480 Biol. Pharm. Bull. 31(3) 480—486 (2008) Vol. 31, No.

Modifications of Aliphatic Side Chain of 20(S)-Ginsenoside Rg3 Cause an

Enhancement or Loss of Brain Na
Channel Current Inhibitions
Jun-Ho LEE,a,# Sun-Hye CHOI,a,# Byung-Hwan LEE,a In-Soo YOON,a Tae-Jun SHIN,a Mi Kyung PYO,a
Sang-Mok LEE,a Hyewhon RHIM,b Myung Hwan PARK,c Tae Yoon PARK,c and Seung-Yeol NAH*,a
a
Ginsentology Research Laboratory and Department of Physiology, College of Veterinary Medicine, Konkuk University;
b
Institute of Biomedical Science and Technology, Konkuk University; Seoul 143–701, Korea: and c Ambo Institute; Seoul
135–535, Korea. Received November 6, 2007; accepted December 14, 2007

A line of evidence has shown that ginsenoside Rg3 (Rg3) could be one of bioactive ligands in brain Na
chan-
nel regulations. Rg3 exists as stereoisomer of 20(R)- or 20(S)-form. Rg3 consists of three different parts; steroid-
like backbone structure, carbohydrate portion, and aliphatic side chain [–CH2CH2CHC(CH3)2], which is cou-
pled to the carbon-20 of backbone structure. In the previous report, we demonstrated that 20(S)- but not 20(R)-
Rg3 and carbohydrate portion of Rg3 play important roles in rat brain NaV1.2 channel regulations. However, lit-
tle is known about the role of aliphatic side chain coupled to the carbon-20 in brain Na
channel regulations. In
the present study, we prepared Rg3 derivatives by modifying the aliphatic side chain of Rg3, remaining with
backbone structure and carbohydrate portion intact, and examined the effects of Rg3 derivatives on Na
channel
activity. We found that reduction of double bond in aliphatic side chain of Rg3 exhibited agonistic actions in Na

channel current inhibitions by shifting concentration–response curve to leftward by three-fold, whereas deletion,
hydroxylation, or oxygenation of aliphatic side chain caused an attenuation or loss of Na
channel current inhibi-
tions. These results provide evidences that the aliphatic side chain of Rg3 is also involved in Na
channel regula-
tions and further show a possibility that the aliphatic side chain of Rg3 could be the target of chemical modifica-
tions for abolishment or potentiation of Rg3 actions in Na
channel regulations.
Key words Panax ginseng; 20(S)-ginsenoside Rg3; aliphatic side chain; NaV1.2 channel

Na
channels are transmembrane proteins that consist of a 20(S)-Rg3-mediated Na
channel regulation using a Xenopus
pore-forming a subunit and auxiliary b 1, b 2 and b 3 sub- oocyte gene expression system. This gene expression system
units.1) The a subunit is composed of four homologous do- has few endogenous ion channels,6) and allows heterologous
mains (I—IV), each composed of six a -helical transmem- expression of ion channels for various biochemical stud-
brane segments (S1—S6), and is responsible for voltage-de- ies.4,7,8) We expressed rat brain Na
channels by intraoocyte
pendent increases in Na
-selective permeability. The inward
Na
current (INa) initiates axonal and somatic action poten-
tials in nerve and muscle fibers, and may also be involved in
axonal intraneuronal or interneuronal information transfer.2)
Na
channels are one of the targets of cardiac- and neuropro-
tective treatments against pathologic conditions including ar-
rhythmia and brain ischemia.
Ginseng, the root of Panax ginseng C. A. MEYER, is well
known in herbal medicine as a tonic and restorative agent.
The main molecular ingredients responsible for the actions
of ginseng are the ginsenosides (also called ginseng
saponins), which are also amphiphilic molecules.3) Ginseno-
sides consist of three major different parts; steroid-like back-
bone structure, carbohydrate portions, and aliphatic side
chain [–CH2CH2CHC(CH3)2], which is coupled to the car-
bon-20 of backbone structure. Some subsets of ginsenosides
also exist as stereoisomer of 20(R)- or 20(S)-form. We re-
cently demonstrated that ginsenoside Rg3 (20-S-protopanaxa-
diol-3-[O-b -D-glucopyranosyl (1→2)-b -glucopyranoside])
20(S)-(Rg3) but not 20(R)-(Rg3) inhibits voltage-dependent
brain Na
channel activity expressed in Xenopus laevis
oocytes.4,5) In the previous report, we also demonstrated that
carbohydrate portion of 20(S)-Rg3 play important roles in rat
brain NaV1.2 channel regulations. However, little is known
about the role of aliphatic side chain coupled to the carbon-
20 in 20(S)-Rg3-induced Na
channel current inhibitions Fig. 1. Chemical Structures of Rg3 and Rg3 Derivatives
Glc, glucopyranoside. Ginsenosides differ in the three side chains attached to a com-
(Fig. 1). mon steroid-like ring. Subscripts indicate the carbons in the glucose rings that link the
We herein identified the role of aliphatic side chain in two carbohydrates.

