop receptors and exist either as homomers composed of
α subunits or heteromers that contain both α and β subunits.9
While four α-subunit isoforms are known (α1−α4), only one β
subunit has been identified. The precise distribution of different
subunits is specific to different parts of the central nervous
system.10,11
Received: May 24, 2016
© XXXX American Chemical Society and
American Society of Pharmacognosy
A
DOI: 10.1021/acs.jnatprod.6b00479
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
C-21−C-22 double bond, and opening of the tetrahydroox-
epine ring, has been reported to decrease the antagonist activity
at GlyRs.16−18 The only structure modification not to impair
the activity of the parent compound is the introduction of an E-
configured hydroxyimino group at C-11 to give (E)-11-
isonitrosostrychnine (1). Similar to strychnine, none of these
analogues exhibited selectivity between homomeric α1 and
heteromeric α1β receptors.
Chart 1
Recently, our groups have become interested in a bivalent
ligand approach for targeting GlyRs. Bivalent ligands are
compounds that comprise two chemical groups (pharmaco-
phores) linked to each other by a spacer of a specific length and
composition and are potentially capable of simultaneous
binding to two distinct binding sites. On the basis of its high
potency and a good understanding of SAR, strychnine
represents an excellent pharmacophore for bivalent ligands
targeting GlyRs. Since the C-11 (E)-hydroxyimino functional
group of compound 1 secured its high antagonistic potency, an
oxime ether linkage appears to be suitable for the attachment of
the spacer. In order to examine the impact of an alkyl
substituent attached to the hydroxy group of 1 on the
antagonist potency, an extensive series of strychnine oxime
ethers have been synthesized and evaluated pharmacologically
in the current work. Additionally, to explore the C-2 amide
group of strychnine as another potential linker for strychnine-
based bivalent ligands, (strychnine-2-yl)propionamide (3) was
synthesized and pharmacologically evaluated. Finally, to extend
the SAR of strychnine at GlyRs, 2-aminostrychnine (2), N-
propylstrychnine bromide (4), and 18-oxostrychnine (5) were
included in the study. The SAR findings are discussed based on
docking experiments using the high-resolution crystal structure
of α3 GlyR with strychnine occupying its orthosteric binding
site.14
Scheme 1
RESULTS AND DISCUSSION
■
The structures of the strychnine analogues studied in this work
are shown in Chart 1 and Scheme 1. (E)-11-Isonitrosos-
trychnine (1) was obtained by nitrosation of strychnine using
tert-butyl nitrite/tert-BuOK as previously reported.19 2-Amino-
strychnine (2) was synthesized in a two-step approach by
nitration of strychnine and subsequent reduction of the nitro
group using SnCl2.20 2-Propionamidostrychnine (3) was
obtained by acylation of 2-aminostrychnine using propionic
acid anhydride. N-Propylstrychnine bromide (4) was prepared
by quaternization of strychnine using n-propyl bromide as
previously reported.21 18-Oxostrychnine (5) was synthesized
from strychnine N-oxide using K2Cr2O7 according to a
modified procedure by Rosenmund and co-workers.22 The
oxime ethers 6a−k were synthesized by O-alkylation of 1 using
NaH as a base and an appropriate alkylating agent.
The functional properties of compounds 2−5 and 6a−k and
the reference ligands strychnine and (E)-11-isonitrosostrych-
nine (1) were characterized at the human α1 and α1β receptor
subtypes in the fluorescence-based FLIPR membrane potential
blue (FMP) assay. Here, the receptors were transiently
expressed in tsA201 cells, and the abilities of the compounds
to inhibit the response induced by glycine EC50 (EC40−EC60)
in the cells were determined.23 The expression of homogeneous
populations of α1 and α1β receptors in the cells was verified on
a routine basis using picrotoxin, which as previously reported
displays ∼100-fold higher antagonist potency at α1 than at α1β
in this assay.24 The IC50 values determined for the compounds
are given in Table 1.
Previous SAR studies have identified the lactam carbonyl
group, the tertiary amino group, and the C-21−C-22 double
bond in strychnine as essential structural elements required for
its strong antagonistic activity at homo- and heteromeric
GlyRs.16−18 These findings are in agreement with the recently
published experimental structures of strychnine bound to the
homomeric α1 and α3 GlyRs.13,14 The residues critical for
strychnine binding are identical in the two receptors, resulting
in the same antagonist binding mode. The orthosteric binding
B
DOI: 10.1021/acs.jnatprod.6b00479
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Table 1. Pharmacological Characterization of the Compounds at α1 and α1β GlyRs
a
b
IC50 [μM] FMP assay
IC50 [μM] patch-clamp assay
α1
α1β
α1
α1β
strychnine
0.093 [7.03 0.09]
0.097 [7.01 0.03]
0.077 [7.11 0.08]
0.54 [6.27 0.05]
10 [4.98 0.07]
0.18 [6.74 0.06]
0.21 [6.67 0.10]
0.27 [6.56 0.05]
0.94 [6.03 0.08]
13 [4.87 0.05]
9.8 [5.00 0.16]
0.21 [6.67 0.05]
0.35 [6.45 0.08]
0.88 [6.06 0.03]
1.10 [5.95 0.06]
0.31 [6.51 0.06]
0.71 [6.15 0.08]
0.78 [6.11 0.08]
0.15 [6.80 0.04]
0.24 [6.62 0.13]
0.67 [6.18 0.06]
1.50 [5.83 0.08]
0.051 0.005
0.082 0.010
0.076 0.009
0.201 0.019
4.59 0.41
0.058 0.003
0.080 0.008
0.097 0.011
0.311 0.039
6.53 0.59
1
2
3
4
5
6.9 [5.16 0.04]
0.10 [6.98 0.09]
0.11 [6.94 0.09]
0.37 [6.42 0.05]
0.26 [6.58 0.07]
0.17 [6.78 0.13]
0.25 [6.61 0.10]
0.69 [6.15 0.07]
0.095 [7.02 0.03]
0.052 [7.29 0.10]
0.22 [6.65 0.09]
0.72 [6.15 0.04]
6a
6b
6c
6d
6e
6f
6g
6h
6i
6j
6k
0.075 0.008
0.109 0.009
0.056 0.011
0.116 0.018
0.049 0.004
0.059 0.008
0.046 0.0049
0.043 0.007
a
Functional properties of the compounds at the human α1 and α1β GlyRs transiently expressed in tsA201 cells in the FLIPR membrane potential
blue (FMP) assay. EC50 (EC40−EC60) concentrations of glycine were used as agonist at the two receptors (EC50 = 42 μM at α1, EC50 = 23 μM at
α1β). The IC50 values for the analogues represent the antagonistic potency and are given in μM with pIC50 SEM in brackets. Data are given as the
b
means based on n = 3, 4. Direct inhibition of glycine-mediated currents was measured using whole-cell patch clamp recordings from HEK293 cells
transiently transfected with GlyR α1 or α1/β cDNA. After determination of EC50, strychnine derivatives in varying concentrations were perfused
over GlyR-expressing cells and inhibition curves constructed from whole-cell current amplitudes measured in the presence and absence of inhibitor
at EC50 concentrations of glycine. IC50 values were determined from inhibition curves constructed from a nonlinear fit to dose−response data using
4−7 cells per inhibitor. Data are given as means SEM.
site of the homomeric α3 receptor with docked strychnine is
shown in Figure 2A. The most important binding interactions
displayed either comparable antagonist potencies at the two
receptors or slightly higher potency at the homomeric receptor
with 2-aminostrychnine (2) and the n-propyl, isobutyl, and n-
pentyl oxime ethers 6b, 6d, and 6f, respectively, showing the
highest preference toward the α1 subtype (3−4-fold).
