Sb

OI

R

sts3tteameosiuussopmtpopm

i

gsdlo(X

0

Biochemical and Biophysical Research Communications 282, 55–59 (2001)

doi:10.1006/bbrc.2001.4502, available online at http://www.idealibrary.com on

keletal Muscle Sodium Channel Is Affectedy an Epileptogenic b1 Subunit Mutation

scar Moran1 and Franco Contistituto di Cibernetica e Biofisica, CNR, Via De Marini, 6. I-16139 Genoa, Italy

eceived February 9, 2001

A pathological condition in humans, generalized ep-i#ictssbieslcfccid(

mcstwbiCvptCc

M

egpwm

The syndrome of generalized epilepsy with febrileeizures plus type 1 (GEFS1) has been associated tohe gene SCN1B coding for the sodium channel b1ubunit (Wallace, R. H. et al. (1998) Nature Genetics 19,66–370). In patients, a mutation of the cysteine 121 torpyptophane (C121W) would cause a lack of modula-ory activity of the b1 subunit on sodium channelsxpressed in the brain, rendering neurons hyperexcit-ble. We have confirmed that the normal b1-odulation of type-IIA adult brain a subunits (BIIA)

xpressed in frog oocytes is defective in C121W. Webserved that the mixture of wild-type and mutant b1ubunits is less effective than wild-type alone, suggest-ng that the mutant b1 subunit does bind the a sub-nit. However, we also observed a similar lack of mod-lation by C121W of the in adult skeletal muscle aubunit (SkM1). This finding is in contrast with theimple idea that the mutational effect observed in theocyte expression system is the principal physio-athological correlate of GEFS1, because no skeletaluscle symptoms have been reported in GEFS1 pa-

ients. We conclude that the manifestation of theathological phenotype is conditioned by the presencef susceptibility genes and/or that the frog oocyte ex-ression system is inadequate for the study of theutant b1 subunit physiopathology. © 2001 Academic Press

Key Words: sodium channel; b1 subunit; GEFS1; ep-lepsy; frog oocytes.

The sodium channel b1 subunit is a membrane inte-ral protein that copurifies with the pore-forming aubunit. From its molecular identification (1), the so-ium channel b1 subunit has been proposed to modu-ate the sodium channel inactivation when it is heter-logously expressed with the adult skeletal muscleSkM1) or brain (BIA, BIIA, and BIIIA) a subunits inenopus oocytes (for review see (2–4)).

1 To whom correspondence should be addressed. Fax: (39)-105636532). E-mail: moran@barolo.icb.ge.cnr.it.

55

lepsy with febrile seizures plus type 1 (GEFS1; OMIM604236), an autosomal dominant hereditary disorder,

s directly associated with a b1 subunit (5). The defectonsists to a C-to-G transversion of nucleotide 387 ofhe SCN1B gene, predicted to resulting in a Cys to Trpubstitution in residue 120. The change of the con-erved Cys 121 most likely disrupts a putative disulfideridge that normally maintains an extracellularmmunoglobulin-like fold of hr b1 polypeptide (6). Co-xpression of the C121W with the BIIA channel aubunit in Xenopus laevis oocytes demonstrated that aack of the normal ability of the b1 subunit to modulatehannel-gating kinetics, consistent with a loss-of-unction allele (5). From these results, it has beenoncluded that the observed functional changes asso-iated with the C121W mutation may cause persistentnward sodium currents in neurons resulting in a moreepolarized membrane potential and hyperexcitabilityfor review see (7–10)).

GEFS1 patients do not exhibit cardiac or skeletaluscle symptoms (5). Since the same gene is known to

ode for b1 subunit that is also assembling heart andkeletal muscle sodium channels (11), we expected theissue-specificity of GEFS1 pathology to be correlatedith a differential effect of C121W b1-mutation on the1 interaction with the brain or the skeletal muscle a

soforms. Therefore, we have analyzed the effect of the121W mutation on the b1-modification of the inacti-ation of adult skeletal muscle sodium channels ex-ressed in frog oocytes. Surprisingly, we demonstratehat, in this preparation, the negative effect of the121W mutation is similar in brain and skeletal mus-le channels.

ETHODS

Mutagenesis sodium channel expression. Site-directed mutagen-sis of the cDNA of rat sodium channel b1 subunit (psPNa-beta,enerous gift from K. Imoto) was accomplished using the Pfu DNAolymerase using the QuikChange kit (Stratagene). The mutationas verified by sequencing 400 bp’s in the flanking region near theutagenic point (M-Medical, Florence).