∗ To whom correspondence should be addressed. e-mail: synah@konkuk.ac.kr © 2008 Pharmaceutical Society of Japan
#
These authors contributed equally to this work.
March 2008
injection of cRNAs encoding the NaV1.2 a and b 1 sub-
units,3,4,9) and examined the changes in INa in response to
20(S )-Rg3 and 20(S)-Rg3 derivatives, which are chemically
modified in the aliphatic side chain. We found that reduction
of double bond in aliphatic side of Rg3 induced more signifi-
cant tonic and use-dependent inhibitions of the peak INa fol-
lowing low- and high-frequency stimulations than Rg3,
whereas deletion, hydroxylation, or oxygenation of the
aliphatic side chain caused an attenuation or loss of Na

channel current inhibitions. These results provide evidences
that the aliphatic side chain of Rg3 is involved in Na
chan-
nel regulation and that the enhancement or loss on Na
chan-
nel current inhibitions by Rg3 depends on chemical struc-
tures of the aliphatic side chain of Rg3.
MATERIALS AND METHODS
Materials 20(S)-Rg3 and Rg3 derivatives such as 2H-
Rg3, 25OH-Rg3, Rg3-C6, protopanaxadiol (PPD), 2H-PPD,
and Rg3-oxide were kindly provided by Ambo Institute
(Seoul, Korea) (Fig. 1). The cDNA for the rat brain Na

channel NaV1.2 a subunit was kindly provided by Dr. A. L.
Goldin (University of California, Irvine, CA, U.S.A.) and
that for the Na
channel b 1 subunit was kindly provided by
Dr. T. Zimmer (Friedrich Schiller University, Jena, Ger-
many). Other agents were purchased from Sigma-Aldrich
(St. Louis, MO, U.S.A.).
Preparation of Xenopus Oocytes and Microinjection
X. laevis frogs were purchased from Xenopus I (Ann Arbor,
MI, U.S.A.). Their care and handling was in accordance with
the highest standards of institutional guidelines. For isolation
of oocytes, frogs were anesthetized with an aerated solution
of 3-amino benzoic acid ethyl ester and ovarian follicles were
removed. The oocytes were separated by treatment with col-
lagenase and agitation for 2 h in Ca2
-free medium contain-
ing 82.5 mM NaCl, 2 mM KCl, 1 mM MgCl2, 5 mM HEPES,
2.5 mM sodium pyruvate, 100 units/ml penicillin and 100
m g/ml streptomycin. Stage V—VI oocytes were collected
and stored in ND96 (96 mM NaCl, 2 mM KCl, 1 mM MgCl2,
1.8 mM CaCl2, and 5 mM HEPES, pH 7.5) supplemented with
0.5 mM theophylline and 50 m g/ml gentamicin. This oocyte-
containing solution was maintained at 18 °C with continuous
gentle shaking and renewed everyday. Electrophysiological
experiments were performed within 5—6 d of oocyte isola-
tion, with chemicals applied to the bath. For Na
channel ex-
periments, 40 nl of cRNAs encoding the NaV1.2 a and b 1
subunits was injected into the animal or vegetal pole of each
oocyte 1 d after isolation, using a 10 m l VWR microdispenser
(VWR Scientific, San Francisco, CA, U.S.A.) fitted with a ta-
pered glass pipette tip (15—20 m m in diameter).4)
Site-Directed Mutagenesis of NaV1.2, and in Vitro
Transcription of Na
Channel Subunit cDNAs The sub-
stitutions mutations of single or triplet amino acids were per-
formed using the QuikChangeTM XL Site-Directed Mutagen-
esis Kit (Stratagene, La jolla, CA, U.S.A.), along with Pfu
DNA polymerase and mutated sense and antisense primers.
Overlap extension of the target domain by sequential poly-
merase chain reaction (PCR) was carried out according to the
manufacturer’s recommended protocol. The final PCR prod-
ucts were transformed to Escherichia coli strain DH5a ,
screened by PCR and confirmed by DNA sequencing of the
481
target region. The mutant DNA constructs were linearized at
the 3 end by NotI digestion, and run-off transcripts were
prepared using methylated cap analog, m7G(5)ppp(5)G.
The cRNAs were prepared using the mMessage mMachine
transcription kit (Ambion, Austin, TX, U.S.A.) with T7 RNA
polymerase. The absence of degraded RNA was determined
by denaturing agarose gel electrophoresis followed by ethi-
dium bromide staining. Similarly, recombinant plasmids con-
taining wild-type NaV1.2 a or b 1 subunit cDNA inserts were
linearized by digestion with the appropriate restriction en-
zymes, and cRNAs were obtained using the mMessage
mMachine in vitro transcription kit (Ambion, Austin, TX,
U.S.A.) with SP6 RNA or T7 polymerases. The final cRNA
products were resuspended at a concentration of 1 m g/m l in
RNase-free water, and stored at 80 °C until use.7)
Data Recording A custom-made Plexiglas net chamber
was used for two-electrode voltage-clamp recordings. The
chamber was constructed by milling two concentric wells
into the chamber bottom (the diameter/height of the upper
well was 8/3 mm and that of the lower well was 6/5 mm) and
gluing plastic meshes (ca. 0.4 mm grid diameter) onto the
bottom of the upper well. A perfusion inlet (ca. 1 mm in di-
ameter) was drilled through the wall of the lower well, and a
suction tube was placed on the edge of the upper well. For
experiments, a single oocyte was placed on the net separating
the upper and lower wells. The net grids helped anchor the
oocyte in place during the electrophysiological recordings.
The oocyte was then impaled with two microelectrodes filled
with 3 M KCl (0.2—0.7 MW) and electrophysiological exper-
iments were carried out at room temperature using an Oocyte
Clamp (OC-725C, Warner Instruments, CT, U.S.A.). Stimu-
lation and data acquisition were controlled with a pClamp 8
(Axon Instruments, Union City, CA, U.S.A.).8) For most of
the electrophysiological experiments on Na
channel activ-
ity, oocytes were clamped at a holding potential of 100 mV
and the membrane potential was depolarized to 10 mV for
100 ms every 5 s. Linear leak currents were corrected by
means of the leak subtraction procedure.4)
The voltage-dependence of Na
channel activation was
calculated by measuring the peak current at test potentials
ranging from 50 to
50 mV evoked in 5 mV increments.
The conductance (gNa) was calculated according to the equa-
tion, gNaINa/(VgVr), where INa is the peak amplitude of the
Na
current, Vg is the test potential, and Vr is the reversal po-
tential for Na
. The conductance–voltage curves were drawn
according to the equation gNa /max gNa1/{1
exp[(Vg0.5
Vg)/kg]}, where max gNa is the maximum value for gNa, Vg0.5 is
the potential at which gNa is 0.5max gNa, and kg is the slope
factor (potential required for an e-fold change). The voltage-
dependence of Na
channel inactivation was determined
using 200 ms conditioning pre-pulses ranging from 60 to