While the lactam group and the C-21−C-22 double bond are
present in all strychnine analogues examined in this study, two
analogues lack the tertiary amino group, namely, the quaternary
n-propyl derivative 4 and 18-oxostrychnine (5), in which N-19
is a part of a lactam function. Consequently, for both
compounds, N-19 cannot exist in a protonated form required
for its action as a hydrogen bond donor, which is reflected in
the antagonist potencies of 4 and 5 (IC50 ≈ 10 μM) being
approximately 100-fold lower than those of strychnine (IC50
0.1 μM) at both α1 and α1β.
≈
2-Aminostrychnine (2) has been previously identified as a
high-affinity ligand for the strychnine binding sites in synaptic
membranes from the brain and spinal cord of rats.15
Radioligand displacement experiments using [3H]strychnine
revealed for 2 only slightly reduced affinity (Ki = 18 nM) when
compared to strychnine (Ki = 12 nM).15 In the FMP assay,
compound 2 exhibited strong inhibition of α1 (IC50 = 77 nM)
and α1β (IC50 = 270 nM) receptors, thus being equipotent to
strychnine at both subtypes. As discussed later, docking
experiments revealed that the orientation of 2-aminostrychnine
in the antagonist binding pocket is highly similar to that of
strychnine, with the primary amino group making an additional
hydrogen bond with the backbone carbonyl group of Gly160
(Figure 2B). For the propionamide 3 (N-acylated 2-amino-
strychnine), the N-acyl substituent is too bulky to be well
accommodated in this region of the binding pocket, resulting in
a significantly lower antagonistic potency at α1 (IC50 = 540
nM) and α1β (IC50 = 940 nM) receptors.
Figure 1. Displacement curves for [3H]strychnine binding experi-
ments of compounds 1, 3, 6a, 6b, and 6h at α1 receptors
recombinantly expressed in HEK293 cells.
are a hydrogen bond between the protonated tertiary amino
group of strychnine and the backbone carbonyl group of
Phe159 and a polar interaction between the lactam oxygen of
strychnine and the guanidinium group of Arg65. Moreover, the
protonated amino group of strychnine is also involved in an
electrostatic interaction with the carboxylate group of the
Glu157 side chain.
The residues essential for strychnine binding are conserved
among the α and β subunits of GlyRs, with the exception that α
Phe207 is replaced by a Tyr in the β subunit.14 Consequently, it
is hardly surprising that neither strychnine nor any of the
analogues in this study displayed substantial selectivity toward
α1 or α1β receptors. In the FMP assay, all compounds
C
DOI: 10.1021/acs.jnatprod.6b00479
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
α1β: IC50 = 350 nM), further increasing the size of the alkyl
substituent (6c−g, R = n-butyl, iso-butyl, benzyl, n-pentyl,
phenethyl) gradually reduced the GlyR activity of the
analogues. For example, 6g, bearing a bulky phenethyl
substituent, exhibited ∼5-fold higher IC50 values at the GlyRs
(690 nM at α1, 780 nM at α1β) than those of the methyl
analogue 6a. A terminal methyl ester group did not significantly
change the antagonist potencies in the oxime ether series, as
indicated by the IC50 values of the methyl acetate 6j (α1: 220
nM; α1β: 720 nM) and the methyl valerate 6k (α1: 720 nM;
α1β: 1500 nM), the latter being less potent because of the
longer spacer between the oxime ether oxygen and the ester
carbonyl group.
Introduction of a double or triple bond into the propyl group
of 6b significantly improved antagonistic potency. The resulting
allyl (6h) and propargyl (6i) analogues are among the most
potent GlyR antagonists published to date, displaying IC50
values of 95 nM (α1) and 150 nM (α1β) and of 52 nM (α1)
and 240 nM (α1β), respectively.
These findings indicate that the strychnine binding site at
GlyRs comprises an additional lipophilic pocket located close to
C-11 of strychnine, and the groups best accommodated in this
pocket are (E)-allyl and (E)-propargyl oxime ethers (Figure
2C; see also later discussion of docking studies).
The analogues displaying the highest antagonist potencies in
the FMP assay (1, 2, 6a, 6b, 6h, 6i) and the less potent
antagonists 3 and 4 were tested in whole-cell patch-clamp
assays at α1 and α1β GlyRs recombinantly expressed in HEK
293 cells. This method records ion-channel-mediated trans-
membrane currents directly and thus allows direct insight into
the inhibition of GlyR function.25 While the FMP assay
generates an equilibrium of all receptor states (free, ligand-
bound, inhibitor-bound, active, desensitized), the patch-clamp
technique observes receptor activation and inhibition in a pre-
equilibrium situation, where receptor desensitization, i.e., the
transition into an inactive refractory state, is excluded.25 The
concentration-dependent reduction of GlyR currents in the
presence of strychnine derivatives was used to construct
concentration-inhibition curves and determine functional IC50
values26,27 (Figure S15, Supporting Information).
The IC50 values (Table 1) obtained in the patch-clamp
recordings generally correlate well with those from the FMP
assay, confirming the SAR conclusions drawn so far and
indicating that the affinity of strychnine analogues for active
and desensitized receptor states is similar. For most analogues,
the IC50 values were slightly lower and none of the ligands
showed preference toward the α1 subtype.
In order to confirm that the antagonist potencies obtained
for the strychnine analogues in the two functional studies
correlate with their binding affinities at the receptors,
compounds 1, 3, 6a, 6b, and 6h were subjected to radioligand
binding experiments at α1 receptors recombinantly expressed
in HEK293 cells using [3H]strychnine as radioligand. The
displacement curves are shown in Figure 1. The binding
affinities for the analogues (Ki values) are given in Table 2. (E)-
11-Isonitrosostrychnine (1), which was found to be equipotent
to strychnine in both functional assays, also exhibitied
comparable binding affinity (Ki = 40 nM) to the parent
compound (Ki = 23 nM). 2-Propionamidostrychnine (3),
which was 3-fold less potent than strychnine in both functional
assays, displayed a 3-fold decreased binding affinity (Ki = 67
nM) compared to strychnine in the binding assay. Surprisingly,
the oxime ethers 6a, 6b, and 6h, which exhibited antagonistic
Figure 2. Graphical representation of the orthosteric binding pocket of
the homomeric GlyR with docked (A) strychnine (magenta), (B) 2-
aminostrychnine 2 (magenta), and (C) propargyl oxime ether 6i
(magenta), showing residues from loop B (coral) and loop C (green)
of the principal subunit together with β5 (cyano), β6 (orange), β2
(yellow), β1 (brown), and β8 (blue) from the complementary subunit.
Only side chain atoms are shown for clarity.
(E)-11-Isonitrosostrychnine (1) has been previously re-
ported to exert strong antagonistic action at GlyRs.17 In the
FMP assay it displayed similar IC50 values (α1: 97 nM; α1β:
210 nM) to strychnine. An equally high inhibition was observed
for its methyl ether 6a (α1: IC50 = 100 nM; α1β: IC50 = 210
nM). The effect of increasing the size of the oxime ether
substituent on the antagonist potency was examined in the
series of ligands 6a−i. While the n-propyl analogue 6b
maintained a high antagonist potency (α1: IC50 = 110 nM;
D
DOI: 10.1021/acs.jnatprod.6b00479
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
the parent compound (Figure 2B). The primary amino group is
involved in a hydrogen bond with the backbone carbonyl of
Gly160 of the principal subunit.
