B R A I N R E S E A R C H 1 2 0 3 ( 2 0 0 8 ) 6 1 – 6 7

ava i l ab l e a t www.sc i enced i rec t . com

www.e l sev i e r. com/ l oca te /b ra in res

Research Report

Involvement of batrachotoxin binding sites inginsenoside-mediated voltage-gated Na+ channel regulation

Jun-Ho Leea,1, Byung-Hwan Leea,1, Sun-Hye Choia, In-Soo Yoona, Tae-Joon Shina,Mi Kyung Pyoa, Sang-Mok Leea, Hyoung-Chun Kimb, Seung-Yeol Naha,⁎aGinsentology Research Laboratory and Department of Physiology, College of Veterinary Medicine, Konkuk University, Seoul Korea, 143-701bNeuropsychopharmacology and Toxicology Program, College of Pharmacy, Kangwon National University, Chunchon 200-701, South Korea

A R T I C L E I N F O

⁎ Corresponding author. Fax: +82 2 450 3037.E-mail address: synah@konkuk.ac.kr (S.-Y

1 These two authors contributed equally to

0006-8993/$ – see front matter © 2008 Elsevidoi:10.1016/j.brainres.2008.01.078

A B S T R A C T

Article history:Accepted 4 January 2008Available online 12 February 2008

Recently, we showed that the 20(S)-ginsenoside Rg3 (Rg3), an active ingredient of Panax ginseng,inhibits rat brain NaV1.2 channel peak currents (INa). Batrachotoxin (BTX) is a steroidal alkaloidneurotoxin and activates NaV channels through interacting with transmembrane domain-I-segment 6 (IS6) of channels. Recent report shows that ginsenoside inhibits BTX binding inrat brain membrane fractions. However, it needs to be confirmed whether biochemicalmechanism is relevant physiologically and which residues of the BTX binding sites areimportant for ginsenoside regulations. Here, we demonstrate that mutations of BTX bindingsites such as N418K and L421K of rat brain NaV1.2 and L437K of mouse skeletal muscle NaV1.4channel reduceorabolishRg3 inhibitionof INaandattenuateRg3-mediateddepolarizing shift ofthe activation voltage and use-dependent inhibition. These results indicate that BTX bindingsites play an important role in modifying Rg3-mediated Na+ channel properties.

© 2008 Elsevier B.V. All rights reserved.

Keywords:Panax ginsengGinsenoside Rg3Batrachotoxin binding sitesNaV channels

1. Introduction

Na+ channels are transmembrane proteins that consist of apore-forming α subunit and auxiliary β1, β2 and β3 subunits(Goldin, 1995). The α subunit is composed of four homologousdomains (I–IV), each composed of six α-helical transmembranesegments (S1–S6), and is responsible for voltage-dependentincreases inNa+-selective permeability. The inwardNa+ current(INa) initiates axonal and somatic action potentials in nerve andmuscle fibers, and may also be involved in axonal intraneur-onal or interneuronal information transfer (Stuart and Sack-mann, 1994). Na+ channels are one of the targets of cardiac- andneuroprotective treatments against pathologic conditionsincluding arrhythmia and brain ischemia. Site-directed muta-genesis has helped to reveal drug/toxin binding sites in the Na+

. Nah).this work.

er B.V. All rights reserved

channel α subunit (Wang and Wang, 1998). For example,batrachotoxin, which is a skin neurotoxin from the SouthAmerican frog Phyllobates terribilis, activates brain NaV1.2 andskeletal muscle NaV1.4 channels. Further studies using site-directed mutagenesis demonstrated that I433, N434 and L437residues of NaV1.4 channel and their homologous residues suchas I417, N418 and L421 of brain NaV1.2 channel in domain-Isegment 6 (IS6) play important roles as BTX interaction sites(Wang and Wang, 1998).

Ginseng, the root of Panax ginseng C.A. Meyer, is well knownin herbal medicine as a tonic and restorative agent. The mainmolecular ingredients responsible for its effects are theginseno-sides, amphiphilic molecules comprising a hydrophobic agly-cone backbone linked to monomeric, dimeric or tetramerichydrophilic carbohydrate side chains (Fig. 1). Ginsenosideshave

.

