C6-ceramide inhibited Na+ currents by intracellular Ca2+ release in rat myoblasts

ORIGINAL ARTICLE 151J o u r n a l o fJ o u r n a l o f

CellularPhysiologyCellularPhysiology
C6-Ceramide Inhibited NaR
Currents by Intracellular Ca2R

Release in Rat Myoblasts
ZHENG LIU,1 JIAN-GUANG XU,2 HUA ZHANG,3 YAN-JIA FANG,1 AND YAN-AI MEI1,3*1School of Life Sciences, Institute of Brain Science, State Key Laboratory of Medical Neurobiology Fudan University,
Shanghai, P.R. China2Hua-Shan Hospital, Fudan University, Shanghai, P.R. China3Institute of Neurobiology, Institute of Brain Science, Fudan University, Shanghai, P.R. China

Ceramides are novel second messengers that may mediate signaling leading to apoptosis and the regulation of cell cycle progression.Moreover, ceramide analogs have been reported to directlymodulateKþ andCa2þ channels in different cell types. In this report, the effectof C6-ceramide on the voltage-gated inward Naþ currents (INa) in cultured rat myoblasts was investigated using whole-cell currentrecording and a fluorescent Ca2þ imaging experiment. At concentrations of 1–100 mM, ceramide produced a dose-independent andreversible inhibition of INa. Ceramide also significantly shifted the steady-state inactivation curve of INa by 16 mV toward thehyperpolarizing potential, but did not alter the steady-state activation properties. C2-ceramide caused a similar inhibitory effect on INa

amplitude. However, dihydro-C6-ceramide, the inactive analog of ceramide, failed to modulate INa. The effect of C6-ceramide on INa wasabolished by intracellular infusion of the Ca2þ-chelating agent BAPTA, but was mimicked by application of caffeine. Blocking the releaseof Ca2þ from the sarcoplasmic reticulumwith xestospongin Cor heparin, an inositol 1,4,5-trisphosphate (IP3) receptor blocker, induced agradual increase in INa amplitude and eliminated the effect of ceramide on INa. In contrast, ruthenium red, which is a blocker of theryanodine-sensitive Ca2þ receptor did not affect the action of C6-ceramide on INa. Intracellular application of the G-protein agonistGTPgS also induced a gradual decrease in INa amplitude, while the G-protein antagonist GDPbS eliminated the effect of C6-ceramideon INa. Calcium imaging showed that C6-ceramide could give rise to a significant elevation of intracellular calcium. Our data show thatincreased calcium release through the IP3-sensitive Ca

2þ receptor, which probably occurred through the G-protein and phospholipase Cpathway, may be responsible for C6-ceramide-induced inhibition of the INa of rat myoblasts.J. Cell. Physiol. 213: 151–160, 2007. � 2007 Wiley-Liss, Inc.

Contract grant sponsor: Committee of Science and Technology,Shanghai;Contract grant number: 04DZ19901.Contract grant sponsor: Chinese National Nature ScienceFoundation;Contract grant number: 30670472.

*Correspondence to: Dr. Yan-Ai Mei, Department of Physiologyand Biophysics, School of Life Sciences, Fudan University, Shanghai200433, P.R. China. E-mail: [email protected]

Received 22 November 2006; Accepted 9 March 2007

DOI: 10.1002/jcp.21106

Ceramides, the products of sphingomyelin turnover, are novellipid second messengers that mediate diverse signalingpathways, including those leading to cell cycle arrest anddifferentiation, apoptosis, and alteration of gene expression indifferent cell types (Schwartz et al., 1997; Yu et al., 1999;Gulbins, 2003). These biological effects of ceramides areassociated with the regulation of intracellular enzymes such asprotein kinase C (Bourbon et al., 2000), tyrosine kinases,diacylglycerol kinase, and phospholipases (Ruvolo, 2003). It hasbeen reported that ceramides can also act on nerve growthfactor (NGF) and nicotinic receptors by the release ofintracellular calcium stores in cerebellar neurons and ratchromaffin cells (Liu et al., 2001; Numakawa et al., 2003).

In addition to the regulation of intracellular enzymes, anumber of studies have revealed important signaling pathwaysin which ceramides regulate ion channels. In rat pinealocytes,Chik et al. (1999, 2001) showed that ceramide selectivelyinhibits L-type Ca2þ channel and Kþ channel activities. Theeffects of ceramide on Kþ channels in other cell types, such as Tlymphocytes and smooth muscle, have been reported (Gulbinset al., 1997; Li et al., 1999). The mechanisms through whichceramides exert their actions on ion channels appear to be cellspecific. Ceramide-induced inhibition of KCa channels invascular smooth muscle cells is mediated by its direct effect onthe cell membrane (Li et al., 1999), whereas ceramide-inducedinhibition of KCa channels in rat pinealocytes involvesintracellular signaling mechanisms (Chik et al., 2001). However,there have been few investigations of the directmodifying effectof ceramides on Naþ channels, which is very pivotal formodulating neuronal and muscle excitability (Zhang et al.,2002).

Myoblasts are myogenic cells that differ from neurons in thatthe former undergo differentiation and fusion during the

� 2 0 0 7 W I L E Y - L I S S , I N C .

development and repair of skeletal muscle. Primary culturedmyoblasts from skeletal muscle contain various ion channels,such as the outward and inward rectifying Kþ channels(Occhiodoro et al., 1998; Fischer-Lougheed et al., 2001). Twopotassium channels are essential to myoblast fusion: the ether ago-go Kþ channel and the inward rectifying Kþ channel. Inaddition, T-type Ca2þ currents can be recorded infusion-competent myoblasts (Bijlenga et al., 2000). Ourprevious studies revealed that a novel Naþ-activated Kþ

current IK(Na) channel is present in myoblasts and that theactivation properties of this IK(Na) are altered duringdevelopment. Thus, the currentmay be involved in ratmyoblastfusion (Zhou et al., 2004). Since IK(Na) are directly correlatedwith Naþ currents (INa), the main current in the pre-fusionmyoblast, the modulation on INa is of great importance toregulate IK(Na) and myoblast fusion.

