Identification of dihydrostilbenes in Pholidota chinensis as a new scaffold for GABAA receptor modulators

Article abstract of DOI:10.1016/j.bmc.2014.01.008

A dichloromethane extract of stems and roots of Pholidota chinensis (Orchidaceae) enhanced GABA-induced chloride currents (I_GABA) by 132.75 ± 36.69% when tested at 100 μg/mL in a two-microelectrode voltage clamp assay, on Xenopus laevis oocytes

Full text of DOI:10.1016/j.bmc.2014.01.008

Bioorganic & Medicinal Chemistry 22 (2014) 1276–1284
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Identification of dihydrostilbenes in Pholidota chinensis as a new
scaffold for GABA receptor modulators
A
a
b
a
a
a
Diana C. Rueda , Angela Schöffmann , Maria De Mieri , Melanie Raith , Evelyn A. Jähne ,
b
a,
⇑
Steffen Hering , Matthias Hamburger
a
Division of Pharmaceutical Biology, University of Basel, Klingelbergstrasse 50, CH-4056 Basel, Switzerland
Institute of Pharmacology and Toxicology, University of Vienna, Althanstrasse 14, A-1090 Vienna, Austria
b
a r t i c l e i n f o
a b s t r a c t
Article history:
A
dichloromethane extract of stems and roots of Pholidota chinensis (Orchidaceae) enhanced
GABA-induced chloride currents (IGABA) by 132.75 ± 36.69% when tested at 100 g/mL in a two-micro-
electrode voltage clamp assay, on Xenopus laevis oocytes expressing recombinant 2S GABA recep-
Received 3 October 2013
Revised 20 December 2013
Accepted 3 January 2014
Available online 10 January 2014
l
a
1
2
b c
A
tors. By means of an HPLC-based activity profiling approach, the three structurally related stilbenoids
coelonin (1), batatasin III (2), and pholidotol D (3) were identified in the active fractions of the extract.
Dihydrostilbene 2 enhanced IGABA by 1512.19 ± 176.47% at 300 lM, with an EC50 of 52.51 ± 16.96 lM,
Keywords:
Pholidota chinensis
HPLC-based activity profiling
while compounds 1 and 3 showed much lower activity. The relevance of conformational flexibility for
receptor modulation by stilbenoids was confirmed with a series of 13 commercially available stilbenes
and their corresponding semisynthetic dihydro derivatives. Dihydrostilbenes showed higher activity in
the oocyte assay than their corresponding stilbenes. The dihydro derivatives of tetramethoxy-piceatannol
A
GABA receptor modulators
Dihydrostilbenes
(
12) and pterostilbene (20) were the most active among these derivatives, but they showed lower effi-
ciencies than compound 2. Batatasin III (2) showed high efficiency but no significant subunit specificity
when tested on the receptor subtypes 2s, and
Dihydrostilbenes represent a new scaffold for GABA
a
1
b
2
c
2s
,
a
2
b
c
2 2s
,
a
3
b
2
c
2s
,
a
4
b
c
2 2s
,
5
a b
2
c
2s
,
a
1
b
1
c
1 3 2s
a b c .
A
receptor modulators.
Ó 2014 Elsevier Ltd. All rights reserved.
1
. Introduction
GABA receptors are ligand-gated chloride channels physiolog-
experimental subunit-specific GABAergic drugs for more than a
decade, no subtype-selective GABA receptor modulators have
been introduced into clinical practice.
In a search for new natural product-derived GABA receptor
A
A
3
,4
A
ically activated by GABA, the major inhibitory neurotransmitter in
the brain. Structurally, they are heteropentameric assemblies
forming a central chloride-selective channel. Up to now 19 differ-
modulators, we screened a library of 880 fungal and plant extracts
in an automated functional two-microelectrode voltage clamp as-
5
ent subunits (
a1–6, b1–3,
c1–3, d,
e
, h,
p,
q1–3) have been identi-
say on Xenopus oocytes transiently expressing GABA
A
receptors
fied. Depending on the nature, stoichiometry, and arrangement of
these subunits, the receptor subtypes exhibit distinct pharmaco-
logical profiles providing the potential for rational drug therapy
1 2
of the a b c2S subtype, the most abundant subunit combination
2
in the human brain. In this screening the dichloromethane extract
of stems and roots of Pholidota chinensis LINDL (Orchidaceae)
showed promising activity.
in several disorders related with impaired GABAergic function,
such as epilepsy, insomnia, anxiety, and mood disorders.1 GABA
–3
Orchidaceae is the largest family of flowering plants, with
around 25,000 species in over 800 genera. The family shows world-
wide distribution, with greatest diversity in tropical and subtropi-
cal climate zones. Apart from their ornamental value, many orchids
have been used as medicinal plants. In traditional Chinese medi-
cine we find the earliest written records for medicinal uses of orch-
A
receptors are the target for many clinically important drugs such as
benzodiazepines (BDZs), barbiturates, neuroactive steroids, anes-
thetics, and certain other CNS depressants. Due to their lack of
subunit specificity, these drugs show a number of adverse side ef-
fects. Hence, there is a need for subtype-selective drugs devoid of
the side effects of the classical BDZs. Despite the availability of
6
–8
ids.
In Chinese folk medicine, the whole plant of P. chinensis (shi
xian tao) has long been used in the treatment of diverse conditions,
such as hypertension, headache, gastroenteritis, and bronchitis.
Previous pharmacological studies on P. chinensis reported sedative
Abbreviations: BDZs, benzodiazepines; GABA, gamma-aminobutyric acid;
GABA Rs, gamma-aminobutyric acid type A receptors.
A
9
–12
⇑
Corresponding author. Tel.: +41 61 2671425; fax: +41 61 2671474.
E-mail address: matthias.hamburger@unibas.ch (M. Hamburger).
and anticonvulsant activities.
The genus Pholidota comprises
approximately 30 species distributed from tropical Asia to tropical
0
968-0896/$ - see front matter Ó 2014 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2014.01.008

D. C. Rueda et al. / Bioorg. Med. Chem. 22 (2014) 1276–1284
1277
Australia. Phytochemical studies on the genus showed the pres-
ence of triterpenes, steroids, lignans, benzoxepines, and stilbe-
(Scharlau Chemie S.A.) and water were used for HPLC separations.