Fig. 1 – Structures of BTX and Rg3. Glc, glucopyranoside.Subscripts indicate the carbons in the glucose rings that linkthe two carbohydrates.

62 B R A I N R E S E A R C H 1 2 0 3 ( 2 0 0 8 ) 6 1 – 6 7

neuroprotective actions in vitro and in vivo (Kim et al., 2005), buttheirmolecular bases remain elusive. The 20(S) but not the 20(R)form of ginsenoside Rg3 inhibits tonic INa expressed in Xenopusoocytes in the resting and open states, and induces a depolariz-ing shift in the activation voltage and use-dependent inhibition(Jeong et al., 2004; Lee et al., 2005). Ginsenoside Rh2 inhibits [3H]BTX-B binding in rat brain membrane fractions (Duan et al.,2006), suggesting that the ginsenoside-induced rat brain NaV1.2channel effectsmay involve the BTX binding sites. However, it isnot clearwhether the interferencebyginsenosidewith [3H]BTX-Bbinding in rat brain membrane fractions is relevant to ginseno-side-inducedNa+ channel regulations, andwhich residues of theBTX binding sites are important for ginsenoside-induced Na+

channel regulations. The aim of the present study was to in-vestigate the relationships betweenginsenoside-mediated inter-ference with Na+ channel function and the BTX binding sitesusing the Xenopus oocyte gene expression system and site-directed mutagenesis. For this, we used rat brain NaV1.2 andmouse skeletal muscle NaV1.4 channels and examined whetherRg3 effects on bothNa+ channel isoforms are influencedwith thesame manner after mutations of BTX binding sites. We presentevidence that BTX binding sites such as N418 and L421 of NaV1.2and L437 of NaV1.4 in IS6 play important roles inmodifying Rg3-mediated Na+ channel properties.

2. Results

2.1. Point mutations of residues in the BTX binding sites ofNaV1.2 and NaV1.4 channels shift the Rg3 concentrationresponse curve rightward and abolish Rg3 inhibition

We showed in a previous report that Rg3 inhibits NaV1.2channel currents (INa) in a stereospecific manner (Jeong et al.,2004). Recently, Duan et al. (2006) reported that ginsenosidesinhibit [3H]BTX-B binding in rat brain membrane fractions,

suggesting the existence of specific binding sites for Rg3 andraising the possibility that ginsenoside affects NaV1.2 channelactivity via the BTX binding sites.We therefore testedwhethersubstitutions of amino acid residues in the BTX binding sitesalter Rg3-mediated tonic inhibition of the channels.

Using the two-electrode voltage-clamp technique werecorded NaV1.2 and NaV1.4 channel currents from Xenopusoocytes, in which cRNA of NaV1.2 α or NaV1.4 αwas co-injectedwith cRNA of β1 subunit. To elicit the currents we appliedvoltage steps (200-msduration) to−10mVat5-s intervals fromaholding potential of −100 mV. The currents evoked were tran-sient inward Na+ currents (INa) that decayed rapidly (Fig. 2) (Leeet al., 2005). As previously reported, Rg3 (100 μM) resulted in aninhibition of INa in wild-type channels averaging 65.1±3.5 % forNaV1.2 and 63.4±4.1 % for NaV1.4. The Rg3 effect was concentra-tion-dependent (Figs. 2A and B), and the IC50 value and Hillcoefficient for two Na+ channels are given in Table 1.

Next, to test whether mutations of the BTX binding sitesaffect Rg3 inhibition of the NaV1.2 and NaV1.4 channels, weconstructed both channels with mutated BTX binding sites(I417K, N418K and L421K for NaV1.2; I433K, N434K and L437K forNaV1.4 channel) (Wang and Wang, 1998). Fig. 2A showsrepresentative traces of peak INa in the absence or presence ofRg3 in oocytes expressing wild-type and L421K of NaV1.2 andwild-type and L437K of NaV1.4 channel. Fig. 2B reproduces rep-resentative concentration–response curves for peak INa inhibi-tion of the wild-type and mutant channels by Rg3. For NaV1.2,L421Kabolished Rg3 inhibition, I417Khadnoeffect on inhibitionandN418Kcausedanapproximately eleven-fold increase in IC50

value from 29.4±3.2 to 329.9±37.6 μM. For NaV1.4, I433K andN434K had no effect on Rg3 inhibition but L437K caused anapproximately two-fold increase in IC50 value from 58.5±6.3 to121.6±20.6 μM (Fig. 2B and Table 1).