0006-291X/01 $35.00Copyright © 2001 by Academic PressAll rights of reproduction in any form reserved.

Plasmids containing the full-length cDNA coding for the adult ratssr(et

lcnc1dmcd

fa(0f1wctNto41p1tEsta

p(trsso

R

tbftcatma(ale

Sinsn

eaawd5ma

pwr

ftodor2

Vol. 282, No. 1, 2001 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

keletal muscle (rSkM1) (12) and the rat brain type-II (rBIIA) (13)odium channel a-subunits were linearised with NheI and SalI,espectively. Plasmids containing the cDNA coding for the wild-typeWT) and mutant (C121W) b1 were linearised with EcoRI. Lin-arised plasmids were transcribed in vitro with T7 polymerase usinghe capped mMessage mMachine kit (Ambion).

Oocytes were surgically extracted from anaesthetized Xenopusaevis frogs, and isolated enzymatically by treatment withollagenase-A (Sigma). Stage IV and V oocytes were injected with 50l cRNA solutions in DEPC-treated water. a-cRNA (rSkM1 or rBIIA)ontained 125 ng/ml of cRNA. b1-cRNA (WT or C121W) contained50 mg/ml of cRNA, or two to fourfold dilutions of the later for b1ose-response measurements. For WT-C121W competition measure-ents, both WT and C121W b1-cRNA were coinjected to the highest

oncentration (150 ng/ml). Injected oocytes were incubated for 5 to 7ays in Barth’s solution at 18°C.

Electrophysiological recording. Sodium currents were measuredrom the whole oocyte using a two-electrode voltage clamp designednd constructed in our laboratory. Borosilicate glass micropipettesHilgemberg) were filled with 3 M KCl and had a resistance of 0.2 to.6 MV. The oocyte bathing solution (normal frog Ringer) had theollowing composition (in mM): NaCl 115, KCl 2.5, CaCl2 1.8, HEPES0, pH 5 7.4. In some cases, when the expression of sodium currentsas too high (.20 mA), compromising the accuracy of the voltage-

lamp, despite a large percentage of series resistance compensation,he currents were reduced by substituting part of the extracellulara1 by K1, to a final NaCl concentration of 30 mM. No differences in

he inactivation properties were detected in low Na1 solutions. Theutput of the voltage-clamp amplifier was filtered by a low-pass-pole Bessel filter with a cut-off frequency of 5 kHz and sampled at0 kHz. The oocytes were normally kept at a holding membraneotential of 2100 mV, Pulse stimulation and data acquisition used6 bit D-A and A-D converters (ITC-16, Instrutech), and were con-rolled by a Macintosh microcomputer with the Pulse software (Hekalectronik). Linear responses were estimated from sub-thresholdtimulations, partially compensated analogically and digitally sub-racted with a standard P/4 protocol. All measurements were done atcontrolled temperature of 15 6 0.5°C.

Data analysis. Data were analyzed with a Macintosh microcom-uter using custom software developed in the IgorPro programWavemetrics). Statistical comparisons were done with a Student’s-test, and statistical significance was defined by P , 0.05. Theesults were expressed as mean 6 sem (number of measures). Aimplex method was used for curve fitting. Data for each a and b1ubunit combination were obtained from 3 to 6 different batches ofocytes.

ESULTS

Following a step depolarization from a holding po-ential, Vh 5 2100 mV, the sodium current mediatedy the expression of a-rSkM1 or a-rBIIA alone shows aast activation and a biphasic inactivation, as illus-rated in Figs. 1A and 2A. The falling phase of theurrent is well fitted by a double-exponential functions I(t) 5 a1exp(2t/t1) 1 a2exp(2t/t2), with fast and slowime constants, t1 and t2, differing by about one orderagnitude, and with a ratio of the two amplitudes a1

nd a2 fairly independent of the depolarizing voltage14, 15). It has been proposed that the two componentsre due to the separate contribution of a mixed popu-ation of channels that at the time of the stimulus areither in a fast or in a slower mode of inactivation.