20 mV from a holding potential of 100 mV in 5 mV in-
crements, followed by a test pulse to 10 mV for 5 ms. The
peak INa was normalized to its respective maximum value
(max INa) and plotted as a function of the pre-pulse potential.
The steady-state inactivation curves were drawn according to
the equation, INa/maxINa1/{1
exp[(VhVh0.5)/kh]}, where Vh
is pre-pulse potential, Vh0.5 is the potential at which INa is
0.5maxINa, and kh is the slope factor.
The frequency-dependent effect of Rg3 was examined
using a protocol in which 50 depolarizing pulses of 10 ms
482 Vol. 31, No. 3

duration and 10 Hz frequency were applied to 10 mV from oxygenation (Rg3-oxide) of aliphatic side chain exhibited a
a holding potential of 100 mV. The protocol was run in the slight or had no effect on the peak INa (Fig. 2A). Thus, the
absence (control) and presence of 2H-Rg3 (10 or 100 m M). percentage inhibition of NaV1.2 channel currents by Rg3, 2H-
The current amplitude of each pulse was normalized to the Rg3, 25OH-Rg3, Rg3-oxide, and Rg3-C6 (100 m M each) were
peak maximal current (pulse number 1) and plotted as a 67.47.1, 79.86.1, 7.53.2, 3.12.1, and 2.81.5%, re-
function of pulse number. spectively, indicating that the aliphatic side chain of Rg3
The kinetics of the Rg3 blockade of IFMQ3, which is the might play important role in Rg3-induced Na
channel regu-
fast inactivation deficient mutant of NaV1.2, was examined lations. In addition, we also tested whether PPD or 2H-PPD,
by clamping oocytes at 100 mV in ND96 solution. A single which has no carbohydrate portion or is reduced in double
500 ms depolarizing pulse to 10 mV was applied and the bond of the aliphatic side chain without carbohydrate portion
INa was recorded. Different concentrations of 2H-Rg3 (10, 30, of Rg3. Interestingly, we also found that PPD and 2H-PPD
100 or 300 m M) were perfused into the bath for 1 min, and a had no effect on peak INa (Table 1). Moreover, Rh2, which
second, single depolarizing pulse from 100 to 10 mV was has only one carbohydrate at the carbon-3, exhibited much
given. The data were individually fit to either a single
[1Aslow · exp(t/t slow)] or double [1Aslow · exp(t/t fast)]