We assessed current–voltage relationships in the absence orpresence of Rg3 with voltage steps from −50 to +50 mV from aholding potential of −100 mV every 5 s. As shown in Fig. 2C, Rg3treatment caused a voltage-dependent reduction in peak INa inwild-type channels with more pronounced inhibition at thelower voltages. However, Rg3 had no significant effect on thecurrent–voltage relationship in oocytes expressing L421K forNaV1.2 and L437K for NaV1.4, indicating that BTX binding sitesare important for Rg3 effects on NaV1.2 and NaV1.4 channelregulation.

2.2. Point mutations of the BTX binding sites abolish theRg3 effect on the steady-state activation of NaV1.2 and NaV1.4

In our previous report we also showed that Rg3 induces adepolarizing shift in the steady-state activation voltage but notin the inactivation voltage, in addition to inhibiting peak INa.However, the mutation of K859 to K859Q in voltage sensorsegment domain II abolished the Rg3-induceddepolarizing shiftin steady-state activation without affecting Rg3-induced peakINa inhibition (Lee et al., 2005). This suggested that K859 residueis another residue that Rg3 interacts with. We next examinedwhether mutations of the BTX binding sites alter the effect ofRg3 on the voltage-dependence of Na+ channel steady-stateactivation and inactivation. The data obtained are summarizedin Table 2. Therewas a significant depolarizing shift of the half-maximal activation voltage in wild-type isoforms (Fig. 3A and

Fig. 2 – Tonic inhibitionof INabyRg3.A, Representative tracesofwild-typeandmutant channels in thepresenceor absenceof Rg3. INa

was then recorded in ND96 using two-electrode voltage clamp as described in the Experimental procedures.B, Concentration–responsecurves for theeffectofRg3onwild-typeandmutantchannels.Dataaremeans±S.E.M. (n=10–13/group).Theelectrophysiological experimentsonNa+ channel activity, oocyteswere clampedat aholdingpotential of−100mVand themembranepotential was depolarized to −10 mV for 200 ms every 5 s. C and D, Representative current – voltage relationship obtained usingvoltagestepsbetween−50and+50mVwith5-mVincrements.Voltagestepswereapplied in thepresenceandabsenceof100μMRg3.

63B R A I N R E S E A R C H 1 2 0 3 ( 2 0 0 8 ) 6 1 – 6 7

Table 1 – Effect of Rg3 of the wild-type and mutant Na+

channels

Type Imax IC50 nH

NaV1.2 Wild-type 78.4±3.5 29.4±3.2 1.1±0.2I417K 79.2±3.7 32.8±4.1 1.3±0.2N418K 69.8±4.7 329.9±37.6⁎ 1.3±0.2L421K ND ND ND

NaV1.4 Wild-type 92.5±21.5 58.5±6.3 1.1±0.4I433K 85.9±12.7 74.2±8.1 1.1±0.2N434K 82.8±12.6 54.4±11.7 0.9±0.2L437K 33.8±2.3⁎ 121.6±20.6⁎ 1.1±0.2

Imax, maximal inhibition by Rg3. Rg3 inhibition was determinedfrom % difference of INa obtained in the absence or presence of Rg3.Values represent the means±S.E.M. (n=10–13/group). ND, notdetermined. ⁎pb0.001 compared with the wild-type Na+ channel.