Fig. 1. C6-ceramide inhibited INa in a concentration-independentmanner. A: Superimposed INa evoked by a 60 msec depolarizing pulsefrom a holding potential of S100 to S20 mV. Current traces wereobtained in the absence and presence of C6-ceramide (C6) atconcentrations of 1–100 mM. B: The blocking effects of variousC6-ceramide concentrations (1–100 mM) on INa. P> 0.05, usingANOVA followed by the Tukey test.

152 L I U E T A L .

In the present study, we first determined whether C6-ceramide has any effect on the INa of myoblasts, and thenexamined the mechanisms of this C6-ceramide-induced effect.

Materials and MethodsCell culture

The experimental procedures were carried out in accordance withEuropean guidelines for the care and use of laboratory animals (CouncilDirective 86/609/EEC). Cells were derived from legs of 3 days oldSprague–Dawley rat pups. Isolated cells were plated onto 35-mm Petridishes coated with poly-L-lysine (10 mg/ml) at a density of 105 cells perdish. The nutrient mixture F-10 Ham culture medium wassupplemented with 10% fetal calf serum and 1% antibiotic–antimycotic.Cultured cells were incubated at 378C, with 5% CO2 and for 2–3 daysafter plating. Most of the myoblasts we used show a spindle-shapedappearance with 15–30 mM in length and 3–6 mM in width.

Patch-clamp recordings

Whole-cell currents of myoblasts within 3-day-in-culture wererecorded using conventional patch-clamp technique. The recordingswere performed at room temperature (20–258C). Prior to INa

recording, the culture medium was replaced with a bath solutioncontaining (mM): NaCl 145, KCl 2.5, HEPES 10, MgCl2 1, glucose 10,and pHwas adjusted to 7.40withNaOH. Soft glass patch pipettes werefilled with an internal solution containing (mM): KCl 10, K-gluconate130, MgCl2 4, HEPES 10; pH was adjusted to 7.40 using KOH. Theresistance of the pipettes was 6–7 MV in the bathing medium whenfilled with the internal solution. Ceramide solutions were preparedextemporaneously and gravity ejected fromMSC-200 Manual SolutionChanger (Bio-Logic-Science Instruments, Claix, France). No effect ofceramide on extracellular and intracellular pH was observed.

Calcium imaging

Intracellular Ca2þ levels in myoblasts were monitored with fura-2 AM(Dojindo, Kumamoto, Japan), amembrane permeable indicator. Fura-2AM (1mM)was dissolved in 20% Pluronic F-127 (w/v, DMSO) and thenadded to Ringer’s solution at dilution of 1:1,000. Fura-2 AM-containingRinger’s was added to a chamber to give a final concentration of 1 mMfura-2 AM. Isolated cells were incubated in the dye solution for 40 minat room temperature and then perfused with dye-free Ringer’s for30min before an experiment. Fluorescence images were acquiredwithan inverted microscope (IX-51; Olympus Optical, Tokyo, Japan)equipped with a digital CCD camera (Hamamatsu Photonics,Hamamatsu, Japan). A high-speed scanning polychromatic light source(Till, Munich, Germany) was used for alternate excitations atwavelengths of 340 and 380 nm. The fluorescence intensities at bothwavelengths (F340 and F380) were measured every 10 sec, and imageswere obtained using SimplePCI software (Hamamatsu Photonics). Theratio between the two images was proportional to [Ca2þ]i of the cellunder study. Before an experiment, a background level of fluorescence(attributable to autofluorescence and camera noise) was determinedand subtracted from all the data obtained.

Data acquisition and analysis

Whole-cell currents were recorded using Axon patch 200B amplifier(Axon Instruments, Foster City, CA) operated in the voltage-clampmode. Data acquisition and analysis were performed with pClamp 8.1software (Axon Instruments, Union City, CA) andOrigin 6.1 (MicrocalSoftware, Northampton, MA). Currents were corrected on-line forleak and residual capacitance transients by a P/4 protocol. Statisticalanalysis was performed using the Student t-test with non-pairedcomparison or paired comparisons where it is relevant. Values weregiven as means� s.e.m. with n as the number of myoblasts tested.P-value <0.05 was used to denote the statistical difference betweengroups.

Drugs and other materials

All drugs were obtained from Sigma-Aldrich (St. Louis,MO) except ruthenium red (Fluka, Neu-Ulm, Germany).C2-ceramide, C6-ceramide, dihydro-C6-ceramide, O9262(ET-18-OCH3) and U6756 (U-73122) were first dissolved in DMSOand then diluted in the bath or pipette solution, with a final DMSOconcentration<1%, which did not produce any effect onNaþ currents

JOURNAL OF CELLULAR PHYSIOLOGY DOI 10.1002/JCP

alone. All other drugs were prepared with Ringer’s solution. F-10 Hamculture medium, fetal calf serum and antibiotic-antimycotic solution wereobtained from Gibco Life Technologies (Grand Island, NY).

Results

Our previous study showed that the tetrodotoxin-sensitive(TTX-S) form of sodium channel predominates in the myoblaststage (Lu et al. 2005), so here we investigated the effect of

C E R A M I D E I N H I B I T E D N a þ C U R R E N T S 153

ceramide on TTX-S sodium channels. INa was evoked using60msec constant depolarizing pulses froma holding potential of�100 to �20 mV at intervals of 10 sec, whereupon themaximum current amplitude was obtained.