Table 2. Binding Affinities of Compounds 1, 3, 6a, 6b, and
6h for the Human α1 GlyRs Expressed in HEK293 Cells
Obtained in Competition Radioligand Binding Assays Using
[3H]Strychnine
As for the oxime ethers, the strychnine core adopted a similar
orientation in the orthosteric site of the α3 GlyR when
compared to strychnine. As shown for the most potent
propargyl oxime ether, 6i (Figure 2C), the propargyl group
protrudes into a groove of a limited size flanked from one end
by Phe44 and Phe63, while the other end is lined by Arg65 and
Asn42 belonging to the β1 and β2 strands of the
complementary subunit. On the other hand, for the least
potent oxime ether, 6g, bearing a phenethyl group, the bulky
phenyl ring fails to fit into this confined area, shifting the
substituent toward the β8 strand (Figure S19, Supporting
Information).
In summary, a series of (E)-11-isonitrosostrychnine oxime
ethers were synthesized and characterized as antagonists at
homomeric and heteromeric glycine receptors in the FMP
assay, in a whole-cell patch-clamp assay, and in [3H]strychnine
binding studies. The most potent antagonists in the series bear
methyl, allyl, and propargyl substituents. While the methyl and
allyl analogues 6a and 6h were equipotent to strychnine, the
propargyl oxime ether 6i was a slightly more potent antagonist
than the parent compound. Docking studies to the strychnine
binding site from the crystal structure of the α3 GlyR revealed
the presence of a hydrophobic region that is able to
accommodate small oxime ether substituents. The findings
provide a valuable extension of the SAR of strychnine at the α1
and α1β receptors and reveal the oxime ether group as a
possible linker for strychnine-based bivalent ligands targeting
glycine receptors, which currently are under investigation in our
laboratories.
Ki [nM] SEM
strychnine
23
40
6
4
1
3
67 17
189 30
241 67
370 116
6a
6b
6h
potencies similar to that of strychnine in both functional assays,
showed significantly reduced binding affinity compared to the
parent compound (6a Ki = 189 nM; 6b Ki = 241 nM; 6h Ki =
371 nM).
To elucidate the binding mode of the strychnine analogues,
molecular docking simulation studies using GOLD 5.3
(Cambridge Crystallographic Data Centre, Cambridge, UK)
were performed. Two experimentally determined structures of
GlyR complexed with strychnine are available: α1 GlyR (PDB
code: 3JAD, resolution 3.9 Å)13 and α3 GlyR (PDB code:
5CFB, resolution 3.0 Å).14 Although the compounds were
tested on α1 GlyR, the lower resolution α1 GlyR structure
appeared to be unsuitable for docking experiments because of a
missing critical side chain of Glu157 in the orthosteric binding
site in addition to a flipped orientation of the backbone
carbonyl group of Phe159, which made it difficult to reproduce
the experimental binding mode of strychnine. Comparing the
sequence of both proteins showed a high sequence identity of
89% and a sequence similarity of 95% with all orthosteric
binding site residues conserved in both proteins (Figure S16,
Supporting Information). Accordingly, the higher resolution
crystal structure of α3 GlyR (5CFB) was used in this study for
docking the strychnine analogues.
To validate the docking protocol, the cocrystallized ligand
strychnine was docked into the orthosteric site of the glycine
receptor (Figure 2A). All the resultant poses converged to a
binding mode similar to that of the experimentally determined
position of strychnine with the best ranking pose having a root-
mean square deviation (RMSD) value of 0.36 Å (Figure S17,
Supporting Information). An overlay of the docked pose and
the crystal structure is shown in the Supporting Information
(Figure S18). The orthosteric site is formed by Ser158, Phe159,
Gly160 from loop B and Tyr202, Thr204, Phe207 from loop C
of the principal unit together with Phe63, Arg65 from β2,
Leu117, Arg119 from β5, and Leu127, Ser129 from β6 of the
complementary unit (Figure 2A).14 The binding interactions
include a hydrogen bond between the protonated tertiary
amine of strychnine and the backbone carbonyl group of
Phe159, a polar interaction between the lactam oxygen of
strychnine and the guanidinium group of Arg65, and an
electrostatic interaction of the protonated amino group of
strychnine with the carboxylate group of the Glu157 side chain.
Moreover, the benzene rings of Phe159, Phe207, and Phe63 are
involved in hydrophobic interactions with hydrophobic parts of
the strychnine molecule, such as the methylene groups CH2-17
and CH2-18 and the C-21−C-22 double bond.
EXPERIMENTAL SECTION
■
General Experimental Procedures. Melting points were
determined using a capillary melting point apparatus (Gallenkamp,
Sanyo) and are uncorrected. A Bruker AV-400 spectrometer was used
to obtain 1H NMR and 13C NMR spectra, respectively. 1H NMR
chemical shifts are referred to CHCl3 (7.26 ppm) and DMSO-d6 (2.50
ppm). 13C NMR chemical shifts are referred to CDCl3 (77.26 ppm)
and DMSO-d6 (39.52 ppm). The NMR resonances were assigned by
means of COSY and HMQC experiments. ESI mass spectra were
determined on an Agilent 1100 MS system. Elemental analyses were
performed by the microanalytical section of the Institute of Inorganic
Chemistry, University of Wurzburg. All reactions were carried out
̈
under an argon atmosphere. Column chromatography was carried out
on silica gel 60 (0.063−0.200 mm) obtained from Merck (Darmstadt,
Germany). 2-Aminostrychnine (2) was synthesized by nitration of
strychnine and reduction of the resulting 2-nitrostrychnine using
SnCl2.20 N-Propylstrychnine bromide (4) was prepared as previously
described.21
N-(Strychnin-2-yl)propionamide (3). Propionic anhydride (4.4
mL) was dropwise added to the solution of 2-aminostrychnine (290
mg, 0.83 mmol) and Et3N (2.53 mL) in dry CH2Cl2 (20 mL). After
stirring for 2 h at room temperature, the solvent was removed under
reduced pressure and the residue was subjected to column
chromatography (CHCl3−MeOH−25% NH3, 100:10:1) to give 2
1
(256 mg, 76%) as a colorless solid: H NMR (400 MHz, CDCl3) δ
7.99 (1H, d, J = 8.6 Hz, H-4), 7.60 (1H, d, J = 1.9 Hz, H-1), 7.42 (1H,
s, NH), 7.14 (1H, dd, J = 8.6, 1.9 Hz, H-3), 5.87 (1H, t, J = 6.0 Hz, H-
22), 4.26 (1H, dt, J = 8.4, 3.3 Hz, H-12), 4.13 (1H, dd, J = 13.8, 6.9
Hz, H-23a), 4.04 (1H, dd, J = 13.8, 6.0 Hz, H-23b), 3.92 (1H, s, H-
16), 3.84 (1H, d, J = 10.5 Hz, H-8), 3.67 (1H, dd, J = 14.8, 1.0 Hz, H-
20a), 3.19−3.06 (3H, m, H-11a, H-14, H-18a), 2.83 (1H, td, J = 10.1,
7.7 Hz, H-18b), 2.72−2.60 (2H, m, H-11b, H-20b), 2.37 (2H, q, J =
7.5 Hz, −CH2CH3), 2.34−2.28 (1H, m, H-17a), 1.95−1.81 (2H, m,
2-Aminostrychnine (2) was equipotent to strychnine in both
functional assays. Docking studies revealed that 2 fitted very
well into the orthosteric site of the α3 GlyR with the same
orientation and interactions of the strychnine skeleton as for
E
DOI: 10.1021/acs.jnatprod.6b00479
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
H-15a, H-15b), 1.42 (1H, d, J = 14.4 Hz, H-17b), 1.25−1.22 (1H, m,
H-13), 1.23 (3H, t, J = 7.5 Hz, −CH2CH3); 13C NMR (100 MHz,
CDCl3) δ 171.76 (C, CO), 168.92 (C, C-10), 140.42 (C, C-5),
138.36 (C, C-21), 134.51 (C, C-6), 133.59 (C C-2), 126.93 (CH, C-
22), 119.83 (CH, C-3), 116.12 (CH,C-4), 114.40 (CH,C-1), 77.45
(CH, C-12), 64.43 (CH2, C-23), 60.21 (CH, C-16), 59.91 (CH, C-8),
52.51 (CH2, C-20), 51.88 (C, C-7), 50.13 (CH2, C-18), 48.03 (CH, C-
13), 42.56 (CH2, C-15), 42.17 (CH2, C-11), 31.40 (CH, C-14), 30.45
(CH2, −CH2CH3), 26.71 (CH2, C-17), 9.50 (CH3, −CH2CH3);
ESIMS m/z 405.8 [M]+ (100%); anal. C 71.00, H 6.86, N 10.15%,
calcd for C24H27N3O3 C 71.09, H 6.71, N 10.36%.