64 B R A I N R E S E A R C H 1 2 0 3 ( 2 0 0 8 ) 6 1 – 6 7

Table 2). However, L421K abolished the Rg3-induced voltageshift for NaV1.2 and L437K greatly attenuated the Rg3-inducedvoltage shift for NaV1.4. We then investigated the Rg3 effecton voltage-dependent Na+ channel inactivation by plottingnormalized peak INa against conditioning pre-pulse voltage(Table 2) and fitting the data to the Boltzmann function. Therewas no significant difference in half-maximal inactivationvoltageandslope factor (kh) betweenbothwild-typeandmutantchannels. This again indicates that Rg3 affects steady-stateactivation but not inactivation of the Na+ channel isoforms, andthat the BTXbinding sites also play an important role in the Rg3-mediated shift of activation voltage.

2.3. Point mutations of the BTX binding sites abolish theRg3 effect on use-dependent blockade of NaV1.2 and NaV1.4

Since we had demonstrated previously that Rg3 displays use-dependent inhibition (Lee et al., 2005), we also examinedwhethermutations of theBTXbinding sites affect this process. In thewild-type, INawas elicited by fifty 20-mspulses from −100 to −10mVat10 Hz. Each peak INa was normalized to the first INa pulse peak. Inthe absence of Rg3 there was a slight reduction in peak INa, whiletreatment with 100 μM Rg3 caused use-dependent inhibition ofpeak INa by 29.9±7.3 and 31.1±4.3% for NaV1.2 and NaV1.4 chan-nels, respectively (n=9 each; Figs. 3C and D). Rg3 caused only 4.2±0.9 and 13.5±3.5% use-dependent inhibition of the peak INa forL421 K of NaV1.2 and L437K of NaV1.4 channels, respectively(Figs. 3C and D). These observations again indicate that BTX

Table 2 – Effect of Rg3 on the voltage-dependence of activation

Type Activa

Vg0.5

NaV1.2 Wild-type(−Rg3) −28.4±0.19Wild-type(+Rg3) −19.1±0.35⁎

L421K (−Rg3) −29.8±0.26L421K (+Rg3) −28.9±0.32

NaV1.4 Wild-type(−Rg3) −27.6±0.64Wild-type(+Rg3) −20.1±0.24⁎

L437K (−Rg3) −26.5±0.58L437K (+Rg3) −23.5±0.39

⁎pb0.001 compared with the absence of Rg3 in wild-type. Values represen

binding sites are involved in Rg3-induced use-dependent inhibi-tion of the NaV1.2 and NaV1.4 channels.

3. Discussion

Ginseng has long been used as a treatment for a wide varietyof ailments, and some of the purported effects of ginseng havebeen documented in laboratory studies (Nah, 1997). Althoughthe beneficial effects andmode of ginsenoside action have notbeen fully elucidated, there is evidence that they can target theNa+ channel involved in neuronal excitability. Ginsenoside Rg3has been shown to affect brain Na+ channel activity (Lee et al.,2005), and its effects are closely coupled to the inhibition ofneurotransmitter release (Tachikawa et al., 1995).

We showed here that Rg3 affects rat brain NaV1.2 andmouseskeletal muscle NaV1.4 channels by interactingwith part of BTXbinding sites. The evidence is as follows: first, Rg3-induced tonicinhibition of peak INa was greatly attenuated or abolished bymutations of BTX binding sites (Fig. 2); second, Rg3-induceddepolarizing shift in steady-state activation and Rg3-induceduse-dependent inhibition were also abolished or attenuated bymutations of BTX binding sites in both Na+ channel isoforms(Fig. 3).

Rg3 is not structurally similar to BTX. The latter is a well-known steroidal alkaloid neurotoxin and persistent Na+ chan-nel activator that binds to receptor site 2 (Catterall et al., 2007). Incontrast, Rg3 does not resemble an alkaloid except for its ste-roidal backbone structure (Fig. 1). Despite the structural dif-ferencebetweenRg3 andBTX,we found that thepointmutationsN418K and L421K but not I417K of NaV1.2 and L437K but notI433K and N434K of NaV1.4 channel strongly shifted theconcentration–response curve rightward, or abolished tonicpeak INa (Fig 2B), as well as abolishing the depolarizing shift ofthe activation voltage and use-dependent inhibition. Thus, theprevious and present studies support that the ginsenoside-in-duced Na+ channel regulations are mediated through BTXbinding sites (Duan et al., 2006). However, it could not be ex-cluded that the Rg3 might regulate Na+ channel activity throughallostericmechanisms. In this case, themutations introduced inBTX binding sites of IS6 might induce the conformationalchanges and interfere the interaction of Rg3 with channel pro-teins, thus attenuating or abolishingRg3-induced INa inhibitions.Thisnotion is supported by previous report that ginsenoside Rh2