The application of different concentrations of C6-ceramideto the bath solution produced a dose-independent reductionin INa amplitude (Fig. 1A). The inhibitory effect of C6-ceramideon INa occurred very rapidly and reached itsmaximumwithin 1–2 min. Currents could recover to near the control levels withinapproximately 1–2 min after the removal of ceramide. Atconcentrations of 1–50 mM, the inhibitory effects of C6-ceramide showed no statistical differences (INa reductions by1 and 50 mM ceramide were 14� 1.7% (n¼ 6) and 17�2.8%(n¼ 9), respectively (P> 0.05, using ANOVA followed by theTukey test); 100mMceramide also induced a reduction of INa ofabout 17�2.4% (n¼ 3)). Statistical analysis of theconcentration-response data is summarized in Figure 1B.

The effects of C6-ceramide on the activation and inactivationproperties of INa were studied using appropriate voltage

Fig. 2. Effect of C6-ceramide on the conductanceSvoltage relationship oftraces) of 50 mM ceramide (C6). The cells were held at S100 mV and depintervalsof10sec.B: Voltage-dependent activation curve for INa obtained fras a function of the command potential in the absence or presence of 50 mequation gNaU INa/(VmSVrev). The data were obtained from seven cells a


protocols. Figure 2A shows that Naþ currents were evoked bya 20 msec depolarizing pulse from a holding potential of�100 mV to potentials between �80 and þ60 mV in steps of10 mV at intervals of 10 sec. The voltage–current curve isshown in Figure 2B. Figure 2C shows a plot of normalizedconductance as a function of command potential in theabsence and presence of 50 mM C6-ceramide. The datapoints were calculated using the equation gNa¼ INa/(Vm�Vrev), where gNa is membrane Naþ conductance, Vm isthe membrane potential, Vrev is the reversal potentialfor Naþ. For both control and C6-ceramide treatment, theratios were fitted to a Boltzmann function of the formgNa/gNa-max¼ 1/{1þ exp[(Vm1/2�Vm)/k]}, where gNa-max isthe maximum inward conductance, Vm1/2 is the half activationpotential of the Naþ currents, and k is the slope factor.According to these curves, the currents were half-activatedat �28.2� 3.3 mV and �31.6� 3.3 mV in the absence andpresence of 50 mM C6-ceramide, respectively (n¼ 7,P> 0.05), suggesting that C6-ceramide treatment did not

INa. A: INa recorded in the absence (upper traces) and presence (lowerolarized to potentials ranging from S80 to R60 mV in 10 mV steps atomthecurrent tracesshown in(A).C:PlotofnormalizedconductanceM C6-ceramide. The data points were calculated using thend are expressed as meanWSEM.

154 L I U E T A L .

significantly shift the voltage dependence of steady-stateactivation of Naþ channels.

We then studied the effect of C6-ceramide on the voltagedependence of steady-state inactivation of INa. Currentswere elicited using 1 sec conditioning pre-pulses ranging from�130 to�20mV in steps of 10 mV before a�20mV test pulse.Figure 3A shows current recordings from a cell obtainedunder control conditions and after exposure to 50 mMC6-ceramide. Consistent with the observations above,application of 50mMC6-ceramide resulted in a reduction of thepeak INa elicited at each conditioning pre-pulse potential, asshown in Figure 3A,B. Figure 3C shows the steady-stateinactivation curve fitted by the Boltzmann equation ofINa/INa-max¼ 1/{1þ exp[(Vm�Vm1/2)/k]}þA, where INa-max isthe maximum INa that elicited by 1 sec conditioning pre-pulsesof�130mV. Among the six cells studied, the Vm1/2 values were�65� 1.1mV and�81.1� 1.2mV in the absence and presenceof 50 mM C6-ceramide, respectively. Application of 50 mMceramide significantly shifted the steady-state inactivation curveof INa toward the hyperpolarizing potential by 16 mV (n¼ 6,P< 0.05).

Fig. 3. EffectofC6-ceramideonthesteady-state inactivationpropertieso50mMC6-ceramide (C6,bottom). Conditioningpre-pulses of1 sec fromS1to S20 mV. The voltage protocol is shown below the current record. B: S50mM C6-ceramide. C: Peak current amplitude normalized to the maximupoints were fitted with a Boltzmann function of INa/INa-maxU 1/{1R exp[(Vexpressed as meanWSEM.


To determine whether the reduction of INa was a specificeffect of ceramide, experiments were performed with10 mM C2-ceramide, another active ceramide analog, and10 mM dihydro-C6-ceramide, an inactive stereoisomer ofC6-ceramides. As shown in Figure 4A, C2-ceramide caused arapid and remarkable reduction of INa by 16.7� 3.6% (n¼ 5),similar to the inhibitory effect of C6-ceramide. In contrast,dihydro-C6-ceramidewas ineffective inmodulating INa (Fig. 4B).Application of 10 and 100mM dihydro-C6-ceramide induced anaverage current reduction of just 3.92� 0.85% (n¼ 6) and1.43� 0.78% (n¼ 6), respectively. The results weresignificantly different from that induced by C6-ceramideand C2-ceramide (P< 0.05).