For analytical separations, HPLC solvents contained 0.1% of HCO H.
NMR spectra were recorded in methanol-d (Armar Chemicals). For
extraction and flash chromatography, distilled technical grade sol-
2
1
0
noids.
Stilbenoids exhibit
a
limited but heterogeneous
4
distribution in the plant kingdom, and have been most widely re-
ported from the Orchidaceae family. Dihydrostilbenes and 9,10-
dihydrophenanthrenes have been previously identified in the
vents were used. Silica gel (63–200
column chromatography.
lm, Merck) was used for open
1
3–17
genus Pholidota.
In this study, batatasin III (2) isolated from a DCM extract of
P. chinensis was identified as a positive GABA receptor modulator
by means of HPLC-based activity profiling,
approach for the rapid identification of new bioactive natural com-
2.2. Plant material
A
18
a miniaturized
Shi Xian Tao (dried stems/roots of P. chinensis Lindl.) was pur-
chased in 2008 from a local herbal market in Kunming and authen-
ticated by Dr. X. Yang (previously Pharmaceutical Biology,
University of Basel, now Kunming Institute of Botany, Chinese
Academy of Science, Kunming, PR China). A voucher specimen
(463) is deposited at the Division of Pharmaceutical Biology, Uni-
versity of Basel.
1
9–22
pounds,
ery of GABA
selectivity of 2 was assessed at GABA
that we have been successfully applying in the discov-
receptor ligands of natural origin.2
receptor subtypes
and Furthermore,
dihydrostilbenes were established as a new scaffold for GABA
3–27
The subunit
A
A
2 2 2s
a b c ,
a
3
b
2
c
2s
,
a
4
b
c
2 2s
,
5
a b
2
c
2s
,
a
1
b
1
c
2s
,
1 3 2s
a b c .
A
receptor modulators, by comparison of the performance of a series
of commercially available stilbenes and their semisynthetic
dihydro derivatives on the oocyte assay.
2.3. Extraction
The plant material was frozen with liquid nitrogen and ground
with a ZM1 ultracentrifugal mill (Retsch). The DCM extract for
screening and HPLC-based activity profiling was prepared with
an ASE 200 extraction system with solvent module (Dionex, Sun-
nyvale CA). Three extraction cycles (5 min each) were performed,
at an extraction pressure of 120 bar and a temperature of 70 °C.
For preparative isolation, 293 g of ground plant material was mac-
erated with DCM (4 ꢀ 1 L, 3 h each, permanent magnetic stirring).
The solvent was evaporated at reduced pressure to yield 10.3 g of
extract. The extracts were stored at 2–8 °C until use.
2
2
. Experimental
.1. General procedures
1
D and 2D NMR spectra were measured at room temperature
on a Bruker Avance III 500 MHz spectrometer (Bruker BioSpin,
Fällanden, Switzerland) operating at 500.13 MHz. Spectra were re-
corded at 291.2 K with a 1 mm TXI probe with z-gradient. The fol-
1
lowing settings were used: 128 scans for H spectra; 8 scans for
1
1
H H-COSY spectra (cosygpqf pulse program); 32 scans and 256
2.4. Microfractionation for activity profiling
increments for HSQC experiments (hsqcedetgp pulse program); 64
scans and 128 increments for HMBC spectra (hmbcgp pulse pro-
gram). Spectra were analyzed by TopSpin 3.0 software (Bruker).
High resolution mass spectra (HPLC–ESITOFMS) were recorded in
positive mode, m/z range 100–800, on a Bruker microTOF ESIMS
system (Bruker Daltonics, Bremen, Germany) connected via a
T-splitter (1:10) to an Agilent HP 1100 system consisting of a deg-
asser, a binary mixing pump, autosampler, column oven, and a
diode array detector (G1315B) (Agilent Technologies, Waldbronn,
Germany). Nitrogen was used as a nebulizing gas at a pressure of
Time-based microfractionation for GABA receptor activity pro-
A
2
3,27,28
filing was performed as previously described,
with minor
modifications: separation was done on a semi-preparative HPLC
column with MeOH (solvent A) and water (solvent B), using a gra-
dient from 50% A to 80% A in 30 min, followed by 80% A to 100% A
in 2 min. The flow rate was 4 mL/min, and 10 mg of extract (in
200 lL of DMSO) were injected. A total of 24 microfractions of
90 s each were collected. After parallel evaporation of solvents,
the dry films were redissolved in 1 mL of methanol, and aliquots
2
.0 bar, and as drying gas at a flow rate of 9.0 L/min (dry gas tem-
perature 240 °C). Capillary voltage was set at 45,000 V; hexapole at
30.0 Vpp. Instrument calibration was done with a reference solu-
tion of sodium formate 0.1% in 2-propanol/water (1:1) containing
mM NaOH. Data acquisition and processing was performed with
of 0.5 mL were dispensed in two vials, dried under N gas, and sub-
mitted to bioassay.
2
2
2.5. Isolation
5
Bruker Daltonics Hystar 3.0 software. Semi-preparative HPLC sep-
arations for activity profiling and purification were performed with
an Agilent HP 1100 series system consisting of a quaternary pump,
autosampler, column oven, and diode array detector (G1315B).