and inactivation of the wild-type and mutant Na+ channels

tion Inactivation

Kg Vh0.5 Kh

4.6±0.17 −2.92±0.44 6.9±0.435.3±0.32 −26.7±0.71 8.4±0.745.2±0.23 −27.7±0.28 7.3±0.284.9±0.28 −27.9±0.39 6.1±0.365.7±0.58 −26.5±0.54 5.4±0.435.5±0.22 −25.6±0.34 5.2±0.625.4±0.53 −25.6±0.54 4.9±0.855.4±0.36 −27.1±0.45 5.9±0.45

t the means±S.E.M. (n=10–11/group).

Fig. 3 – Effects of Rg3 on steady-state activation of INa and on use-dependent blockade of the wild-type, NaV1.2, and NaV1.4channels. A and B, Effect of Rg3 on steady-state activation of INa in the wild-type and mutant channels was examined asdescribed (Lee et al., 2005). The voltage-dependence of the conductance of thewild-type andmutant channelswas examined inthe presence and absence of 100 μM Rg3. Data are means±S.E.M. (n=10–11/group). The curves described by the smooth solidlines are two-state Boltzmann functions as described in Experimental procedures. C and D, Fifty 10-ms depolarizing pulses to−10 mVwere applied from a holding potential of −100 mV at 10 Hz in the presence and absence of 100 μM Rg3 in the wild-typeand mutant channels. Insets, only sweeps 1 and 50 are shown for both control and Rg3 treated channels. The other sweeps areomitted for clarity. Data are means±S.E.M. (n=9–10/group).

65B R A I N R E S E A R C H 1 2 0 3 ( 2 0 0 8 ) 6 1 – 6 7

and PPD induced the inhibition of BTX binding via allostericmanner, indicating that these ginsenosides might bind to sitesthat are distinct from but closely related with BTX binding sites(Duan et al., 2006). Thus, BTX binding sites might play a role inRg3-mediated Na+ channel inhibitions and the conforma-tional changes to IS6 resulting from these mutations might bemore relevant. However, further studies will be necessary tocharacterize the exact Rg3 interaction sites in Na+ channelregulations.

In the present study we also found that mutations of BTXbinding sites of NaV1.2 and NaV1.4 channel caused differenteffects in Rg3-induced Na+ channel isoform inhibitions. Wefound that among I417, N418 and L421 residues of BTX bindingsites in brain NaV1.2 channels, mutations of two amino acidssuchasN418 and L421 toN418Kand L421Kgreatly attenuated orabolished Rg3 actions, whereas among I433, N434 and L437residues ofmuscle NaV1.4 channel, mutation of one amino acidsuch as L437 to L437K attenuated Rg3 actions and significantlyshifted concentration–response curve rightward. From these

results, it appears that mutations of the homologous L421 andL437 residues of NaV1.2 and NaV1.4 channels, respectively,strongly affect Rg3 actions and these results further show apossibility that the amino acid leucine in IS6 of both channelsmight be the common interaction residue for Rg3 with Na+

channel isoforms and that plays a key role in modifying Rg3-mediated Na+ channel properties.

Previousbiochemical andelectrophysiological studies showedthat BTX binding is also inhibited allosterically by local anes-thetics and related drugs, indicating that these compounds alsointeract with the BTX binding sites (Postma and Catterall, 1984).Linford et al (1998) further demonstrated that point mutations ofthe lidocaine binding sites such as F1764A not only reduced BTXbinding but also affected the BTX-induced hyperpolarizing shiftandnon-inactivatingsodiumcurrents.However, it isunlikely thatRg3 affects NaV1.2 channel activity through F1764 residue, sincewe have shown that mutations of these residues do not alter Rg3inhibition of INa (Lee et al., 2005). This indicates that Rg3 regulatesthe NaV1.2 channel activity via interaction with IS6 but not IVS6.