Since Ca2þ-free solution was used in our initial experiments,and previous reports suggest that L-type Ca2þ channels appearin fusion-competent myoblasts (Bijlenga et al., 2000), we couldexclude the possibility that extracellularCa2þmight be involvedin the effect of C6-ceramide on myoblast INa. However,whether calcium release from intracellular stores was involvedin the inhibitory effect of ceramide on INa required furtherinvestigation. To remove intracellular calcium, the recording

f INa.A:Control currents (top)andcurrents followingtheapplication of30 toS20mV in incrementsof 10mV were applied before the testpulseteady-state inactivation curves of INa in the absence and presence ofm current plotted against the pre-pulse potential. Normalized current

mSVm1/2)/k]}RA. The data were obtained from six cells and are

Fig. 4. Effect of C6-ceramide analogs and intracellular Ca2R changeon INa amplitude. A,B: Superimposed INa evoked by a 60 msecdepolarizing pulse from a holding potential of S100 to S20 mV.Current traces were obtained in the absence and presence of50 mM C2-ceramide (C2) and 100 mM dihydro-C6-ceramide (DH-C6).C,D: Superimposed INa obtained from application of C6-ceramide tomyoblasts in the intercellular presence of 10 mM BAPTA andapplication of 1 mM caffeine alone. E: Current inhibition induced byvarious ceramides, caffeine, and BAPTA. MP< 0.05, comparedwith C6-ceramide.


pipette was loaded with Ca2þ-free solution containing 10 mMBAPTA [1,2-bis (2-aminophenoxy) ethane-N,N,N9,N9-tetraacetic acid], a fast Ca2þ-chelating agent. Calculations madeusing MaxChelator version 6.81 (Bers et al., 1994) indicatedthat the Ca2þ intracellular concentration fell to 1 pM with thistreatment. In the first 1–2 min after the establishment of thewhole cell configuration in the seven cells tested, the INa

amplitude gradually increased from an initial value of0.99� 0.16 nA, then reached a steady level of 1.5� 0.3 nA, anincrease of 52.7� 11%. Under these conditions, application of50 mM C6-ceramide failed to inhibit INa appreciably andreduced INa slightly by 3.3� 0.9% (n¼ 7), as shown inFigure 4C. This differed significantly from the effect of


C6-ceramide in the absence of BAPTA (P< 0.05). Applicationof caffeine (1mM) alone into the bath solution and no BAPTA inthe pipette solution stimulated release of Ca2þ from theendoplasmic reticulum and caused a rapid and markedreduction of INa by 15� 1.9% (85� 1.9% of control; n¼ 6,P< 0.05), which mimicked the inhibitory effect of C6-ceramideon INa (Fig. 4D). These results strongly suggest that changes ofintracellular Ca2þ were involved in the C6-ceramide effect.Statistical analysis of the above data is shown in Figure 4E.

It is well established that calcium is released fromintracellular stores via inositol 1,4,5-trisphosphate (IP3)- and/orryanodine-sensitive receptors. Therefore, we investigated thepathway through which Ca2þ was released from thesarcoplasmic reticulum during the action of ceramide. Internalinfusion of xestospongin C (20 mM), a potent IP3 receptorantagonist (Gafni et al., 1997), resulted in a progressive increaseof INa after membrane rupture (Fig. 5A, right) from 0.9� 0.3 nAat the control level to 1.3� 0.3 nA at the final plateau, asignificant increase of 51� 17% (n¼ 5, P< 0.05). Moreover,the inhibitory effect of C6-ceramide on INa was significantlyreduced by intracellular xestosponginC, to 3.4� 0.8% (n¼ 10).The effect of heparin, another potent IP3 receptor antagonist(Gafni et al., 1997), was similar to that of xestospongin C(Fig. 5A, left). In the presence of intracellular heparin (5 mM),the INa amplitude was increased by 53� 20% and nosignificant C6-ceramide-induced reduction was observed(2.5� 0.5%; n¼ 8). The effect of C6-ceramide in the presenceof intracellular xestospongin C or heparin was significantlydifferent to that in their absence (Fig. 5C,D; P< 0.05). Incontrast, intracellular loading with ruthenium red (20 mM), anantagonist of the ryanodine-sensitive Ca2þ receptor, did notsignificantly increase INa amplitude after membrane rupture(Fig. 5B, 5� 4.9% of control; n¼ 6). C6-ceramide induced anaverage INa reduction of 14.6� 0.7% (n¼ 6), which was notsignificantly different from that induced in the absence ofintracellular ruthenium red (Fig. 5D, P> 0.05). These resultssuggest that IP3 receptor-stimulated calcium release was likelyto be involved in the C6-ceramide-induced reduction of INa.

Kobrinsky et al. (1999) found that ceramide causeddose-dependent elevation in the second messenger IP3 via theactivation of G-protein and phospholipase C (PLC) in Xenopuslaevis oocytes. In the present study, whether G-proteinscontributed to the inhibitory effect of C6-ceramide on INa wasfurther tested by adding GTPgS (100 mM) or GDPbS (100mM)to the pipette solution. Dialysis of myoblasts with GDPbSabolished their response to C6-ceramide (Fig. 6A). Theinhibitory effect of C6-ceramide (50 mM) on INa was severelyimpaired in the presence of GDPbS (Fig. 6A). In the presence of100 mM or 1 mM of GDPbS, C6-ceramide induced averagereductions in INa of 4.2� 0.4% (n¼ 6) and 4� 1.1% (n¼ 8),respectively. In the presence of intracellular GDPbS, thereduction of INa induced by C6-ceramide was significantlydecreased compared with that induced by C6-ceramide alone(P< 0.05). In contrast, internal infusion of GTPgS (100 mM)resulted in progressive suppression of INa after membranerupture (Fig. 6B). The amplitude declined significantly from1.1� 0.2 nA at the control level to 0.7� 0.18 nA (P< 0.05,analysis by a paired t-test), a decrease of 32.8� 4.8% (n¼ 6).