18-Oxostrychnine (5). A solution of K2Cr2O7 (6 mg, 0.02 mmol)
in 2 mL of H2O was added to a solution of strychnine N-oxide (200
mg, 0.57 mmol) in 10 mL of H2O and 5 mL of dioxane, and the
reaction mixture was heated under reflux for 2 h. After cooling to room
temperature, the solvent was removed under reduced pressure. Water
(4 mL) and warm (45 °C) 1 N HCl (10 mL) were added to the
residue, and the insoluble brown material was filtered off and subjected
to column chromatography (CH2Cl2−MeOH−25% NH3, 100:5:0.5)
to give compound 5 (70 mg, 35.3%) as a colorless solid: mp 263 °C
(lit. 263 °C);22 1H NMR (400 MHz, CDCl3) δ 8.09−8.01 (1H, m, H-
4), 7.82 (1H, dd, J = 7.8, 0.9 Hz, H-1), 7.23−7.15 (1H, m, H-3), 7.00
(1H, dd, J = 5.1, 2.5 Hz, H-2), 5.88 (1H, t, J = 6.2 Hz, H-22), 4.21
(1H, dt, J = 8.4, 3.6 Hz, H-12), 4.09 (1H, dd, J = 13.9, 6.9 Hz, H-23a),
3.98 (1H, dd, J = 13.9, 6.1 Hz, H-23b), 3.85 (1H, dd, H-20a), 3.77
(1H, d, J = 10.6 Hz, H-8), 3.20 (1H, d, J = 11.0 Hz, H-14), 3.18−3.13
(1H, m, H-18a), 3.08−3.00 (1H, m, H-11a), 2.85−2.72 (2H, m, H-
18b, H-20b), 2.57 (1H, dd, J = 17.4, 3.6 Hz, H-11b), 2.301−2.162
(2H, m, H-15a, H-17a), 1.831−1.680 (2H, m, H-15b, H-17b), 1.31
(1H, ddd, J = 13.0, 8.2, 4.8 Hz, H-13); 13C NMR (101 MHz, CDCl3)
δ 170.3 (C, CO), 169.0 (C, C-10), 142.0 (C, C-5), 138.9 (C, C-21),
131.8 (C, C-6), 128.7 (CH, C-2), 127.4 (CH, C-3), 124.3 (CH, C-22),
115.7 (CH, C-4), 92.5 (C, C-16), 77.3 (CH, C-12), 65.2 (CH2, C-23),
60.4 (CH, C-8), 56.6 (C, C-7), 52.8 (CH2, C-20), 48.7 (CH, C-13),
48.4 (CH2, C-18), 42.7 (CH2, C-11), 39.8 (CH2, C-17), 35.0 (CH2, C-
15), 33.4 (CH, C-14); ESIMS m/z = 350.2 [M]+.
General Procedure for the Synthesis of the Oxime Ethers
6a−k. A solution of 1 (500 mg, 1.4 mmol) in dry DMF (4 mL) was
added dropwise to a stirred suspension of NaH (0.03 g, 1.4 mmol, 60%
dispersion in mineral oil) in dry DMF (3 mL) at 0 °C. After stirring at
0 °C for 30 min, the respective alkyl halide (0.32 mmol) was dropwise
added. After stirring at room temperature for 16 h, the mixture was
concentrated under reduced pressure and the residue was subjected to
column chromatography using CH2Cl2−MeOH−25% NH3
(100:5:0.5) as eluent unless otherwise stated.
(E)-11-(Methoxyimino)strychnine (6a). 6a (60 mg, 12%) was
obtained from 1 and CH3I (0.02 mL) as a pale yellow solid: 1H NMR
(400 MHz, DMSO-d6) δ 8.03 (1H, d, J = 7.7 Hz, H-4), 7.39 (1H, d, J
= 7.0 Hz, H-2), 7.34−7.28 (1H, m, H-3), 7.17 (1H, td, J = 7.5, 0.9 Hz,
H-1), 5.73 (1H, t, J = 4.9 Hz, H-22), 4.85 (1H, d, J = 2.4 Hz, H-12),
4.092−4.013 (2H,m, H-23a, H-23b), 3.97 (3H, s, CH3), 3.93 (1H, d, J
= 10.8 Hz, H-8), 3.80 (1H, dd, J = 12.1 Hz, H-16), 3.49 (1H, d, J =
14.2 Hz, H-20a), 2.96−2.96 (1H, m, H-14), 2.93−2.87 (1H, m, H-
18a), 2.63−2.57 (1H, m, H-18b), 2.50 (1H, d, J = 14.2 Hz, H-20b),
2.26−2.14 (1H, m, H-15a), 1.79 (1H, m, H-17a), 1.64−1.57 (1H, m,
H-13), 1.51 (ddd, J = 10.8, 2.6 Hz, H-17b), 1.27 (1H, ddd, J = 14.3
Hz, H-15b); 13C NMR (100 MHz, DMSO-d6) δ 158.2 (C, CO),
149.5 (C, CN), 142.2 (C, C5), 140.4 (C, C21), 134.1 (C, C6),
128.2 (CH, C-3), 126.6 (CH, C-22), 124.6 (CH, C-1), 122.8 (CH, C-
2), 114.8 (CH, C-4), 75.4 (CH, C-12), 65.3 (CH2, C-23), 63.6 (CH3,
O−CH3), 59.5 (CH, C-16), 58.8 (CH, C-8), 52.4 (CH2, C-20), 51.2
(C, C7), 49.4 (CH2, C-18), 44.5 (CH, C-13), 42.8 (CH2, C-17), 31.3
(CH, C-14), 26.4 (CH2, C-15); ESIMS m/z 377.6 [M]+; mp 181 °C;
anal. C 69.42, H 6.30, N 10.64%, calcd for C22H23N3O3 × 1/3MeOH
C 69.11, H 6.32, N 10.83%.
hydrogen atoms coincide with the values for the corresponding atoms
of 6a within 0.04 ppm and 0.1 Hz, respectively; 13C NMR (100
MHz, DMSO-d6) 77.0 (OCH2CH2CH3), 22.1 (OCH2CH2CH3), 10.0
(CH3, OCH2CH2CH3); chemical shifts for all other carbon atoms
coincide with the δ values for the corresponding atoms of 6a within
0.1 ppm; ESIMS m/z 405.4 [M]+; mp 182 °C; anal. C 69.63, H 6.41,
N 9.98%, calcd for C24H27N3O3 × 0.5MeOH C 69.81, H 6.93, N
9.97%.