An aliquot of the DCM extract (450 mg) was dissolved in CHCl3
and adsorbed onto silica gel (3 g), prior to packing into a Buchi Prep
Elut adapter. Separation was performed on a Sepacore silica gel
Ò
cartridge eluted with an n-hexane (solvent A) and EtOAc (solvent
B) gradient: 0% B to 30% B in 90 min, followed by 30% B to 50% B
in 30 min, and 50% B to 80% B in 30 min. The flow rate was set at
15 mL/min. Fractions of 15 mL were collected and later combined
into 18 fractions (1–18) on the basis of a TLC analysis (detection
at 254, 366, and at daylight after staining with anisaldehyde–sulfu-
ric acid reagent). Fractions 1–18 were submitted to analytical
HPLC–PDA–ESIMS with MeOH (solvent C) and water (solvent D),
using a gradient from 50% C to 100% C in 30 min, hold for
Waters SunFire™ C18 (3.5
C18 (5
l
m, 3.0 ꢀ 150 mm) and SunFire™ Prep
l
m, 10 ꢀ 150 mm) columns were used for analytical and
semi-preparative HPLC analysis, respectively (Waters, Wexford,
Ireland). Parallel evaporation of semi-preparative HPLC fractions
was performed with a Genevac EZ-2 plus vacuum centrifuge (Gen-
evac Ltd, Ipswich, United Kingdom). Flash chromatography was
performed with pre-packed Buchi SepacoreÒ silica flash cartridges
Ò
(
40–63
l
m, 40 ꢀ 150 mm) on a Buchi Sepacore system consisting
of two C-605 pumps, a C-620 control unit, and a C-660 fraction col-
lector (Buchi, Flawil, Switzerland). Sample introduction was car-
ried out with a Buchi Prep Elut adapter filled with the sample
adsorbed onto silica gel. The separation was monitored by TLC. Pre-
parative HPLC separations were performed with a Waters SunFire
15 min. The flow rate was 0.4 mL/min, and 10
lg of each fraction
(in 10 L of DMSO) were injected. Fractions 13 and 14 were found
l
to contain the compounds of interest and were submitted to semi-
preparative HPLC using solvents C and D as eluents. A gradient of
50% C to 60% C in 30 min was used for fraction 13, and isocratic
conditions (50% C, 30 min) for fraction 14. The flow rate was
4 mL/min. Stock solutions in DMSO (50 mg/mL) were prepared
Prep C18 OBD column (5
Shimadzu LC-8A preparative HPLC with SPD-M10A VP diode array
l
M, 30 ꢀ 150 mm) connected to a
detector. HPLC-grade MeOH (Scharlau Chemie S.A.), acetonitrile
and repeatedly injected in portions of 50–100 lL. A portion

1
278
D. C. Rueda et al. / Bioorg. Med. Chem. 22 (2014) 1276–1284
(
17 mg) of fraction 13 (25 mg) afforded compounds 1 (2.3 mg) and
2.6.2. General procedure
2
1
(6.5 mg). Compound 3 (2 mg) was isolated from 10 mg of fraction
4 (16 mg).
Compounds 1–3 were identified by comparison of their physio-
Dihydro derivatives of compounds 7, 9, 11, 13, 15, 17, 19, 21,
and 23 were prepared by hydrogenation of corresponding stilb-
enes. A standard protocol was followed, with minor modifica-
tions. Solutions of each stilbene (10 mg) in absolute EtOH (5 ml)
3
2
chemical data (NMR, ESI-TOFMS, and UV–vis) with published val-
ues.
1
4,29–31
1
The purity was >95% (purity check by H NMR).
2
were stirred under H for 3 h in the presence of 10% Pd/C. The reac-
tion mixtures were filtered over Celite to remove the catalyst, and
evaporated to dryness. The resulting residues were purified by
flash column chromatography, using a hexane/EtOAc gradient, to
afford target compounds 8, 10, 12, 14, 16, 18, 20, 22, and 24,
respectively, in yields of 85–95%. The spectroscopic data of com-
pounds were in agreement with the literature, except for com-
2
.5.1. Coelonin (1)
1
4 H
H NMR (methanol-d , 500.13 MHz) d (ppm): 8.13 (1H, d,
J = 8.4 Hz, H-5), 6.62 (1H, dd J = 8.3 and 2.7 Hz, H-6), 6.61 (1H, d,
J = 2.6 Hz, H-8), 6.30 (1H, d, J = 2.5 Hz, H-3), 6.26 (1H, d,
J = 2.5 Hz, H-1), 3.67 (3H, s, 4-OCH
3
), 2.59 (4H, s, H-9 and H-10);
C shifts (derived from multiplicity-edited HSQC and HMBC spec-
tra), d (ppm): 158.3 (C, C-4), 155.4 (C, C-2), 154.8 (C, C-7), 139.8 (C,
1
3
1
pound 24, for which no report was found ( H NMR spectrum is
3
2–41
C
provided as Supporting information).
C-8a), 138.7 (C, C-10a), 128.6 (CH, C-5), 125.2 (C, C-4b), 114.8 (C,
C-4a), 113.8 (CH, C-8), 112.2 (CH, C-6), 104.8 (CH, C-1), 100.1
0
2.6.3. trans-2-Fluoro-4 -methoxy-dihydrostilbene (24)
1
(
CH, C-3), 54.2 (4-OCH
3
), 30.8 (CH
2 2
, C-10) 30.1 (CH , C-9).
H NMR (chloroform-d
m), 7.08–7.98 (2H, m), 6.90–6.80 (2H, m), 3.81 (3H, s), 3.05–2.84
4H, m). HRESI-MS m/z 253.1589 [M+Na] (calcd formula weight
4 H
, 500.13 MHz) d (ppm): 7.26–7.08 (4H,
+
HR-ESIMS m/z 243.1016 [M+H] (calcd for C15
15 3
H O
, 243.1016).
+
(
2
.5.2. Batatasin III (2)
for C15 15FO, 230.2774).
H
1
H NMR (methanol-d
4
, 500.13 MHz) d
H
0
(ppm): 7.03 (1H, dd,
0
0
0
J = 7.9 and 7.5 Hz, H-5 ), 6.64 (3H, m, H-2 , H-4 , and H-6 ), 6.29
2.7. Expression of GABA
A
receptors
(
2
1H, br s, H-2), 6.23 (2H, m, H-4 and H-6), 3.63 (3H, s, 5-OCH
3
),
.75 (4H, m, H-
a
and H-b); 13C shifts (derived from multiplicity-
Stage V–VI oocytes from Xenopus laevis were prepared, and
2
3
edited HSQC and HMBC spectra), d
C
(ppm): 160.9 (C, C-5), 157.6
cRNA injected as previously described. Female Xenopus laevis
(NASCO, Fort Atkinson, WI) were anesthetized by exposing them
for 15 min to a 0.2% MS-222 (methanesulfonate salt of 3-amino-
benzoic acid ethyl, Sigma) solution before surgically removing
parts of the ovaries. Follicle membranes from isolated oocytes were
enzymatically digested with 2 mg/mL collagenase from Clostridium
histolyticum (Type 1A, Sigma). Synthesis of capped runoff poly(A+)
cRNA transcripts was obtained from linearized cDNA templates
(pCMV vector). Directly after enzymatic isolation, the oocytes were
injected with 50 nL of DEPC-treated water (Sigma) containing dif-
ferent cRNAs at a concentration of approximately 300–3000 pg/nL
per subunit. The amount of injected cRNA mixture was determined
by means of a NanoDrop ND-1000 (Kisker Biotech). To ensure
0
0
(
C, C-3), 156.7 (C, C-3 ), 144.9 (C, C-1), 143.3 (C, C-1 ), 129.0 (CH,
0
0
0
0
C-5 ), 119.8 (CH, C-6 ), 115.3 (CH, C-2 ), 112.4 (CH, C-4 ), 108.1
CH, C-2), 105.5 (CH, C-6), 98.7 (CH, C-4), 54.3 (CH , 5-OCH ),
7.6 (CH , C-b), 37.0 (CH , C- ). HR-ESIMS m/z 245.1176 [M+H]
calcd for C15 , 245.1172).