The effects of PLC blockade were also tested todetermine whether the PLC pathway was involved. TwoPLC blockers were added to the pipette solution.Application of 20 mM ET-18-OCH3, a potent PLC blockerabolished the inhibitory effect of C6-ceramide (50 mM) onINa, with an average inhibition of 2.6� 1.4% (n¼ 5). The effectof U-73122was similar (Fig. 6B). In the presence of 10 or 30mMintracellular U-73122, the C6-ceramide-induced reductionsof INa were 7.8� 1.1% (n¼ 5) and 3.7� 0.9% (n¼ 6),respectively, which were significantly different from thatinduced by C6-ceramide without U-73122 or ET-18-OCH3

Fig. 5. Effects of blockade of IP3- and/or ryanodine-sensitive receptors on C6-ceramide-induced reduction of INa. A: Time course of the effect ofinternal infusion of xestospongin C (XC, 20mM) and heparin (Hepa, 5 mM) on INa amplitude and C6-ceramide-induced inhibitory effect on INa. Theinsets inthegraphsshowthesuperimposedINatracestakenfromtheinitialcontrol levels (I0,atmembranerupture,0min),after internal infusionofxestosponginCorheparin for2min(2min)andapplicationofC6-ceramide(C6),respectively.B:SuperimposedINatracesobtainedfrommyoblastswith internally infused ruthenium red (RuR, 20 mM) in the presence and absence of C6-ceramide. C: Statistical analysis of internal infusion ofxestospongin-C and heparin increased the INa amplitude (Imax), compared with the initial control levels (I0). MP< 0.05, compared with the initialcontrol levels, using the paired t-test. D: Statistical analysis of the inhibitory effect induced by C6-ceramide on INa amplitude in the present ofintracellular xestospongin-C, heparin, or ruthenium red. MP< 0.05, compared with the control.

156 L I U E T A L .

(P< 0.05). Statistical analysis of the above data is shown inFigure 6C.

Previous investigations showed that Ca2þ-inducedmodulation of ion channels might be mediated by the activationof Ca2þ-dependent enzymatic processes. Calmodulin (CaM) is aubiquitous intracellular protein that regulates the activity ofvarious enzymes and channel proteins in a Ca2þ-dependentmanner (Cheung, 1980). Ca2þ/CaM can modulatevoltage-gatedNaþ channels in neurons andmuscles (Dolmetschet al., 2001; Young andCaldwell, 2005).We therefore examinedthe effects of a CaM blocker, W-7 (Yu et al., 2006), on theC6-ceramide-induced inhibitory effect on INa. Internalapplication of 100 nMW-7 resulted in a progressive increase of


the INa amplitude in the first several minutes after the whole cellconfiguration was established (Fig. 7B). When the currentreached a steady level (146� 9% of control; n¼ 5), 50 mMC6-ceramide did not affect the INa (3.3� 1.1%), suggesting thatthe Ca2þ/CaM pathway was probably involved in the effectof C6-ceramide on INa. In the presence of W-7, the INa

amplitude and the C6-ceramide-induced inhibitory effect weresignificantly different from that induced by C6-ceramide alone(P< 0.05). Statistical analysis of the above data is shown inFigure 7C,D.

To confirm that intracellular Ca2þ was associated with theeffect of C6-ceramide on INa under our experimentalconditions, we monitored C6-ceramide-induced changes of

Fig. 6. Involvement of the G-protein/PLC pathway in C6-ceramide-induced reduction of INa. A: Superimposed INa traces obtained frommyoblasts with internally infused GDPbS or GTPgS. C6-ceramide(C6) failed to inhibit INa while the myoblast was dialyzed with GDPbS(1 mM) in the pipette solution (left), and intracellular application ofGTPgS (100 mM) mimicked the inhibitory effect on INa induced byC6-ceramide (right). B: Superimposed INa traces obtained from themyoblast with internal infusion of ET-18-OCH3 (ET-18, 20 mM)or U-73122 (20 mM) in the absence or presence of C6-ceramide.Applying PLC blockers abolished the inhibitory effect of C6-ceramide.C: Bar chart showing the INa inhibition induced by C6-ceramide in thepresence of the intercellular PLC blockers, GDPbS and GTPgS.MP< 0.05, compared with the control.


intracellular Ca2þ by calcium imaging. Figure 8A shows theresults obtained from one myoblast. With addition of 50 mMC6-ceramide to the bath solution for 1 min, theintracellular Ca2þ level of the cell, represented as a percentageof the control calculated from the ratio F340/F380, increasedmoderately with time and reached amaximum at about the endof the drug application. When the washed out with normalRinger’s solution, the intracellular Ca2þ declined slowly to the


control level in about 5 min. Figure 8B shows two CCD imagesof the same myoblast, showing the change of Ca2þ induced by50 mM C6-ceramide. Intracellular Ca2þ was clearly increasedafter the application of C6-ceramide. For each cell, the maximalintracellular Ca2þ level obtained with C6-ceramide wasrespectively normalized to that obtained in Ringer’s solution(control). As shown in Figure 8C, ceramide caused a potentincrease of intracellular Ca2þ in all of the cells, with an averageincrease of 13.4� 7.2% (n¼ 10, P<0.05).

Discussion

In this report, we showed that C6-ceramide induced inhibitionof the INa of rat myoblasts through IP3 receptor-mediated Ca2þ

release that seemed to be downstream of the G-protein/PLC pathway.