(E)-11-(Butyloxyimino)strychnine (6c). 6c (40 mg, 7.3%) was
obtained from 1 and n-butyl bromide (0.03 mL) in the presence of KI
(1 mg) as a brown solid: 1H NMR (400 MHz, CDCl3) δ 8.17 (1H, d, J
= 8.0 Hz, H-3), 8.00 (1H, s, H-1), 7.30−7.23 (1H, m, H-4), 7.11 (1H,
t, J = 7.1 Hz, H-2), 5.89 (1H, s, H-22), 4.95 (1H, d, J = 2.4 Hz, H-12),
4.46−4.36 (2H, m, OCH2(CH2)2CH3), 4.28 (1H, dd, J = 13.9, 7.1 Hz,
H-23a), 4.17 (1H, d, J = 10.6 Hz, H-8), 4.11 (1H, dd, J = 13.9, 5.7 Hz,
H-23b), 3.96 (1H, s, H-16), 3.70 (1H, d, J = 14.6 Hz, H-20a), 3.14
(1H, dd, J = 16.8, 8.2 Hz, H-18a), 3.06 (1H, s, H-14), 2.84 (1H, d, J =
5.0 Hz, H-18b), 2.68 (1H, d, J = 14.6 Hz, H-20b), 2.39 (1H, dt, J =
14.4, 4.2 Hz, H-15a), 1.92−1.85 (2H, m, H-17), 1.79−1.70 (2H, m,
OCH2CH2CH2CH3), 1.56−1.38 (4H, m, O(CH2)2CH2CH3, H-13, H-
15b), 0.99−0.90 (3H, m, CH3); 13C NMR (101 MHz, CDCl3) δ 159.2
(C, CO), 149.5 (C, CN), 142.7 (C, C-5), 139.5 (C, C-21), 133.1
(C, C-6), 129.4 (C, C-3), 127.4 (CH, C-22), 125.2 (C, C-1), 122.1 (C,
C-4), 116.6 (C, C-1), 77.3 (CH2, OCH2(CH2)2CH3), 76.3 (CH, C-
12), 66.5 (CH2, C-23), 60.1 (CH, C-16), 58.6 (CH, C-8), 53.5 (C, C-
7), 52.8 (CH2, C-20), 49.9 (CH2, C-18), 45.8 (CH, C-13), 43.3 (CH2,
C-17), 32.8 (CH, C-14), 30.6 (CH2, CH2CH2CH2CH3), 27.1 (CH2,
C-15), 19.5 (CH2, O(CH2)2CH2CH3), 14.1 (CH3, CH3); ESIMS m/z
419.4 [M]+; mp 121 °C; anal. C 71.32, H 6.52, N 9.72%, calcd for
C25H29N3O3 C 71.58, H 6.97, N 10.02%.
(E)-11-(Isobutyloxyimino)strychnine (6d). 6d (30 mg, 5.5%) was
obtained from 1 and 1-bromo-2-methylpropane (0.04 mL)) in the
presence of KI (1 mg) as a white solid: 1H NMR (400 MHz, CDCl3) δ
4.13−4.08 [2H, m, OCH2 CH(CH3 )2 ], 2.02 [1H, m,
OCH2CH(CH3)2)], 0.91 [6H, dd, J = 6.6, 5.8 Hz, OCH2CH(CH3)2)],
δ and J values for all other hydrogen atoms coincide with the values for
the corresponding atoms of 6c within 0.04 ppm and 0.1 Hz,
respectively; 13C NMR (100 MHz, CDCl3) δ 83.0 (CH2, OCH2CH-
(CH3)2), 28.7 (CH, OCH2CH(CH3)2), 19.2 (CH3, OCH2CH-
(CH3)2); chemical shifts for all other carbon atoms coincide with
the δ values for the corresponding atoms of 6c within 0.1 ppm;
ESIMS m/z 419.5 [M]+; mp 121 °C; anal. C 71.42, H 6.66, N 9.78%,
calcd for C25H29N3O3 C 71.58, H 6.97, N 10.02%.
(E)-11-(Benzyloxyimino)strychnine (6e). 6e (45 mg, 9%) was
obtained from 1 and benzyl bromide (0.04 mL)) as a white solid;
eluate for column chromatography CH2Cl2−MeOH (100:7): 1H
NMR (400 MHz, CDCl3) δ 7.43−7.32 (5H, m, CH2C6H5), 5.57 (1H,
d, J = 12.3 Hz, HCHC6H5), 5.43 (1H, d, J = 12.3 Hz, HCHC6H5); δ
and J values for all other hydrogen atoms coincide with the values for
the corresponding atoms of 6c within 0.04 ppm and 0.1 Hz,
respectively; 13C NMR (100 MHz, CDCl3) δ 136.9 (C, CH2Car),
128.7 (CH, 2× CH2CCHar), 128.2 (C, CH2CCHCHCHar), 128.1
(CH, 2× CH2CCHCHar); chemical shifts for all other carbon atoms
coincide with the δ values for the corresponding atoms of 6c within
0.1 ppm; ESIMS m/z 453.4 [M]+; mp 144 °C; anal. C 73.53, H 5.88,
N 8.77%, calcd for C28H27N3O3 × 1/3MeOH C 73.31, H 6.15, N
9.05%.
(E)-11-(Pentyloxyimino)strychnine (6f). 6f (100 mg, 18%) was
obtained from 1 and n-pentyl bromide (0.04 mL) in the presence of
KI (1 mg) as a white solid: 1H NMR (400 MHz, CDCl3) δ 4.43−4.32
[2H, m, OCH2(CH2)3CH3], 1.73 [2H, m, OCH2CH2(CH2)2CH3],
1.40−1.31 [4H, m, OCH2CH2(CH2)2CH3], 0.92−0.86 [3H, m,
O(CH2)4CH3]; δ and J values for all other hydrogen atoms coincide
with the values for the corresponding atoms of 6c within 0.04 ppm
and 0.1 Hz, respectively; 13C NMR (100 MHz, CDCl3) δ 76.9 (CH2,
OCH2(CH2)3CH3), 29.0 [CH2, OCH2CH2(CH2)2CH3], 28.1 [CH2,
O(CH2)2CH2CH2CH3], 22.4 [CH2, O(CH2)3CH2CH3], 14.1 (CH3,
CH3); chemical shifts for all other carbon atoms coincide with the δ
values for the corresponding atoms of 6c within 0.1 ppm; ESIMS m/
(E)-11-(Propyloxyimino)strychnine (6b). 6b (130 mg, 25%) was
obtained from 1 and n-propyl bromide (0.03 mL) in the presence of
1
KI (1 mg) as a white solid: H NMR (400 MHz, DMSO-d6) δ 4.28
(2H, t, J = 6.4 Hz, OCH2CH2CH3), 1.74 (2H, m, OCH2CH2CH3),
0.99 (3H, t, J = 7.4 Hz, −OCH2CH2CH3); δ and J values for all other
F
DOI: 10.1021/acs.jnatprod.6b00479
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
z 433.5 [M]+; mp 109 °C; anal. C 73.53, H 5.88, N 8.77%, calcd for
C26H31N3O3 C 73.40, H 5.80, N 8.69%.