[1Aslow · exp(t/t slow)] exponential equation, in which Afast
and Aslow represent the proportion of current decaying with
time constants t fast and t slow, respectively, and t is the time in-
terval. If we assume that a first order relationship describes
the dependence of the blocking rate on the concentration of
the blocking chemical, the apparent rate constant for binding
(kon) can be obtained by fitting the t fast values with the equa-
tion: 1/t fastkon · [2H-Rg3]
l.10) Recovery from open channel
blockade was measured at a holding potential of 100 mV
with a 10 ms depolarizing pre-pulse to 10 mV (P1) fol-
lowed by a variable recovery period from 10 to 500 ms, sub-
sequently followed by a 10 ms test pulse to 10 mV (P2).
Data Analysis To obtain the concentration–response
curve of the effect of Rg3 and Rg3 derivatives on INa, the peak
amplitudes at different concentrations of Rg3 and Rg3 deriva-
tives were plotted, and then fitted to the following Hill equa-
tion using the Origin software (Origin, Northampton, MA,
U.S.A.): y/ymax[A]nH/([A]nH
[EC50]n), where y is the peak
INa at given concentration of Rg3 and Rg3 derivatives, ymax is
the maximal peak INa, EC50 is the concentration of Rg3 and
Rg3 derivatives producing a half-maximum effect, [A] is the
concentration of Rg3 and Rg3 derivatives, and nH is the inter-
action coefficient. All values are presented as meansS.E.M.
The differences between the means of control and treatment
values were determined using an unpaired Student’s t-test. A
value of p0.05 was considered statistically significant. Fig. 2. Tonic Inhibition of INa by Rg3 and Rg3 Derivatives
(A) Representative traces of NaV1.2 in the presence or absence of Rg3 and Rg3 deriv-
atives. Oocytes were injected with wild-type NaV1.2 a and b 1 subunit cRNAs and held
RESULTS for 2—4 d; INa was then recorded in ND96 using two-electrode voltage clamps as de-
scribed in Materials and Methods. (B) Concentration–response curves for the effect of
Rg3 and Rg3 derivatives on NaV1.2. Data are meansS.E.M. (n10—13/group). (C)
Effects of Rg3 and Rg3 Derivatives, Which Are Modi- Representative current–voltage relationship obtained using voltage steps between 50
fied in Aliphatic Side Chain, on Peak INa in Oocytes Ex- and
50 mV with 5-mV increments. Voltage steps were applied in the presence and ab-
pressing NaV1.2 We examined effects of Rg3 and Rg3 de- sence of Rg3 and Rg3 derivatives (100 m M).

rivatives, which are modified in aliphatic side chain (Fig. 1),

on peak INa in oocytes expressing NaV1.2. We recorded INa by Table 1. Effects of Rg3 and Rg3 Derivatives on NaV1.2 Expressed in Xeno-
two-electrode voltage clamping of oocytes injected with pus laevis Oocytes
cRNAs encoding NaV1.2 a and b 1 subunits. Oocytes were Vmax IC50 nH
held at 100 mV, and INa was elicited by depolarization to
10 mV at a low frequency (0.2 Hz). This procedure mini- Rg3 82.13.4 28.63.4 1.10.1
mized the use-dependent blockade and allowed evaluation of Rh2 39.31.0 52.43.1 1.30.1
whether Rg3 or Rg3 derivatives produced a tonic blockade of PPD 13.13.4* ND ND
2H-PPD 4.50.2* ND ND
peak INa.3,4,11,12) As shown in Fig. 2A, the depolarizing volt- 2H-Rg3 88.95.4 10.32.3* 1.00.1
age step induced a large inward INa with rapid inactivation. 25OH-Rg3 14.11.3* 82.825.1* 1.20.1
The representative traces showed that 2H-Rg3, which is re- Rg3-oxide 9.13.3* ND ND
duced in double bond of aliphatic side chain, caused more Rg3-C6 8.16.9* ND ND
significant inhibitions of peak INa than Rg3, whereas deletion Currents were elicited by single-step voltage pulses from 100 to 10 mV. ∗ p
(Rg3-C6), hydroxylation of dimethyl group (25OH-Rg3), 0.01 compared with control.
March 2008 483