66 B R A I N R E S E A R C H 1 2 0 3 ( 2 0 0 8 ) 6 1 – 6 7

Na+ channel blocking agents have significant cardio- andneuro-protective effects (Alzheimer, 2002). We may askwhether Rg3 inhibition of NaV1.2 channel through the BTXbinding site as seen in vitro also applies to its in vivo phar-macological effects. Bae et al. (2004) and Tian et al. (2005)showed in rats that oral (100 mg/kg) or intravenous adminis-tration (5 and 10 mg/kg) of Rg3 had a neuroprotective effectagainst focal cerebral ischemic injury, decreasing neurologicaldeficit scores and reducing the infarct area. Rg3 also improvedmitochondrial energy metabolism, and opposed decreases insuperoxide dismutase and glutathione-peroxidase activitiesin a cerebral ischemia model (Tian et al., 2005). Similarly, then-3 polyunsaturated fatty acids (PUFAs) exhibit antiarrhyth-mic effects by inhibiting Na+ channel activity in animals andhumans (Leaf et al., 1999). Xiao et al (2001) demonstrated thatoneof polyunsaturated fatty acids, EPA, regulateshumanheartNa+ channel activity via one of residues in the BTXbinding site,namely N406, which is homologous to N418 of the rat brainNaV1.2 channel. Thus, Rg3-mediated affects on brain Na+

channels may well be related to their in vivo pharmacologicaleffects. However, we do not yet have any direct evidence thatRg3-mediated Na+ channel regulation can be used prophylac-tically or therapeutically against ischemia or arrhythmia.

In summary, we have used site-directed mutagenesis toidentify the residues of rat brain NaV1.2 and mouse skeletalNaV1.4 channels with which Rg3 interacts. We found that theN418 and L421 residues of the NaV1.2 and the L437 residue ofNaV1.4 channel play important roles in the Rg3-mediatedNaV1.2 and NaV1.4 channel regulations, respectively. Thisfinding may contribute to increased understanding of thepharmacological effects of P. ginseng.

4. Experimental procedures

4.1. Materials

Fig. 1 gives the chemical structure of 20(S)-ginsenoside Rg3 (Rg3).Rg3 was kindly provided by the Korea Ginseng Cooperation(Taejon, Korea). Rat brain Na+ channel NaV1.2 α, and mouseskeletal muscle NaV1.4 α and β1 subunit cDNAs were gifts fromDr. Goldin (University of California at Irvine) and Dr. Zimmer(Friedrich Schiller University, Jena), respectively. Other reagentswere purchased from Sigma-Aldrich (St. Louis, MO).

4.2. Preparation of Xenopus oocytes and microinjection

Xenopus laevis frogs were purchased from Xenopus I (AnnArbor, MI). Their care and handling were in accordance withthe most rigorous institutional guidelines. Preparation of Xe-nopus oocytes andmicroinjectionwere performed as describedpreviously (Lee et al., 2005).

4.3. Site-directed mutagenesis of NaV1.2 or NaV1.4 and invitro transcription of Na+ channel subunit cDNAs

Single and triple amino acid substitutions were generatedwith Pfu DNA polymerase (QuikChange™ XL Site-DirectedMutagenesis Kit, STRATAGENE) using appropriate sense andantisense primers, as described (Lee et al., 2005).