(
3
(
3
3
+
2
2
a
17 3
H O
2
.5.3. Pholidotol D (3)
1
H NMR (methanol-d
4
, 500.13 MHz) d
H
(ppm): 7.17 (1H, dd,
0
0
0
J = 7.9 and 7.8 Hz, H-5 ), 7.00–6.95 (4H, m, H-2 , H-6 , H-
a and H-
0
b), 6.69 (1H, dd, J = 8.2 and 2.2 Hz, H-4 ), 6.58 (2H, m, H-2 and H-
13
6
3
), 6.31 (1H, t, J = 2 Hz, H-4), 3.76 (3H, s, 5-OCH ); C shifts (de-
rived from multiplicity-edited HSQC and HMBC spectra), d
C
0
(
1
a
ppm): 160.8 (C, C-5), 157.7 (C, C-3), 156.0 (C, C-3 ), 139.7 (C, C-
1 2
expression of the gamma subunit in a b c2S receptors, rat cRNAs
0
0
), 138.5 (C, C-1 ), 129.4 (CH, C-5 ), 128.6 (CH, C-b), 128.4 (CH, C-
were mixed in a 1:1:10 ratio. Oocytes were then stored at 18 °C
in ND96 solution containing 1% of penicillin–streptomycin solution
(Sigma–Aldrich). Voltage clamp measurements were performed
between days 1 and 5 after cRNA injection.
0
0
0
), 117.8 (CH, C-6 ), 114.4 (CH, C-4 ), 112.4 (CH, C-2 ), 105.8 (CH,
, 5-OCH ). HR-
, 243.1016).
C-2), 103.4 (CH, C-6), 100.3 (CH, C-4), 54.4 (CH
ESIMS m/z 243.1017 [M+H] (calcd for C15H O
15 3
3
3
+
Further purification of compound 2 for subunit specificity tests
was achieved by separating a portion of the extract (7.3 g) by open
column chromatography (6 ꢀ 69 cm, 700 g of silica gel), using a
step gradient of n-hexane–EtOAc (100:0, 95:5, 90:10, 85:15,
2.8. Positive control
Diazepam (7-chloro-1,3-dihydro-1-methyl-5-phenyl-2H-1,4-
8
2
1
0:20, 75:25, 70:30, 65:35, 60:40, 55:45, 50:50, 40:60, 30:70,
0:80, 10:90, 0:100, 1 L each) and washing in the end with MeOH
00% (1.5 L). The flow rate was ca. 50 mL/min. The effluent was
benzodiazepin-2-one, Sigma, purity not less than 98%) was used
as positive control. At 1 lM diazepam enhanced IGABA up to
231.3 ± 22.6% (n = 3). See also Figure S1, Supporting information.
combined to 15 fractions (1–15) based on TLC patterns. After
HPLC–PDA–MS analysis, fractions 7 and 8 were selected for isola-
tion of compound 2 by preparative HPLC, with acetonitrile (solvent
A) and water (solvent B), using a gradient from 40% A to 50% A in
2.9. Two-microelectrode voltage clamp studies
Electrophysiological experiments were performed by the two-
microelectrode voltage clamp method making use of a TURBO
TEC 03X amplifier (npi electronic GmbH) at a holding potential
of ꢁ70 mV and pCLAMP 10 data acquisition software (Molecular
Devices). Currents were low-pass-filtered at 1 kHz and sampled
at 3 kHz. The bath solution contained 90 mM NaCl, 1 mM KCl,
3
0 min, followed by 50% A to 100% A in 5 min, hold for 10 min.
The flow rate was 20 mL/min. Stock solutions in THF (100 mg/
mL) were prepared and repeatedly injected in portions of 300–
4
00
lL. The separation of fractions 7 (129 mg) and 8 (132 mg),
yielded compound 10.8 mg of 2.
1
mM MgCl
ing solution contained 2 M KCl. Oocytes with maximal current
amplitudes >3 A were discarded to exclude voltage clamp errors.
2 2
, 1 mM CaCl , and 5 mM HEPES (pH 7.4). Electrode fill-
2
2
.6. Synthesis of dihydrostilbenes
l
.6.1. Stilbenes
Compounds 4–7, 9, 11, 13, 15, 17, and 19 were purchased from
2.10. Fast solution exchange during IGABA recordings
TCI Europe N.V. Compounds 21 and 25 were purchased from Sig-
ma–Aldrich Co. Compound 23 was purchased from Santa Cruz Bio-
technology, Inc.
Test solutions (100
lL) were applied to the oocytes at a speed of
300 L/s by means of the ScreeningTool (npi electronic, Tamm,
l

D. C. Rueda et al. / Bioorg. Med. Chem. 22 (2014) 1276–1284
1279
5
Germany) automated fast perfusion system. In order to determine
GABA EC3–10 (typically between 3 and 10 M for receptors of sub-
unit composition 2s), a dose–response experiment with GABA
concentrations ranging from 0.1 M to 1 mM was performed. Stock
solution of the DCM extract (10 mg/mL in DMSO) was diluted to a
concentration of 100 g/mL with bath solution containing GABA
l
1 2
a b c
l
l
23
EC3–10 according to a validated protocol. As previously described,
microfractions collected from the semi-preparative HPLC separa-
tions were dissolved in 30
with 2.97 mL of bath solution containing GABA EC3–10
solution of each pure compound tested (100 mM in DMSO) was di-
luted to concentrations of 0.1, 1.0, 3.0, 10, 30, 100, 300, and 500
lL of DMSO and subsequently mixed
2
3
.