Ceramide is a family of highly hydrophobic molecules thatcontain a variable length of fatty acid (containing 2–28 carbons)linked to sphingosine. Ceramides can modify membrane ionchannels by altering the permeability of the cell membrane(Ruiz-Arguello et al., 1996) or by a non-specific lipid effect onthe channels (Li et al., 1999; Bock et al., 2003). However, thiswas unlikely in our study, because C2- and C6-ceramideproduced similar inhibitory effects on INa, but dihydroceramide,an inactive analog of ceramide, was impotent. Furthermore, theeffect of C6-ceramide was blocked by PLC inhibitors. Theobserved effect of ceramide on INa may therefore illustrate theuniversal significance of ceramide in cells generally, besides itscontribution to myoblasts. Other studies have shown thatabout 1% of applied ceramide is metabolized during the first30 min after application (Bielawska et al., 1993), suggestingthat C2-ceramide was poorly metabolized. In our currentrecording and Ca2þ imaging experiments, inhibitory effectson INa and Ca2þ release from intracellular stores wereobserved within 1–2 min after application of C6-ceramideor C2-ceramide. The effect of ceramide on INa might be due toceramide directly, but not its metabolized products. However,Zhang et al.’s (2002) investigations in the rat sensory neuronsrevealed that ceramide modulated neuronal excitability byfacilitating the amplitude of INa, and that intracellularsphingosine 1-phosphate (S1P) which derived from ceramidewas involved in modulating neuronal excitability (Zhang et al.,2006). Therefore, what induce the opposite ceramide’s effecton INa and whether S1P is involved in our cell model need to bedetermined in the future.

Mediation of the action of NGF and nicotinic receptors byceramide via the release of intracellular calcium stores has beenreported (Liu et al., 2001; Numakawa et al., 2003). UsingaCa2þ-free bath solution, the fastCa2þ-chelating agent BAPTA,and Ca2þ imaging, we found that intracellular Ca2þ release wasnecessary for C6-ceramide-induced reduction of INa. It is wellestablished that calcium release from intracellular stores isnormally through the IP3 and/or ryanodine receptor pathway(Berridge, 1997). Moreover, expression of IP3 receptors alone,or of both IP3 and ryanodine receptors, is dependent on celltype (Kobrinsky et al., 1995). It is interesting that, in our study,blocking the IP3 receptor pathway evoked an increase of INa andabolished its ceramide-induced reduction, suggesting that therewas a tonic release of Ca2þ by the IP3 receptor pathwaythat might maintain INa at a lower level. A ryanodinereceptor antagonist was ineffective on both tonic Ca2þ

release and C6-ceramide-induced reduction of INa.However, caffeine, an activator of ryanodine receptors,mimicked the ceramide-induced inhibitory effect on INa andsignificantly reduced INa. By contrast, while 20 mM rutheniumred was applied in the pipette solution, the inhibitory effectinduced by caffeine was significantly disturbed (results notshown). Taken together, these data show that both IP3 andryanodine receptor pathways exist in rat myoblasts to regulate

158 L I U E T A L .

intracellular Ca2þ, but only the IP3-sensitive receptor pathwaycontributed to the basic tonic Ca2þ release and the modulationby ceramide.

Compared with previous reports, our research is distinct inthat the inhibitory effect of C6-ceramide on INa wasconcentration independent, and that the inhibition ratewas lessthan 20%. The concentrations of ceramide (1–50 mM) used inour study were consistent with those used in previous studiesinvolving the modulation of ion channels by ceramide (Yu et al.,

Fig. 7. Involvement of CaM in C6-ceramide-induced reductionof INa. A: Time course of changes of INa amplitude obtained from cellswith normal intracellular solution in the absence or presence ofC6-ceramide (C6). Superimposed INa evoked by a 60 msec constantdepolarizing pulse from a holding potential of S100 to S20 mV.B: Time course of changes of INa amplitude obtained from cells withintracellular W-7, and in the absence or presence of C6-ceramide.Internal infusion of W-7 (100 mM) increased INa amplitude, andC6-ceramide no longer suppressed INa in the presence of W-7.C: Statistical analysis of the effect of internal infusion of W-7 on INa

amplitude (Imax). MP< 0.05, compared with the initial control levels(I0), using the paired t-test. D: Statistical analysis of the inhibitoryeffect induced by C6-ceramide on INa amplitude in the presence ofintracellular W-7. MP< 0.05, compared with the control.


1999; Chik et al., 2001). Increasing the C6-ceramideconcentration to 100mM failed to augment the inhibitory effectof ceramide on INa. Bielawska et al. (1993) showed that only10% of externally applied C2-ceramide was taken up by cells;thus, it was possible that, in our study, the concentration ofC6-ceramide inside the cell was not equal to the concentrationof C6-ceramide in the external solution. In addition, the tonicrelease of intracellular Ca2þ mentioned above resulted in ahigher basic level of cytoplast Ca2þ, which may have impededmass Ca2þ release from intracellular stores under the higherconcentrations of C6-ceramide used.

Based on evidence from other cell types, the downstreamsignaling elements associated with ceramide can be diverse andinclude multiple kinases, phosphatases, and caspases (Mathiaset al., 1998; Huwiler et al., 2000). Ceramide has also beensuggested to activate phosphokinase C and modulatea-adrenergic-stimulated cAMP in smooth muscle and ratpinealocytes (Negishi et al., 1998; Bourbon et al., 2000). Ourpreliminary observations showed that forskolin and PMA,agonists of phosphokinases A and C, did not mimic theinhibitory effect of ceramide on INa (results not shown). Incontrast, two blockers of the PLC pathway, U-73122 and ET-18-OCH3, eliminated the inhibitory effect of C6-ceramide onINa. GTPgS mimicked the inhibitory effect of C6-ceramide, andblocking G-protein by GDPbS abolished the ceramide-inducedinhibition of INa. These results are in accord with Kobrinskyet al. (1999), who found that ceramide caused an elevation of IP3via activation of G-protein and PLC. It is noticed that in ourinvestigation, the response of INa to GTPgs is much larger thanthat of the highest dose of ceramide. Considering that GTPgS isa potent activator of many downstream reactions, and PLCpathway is just one of them, we suppose that there might besome other pathways that simultaneously influence theinward Naþ currents. Taken together, these results suggestthat C6-ceramide-induced inhibition of INa involved theG-protein/PLC pathway. However, as the tyrosinekinase and Ras/Raf pathway has also been implicated inceramide-dependent inhibition of potassium channels andCa2þ

channels in pinealocytes and cortical neurons (Chik et al., 1999,2001; Yu et al., 1999), whether other pathways are involved inthe inhibitory effect of ceramide on INa needs to be explored infurther studies.