4.4. Data recording

Oocyteswere placed on the net that separates upper and lowerwells (Tachikawa et al., 1995) and were impaled with twomicroelectrodes filled with 3 M KCl (0.2–0.7 MΩ). The electro-physiological experiments were performed at room tempera-ture with an Oocyte Clamp (OC-725C,Warner Instrument, CT),and stimulation and data acquisition were controlled withpClamp 8 (Axon Instruments, Union City, CA). In most of theelectrophysiological experiments data were recorded asdescribed (Lee et al., 2005). In brief, a custom-made Plexiglasnet chamber was used for two-electrode voltage-clamp re-cordings. The chamber was constructed by milling two con-centric wells into the chamber bottom (the diameter/heightof the upper well was 8/3 mm and that of the lower well was6/5 mm) and gluing plastic meshes (~0.4 mm grid diameter)onto the bottom of the upper well. A perfusion inlet (~1 mm indiameter) was drilled through the wall of the lower well, and asuction tube was placed on the edge of the upper well. Forexperiments, a single oocyte was placed on the net separatingthe upper and lower wells. The net grids helped anchor theoocyte in place during the electrophysiological recordings. Theoocyte was then impaled with two microelectrodes filled with3 M KCl (0.2–0.7 MΩ) and electrophysiological experimentswere carried out at room temperature using an Oocyte Clamp(OC-725C,Warner Instruments, CT, USA). Stimulationanddataacquisition were controlled with a pClamp 8 (Axon Instru-ments, Union City, CA, USA). For most of the electrophysiolo-gical experiments on Na+ channel activity, oocytes wereclamped at a holding potential of −100 mV and themembranepotential was depolarized to −10 mV for 200 ms every 5 s.Linear leak currents were corrected by means of the leak sub-traction procedure.

The voltage-dependence of Na+ channel activation wascalculated by measuring the peak current at test potentialsranging from −50 mV to +50 mV evoked in 5 mV increments.The conductance (gNa)was calculatedaccording to theequation,gNa= INa/(Vg−Vr), where INa is the peak amplitude of the Na+

current, Vg is the test potential, and Vr is the reversal potentialfor Na+. The conductance–voltage curves were drawn ac-cording to the equation gNa/maxgNa=1/{1+exp [(Vg0.5−Vg)/kg]},wheremaxgNa is themaximumvalue for gNa,Vg0.5 is the potentialat which gNa is 0.5maxgNa, and kg is the slope factor (potentialrequired for an e-fold change). The voltage-dependence of Na+

channel inactivationwasdeterminedusing200msconditioningpre-pulses ranging from −60 mV to +20 mV from a holdingpotential of −100 mV in 5 mV increments, followed by a testpulse to −10 mV for 5 ms. The peak INa was normalized to itsrespective maximum value (maxINa) and plotted as a function ofthe pre-pulse potential. The steady-state inactivation curvesweredrawnaccording to theequation INa/maxINa=1/{1+exp [(Vh−Vh0.5)/kh]}, where Vh is pre-pulse potential, Vh0.5 is the potentialat which INa is 0.5 maxINa, and kh is the slope factor.

The frequency-dependent effect of Rg3 was examinedusing a protocol in which 50 depolarizing pulses of 10 msduration and 10 Hz frequency were applied to −10 mV from aholding potential of −100 mV. The oocytes were preincubatedwith Rg3 for 2 min before application of high frequency stimu-lation. The protocol was run in the absence (control) and pre-sence of Rg3 (100 μM). The current amplitude of each pulsewas

67B R A I N R E S E A R C H 1 2 0 3 ( 2 0 0 8 ) 6 1 – 6 7

normalized to the peakmaximal current (pulse number 1) andplotted as a function of pulse number.

4.5. Data analysis

To obtain concentration–response curve of Rg3 effect on INa, thepeak amplitudes at different concentrations of Rg3were plotted,and then fitted to the following Hill equation using Originsoftware (Northampton, MA); y/ymax=[A]nH/([A]nH+[EC50]n),where y is the peak INa at given concentration of Rg3, ymax isthemaximal peak INa, IC50 is the concentration of Rg3 producinga half-maximumeffect, [A] is the concentration of Rg3, and nH isthe interaction coefficient. All values are presented as means±S.E.M. The differences betweenmeans of control and treatmentdata were determined using an unpaired Student’s t-test. Avalue of Pb0.05 was considered statistically significant.

Acknowledgments

This work was supported by grants to S. Y. Nah from the BK21project, Korean Research Foundation Grant funded byMOEHRD(KRF-2005-015-E00222), and theNeurobiologyResearchProgramfromMOST.

Appendix A. Supplementary data

Supplementary data associated with this article can be found,in the online version, at doi:10.1016/j.brainres.2008.01.078.