A stock
lM
with bath solution for measuring direct activation, or with bath
solution containing GABA EC3–10 for measuring modulation of
GABA
including the GABA control samples was adjusted to 1% to avoid
solvent effect at the GABA receptor.
A
receptors. The final DMSO concentration in all the samples
A
2
.11. Data analysis
Enhancement of the IGABA was defined as I(GABA+Comp)/IGABA ꢁ 1,
where I(GABA+Comp) is the current response in the presence of a given
compound, and IGABA is the control GABA-induced chloride current.
Data were analyzed using the ORIGIN 7.0 SR0 software (OriginLab
Corporation) and are given as mean ± SE of at least two oocytes and
P2 oocyte batches.
3
3
. Results and discussion
.1. Isolation and structure elucidation of active compounds
Figure 1. HPLC-based activity profiling of a DCM extract of stems and roots of P.
chinensis, for GABA
A
receptor modulatory activity. (B) HPLC chromatogram (210–
7
00 nm) of a semipreparative separation of 10 mg of extract. The numbers above
Screening for GABA
Xenopus laevis oocytes transiently expressing GABA
the subtype 2s. In an automated fast-perfusion system used
for two-microelectrode voltage clamp measurements, a dichloro-
methane extract (100 g/mL) of P. chinensis roots enhanced the
A
modulating activity was performed with
peaks designate compounds 1–3. The 24 time-based fractions of 90 s each are
indicated with dashed lines. (A) Potentiation of the IGABA by each microfraction
A
receptors of
1 2
a b c
(
error bars correspond to SE).
5
l
GABA-induced chloride ion current (IGABA) by 132.8 ± 36.7%. To
track the activity in the extract, we used HPLC-based activity pro-
filing with a validated protocol.23 The chromatogram (210–
A
3.2. Modulation of GABA receptors
For a preliminary activity profile at GABA
A
Rs of the subtype
M in the Xeno-
7
00 nm) of a semipreparative HPLC separation (10 mg of extract)
1
a b
2
c
2s, 1–3 were tested at a concentration of 100 l
and the corresponding activity profile of the time-based fraction-
ation (24 microfractions of 90 s each) are shown in Figure 1B and
A, respectively. The major peak of activity was found in fraction
pus oocyte assay. Batatasin III (2) was the most efficient among the
three compounds. It potentiated IGABA by 628.3 ± 87.1%, while com-
pounds 1 and 3 exhibited weaker enhancements (139.5 ± 14.4%
and 192.0 ± 64.1%, respectively) (Fig. 3A). Further concentration–
9
, which potentiated IGABA by 119.1 ± 19.1%. Fraction 8 showed
marginal activity (enhancement of IGABA by 26.5 ± 4.7%). All the
remaining fractions showed minimal activity and were not consid-
ered further.
response experiments on
compounds 1–3, at concentrations ranging from 1 to 300
(500 M for compound 3). As shown in Figure 3B, all stilbenoids
enhanced IGABA at a GABA EC3–10 in a concentration-dependent
manner. The bibenzyl batatasin III (2) displayed strong GABA
1 2
a b c2s receptors were performed with
l
M
l
Isolation of the active compounds was achieved by flash chro-
matography and subsequent purification by semi-preparative
HPLC. Compounds were tracked with the aid of TLC and HPLC–
ESIMS. The three structurally related stilbenoids coelonin (1),
batatasin III (2), and pholidotol D (3) (Fig. 2) were identified by
ESI-TOF-MS, 1D and 2D microprobe NMR, and comparison with
A
receptor modulatory activity, with an efficiency (maximal stimula-
tion of IGABA) of 1512.9 ± 176.5% and a potency (higher concentra-
tion for half-maximal stimulation of
52.5 ± 17.0 M. The structurally related stilbene pholitodol D (3)
showed much lower activity, with an efficiency of 786.8 ± 72.1%
and potency of 175.5 ± 25.5 M. The dihydrophenanthrene coelo-
I
GABA
,
or EC50
)
of
l
1
4,29–31
published data.
roborated by proton NMR, using the chemical shifts and coupling
constant of the two olefinic protons at d 6.95 (2H, d, J = 6.0 Hz,
H-b and ), which discards the presence of the trans-stereoisomer
The Z configuration in compound 3 was cor-
l
H
nin (1) showed activity similar to compound 2, but no saturation
of the receptors was reached at the highest concentration tested
a
thunalbene. Detailed spectroscopic data of compounds 1–3 are
available as Supporting information.
(300
the receptors when applied prior to GABA, at concentrations lower
than 100 M. This was indicative of an allosteric modulation of the
lM). None of the compounds induced direct activation of
Stilbenoids are the major secondary metabolites in the genus
l
1
0
Pholidota, and the identification of compounds 1–3 in the ac-
tive fractions of P. chinensis DCM extract was not surprising.
The three compounds have been previously isolated from the
receptor with the subunit composition
1 2
a b c2s, rather than direct
agonistic activity (Fig. 3C).
Compared to other natural products tested in the same in vitro
1
0,12,14
24,27,28,33
species,
but they have not been reported as GABA
A
recep-
model and GABA
A
receptor subtype,
batatasin III (2) exhib-
Rs of the
tor modulators.
ited much higher efficiency. The efficiency of 2 in GABA
A

1
280
D. C. Rueda et al. / Bioorg. Med. Chem. 22 (2014) 1276–1284
Figure 2. Chemical structures of compounds 1–3.
Figure 3. (A) Potentiation of IGABA by the DCM extract of P. chinensis stems and roots (100
response curve for compounds 1–3 on GABA receptors of the subunit composition 2S. (C–E) Typical traces for modulation of IGABA by compounds 1–3, respectively. The
flat segments in the currents indicate the absence of direct activation of the receptors. All experiments in A–E were carried out using a GABA EC3–10
lg/mL), by microfraction 9, and compounds 1–3 (100 lM). (B) Concentration–
A
1 2
a b c
.
subtype
a
1
b
2
c2s was also significantly higher than that of classical
flexibility as in batatasin III (2) appeared to be critical for the mod-
BDZs, with a potentiation of IGABA at least fourfold that of triazolam,
clotiazepam, and midazolam.34 However, its EC50 value was signif-
icantly higher than that of BDZs and indicated a much lower bind-
ing affinity.