Several lines of evidence suggest that voltage-gated Naþ

channels might be modulated by CaM within neurons andmuscles (Dolmetsch et al., 2001; Young and Caldwell,2005). A large number of CaM-binding proteins have anisoleucine-glutamine (IQ) CaM-binding motif in the C-terminus(Bahler and Rhoads, 2002). Studies suggest that CaM can affectthe activity of NaV1.4 (Deschenes et al., 2002) and NaV1.5 (Tanet al., 2002) muscle Naþ channels by binding to their IQ motifs,and regulate the kinetic properties of NaV1.6 current in acalcium-dependent manner (Herzog et al., 2003). Usingreverse-transcription PCR, expression of the NaV1.2-NaV1.6 subunit was detected in cultured rat myoblasts

Fig. 8. Effect of C6-ceramide on intracellular Ca2R concentrationmeasured by Ca2R imaging. A: Continuous recording of intracellularcalcium concentration given by the ratio of fura-2 AM fluorescence at340 and 380 nm (340/380) from a myoblast using a fluorescencemicroscope equipped with a CCD camera. C6-ceramide (50 mM) wasapplied for 1 min and markedly increased intracellular Ca2R in areversible manner. B: CCD pictures showing the change of Ca2R

caused by C6-ceramide in a myoblast. The scale bar represents 10mm.C: C6-ceramide-induced changes of intracellular Ca2R obtained fromten cells are depicted. For each cell, the average steady value obtainedduring ceramide application was normalized to the value in normalRinger’s solution (control), and the normalized data were thenaveraged. Error bars represent SEM. MP< 0.05, compared with thecontrol.


(results not shown). Although we have no direct evidence forCaM binding to IQmotifs inNaþ channel subunits of myoblasts,the observation that W-7, a CaM blocker, significantlyabolished C6-ceramide-induced reduction of INa is an indirectevidence for CaM involvement. In addition, the contribution of


CaM to the effect of C6-ceramide on INa was highlighted by thedecrease of amplitude and the modulation of inactivationproperties in our study. This observation is consistent withYoung and Caldwell’s (2005) investigation of the CHOexpression system, in which voltage-dependent inactivationwas shifted to the left with co-expression of NaV1.4or NaV1.5 with CaM. Our results therefore suggest thatthe Ca2þ/CaM pathway was probably involved in theC6-ceramide-induced reduction of INa.

Our results here are the first to show that ceramide caninhibit INa by activating cellular signaling pathways. Thisinvestigation will be important for understanding themechanisms of the physiological, pathological, andpharmacological stimuli that trigger intracellular hydrolysisof sphingomyelin via sphingomyelinases and result in thegeneration of ceramide. In addition, the physiologicalsignificance of ceramide-induced reduction of INa inmyoblasts will be explored in the future.

Literature Cited

Bahler M, Rhoads A. 2002. Calmodulin signaling via the IQ motif. FEBS Lett 513:107–113.Berridge MJ. 1997. Elementary and global aspects of calcium signalling. J Physiol499:291–306.

BersDM, PattonCW,Nuccitelli R. 1994. A practical guide to the preparation ofCa2þ buffers.Methods Cell Biol 40:3–29.

Bielawska A, Crane HM, Liotta D, Obeid LM, Hannun YA. 1993. Selectivity of ceramide-mediated biology. J Biol Chem 268:26226–26232.

Bijlenga P, Liu JH, Espinos E, Haenggeli CA, Fischer-Lougheed J, Bader CR, Bernheim L. 2000.T-type a1H Ca2þ channels are involved in Ca2þ signaling during terminal differentiation(fusion) of human Myoblasts. Proc Natl Acad Sci USA 97:7627–7632.

Bock J, Szabo I, Gamper N, Adams C, Gulbins E. 2003. Ceramide inhibits the potassiumchannel Kv1.3 by the formation of membrane platforms. Biochem Biophys Res Commun305:890–897.

BourbonNA, Yun J, KesterM. 2000. Ceramide directly activates protein kinase C to regulatea stress-activated protein kinase signaling complex. J Biol Chem 275:35617–35623.

CheungWY. 1980. Calmodulin plays a pivotal role in cellular regulation. Science 207:19–27.Chik CL, Li B, Negishi T, Karpinski E, Ho AK. 1999. Ceramide inhibits L-type Ca21 channelcurrents in rat pinealocytes. Endocrinology 140:5682–5690.

Chik CL, Li B, Karpinski E, Ho AK. 2001. Ceramide inhibits the outward potassium current inrat pinealocytes. J Neurochem 79:339–348.

Deschenes I, Neyroud N, DiSilvestre D, Marban E, Yue DT, Tomaselli GF. 2002.Isoform-specific modulation of voltage-gated Naþ channels by calmodulin. Circ Res90:E49–E57.

DolmetschRE, PajvaniU, FifeK, Spotts JM,GreenbergME. 2001. Signaling to the nucleus by anL-type calcium channel-calmodulin complex through the MAP kinase pathway. Science294:333–339.

Fischer-Lougheed J, Liu JH, Espinos E, Mordasini D, Bader CR, Belin D, Bernheim L. 2001.Humanmyoblast fusion requires expression of functional inward rectifier Kir2.1 channels.J Cell Biol 153:677–685.