R E F E R E N C E S

Alzheimer, C., 2002. Na+ channels and Ca2+ channels of the cellmembrane as targets of neuroprotective substances. Adv. Exp.Med. Biol. 513, 161–181.

Bae, E.A., Hyun, Y.J., Choo, M.K., Oh, J.K., Ryu, J.H., Kim, D.H., 2004.Protective effect of fermented red ginseng on a transient focalischemic rats. Arch. Pharm. Res. 27, 1136–1140.

Catterall, W.A., Cestele, S., Yarov-Yarovoy, V., Yu, F.H., Konoki, K.,Scheuer, T., 2007. Voltage-gated ion channels and gatingmodifier toxins. Toxicon 49, 124–141.

Duan, Y., Zheng, J., Law, V., Nicholson, R., 2006. Natural productsfrom ginseng inhibit [3H]batrachotoxinin A 20-alpha-benzoatebinding to Na+ channels in mammalian brain. Eur. J.Pharmacol. 530, 9–14.

Goldin, A.L., 1995. Voltage-gated Na+ channels. In: North, R.A. (Ed.),Ligand- and voltage-gated ion channels. CRC Press, Boca Raton,FL, pp. 73–89.

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., 2004. Stereospecificityof ginsenosideRg3 action on ion channels. Mol. Cells 18, 383–389.

Kim, J.H., Kim, S., Yoon, I.S., Lee, J.H., Jang, B.J., Jeong, S.M., Lee, J.H.,Lee, B.H.,Han, J.S., Oh, S., Kim,H.C., Park, T.K., Rhim,H.,Nah, S.Y.,2005. Protective effects of ginseng saponins on 3-nitropropionicacid-induced striatal degeneration in rats. Neuropharmacology48, 743–756.

Leaf, A., Kang, J.X., Xiao, Y.F., Billman, G.E., Voskuyl, R.A., 1999. Theantiarrhythmic and anticonvulsant effects of dietary N-3 fattyacids. J. Membr. Biol. 172, 1–11.

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., 2005. Characteristics of ginsenoside Rg3-mediated brainNa+

current inhibition. Mol. Pharmacol. 68, 1114–1126.Linford, N.J., Cantrell, A.R., Qu, Y., Scheuer, T., Catterall, W.A.,

1998. Interaction of batrachotoxin with the local anestheticreceptor site in transmembrane segment IVS6 of thevoltage-gated sodium channel. Proc. Natl. Acad. Sci. U. S. A. 95,13947–13952.

Nah, S.Y., 1997. Ginseng, recent advances and trend. Korea J.Ginseng Sci. 21, 1–12.

Stuart, G.J., Sakmann, B., 1994. Active propagation of somaticaction potentials into neocortical pyramidal cell dendrites.Nature 367, 60–72.

Tachikawa, E., Kudo, K., Kashimoto, T., Takahashi, E., 1995.Ginseng saponins reduce acetylcholine-evoked Na+ influx andcatecholamine secretion in bovine adrenal chromaffin cells.J. Pharm. Exp. Ther. 273, 629–636.

Tian, J., Fu, F., Geng, M., Jiang, Y., Yang, J., Jiang, W., Wang, C., Liu,K., 2005. Neuroprotective effect of 20(S)-ginsenoside Rg3 oncerebral ischemia in rats. Neurosci. Lett. 374, 92–97.

Postma, S.W., Catterall, W.A., 1984. Inhibition of binding of [3H]batrachotoxinin A 20-alpha-benzoate to sodium channels bylocal anesthetics. Mol. Pharmacol. 25, 219–227.

Wang, S.Y., Wang, G.K., 1998. Point mutations in segment I-S6render voltage-gated Na+ channels resistant to batrachotoxin.Proc. Natl. Acad. Sci. U. S. A. 95, 2653–2658.

Xiao, Y.F., Ke, Q., Wang, S.Y., Yang, Y., Wang, G.K., Morgan, J.P.,Leaf, A., 2001. Point mutations in alpha-subunit of humancardiac Na+ channels alter Na+ current kinetics. Biochem.Biophys. Res. Commun. 281, 45–52.