Despite the small number of compounds, preliminary struc-
ture–activity considerations could be derived. Conformational
ulatory activity of stilbenoids, since introduction of a double bond
b,a
D
in pholidotol D (3) drastically decreased potency and effi-
ciency. The importance of flexibility was confirmed by the weak
activity of coelonin (1) in which the dihydrophenanthrene ring
conferred additional rigidity to the structure. Although stilbenoids
such as resveratrol have been described as neuroprotective

D. C. Rueda et al. / Bioorg. Med. Chem. 22 (2014) 1276–1284
1281
agents,17,35–38 none of them has been reported as GABA
ligand so far. Batatasin III (2) is thus the first representative of a
new scaffold for GABA receptor modulators. It is noteworthy that
compounds with biosynthetically related scaffolds such as flavo-
receptor
lower potency on
a
b
c
2s receptors compared to
a
b
c
2s was sta-
A
2
2
4
2
tistically significant, while there were no significant differences
in efficiency among the other -containing receptor subtypes. On
GABA receptors comprising different b subunits, almost no differ-
ences in potency and efficiency were observed. Thus, batatasin III
A
a
A
2
5,39
24
40
noids,
coumarins, and lignans have been previously shown
receptor modulatory properties.
to possess GABA
A
A
(2) was a positive allosteric modulator of GABA Rs, devoid of sig-
nificant subtype specificity.
3
A
.3. GABA receptor subtype selectivity
3
A
.4. GABA R modulatory activity of dihydrostilbenes
Batatasin III (2) was tested for potential
replacing the subunit in the receptor subtype
, and . Likewise, b subunit specificity was evaluated by
replacing b with b and b3. Concentration-dependent IGABA modu-
lation of compound 2 was evaluated on receptor subtypes
2s, and 2s (Table 1).
As shown in Figure 4 and summarized in Table 1, compound 2
a
subunit specificity, by
a
a
5
1
a
b
1 2
c2s with
a
2
,
Flexibility appeared to be a critical factor for the GABA
A
R mod-
a
3
,
a
4
ulatory activity of stilbenoids. To confirm the influence of the dou-
b,a
2
1
ble bond D , 13 commercially available stilbenoids and their
a
2
2
b c
2s
,
corresponding dihydro derivatives (compounds 4–25; Fig. 5) were
tested in the Xenopus oocyte assay. Compounds were initially
a
3
b
2
c
2s
,
a
4
b
2
c
2s
,
a
5
b
2
c
2s
,
a
1
b
1
c
1 3
a b c
tested at a concentration of 100
A
lM on GABA Rs of the subtype
did not exhibit subtype specificity, as reflected by comparable EC50
1 2
a b c2s. As expected, dihydrostilbenes showed higher activity than
values with all receptor subtypes studied (p > 0.05). The order of
the corresponding stilbenes (Table 2, Fig. 6A). These differences in
the activity of stilbenes and their dihydro derivatives were statisti-
cally significant in almost every case, with the exception of the
pairs 4 and 5/6, 9/10, 13/14, and 23/24 (p > 0.05).
potency of batatasin III (2) in receptor composed by different
a
subunits was
a
4
2
b c
2s
>
a
b
5 2
c
2s
>
1
a b
2
c
2s
>
3 2 2s 2 2
a b c > a b c2s. The
Among the stilbenes, tetramethoxy-piceatannol (11), resvera-
trol (13), pterostilbene (19), and resveratrol triacetate (21), dis-
played the highest activity, potentiating IGABA between 100% and
Table 1
Potencies and efficiencies of batatasin III (2) for GABA
compositions
A
receptors of different subunit
2
00%. Their corresponding dihydro derivatives showed the highest
Subtype
EC50
(
l
M)
Max. potentiation of IGABA
EC3–10) (Imax) (%)
Hill coeff. (n
H
)
na
activity among dihydrostilbenes, but only compounds 12 and 20
showed efficiencies comparable to that of batatasin III (2)
(544.5 ± 104.4% and 660.6 ± 100.2% respectively). A comparison
(
a
1
a
2
a
3
a
4
a
5
a
1
a
1
b
2
b
2
b
2
b
2
b
2
b
1
b
3
c
c
c
c
c
c
c
2s
2s
2s
2s
2s
2s
2s
52.5 ± 17.0
80.8 ± 22.1
67.3 ± 18.6
26.2 ± 3.6
46.7 ± 9.0
66.7 ± 21.0
67.2 ± 10.5
1512.9 ± 176.5
1026.5 ± 139.2
1694.2 ± 229.0
1588.2 ± 97.5
1375.7 ± 76.5
1251.3 ± 157.0
1252.9 ± 79.9
1.4 ± 0.3
1.2 ± 0.1
1.2 ± 0.1
1.5 ± 0.1
1.3 ± 0.1
1.8 ± 0.4
1.4 ± 0.1
5
6
5
6
5
5
5
of the activity of the dihydrostilbenes at 100
the bibenzyl scaffold alone (6) does not possess any GABA
l
M revealed that
R mod-
A
ulatory activity. In general, substituents at C-3 and C-5 (12, 14, 16,
18, 20, and 22) resulted in an enhancement of the activity. Increas-
ing the lipophilicity by replacing the hydroxy groups at C-3 and C-5
with bulkier oxygenated functions (12, 20, and 22) enhanced the
a
Number of experiments.
Figure 4. (A)
a
-Subunit dependency of batatasin III (2), depicted as concentration–response curves, with GABA
2s, and 2s. (B) b-Subunit dependency of batatasin III (2), depicted as concentration–response curves with GABA
2s, and 2s. (C and D) Typical traces for modulation of IGABA by compound 2, in receptors with different and b subunit composition, respectively. All
A
receptors of the subunit compositions
1 2 2s 2 2 2s
a b c , a b c ,
a
a
3
b
b
2
c
c
2s
,
,
a
a
4
b
2
c
a
a
5
b
2
c
A
receptors of the subunit compositions
1
1
2s
1
b
2
c
1
b
3
c
a
experiments were performed using a GABA EC3–10
.