Gafni J, Munsch JA, Lam TH, Catlin MC, Costa LG, Molinski TF, Pessah IN. 1997.Xestospongins: Potent membrane permeable blockers of the inositol 1,4,5-trisphosphatereceptor. Neuron 19:723–733.

Gulbins E. 2003. Regulation of death receptor signaling and apoptosis by ceramide. PharmacolRes 47:393–399.

Gulbins E, Szabo I, Baltzer K, Florian L. 1997. Ceramideinduced inhibition of T lymphocytevoltage-gated potassium channel is mediated by tyrosine kinases. Proc Natl Acad Sci USA94:7661–7666.

Herzog RI, Liu C, Waxman SG, Cummins TR. 2003. Calmodulin binds to the C Terminus ofsodium channels Nav1.4 and Nav1.6 and differentially modulates their functionalproperties. J Neurosci 23:8261–8270.

Huwiler A, Kolter T, Pfeilschifter J, Sandhoff K. 2000. Physiology and pathophysiology ofsphingolipid metabolism and signaling. Biochim Biophys Acta 1485:63–99.

Kobrinsky E, Ondrias K, Andrew R. 1995. Marks1 expressed ryanodine receptor cansubstitute for the inositol 1,4,5-trisphosphate receptor in xenopus laevis oocytes duringprogesterone-induced maturation. Dev Biol 172:531–540.

Kobrinsky E, Spielman AI, Rosenzweig S, Marks AR. 1999. Ceramide triggers intracellularcalcium release via the IP3 receptor in Xenopus laevis oocytes Am J Physiol Cell Physiol.277:665–672.

Li PL, ZhangDX,ZouAP,CampbellWB. 1999. Effect of ceramide onKCa channel activity andvascular tone in coronary arteries. Hypertension 33:1441–1446.

Liu J, Jorgensen MS, Adams JM, Titlow WB, Nikolova-Karakashian M, Jackson BA. 2001.Ceramide modulates nicotinic receptor-dependent Ca2þ signaling in rat chromaffin cells.J Neurosci Res 66:559–564.

Lu BX, Liu LY, Liao L, Zhang ZH, Mei YA. 2005. Inhibition of Naþ channel currents in ratmyoblasts by 4-aminopyridine. Toxicol Appl Pharmacol 207:275–282.

Mathias S, Pena LA, Kolesnick RN. 1998. Signal transduction of stress via ceramide. Biochem J335:465–480.

Negishi T, Chik CL, Ho AK. 1998. Ceramide selectively inhibits calcium-mediatedpotentiation of beta-adrenergic-stimulated cyclic nucleotide accumulation in ratpinealocytes. Biochem Biophys Res Commun 244:57–61.

Numakawa T, Nakayama H, Suzuki S, Kubo T, Nara F, Numakawa Y, Yokomaku D, Araki T,Ishimoto T, Ogura A, Taguchi T. 2003. Nerve growth factor-induced glutamate release isvia p75 receptor, ceramide, and Ca2þ from ryanodine receptor in developing cerebellarneuron. J Biol Chem 278:41259–41269.

Occhiodoro T, Bernheim L, Liu JH, Bijlenga P, Sinnreich M, Bader CR, Fischer-Lougheed J.1998. Cloning of a human ether-a-go-go potassium channel expressed in myoblasts at theonset of fusion. FEBS Lett 434:177–182.

160 L I U E T A L .

Ruiz-Arguello MB, Basanez G, Goni FM, Alonso A. 1996. Different effects of enzyme-generated ceramides and diacylglycerols in phospholipid membrane fusion and leakage.J Biol Chem 271:26616–26621.

Ruvolo PP. 2003. Intracellular signal transduction pathways activated by ceramide and itsmetabolites. Pharmacol Res 47:383–392.

Schwartz PM, Borghesani PR, Levy RL, Pomeroy SL, Segal RA. 1997. Abnormal cerebellardevelopment and foliation in BDNF�/� mice reveals a role for neurotrophins in CNSpatterning. Neuron 19:269–281.

Tan HL, Kupershmidt S, Zhang R, Stepanovic S, Roden DM,Wilde AA, Anderson ME, BalserJR. 2002. A calcium sensor in the sodium channel modulates cardiac excitability. Nature415:442–447.

Young KA, Caldwell JH. 2005. Modulation of skeletal and cardiac voltage-gated sodiumchannels by calmodulin. J Physiol 565:349–370.


Yu SP, YehCH,Gottron F,WangX,GrabbMC,ChoiDW. 1999. Role of theoutward delayedrectifier k1 current in ceramide-induced caspase activation and apoptosis in culturedcortical neurons. J Neurochem 73:933–941.

Yu YC, Cao LH, Yang XL. 2006. Modulation by brain natriuretic peptide of GABA receptorson rat retinal ON-type bipolar cells. J Neurosci 26:696–707.

Zhang YH, Vasko MR, Nicol GD. 2002. Ceramide, a putative second messenger for nervegrowth factor, modulates the TTX-resistant Na þ current and delayed rectifier Kþcurrent in rat sensory neurons. J Physiol 544:385–402.

Zhang YH, Vasko MR, Nicol GD. 2006. Intracellular sphingosine 1-phosphate mediates theincreased excitability produced by nerve growth factor in rat sensory neurons. J Physiol575:101–113.

Zhou MO, Liu Z, Hu CL, Zhang ZH, Mei YA. 2004. Developmental regulation of a Naþ-activated fast outward Kþ current in rat myoblasts. Cell Physiol Biochem 14:225–230.

C6-ceramide inhibited Na+ currents by intracellular Ca2+ release in rat myoblasts

Documents

Transcript of C6-ceramide inhibited Na+ currents by intracellular Ca2+ release in rat myoblasts