less effects on the peak INa compared with Rg3 (Table 1). The and the slope factor (kh) between control and modified Rg3
Rg3, 2H-Rg3, and Rg3-oxide effects on peak INa was concen- treatment groups (Table 2). These findings indicate that Rg3
tration-dependent (Fig. 2B) and the IC50 value and Hill coef- and 2H-Rg3 affect the steady-state activation but not inactiva-
ficient for Na
channels are given in Table 1. 2H-Rg3 caused tion of the Na
channel.
an approximately three-fold decrease in IC50 value from Use-Dependent Blockade of NaV1.2 by Rg3 and Rg3 De-
28.63.4 to 10.32.3 m M, whereas Rh2 caused an approxi- rivatives Since Na
channel blockers such as lidocaine and
mately two-fold increase in IC50 value from 28.63.4 to other antiarrhythmic drugs exhibit use-dependent inhibi-
52.43.1 m M. These results indicate that reduction of the tion13) and in previous report we showed that Rg3 also exhib-
aliphatic side chain of Rg3 exhibits inhibitory effects without ited use-dependent inhibition,4) we tested whether 2H-Rg3
hydroxylation of the carbon-25. In addition to the aliphatic and other Rg3 derivatives behaved in the same way. INa was
side chain, two carbohydrates also play a crucial role in Rg3- elicited by 20-ms pulses from 100 to 10 mV for 50 times
induced Na
channel regulations as previously reported.9) at 10 Hz. Each peak INa was normalized to the first pulse
The current–voltage relationships were assessed in the ab- peak INa. Under control conditions, there was a slight reduc-
sence or presence of Rg3, 2H-Rg3, 25OH-Rg3, Rg3-oxide, tion in peak INa, while treatment with 2H-Rg3 but not other
and Rg3-C6 with voltage steps ranging from 50 to
50 mV Rg3 derivatives induced further use-dependent inhibitions of
evoked from a holding potential of 100 mV every 5 s. As peak INa compared with Rg3 (Fig. 4). Thus, application of
shown in Fig. 2C, addition of Rg3 and 2H-Rg3 caused a volt- Rg3 and 2H-Rg3 produced the inhibition of peak INa values
age-dependent reduction in peak INa, with a more pro- by 32.56.7 and 47.47.4% (p0.005, compared with
nounced reduction noted at lower voltage ranges, whereas Rg3). These results show that 2H-Rg3 is a better use-depend-
other Rg3 derivatives exhibited only a slight inhibition of ent inhibitor of INa than Rg3.
peak INa in tested voltage steps. Effects of Rg3 and Rg3 Derivatives on a Fast Inactiva-
Effects of Rg3 and Rg3 Derivatives on the Activation tion Gating-Deficient Mutant NaV1.2 In our previous re-
and Inactivation of NaV1.2 Since Rg3 treatment was port, we demonstrated that Rg3 also inhibits resting and
shown to induce a depolarizing shift in the activation voltage open-state Na
currents using a fast inactivation gating-defi-
but not inactivation voltage in rat brain NaV1.2 channels,4) we cient mutant channel (IFMQ3) by replacing residues IFM
next examined the effects of Rg3 and Rg3 derivatives on the with three glutamines.4) As previously reported, IFMQ3 mu-
voltage-dependence of Na
channel steady-state activation tant channels exhibited two components of inward Na
cur-
and inactivation. First, the effects of Rg3, 2H-Rg3, Rg3-C6, rents; one is peak and the other is a steady-state inward cur-
and 25OH-Rg3 on Na
channel activation was determined by rents. In this study, we examined whether 2H-Rg3 exhibits
a conductance transformation of the peak current–voltage re- stronger inhibitions in peak and steady-state Na
currents
lationship (Fig. 3A), with the curves representing the best
data fit using the Boltzmann function. There was a significant Table 2. Effect of Rg3 and Rg3 Derivatives on the Voltage-Dependence of
depolarizing shift of the half-maximal activation voltage Activation and Inactivation of NaV1.2
(Vg0.5) by treatment of 2H-Rg3 and Rg3 but not Rg3-C6, and
25OH-Rg3. We could not observe additional depolarizing Activation Inactivation
shift of the half-maximal activation voltage by 2H-Rg3 com- Type
Vg0.5 kg Vh0.5 kh
pared with Rg3. However, the slope factor (kg) was not sig-
nificantly different (Table 2). We then investigated the effect Con 28.90.16 4.60.51 29.30.41 6.60.36
of Rg3 and Rg3 derivatives on voltage-dependent Na
chan- Rg3-C6 28.50.57 5.10.43 26.90.61 8.10.68
nel inactivation by plotting the normalized peak INa against 25OH-Rg3 29.50.59 5.90.53 27.70.81 8.20.43
the conditioning pre-pulse voltage (Fig. 3B), and then fitting Rg3 17.80.22* 4.90.19 26.90.27 6.40.25
2H-Rg3 16.80.34* 4.80.31 28.40.38 6.80.36
the data to the Boltzmann function. There was no significant
difference in the half-maximal inactivation voltage (Vh0.5) * p0.01 compared with control.