FLIPR membrane potential blue assay (Molecular Devices, USA).23
The tsA201 cells were maintained in Dulbecco’s modified Eagle
medium + Glutamax-I, supplemented with 10% fetal bovine serum
(FBS), 100 units/mL penicillin, and 100 μg/mL streptomycin at 37 °C
in a humidified 5% CO2 atmosphere. The cells were split into 10 cm
(2 × 106 cells) tissue culture plates and transfected the following day
with a total of 8 μg of cDNA using Polyfect transfection reagent
according to the protocol of the manufacturer (Qiagen, Hilden,
Germany). The following day, the cells were split into poly-D-lysine-
coated black 96-well plates with a clear bottom (BD Biosciences,
Bedford, MA, USA). Sixteen to 24 h later the medium was aspirated,
and the cells were washed with 100 μL of Krebs buffer [140 mM
NaCl/4.7 mM KCl/2.5 mM CaCl2/1.2 mM MgCl2/11 mM HEPES/
10 mM D-glucose, pH 7.4]. Then 50 μL of Krebs buffer was added to
the wells (in the antagonist experiments, various concentrations of the
antagonist were dissolved in the buffer), and an additional 50 μL of
Krebs buffer supplemented with the FMP assay dye (1 mg/mL) was
added to each well. The plate was incubated at 37 °C in a humidified
5% CO2 incubator for 30 min and assayed in a FlexStation3 benchtop
multimode microplate reader (Molecular Devices) measuring emission
at 565 nm (in fluorescence units) caused by excitation at 525 nm a
total of 90 s before and after addition of 33 μL of agonist solution in
Krebs buffer. The experiments were performed in duplicate at least
three times for each compound at each receptor using an EC50 (EC40−
EC60) concentration of glycine as agonist.
Single-Cell Patch-Clamp Assay. The direct activity of strychnine
derivatives on GlyR ion channel function was assessed using whole-cell
patch-clamp recording techniques. HEK293 cells were grown in 10 cm
tissue culture Petri dishes in MEM (Sigma, Deisenhofen, Germany)
supplemented with 10% FBS (Invitrogen, Karlsruhe, Germany) and
5000 IU penicillin/streptomycin at 5% CO2 and 37 °C in a water-
saturated atmosphere. For experiments, cells were plated on poly-L-
lysine-treated glass coverslips in 6 cm dishes. Transfection with human
GlyR alpha1 and beta subunits, each in the pRK5 vector, was
performed 1 day after cell passage using Gen-Carrier (Epoch
Lifesciences, Sugarland, TX, USA): 1.3 μg of receptor cDNA, 1.3 μg
of green fluorescence protein cDNA, and 2.6 μL of GenCarrier were
used, following the manufacturer’s instructions. Measurements were
performed 2 to 4 days after transfection.
(E)-11-[(Phenylethyl)imino]strychnine (6g). 6g (33 mg, 6%) was
obtained from 1 and 2-phenylethyl bromide (0.044 mL) as a white
solid; eluate for column chromatography CH2Cl2−MeOH (100:7): 1H
NMR (400 MHz, CDCl3) δ 7.30−7.09 (5H, m, CH2C6H5), 4.69−4.55
(2H, m, OCH2CH2C6H5), 3.12−3.04 (2H, m, OCH2CH2C6H5); δ
and J values for all other hydrogen atoms coincide with the values for
the corresponding atoms of 6c within 0.02 ppm and 0.1 Hz,
respectively; 13C NMR (100 MHz, CDCl3) δ 138.6 (C, CH2Car),
129.3 (CH, 2× CH2CCHar), 129.2 (C, CH2CCHCHCHar), 128.6
(CH, 2× CH2CCHCHar); 77.6 (CH2, OCH2CH2C6H5), 35.7 (CH2,
OCH2CH2C6H5); chemical shifts for all other carbon atoms coincide
with the δ values for the corresponding atoms of 6c within 0.1 ppm;
ESIMS m/z 467.6 [M]+; mp 189 °C; anal. C 73.74, H 6.61, N 8.58%,
calcd for C29H29N3O3 × 1/3MeOH, C 73.67, H 6.39, N 8.79%.
(E)-11-(Allyloxyimino)strychnine (6h). 6h (45 mg, 9%) was
obtained from 1 and allyl bromide (0.03 mL) as a pale yellow solid:
1H NMR (400 MHz, CDCl3) δ 6.10−6.00 (1H, m, CHCH2), 5.39
(1H,ddd, J = 17.3, 3.2, 1.6 Hz, CHCHH), 5.27 (1H, dd, J = 10.5, 1.4
Hz, CHCHH), 4.98−4.86 (2H, m, CH2CHCH2); δ and J values
for all other hydrogen atoms coincide with the values for the
corresponding atoms of 6c within 0.04 ppm and 0.1 Hz,
respectively; 13C NMR (100 MHz, CDCl3) δ 133.2 (CH, CH
CH2), 118.6 (CH2, CHCH2), 77.9 (CH2, CH2CHCH2);
chemical shifts for all other carbon atoms coincide with the δ values
for the corresponding atoms of 6c within 0.1 ppm; ESIMS m/z
402.4 [M − 1]+; mp 161 °C; anal. C 70.42, H 6.30, N 9.70%, calcd for
C24H25N3O3 × 1/3MeOH, C 70.57, H 6.41, N 10.15%.
(E)-11-(Propargyloxyimino)strychnine (6i). 6i (130 mg, 25%) was
obtained from 1 and propargyl chloride (0.023 mL) in the presence of
1
KI (1 mg) as a pale brown solid: H NMR (400 MHz, CDCl3) δ
4.95−4.92 (3H, m, CH2CCH, H-12), 2.47 (1H, t, J = 2.4 Hz, C
CH); δ and J values for all other hydrogen atoms coincide with the
values for the corresponding atoms of 6c within 0.04 ppm and 0.1
Hz, respectively; 13C NMR (100 MHz, CDCl3) δ 78.5 (CH, CCH),
63.9 (CH2, CH2CCH); chemical shifts for all other carbon atoms
coincide with the δ values for the corresponding atoms of 6c within
0.1 ppm; ESIMS m/z 401.4 [M]+; mp 148 °C; anal. C 71.42, H 6.12,
N 10.06%, calcd for C24H23N3O3, C 71.80, H 5.77, N 10.47%.