1
282
D. C. Rueda et al. / Bioorg. Med. Chem. 22 (2014) 1276–1284
Figure 5. Chemical structures of compounds 4–25.
activity of dihydrostilbenes. The role of substituents in ring B was
Table 2
Potentiation of IGABA in
of 100 lM
less clear within this compounds series. In the case of compounds
12 and 20, different substitution patterns in ring B did not influ-
ence the activity. In contrast, when comparing compounds 14,
a
1
b
2
c
2s receptors by compounds 4–25, at a test concentration
Stilbenes
Dihydrostilbenes
0
0
1
6, and 18, addition of a hydroxy group in C-4 or C-6 led to a sig-
nificant decrease of activity. Introduction of a halogen atom as in
5 induced slight negative receptor modulation, and substitution
at C-4 (compound 10) decreased activity. Since we had only one
pair of cis and trans isomers (4 and 5, both inactive at 100 M),
Compound
Max. potentiation na Compound Max. potentiation na
of IGABA
of IGABA
2
4
5
7
9
12.6 ± 9.2
ꢁ11.3 ± 12.3
ꢁ7.3 ± 0.1
3
3
3
3
3
3
3
3
3
3
3
3
6
8.3 ± 22.9
3
l
8
51.6 ± 1.1
ꢁ20.9 ± 3.7
544.5 ± 140.4
162.2 ± 17.5
86.3 ± 9.9
44.1 ± 19.7
660.6 ± 100.2
227.7 ± 1.3
ꢁ16.8 ± 7.9
ꢁ12.9 ± 0.4
3
3
3
3
3
3
3
2
3
3
the role of geometric isomerism could not be assessed in more
detail.
The dihydro derivatives of tetramethoxy-piceatannol and ptero-
stilbene (compounds 12 and 20, respectively) were submitted to
ꢁ24.8 ± 4.8
101.3 ± 0.9
121.9 ± 21.8
ꢁ35.4 ± 9.8
ꢁ19.7 ± 3.9
212.4 ± 10.9
122.8 ± 18.6
ꢁ22.9 ± 7.5
231.3 ± 22.6
10
12
14
16
18
20
22
24
25
11
13
15
17
19
21
23
further concentration–response experiments on
1 2
a b c2s receptors.
Both compounds enhanced IGABA at a GABA EC3–10 in a concentra-
tion-dependent manner (Fig. 6B). Compounds 12 and 20 had lower
efficiency than the natural dihydrostilbene 2 (Table 3), with max-
imal stimulations of IGABA of 870.7 ± 106.8% and 694.2 ± 86.0%,
respectively. In terms of potency, 20 was comparable to 2 (EC50
b
Diazepam
M)
(1 l
a
Number of experiments.
Positive control.
b
5
4.5 ± 13.4 lM), whereas 12 was twice as potent (EC50 20.2 ±

D. C. Rueda et al. / Bioorg. Med. Chem. 22 (2014) 1276–1284
1283
Figure 6. (A) Potentiation of IGABA by compounds 4–25 (100
receptors of the subunit composition 2S. (C and D) Typical traces for modulation of IGABA by compounds 12 and 20, respectively. The inward currents induced in the
absence of GABA (C and D) indicate direct activation of the receptors. All experiments shown in A–D were performed using a GABA EC3–10
lM). (B) Potentiation of IGABA by compounds 2, 12 and 20. Concentration–response curves are shown for GABA
A
1 2
a b c
.
Table 3
products tested up to now, but its EC50 value was significantly
higher than that of BDZs. Dihydrostilbenes represent a new scaf-
Potencies and efficiencies of compounds 2, 3, 12 and 20 for
1
a b
2 A
c2s GABA receptors
Hill coeff. na
(n )
H
fold for GABA
The conformational flexibility of dihydrostilbenoids appeared
critical for GABA R modulatory properties. For a further explora-
tion of this scaffold, conformationally restricted derivatives should
be synthesized in order to explore in more detail the optimal ori-
entation of the aromatic rings and substituents.
A
receptor modulators.
Compound EC50
(
l
M)
Max. potentiation of IGABA
EC3–10) (Imax) (%)
(
2
3
1
2
52.5 ± 17.0 1512.9 ± 176.5
175.5 ± 25.5
20.2 ± 6.4
1.2 ± 0.1
1.5 ± 0.2
2.3 ± 1.0
1.6 ± 0.2
5
5
4
4
A
786.8 ± 72.1
870.7 ± 106.8
694.2 ± 86.0
2
0
54.5 ± 13.4
a
Number of experiments.
Acknowledgements
6
.4
l
M). This suggests that increased lipophilicity of ring B may
Financial support was provided by the Swiss National Science
Foundation through project 205320_126888 (M.H.). D.C.R. thanks
the Swiss Federal Commission for Scholarships for Foreign Stu-
dents (FCS) and the Department of Education of Canton Basel
have a positive effect on the potency of dihydrostilbenes. However,
further studies with a larger series of compounds are needed for
confirmation. None of the compounds induced direct activation
of the receptors when applied prior to GABA, at concentrations
(
Erziehungsdepartement des Kantons Basel-Stadt) for fellowships
lower than 100
l
M (Fig. 6C).
granted in 2011 and 2012, respectively.
Stilbenoids have attracted significant attention in recent years
due to their wide range of useful properties, including applications
Supplementary data
in optics, biochemistry, and chemotherapy.1
3,41
The stilbenoid
scaffold can be considered as a priviledged structure.4
2,43
However,
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmc.2014.01.008.
there have been no reports on GABA receptor modulatory activity
A
of stilbenoids up to now, despite a significant number of publica-
tions on biological activities of natural stilbenoids, and in particu-
lar, on resveratrol. Dihydrostilbenes such as 2 may thus be an
References and notes
interesting starting point for the synthesis of new GABA
modulators.
A
receptor
1. Olsen, R. W.; Sieghart, W. Pharmacol. Rev. 2008, 60, 243.
2
3
4
.
.
.
Rudolph, U.; Knoflach, F. Nat. Rev. Drug Disc. 2011, 10, 685.
D’Hulst, C.; Atack, J. R.; Kooy, R. F. Drug Discovery Today 2009, 14, 866.
Tan, K. R.; Rudolph, U.; Lüscher, C. Trends Neurosci. 2011, 34, 188.
4
. Conclusions
5. Baburin, I.; Beyl, S.; Hering, S. Pflug. Arch. Eur. J. Physiol. 2006, 453, 117.
6
7
.
.