Fig. 3. Effects of Rg3 and Rg3 Derivatives on Steady-State Activation and Inactivation of INa
(A, B) Effect of Rg3 and Rg3 derivatives on steady-state activation and inactivation of NaV1.2 was examined as described.4) The voltage-dependence of the conductance of
NaV1.2 was examined in the presence and absence of Rg3 and Rg3 derivatives (100 m M). Data are meansS.E.M. (n10—11/group). The curves described by the smooth solid
lines are two-state Boltzmann functions as described in Materials and Methods. Data are meansS.E.M. (n9—10/group).
484 Vol. 31, No. 3

than Rg3. We found that 2H-Rg3 exhibited more potent inhi-
bition of peak and steady-state Na
currents than Rg3. The
presence of 2H-Rg3 caused a leftward shift in the concentra-
tion–response curve compared with Rg3 by two-fold. The
IC50 values of Rg3 acting on the peak and steady-state
IFMQ3 mutant currents were 40.79.6 and 15.93.9 m M, re-
spectively,4) whereas the IC50 values of 2H-Rg3 acting on the
peak and steady-state IFMQ3 mutant channel currents were
30.41.7 and 7.61.3 m M, respectively (Table 3). However,
Rg3-C6 almost had no effects on the peak and steady-state
IFMQ3 mutant currents (Fig. 5B). Figure 5C shows repre-
sentative IFMQ3 mutant channel current traces normalized
in the absence and presence of various concentrations of 2H-
Rg3. The percent inactivation at 1 s was used to describe the
extent of inactivation.14) In the absence of Rg3 (control), the
percent inactivation at 1 s was 12.33.6% in the peak cur-
rents. The percent inactivation at 1 s increased to 14.44.6,
24.44.5, 46.26.2, and 63.75.1% compared to control
currents by 10, 30, 100, and 300 m M, respectively. In the ab-
sence of 2H-Rg3, the percent inactivation at 1 s was
13.33.6%. The percent inactivation at 1 s enhanced to
20.44.6, 46.97.1, 68.86.5, and 83.36.2% compared to
control currents by 10, 30, 100, and 300 m M, respectively
(n9; ∗ p0.05, ∗∗ p0.01 compared with Rg3). In the ab-
sence of Rg3-C6, the percent inactivation at 1 s was also
11.61.6 % (n9; Fig. 5D). The percent inactivation at 1 s
only slightly increased to 14.41.6, 15.93.1, 17.82.5,
and 21.36.2% compared to control currents by 10, 30, 100,

Table 3. Effects of Rg3 and Rg3 Derivatives on a Fast Inactivation Gating-

Deficient Mutant NaV1.2

Vmax IC50 nH

2H-Rg3 peak 72.71.5* 30.41.7* 1.10.1

2H-Rg3 steady-state 101.02.6* 7.61.3* 1.20.2
Rg3 peak 69.95.9 40.79.6 1.10.1
Fig. 4. Effects of Rg3 and Rg3 Derivatives on Use-Dependent Blockade of Rg3 steady-state 90.25.9 15.93.9 1.30.1
NaV1.2 Rg3-C6 peak ND ND ND
Fifty 10-ms depolarizing pulses to 10 mV were applied from a holding potential of Rg3-C6 steady-state ND ND ND
100 mV at 10 Hz in the presence and absence of Rg3 and Rg3 derivatives (100 m M) in
NaV1.2. Data are meansS.E.M. (n9—10/group). * p0.05 compared with control.