(E)-11-[(Methoxycarbonylmethyl)oxyimino]strychnine (6j). 6j
(130 mg, 23%) was obtained from 1 and methyl bromoacetate (0.03
Current responses from GlyR-transfected HEK293 cells were
measured at room temperature (21− 23 °C) at a holding potential
of −50 mV. Whole-cell recordings were performed using a HEKA
EPC10 amplifier (HEKA Electronics, Lambrecht, Germany) con-
trolled by Pulse software (HEKA Electronics). Recording pipets were
pulled from borosilicate glass tubes (World Precision Instruments,
Berlin, Germany) using a Sutter P-97 horizontal puller (Sutter,
Novato, CA, USA). Solutions were applied using an Octaflow system
(NPI Electronics, Tamm, Germany), where cells were bathed in a
laminar flow of buffer, giving a time resolution for solution exchange
and re-equilibration of about 100 ms. The external buffer consisted of
135 mM NaCl, 5.5 mM KCl, 2 mM CaCl2, 1.0 mM MgCl2, and 10
mM Hepes (pH adjusted to 7.4 with NaOH); the internal buffer was
140 mM CsCl, 1.0 mM CaCl2, 2.0 mM MgCl2, 5.0 mM EGTA, and 10
mM Hepes (pH adjusted to 7.2 with CsOH). Dose−response curves
for GlyR activation by glycine were calculated from a fit to the Hill
1
mL) as a pale brown solid: H NMR (400 MHz, CDCl3) δ 4.99 (1H,
d, J = 16.5 Hz, CHHCO2CH3), 4.90 (1H, d, J = 16.5 Hz,
CHHCO2CH3), 3.74 (3H, s, OCH3); δ and J values for all other
hydrogen atoms coincide with the values for the corresponding atoms
of 6c within 0.02 ppm and 0.1 Hz, respectively; 13C NMR (100
MHz, CDCl3) δ 169.3 (C, COOCH3), 72.2 (CH2, CH2CO2CH3),
52.2 (CH3, OCH3); chemical shifts for all other carbon atoms coincide
with the δ values for the corresponding atoms of 6c within 0.1 ppm;
ESIMS m/z 435.4 [M]+; mp 233 °C; anal. C 65.48, H 6.24, N 9.58%,
calcd for C24H25N3O5 × 1/3MeOH, C 65.51, H 5.95, N 9.42%.
(E)-11-[(Methoxycarbonylbutyl)oxyimino]strychnine (6k). 6k
(100 mg, 16%) was obtained from 1 and methyl 5-bromovalerate
(0.05 mL) as a white solid: 1H NMR (400 MHz, CDCl3) δ 4.42−4.30
[2H, m, OCH2(CH2)3CO2CH3], 3.64 (3H, s, OCH3), 2.39−2−33
[2H, m, O(CH2 )3 CH2 CO2 CH3 ], 1.78−1.72 [2H, m,
OCH2 CH2 (CH2 )2 CO2 CH3 ], 1.48−1.44 [2H, m, O-
(CH2)3CH2CO2CH3]; δ and J values for all other hydrogen atoms
coincide with the values for the corresponding atoms of 6c within
0.02 ppm and 0.1 Hz, respectively; 13C NMR (100 MHz, CDCl3) δ
174.0 (C, COOCH3), 76.9 [CH2, OCH2(CH2)3CO2CH3], 51.5 (CH3,
O C H 3 ) , 4 3 . 4 [ C H 2 , O ( C H 2 ) 3 C H 2 C O 2 C H 3 ] , 2 7 . 2
[OCH2CH2(CH2)2CO2CH3], 21.5 [CH2, O(CH2)3CH2CO2CH3];
chemical shifts for all other carbon atoms coincide with the δ values
for the corresponding atoms of 6c within 0.1 ppm; ESIMS m/z
477.5 [M]+; mp 115 °C; anal. C 67.81, H 6.43, N 9.58%, calcd for
C24H25N3O5, C 67.91, H 6.54, N 9.48%.
n
Hill
Iglycine
[Glycine]
equation
=
using a nonlinear algorithm in
Hill
n
n
Hill
Isat
[Glycine]
+ EC50
Microcal Origin (Additive, Friedrichsdorf, Germany). Here, Iglycine is
the current amplitude at a given glycine concentration, Isat is the
maximum current amplitude at saturating concentrations of glycine,
EC50 is the glycine concentration at half-maximal current responses,
and nHill is the Hill coefficient. Currents from each individual cell were
normalized to the maximum response at saturating glycine
concentrations. Inhibition curves were determined from varying the
concentration of strychnine derivatives in the presence of glycine near
EC50, i.e., 40 μM glycine for α1 receptors and 50 μM glycine for α1β
receptors using 4 to 7 cells per each inhibitor. IC50 values were
determined from inhibition curves that were constructed using a
nonlinear fit to the equation Iobs = Imax/[1 + ([I]/IC50)], where Iobs is
the observed current at any given concentration of inhibitor, Imax is the
FLIPR Membrane Potential Blue Assay. The functional
characterization of the strychnine analogues was performed in the
G
DOI: 10.1021/acs.jnatprod.6b00479
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
1H and 13C NMR spectra of compounds 2, 3, 5, and 6a−
k, traces of patch-clamp recordings and concentration
response curves for strychnine and compounds 1−4, 6a,
6b, 6h, and 6i, alignment of GlyR sequences from
subtype α3 (PDB: 5CFB) and α1 (PDB: 3JAD), pose-
retrieval docking data of the cocrystallized strychnine,
overlay of the docked pose and PDB coordinates (5CFB)
of strychnine, overlay of compounds 6i and 6g docked to
the strychnine binding site of the α3 GlyR (PDF)
maximum current amplitude observed in the absence of inhibitor, and
[I] is the concentration of inhibitor. In all experiments EC50 values
were determined for each individual cell from a nonlinear fit of dose
response data to the logistic equation (above). All data are given as
means SEM.
Radioligand Binding Studies. Cell Line and Transfection.
Transiently transfected tsA201 (=HEK293) cells in Dulbecco’s
modified Eagle medium + Glutamax-I (2 mM), supplemented with
10% fetal bovine serum, 1 mM sodium pyruvate, and 1% Pen/Strep at
37 °C in a humidified 5% CO2 atmosphere were used for the
experiments. Cells were transfected with a modified Ca3(PO4)3
precipitation method (10 μg plasmid GlyRα1 DNA, 50 μL of 2.5 M
CaCl2, 440 μL of 0.1× TE-buffer, 500 μL of HBS-transfection buffer
(50 mM HEPES, 12 mM dextrose, 10 mM KCl, 28 mM NaCl, 1.5 mM
Na2HPO4, pH 6.95), mixed, incubated for 20 min at room
temperature, dropwise given to cells) and washed with medium 12
h post-transfection. Following 48 h post-transfection, cells were
harvested.
Membrane Protein Preparation. Cell pellets from transfected cells
were homogenized in K3PO4 buffer (buffer H = 10 mM K3PO4, pH
7.4, protease inhibitor tablet complete, Roche, Mannheim, Germany).
Following homogenization, the protein suspension was centrifuged at
30000g for 20 min. After a repeating step of homogenization and
centrifugation, the membrane pellet was resuspended in buffer B (25
mM K3PO4, 200 mM KCl). GlyRα1 expression was verified using
immunostaining with a pan-α antibody following Western blotting.
Radioligand Binding Assay. GlyR protein (80 μg) was used for
sample preparation in a radioligand binding assay. In order to
determine the binding affinities of the strychnine analogues, all
compounds were diluted to concentrations of 10 000, 3000, 1000, 300,
100, 30, 10, 1, and 0.3 nM and incubated with 80 μg of GlyR protein
for 30 min on ice (all data points were done in triplicates). [3H]
strychnine (9.6 nM) was added and incubated with the suspension for
an additional 30 min on ice. The suspension was filtered through GC/
F glass microfiber filters (preincubated in buffer B + 0.5% BSA) and
washed twice with buffer B, dried, and resuspended in scintillation
liquid. All samples were analyzed in a Tri-Carb liquid scintillation
counter (PerkinElmer, Rodgau, Germany).
AUTHOR INFORMATION
■
Corresponding Author
*E-mail (D. Zlotos): darius.zlotos@guc.edu.eg.