Pérez Gutiérrez, R. M. J. Med. Plants Res. 2010, 4, 592.
Williams, R.; Martin, S.; Hu, J.-F.; Garo, E.; Rice, S.; Norman, V.; Lawrence, J.;
Hough, G.; Goering, M.; O’Neil-Johnson, M.; Eldridge, G.; Starks, C. Planta Med.
2011, 78, 160.
With the aid of an HPLC-based profiling approach, we identified
batatasin III (2) as the major compound responsible for GABA
modulatory activity of the dichloromethane extract of P. chinensis.
This dihydrostilbene showed allosteric modulation in
GABA receptors with a higher efficiency than any other natural
A
R
8
9
.
.
Bulpitt, C. J.; Li, Y.; Bulpitt, P. F.; Wang, J. J. R. Soc. Med. 2007, 100, 558.
Wang, J.; Matsuzaki, K.; Kitanaka, S. Chem. Pharm. Bull. (Tokyo) 2006, 54, 1216.
1 2 2s
a b c
10. Bandi, A. K. R.; Lee, D.-U. Chem. Biodiversity 2011, 8, 1400.
A
11. Wu, B.; Qu, H.; Cheng, Y. Chem. Biodiversity 1803, 2008, 5.

1
284
D. C. Rueda et al. / Bioorg. Med. Chem. 22 (2014) 1276–1284
1
1
1
2. Yao, S.; Tang, C.-P.; Li, X.-Q.; Ye, Y. Helv. Chim. Acta 2008, 91, 2122.
3. Rivière, C.; Pawlus, A. D.; Mérillon, J.-M. Nat. Prod. Rep. 2012, 29, 1317.
4. Majumder, P. L.; Roychowdhury, M.; Chakraborty, S. Phytochemistry 1998, 49,
28. Zaugg, J.; Baburin, I.; Strommer, B.; Kim, H. J.; Hering, S.; Hamburger, M. J. Nat.
Prod. 2010, 73, 185.
29. Wang, J.; Wang, L.; Kitanaka, S. J. Nat. Med. 2007, 61, 381.
30. Majumder, P.; Laha, S.; Datta, N. Phytochemistry 1982, 21, 478.
31. Hashimoto, T.; Hasegawa, K.; Yamaguchi, H.; Saito, M.; Ishimoto, S.
Phytochemistry 1974, 13, 2849.
32. Oh, K. B.; Kim, S. H.; Lee, J.; Cho, W. J.; Lee, T.; Kim, S. J. Med. Chem. 2004, 47,
2418.
33. Zaugg, J.; Eickmeier, E.; Ebrahimi, S. N.; Baburin, I.; Hering, S.; Hamburger, M. J.
Nat. Prod. 2011, 74, 1437.
34. Khom, S.; Baburin, I.; Timin, E. N.; Hohaus, A.; Sieghart, W.; Hering, S. Mol.
Pharmacol. 2006, 69, 640.
35. Richard, T.; Pawlus, A. D.; Iglésias, M.-L.; Pedrot, E.; Waffo-Teguo, P.; Mérillon,
J.-M.; Monti, J.-P. Ann. N.Y. Acad. Sci. 2011, 1215, 103.
36. Vingtdeux, V.; Dreses-Werringloer, U.; Zhao, H.; Davies, P.; Marambaud, P. BMC
Neurosci. 2008, 9, S6.
2
375.
1
1
1
1
1
5. Kovacs, A.; Vasas, A.; Hohmann, J. Phytochemistry 2008, 69, 1084.
6. Shen, T.; Wang, X.-N.; Lou, H.-X. Nat. Prod. Rep. 2009, 26, 916.
7. Greger, H. J. Nat. Prod. 2012, 75, 2261.
8. Potterat, O.; Hamburger, M. Curr. Org. Chem. 2006, 10, 899.
9. Danz, H.; Stoyanova, S.; Wippich, P.; Brattstrom, A.; Hamburger, M. Planta Med.
2
001, 67, 411.
0. Dittmann, K.; Gerhauser, C.; Klimo, K.; Hamburger, M. Planta Med. 2004, 70,
09.
2
2
2
2
9
1. Adams, M.; Christen, M.; Plitzko, I.; Zimmermann, S.; Brun, R.; Kaiser, M.;
Hamburger, M. J. Nat. Prod. 2010, 73, 897.
2. Adams, M.; Zimmermann, S.; Kaiser, M.; Brun, R.; Hamburger, M. Nat. Prod.
Commun. 2009, 4, 1377.
3. Kim, H. J.; Baburin, I.; Khom, S.; Hering, S.; Hamburger, M. Planta Med. 2008, 74,
37. Sun, A. Y.; Wang, Q.; Simonyi, A.; Sun, G. Y. Mol. Neurobiol. 2010, 41, 375.
38. Zhang, F.; Liu, J.; Shi, J. S. Eur. J. Pharmacol. 2010, 636, 1.
39. Hanrahan, J. R.; Chebib, M.; Johnston, G. A. Br. J. Pharmacol. 2011, 163, 234.
40. Zaugg, J.; Ebrahimi, S. N.; Smiesko, M.; Baburin, I.; Hering, S.; Hamburger, M.
Phytochemistry 2011, 72, 2385.
5
21.
2
2
4. Li, Y.; Plitzko, I.; Zaugg, J.; Hering, S.; Hamburger, M. J. Nat. Prod. 2010, 73, 768.
5. Yang, X.; Baburin, I.; Plitzko, I.; Hering, S.; Hamburger, M. Mol. Diversity 2011,
15, 361.
2
2
6. Zaugg, J.; Khom, S.; Eigenmann, D.; Baburin, I.; Hamburger, M.; Hering, S. J. Nat.
Prod. 2011, 74, 1764.
7. Zaugg, J.; Eickmeier, E.; Rueda, D. C.; Hering, S.; Hamburger, M. Fitoterapia
41. Gray, E. E.; Rabenold, L. E.; Goess, B. C. Tetrahedron Lett. 2011, 52, 6177.
42. Welsch, M. E.; Snyder, S. A.; Stockwell, B. R. Curr. Opin. Chem. Biol. 2010, 14, 1.
43. Breinbauer, R.; Vetter, I. R.; Waldmann, H. Angew. Chem., Int. Ed. 2002, 41, 2878.
2
011, 82, 434.