Fig. 5. Effects of Rg3 and Rg3 Derivatives on a Fast Inactivation Gating-Deficient Mutant Channel of NaV1.2
(A) Traces evoked from 100 to 10 mV in the absence and presence of various concentrations of 2H-Rg3 showing open channel blockade of the fast inactivation gating-defi-
cient mutant channel (IFMQ3). IFMQ3 mutant channel was constructed as described in the Materials and Methods. INa was elicited in IFMQ3 mutant channel by a 500 ms depolar-
ization to 10 mV from a holding potential of 100 mV, evoked every 5 s. The evoked current exhibited a slower decay and sizable persistent non-inactivating or plateau current.
2H-Rg3 treatment dose-dependently inhibited the peak and plateau INa values. (B) Concentration–response curves for the effect of Rg3 and Rg3 derivatives on the IFMQ3 peak and
plateau INa values. The solid lines were fit by the Hill equation. (C) Normalized current of typical IFMQ3 mutant channel in the absence and presence of various concentrations of
2H-Rg3. (D) The percent inactivation at 1 s was used to describe the extent of inactivation on various concentrations of Rg3 and Rg3 derivatives. Data represent the meansS.E.M.
(n9—10/group; ∗ p0.05, ∗∗ p0.01 compared with Rg3).
March 2008
and 300 m M, respectively.
DISCUSSION
Ginseng has long been used as a treatment for a wide vari-
ety of ailments, and some of the purported effects of ginseng
have been documented in laboratory studies.5) Although the
beneficial effects and modes on ginsenoside actions have not
been fully elucidated, there are evidences that they can target
the Na
channels involved in neuronal excitability. Thus, Rg3
has been shown to inhibit brain Na
channel activity,3,4) and
its effects are closely coupled to the inhibition of neurotrans-
mitter release.18) However, relatively little is known on the
structure–activity relationships in Rg3-induced Na
channel
regulations.
Ginsenosides consist three different parts; steroid-like
backbone structure, carbohydrate portions, and aliphatic side
chain, which are coupled to carbon-20 of backbone structure
(Fig. 1). In the present study, we found that the aliphatic side
of Rg3 plays a differential role in NaV1.2 channel regulations.
The main evidences are as follows: first, reduction of double
bond of aliphatic side chain significantly increased tonic in-
hibition of peak INa, use-dependent inhibition, and a fast in-
activation in inactivation gating-deficient mutant channels
compared with Rg3 (Fig. 2); second, hydroxylation or oxy-
genation of aliphatic side chain of Rg3 greatly attenuated Rg3
effects on Na
channel; third, deletion of aliphatic side chain
of Rg3 caused a loss of Rg3 effects on Na
channel regula-
tion. These results indicate that the aliphatic side chain of
Rg3 is also involved in Na
channel regulation and show pos-
sibility that the aliphatic side chain of Rg3 could be the main
target in producing Rg3 analogs (Fig. 2).
The hydroxyl group at the carbon-20 of backbone struc-
ture of ginsenosides provides stereoisomers, (R)- and (S)-
epimers.15) In the previous study, we have shown that a slight
difference in chemical structure between Rg3 epimers pro-
duces a large difference in Na
channel regulation. Thus, we
demonstrated that voltage-dependent Na
channel distin-
guishes 20(S)-Rg3 from 20(R)-Rg3. Only 20(S)-Rg3 shows
an inhibitory effect on Na
channel currents. As the evi-
dences, we showed that 20(S)-Rg3 treatment induced a large
attenuation of Na
channel activity with reversible and dose-
dependent manner, whereas 20(R)-Rg3 had a slight effect, in-
dicating that the effect of ginsenoside Rg3 on Na
channel
activity is stereoselective.
On the other hand, Rg3 including other ginsenosides is also
amphiphilic compound, since aglycone (PPD) is the back-
bone of the Rg3, with a hydrophobic four-ring steroid-like
structure that may be non-polar, whereas the two carbohy-
drates on the carbon-3 of backbone are polar (Fig. 1). Our
previous report showed that modifications of the carbohy-
drate portion of Rg3 abolish its ability to inhibit neuronal
Na
channels. Furthermore, neither the aglycone (PPD) por-
tion of Rg3, nor sophorose, a disaccharide of the carbohy-
drate portion of Rg3, had any effect on peak INa, indicating
that the amphiphilic property of Rg3 is also necessary for
Na
current inhibitions.9)
However, it is not yet known about the physiological role
of the aliphatic side chain of Rg3, which is coupled to the
carbon-20 of backbone structure. In the present study, we
further demonstrated that the aliphatic side chain of Rg3 also
485
participates in Rg3-mediated Na
channel current inhibi-
tions. Interestingly, the role of the aliphatic side chain in Na

channel regulation seems to be different from that of Rg3
epimers or carbohydrate portion of Rg3, since modifications
of the aliphatic side chain exhibited differential effects in
Na
channel current inhibitions. Thus, reduction of double
bond of aliphatic side chain significantly increased Rg3-in-
duced Na
current inhibition, whereas hydroxylation of di-
methyl group, oxygenation or deletion of aliphatic side chain
causes an attenuation or loss of Rg3 effects. One notion for
the enhancement of Na
current inhibitions by reduction of
aliphatic side chain is as follows: single rather than double
bond of aliphatic side chain might be less rigid for geometri-
cally better alignment in the interaction of Rg3 with Na

channel proteins by reducing steric hindrance induced by
double bond. In addition, since deletion of aliphatic side
chain might remove the counterpart of Rg3 for interaction
with Na
channel proteins, Rg3-C6 had no effects on Na