ORCID
Hans-Georg Breitinger: 0000-0003-0111-3528
Christoph Sotriffer: 0000-0003-4713-4068
Darius P. Zlotos: 0000-0002-1193-3119
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was partially funded by Bundesministerium fur
̈
Bildung und Forschung (BMBF) and Deutscher Akademischer
Austauschdienst (DAAD). A.A.J. thanks the Novo Nordisk
Foundation for financial support.
REFERENCES
■
(1) Lynch, J. W. Physiol. Rev. 2004, 84, 1051−1095.
(2) Laube, B.; Maksay, G.; Schemm, R.; Betz, H. Trends Pharmacol.
Sci. 2002, 23, 519−527.
(3) Jensen, A. A.; Kristiansen, U. Biochem. Pharmacol. 2004, 67,
1789−1799.
(4) Lynch, J.; Callister, R. Curr. Opin. Invest. Drugs 2006, 7, 48−53.
(5) Breitinger, H.-G. eLS; 2014, DOI: 10.1002/
9780470015902.a0000236, pub2.
Molecular Modeling. The crystal structure of the human glycine
receptor α3 bound to strychnine (PDB code: 5CFB)14 was
downloaded from the PDB database. The protein was prepared with
the structure preparation wizard in MOE (version 2015.1028) and
saved as a mol2 file. The 3D structures of compounds 1, 2, 6i, and 6g
were built based on the X-ray crystallographic structure of
strychnine.29 The compounds were minimized using the MMFF94x
force field in MOE using a gradient of 0.0001 kcal/(mol Å), and their
protonation and tautomeric states at pH 7.0 were generated using
(6) Breitinger, H.-G.; Becker, C.-M. Curr. Pharm. Design 1998, 4,
315−334.
(7) Bode, A.; Lynch, J. Mol. Brain 2014, 7, 2.
(8) Schaefer, N.; Langlhofer, G.; Kluck, C. J.; Villmann, C. Br. J.
Pharmacol. 2013, 170, 933−952.
(9) Langosch, D.; Thomas, L.; Betz, H. Proc. Natl. Acad. Sci. U. S. A.
1988, 85, 7394−7398.
Epik30,31 (version 3.6, Schrodinger) and Spores.32,33
̈
Docking was performed using GOLD (version 5.3).34,35 Binding
site residues were defined by specifying the crystal structure ligand
coordinates bound to protein chains A and B and using the default
cutoff radius of 6 Å, with the “detect cavity” option enabled. GOLD
docking experiments were performed using the GoldScore scoring
function. The search efficiency of the genetic algorithm was at 200%
setting with the receptor kept rigid. For each compound, 50 complexes
were generated and clustered based on their RMSD with the threshold
set at 0.75 Å using the complete linkage method. Complexes were
rescored using DrugScore (version 0.9)36−38 utilizing the DrugScor-
(10) Watanabe, E.; Akagi, H. Neurosci. Res. 1995, 23, 377−382.
(11) Lynch, J. W. Neuropharmacology 2009, 56, 303−309.
(12) Yu, R.; Hurdiss, E.; Greiner, T.; Lape, R.; Sivilotti, L.; Biggin, P.
C. Biochemistry 2014, 53, 6041−6051.
(13) Du, J.; Lu, W.; Wu, S.; Cheng, Y.; Gouaux, E. Nature 2015, 526,
̈
224−229.
(14) Huang, X.; Chen, H.; Michelsen, K.; Schneider, S.; Shaffer, P. L.
Nature 2015, 526, 277−280.
(15) Mackerer, C. R.; Kochman, R. L.; Shen, T. F.; Hershenson, F.
M. J. Pharmacol. Exp. Ther. 1977, 201, 326−331.
(16) Iskander, G. M.; Bohlin, L. Acta Pharm. Suec. 1978, 15, 431−
438.
e
CSD potential, and the best-ranked pose from each cluster was visually
examined to select the final pose. The quality of pose prediction was
assessed by calculating the heavy atom RMSD between the docked
poses and the original PDB coordinates of strychnine (Figure S18,
Supporting Information). The figures were prepared using PyMol.39
(17) Jensen, A. A.; Gharagozloo, P.; Birsdall, N. J. M.; Zlotos, D. P.
Eur. J. Pharmacol. 2006, 539, 27−33.
(18) Mohsen, A. M. Y.; Heller, E.; Holzgrabe, U.; Zlotos, D. P. Chem.
Biodiversity 2014, 11, 1256−1262.
(19) Zlotos, D. P.; Buller, S.; Stiefl, N.; Baumann, K.; Mohr, K. J.
Med. Chem. 2004, 47, 3561−3571.
ASSOCIATED CONTENT
■
S
* Supporting Information
(20) Tang, P.; Furuya, T.; Ritter, T. J. Am. Chem. Soc. 2010, 132,
12150−12154.
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jnat-
prod.6b00479.
(21) Gharagozloo, P.; Lazareno, S.; Popham, A.; Birdsall, N. J. M. J.
Med. Chem. 1999, 42, 438−445.
H
DOI: 10.1021/acs.jnatprod.6b00479
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
(22) Rosenmund, P.; Schmitt, M.-P.; Franke, H. Liebigs Ann. Chem.
1980, 6, 895−907.
(23) Jensen, A. A. Eur. J. Pharmacol. 2005, 521, 39−42.
(24) Jensen, A. A.; Bergmann, M. L.; Sander, T.; Balle, T. J. Biol.
Chem. 2010, 285, 10141−10153.
(25) Hess, G. P. Biochemistry 1993, 32, 989−1000.
(26) Breitinger, H. G.; Geetha, N.; Hess, G. P. Biochemistry 2001, 40,
8419−8429.
(27) Raafat, K.; Breitinger, U.; Mahran, L.; Ayoub, N.; Breitinger, H.-
G. Toxicol. Sci. 2010, 118, 171−182.
(28) Molecular Operating Environment (MOE), version 2013.08;
Chemical Computing Group Inc: Montreal, QC, Canada, 2016.
(29) Mostad, A. Acta Chem. Scand. 1985, B39 (9), 705−716.
(30) Greenwood, J. R.; Calkins, D.; Sullivan, A. P.; Shelley, J. C. J.
Comput.-Aided Mol. Des. 2010, 24, 591−604.
(31) Shelley, J. C.; Cholleti, A.; Frye, L.; Greenwood, J. R.; Timlin,
M. R.; Uchimaya, M. J. Comput.-Aided Mol. Des. 2007, 21, 681−691.
(32) ten Brink, T.; Exner, T. E. J. Chem. Inf. Model. 2009, 49, 1535−
1546.
(33) ten Brink, T.; Exner, T. E. J. Comput.-Aided Mol. Des. 2010, 24,
935−942.
(34) Jones, G.; Willett, P.; Glen, R. C. J. Mol. Biol. 1995, 245, 43−53.
(35) Jones, G.; Willett, P.; Glen, R. C.; Leach, A. R.; Taylor, R. J. Mol.
Biol. 1997, 267, 727−748.
(36) Gohlke, H.; Hendlich, M.; Klebe, G. J. Mol. Biol. 2000, 295,
337−356.
(37) Neudert, G.; Klebe, G. J. Chem. Inf. Model. 2011, 51, 2731−
2745.
(38) Velec, H. F.; Gohlke, H.; Klebe, G. J. Med. Chem. 2005, 48,
6296−303.
(39) PyMOL Molecular Graphics System, Version 1.8; Schrodinger,
̈
LLC; 2015.
I
DOI: 10.1021/acs.jnatprod.6b00479
J. Nat. Prod. XXXX, XXX, XXX−XXX