channel activity. Taken together, the previous and present
studies show that the aliphatic side chain as well as carbohy-
drate portions and position of hydroxyl group at the carbon-
20 of Rg3 are involved in Na
channel regulations. We also
tested the effects of Rg3 analogs, which are modified in
aliphatic side chain, on 5-HT3A receptor and Kv1.4 channel
activity. We found that these Rg3 analogs were not selective
for Na
channel rather than they exhibited differential regu-
lations between ligand-gated ion channels and voltage-gated
ion channels (data not shown). Currently, we are doing fur-
ther investigations to elucidate the role of aliphatic side chain
in ginsenoside-induced voltage- and ligand-gated ion channel
regulations.
On the other hand, recent reports showed that Rg3, which
is in vivo administered into rat, could be metabolized into
several components. First, Rg3 loses one carbohydrate at the
carbon-3 and then further is metabolized into aglycone
(PPD) in rat plasma.16) Qian et al. showed in rat gastrointesti-
nal tract that Rg3 also undergoes mono- or di-oxygenation
processes by adding one or two molecules of oxygen to
aliphatic side chain, resulting in production of epoxy com-
pounds.17) Furthermore, they also showed that these epoxy
products of aliphatic side chain of Rg3 was removed with
Rg3-C6 or Rh2-C6 formation. Interestingly, in the present
study, we showed that Rg3 derivatives such as Rh2, PPD, oxy-
genated products of Rg3 and Rg3-C6 exhibited only a slight
inhibition of peak INa or had no effect on peak INa, whereas
reduction of aliphatic side chain of Rg3 increased the in-
hibitory effect of Rg3 on peak INa. These results show a pos-
sibility that the regulatory effects of Rg3 on Na
channel ac-
tivity might be lost if Rg3 is metabolized after in vivo admin-
istration as other drugs do.
In summary, we have used Rg3 and Rg3 derivatives, which
are modified in the aliphatic side chain, to further character-
ize structure–activity relationship of Rg3 in regulation of rat
brain Na
channel activity. We found that reduction of the
double bond of aliphatic side chain increased Rg3-mediated
Na
current inhibitions, whereas hydroxylation, oxygenation
or deletion of aliphatic side chain attenuated or abolished
Rg3 action on Na
channel regulation. These novel findings
might provide an insight in identifying the functional groups
of Rg3 in Rg3-mediated Na
channel regulations.
486
Acknowledgements This work was supported by grants
to S. Y. Nah from the BK21 project, Korean Research Foun-
dation Grant funded by MOEHRD (KRF-2005-015-E00222),
Korea Institute of Science and Technology Core-Competence
Program to H.R. and the Neurobiology Research Program
from MOST.
REFERENCES
1) Goldin A. L., Voltage-gated Na channels, “Ligand- and Voltage-Gated
Ion Channels,” CRC Press, Boca Raton, FL, 1995, pp. 73—89.
2) Stuart G. J., Sakmann B., Nature (London), 367, 60—72 (1994).
3) Jeong S. M., Lee J. H., Kim J. H., Lee B. H., Yoon I. S., Lee J. H., Kim
D. H., Rhim H., Kim Y., Nah S.Y., Mol. Cells, 18, 383—389 (2004).
4) Lee J. H., Jeong S. M., Kim J. H., Lee B. H., Yoon I. S., Lee J. H.,
Choi S. H., Kim D. H., Rhim H., Kim S. S., Kim J. I., Jang C. G.,
Song J. H., Nah S. Y., Mol. Pharmacol., 68, 1114—1126 (2005).
5) Nah S. Y., Korea J. Ginseng Sci., 21, 1—12 (1997).
6) Dascal N., CRC Crit. Rev. Biochem., 22, 317—387 (1987).
7) Choi S., Jung S. Y., Ko Y. S., Koh S. R., Rhim H., Nah S. Y., Mol.
Pharmacol., 61, 928—935 (2002).
8) Choi S., Lee J. H., Kim Y. I., Kang M. J., Rhim H., Lee S. M., Nah S.
Vol. 31, No. 3
Y., Eur. J. Pharmacol., 468, 83—92 (2003).
9) Kim J. H., Lee J. H., Jeong S. M., Lee B. H., Yoon I. S., Lee J. H.,
Choi S. H., Kim D. H., Park T. K., Kim B. K., Nah S. Y., Biol. Pharm.
Bull., 29, 365—370 (2006).
10) Lansman J. B., Hess P., Tsien R. W., J. Gen. Physiol., 88, 321—347
(1986).
11) Pugsley M. K., Goldin A. L., Eur. J. Pharmacol., 342, 93—104
(1998).
12) Pugsley M. K., Yu E. J., Goldin A. L., Gen. Pharmacol., 34, 417—427
(2000).
13) Hondeghem L. M., Katzung B. G., Annu. Rev. Pharmacol. Toxicol.,
24, 387—423 (1984).
14) Claydon T. W., Boyett M. R., Sivaprasadarao A., Orchard C. H., Am. J.
Physiol. Cell. Physiol., 283, C1114—C1121 (2002).
15) Kang D. I., Lee J. Y., Yang J. Y., Jeong S. M., Lee J. H., Nah S. Y.,
Kim Y., Biochem. Biophys. Res. Commun., 333, 1194—1201 (2005).
16) Xie H. T., Wang G. J., Sun J. G., Tucker I., Zhao X. C., Xie Y. Y., Li
H., Jiang X. L., Wang R., Xu M. J., Wang W., J. Chromatogr. B, 818,
167—173 (2005).
17) Qian T., Cai Z., Wong R. N., Mak N. K., Jiang Z. H., J. Chromatogr. B,
816, 223—232 (2005).
18) Tachikawa E., Kudo K., Kashimoto T., Takahashi E., J. Pharm. Exp.
Ther., 273, 629—636 (1995).