56

ingle channel recordings have shown that in the fastnactivating mode, which we shall call M1, the chan-els open once and briefly during the depolarizingtimulus, while in the slower mode, called M2, chan-els have late reopenings during the sweep (16–19).To quantify the propensity of the channels to gate in

ither mode, we defined the parameter PM2 5 [a 2/(a 1 12)] 3 100, as a measure of the probability of findingchannel in mode M2 during any test stimulus. PM2

as measured from the double exponential fit of theecaying phase of the sodium current evoked by 30 to0 ms depolarization to voltages between 220 and 0V. We previously reported that the estimates of PM2

re fairly independent of the test voltage (14, 15).In the series of experiments reported here the ex-

ression of a-rBIIA cRNA yielded sodium currentsith PM2 5 16.7 6 2.8 (n 5 30) (Fig. 1A). As previously

eported (1, 3, 6, 20, 21), the propensity of the channels

FIG. 1. Adult rat brain sodium channels (rBIIA) expressed inrog oocytes. The rBIIA a subunit alone (A) has a slow component inhe inactivation phase, which is strongly reduced by the coexpressionf WT b1 subunit (B). Differently, coexpression of C121W b1 subunitoes not reduce the slow inactivation component (C). Traces werebtained from two-electrode voltage-clamp experiments, and cur-ents were evoked by a 210 mV pulse from a holding potential of100 mV.

foiPi

for increasing amounts of coinjected b1-cRNA, ap-pbr

smc1sbdtaM

Cmysw

sce((wTssb

pcttnowor2

Vol. 282, No. 1, 2001 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

or the mode M2 is strongly reduced (P , 0.0002) inocytes coinjected with WT b1-cRNA (Fig. 1B), result-ng in a PM2 5 4.8 6 2.5 (n 5 27). The dependence of

M2 on the amount of coinjected WT b1-cRNA is shownn Fig. 3A (filled symbols). It is seen that PM2 decreases

FIG. 2. Adult rat skeletal muscle sodium channels (rSkM1) ex-ressed in frog oocytes. The rSkM1 a subunit alone (A) has a slowomponent in the inactivation phase, which is strongly reduced byhe coexpression of WT b1 subunit (B). Differently, coexpression ofhe C121W b1 subunit does not reduce the slow inactivation compo-ent (C). The coexpression of the rSkM1 a subunits with a mixturef WT and C121W b1 subunits produce an intermediate phenotype,here the slow component is only partially reduced. All traces werebtained from two-electrode voltage-clamp experiments, and cur-ents were evoked by a 210 mV pulse from a holding potential of100 mV.

57

roaching a plateau at the maximum concentration of1-cRNA that have used, corresponding to a molaratio of b1:a cRNA larger than 10:1.Similarly, injection of a-rSkM1 cRNA also expresses

odium currents with a relatively high propensity forode M2 (PM2 5 11.8 6 1.3; n 5 53), that is signifi-

antly reduced (P , 0.0001) by the coexpression of50 ng/ml of WT b1-cRNA (20, 22, 23) (Fig. 2B). Ashown in Fig. 3B (filled symbols), the effect of the WT1 subunit on the modal inactivation of a-rSkM1 alsoepends on the cRNA concentration (22), and also inhis case the injection of 50 nl of b1-cRNA 150 ng/mlppears to produce a near maximal depression of mode2 (PM2 5 4.8 6 0.9; n 5 27).As reported by others (5), we also observed that the121W mutation of b1 apparently fails to modulate theodal inactivation of the a-rBIIA channels (Fig. 1C),

ielding a PM2 5 22.5 6 2.5 (n 5 38), that is nottatistically different (P . 0.12) from control levelsithout the b1 subunit (Fig. 3A, open symbols). This

FIG. 3. The propensity of the sodium current to inactivate in thelow mode, PM2, is plotted as a function of the cRNA concentrationoding for the b1 subunit injected in the oocyte. It is shown by theffect of coexpression of WT (filled symbols) and C121W b1 subunitempty symbols) on the currents expressed by rBIIA (A) and rSkM1B) a subunits. The broken lines are empirical interpolations of dataith PM2 5 A 1 [B/(1 1 [cRNA]/k)], for clarity of the presentation.he triangle in panel B represents data obtained from the coexpres-ion of the rSkM1 a subunit with a mixture of WT and C121W b1ubunit, both injected as 150 ng/ml cRNA. Data represent means andars are sem.

result is consistent with the simple interpretation oftaipttooco

ftdibCtopbwmiytaw

D

srg(pmtabttguirocplt

iob

muscle (11) and likely functionally assembled with thes(ndtctettas

fmtdibmtotnlposoiuccwt

A

Gw

R

Vol. 282, No. 1, 2001 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

he physiopathological effects of the GEFS1 mutations due to lack of inhibition of the M2 mode of the brainsoforms of the sodium channel a subunit. Quite sur-risingly, however, we discovered that the C121W mu-ation loses also the ability to inhibit the M2 mode ofhe muscle isoform a-rSkM1 (Fig. 2C). The expressionf C121W with the later phenotype shows a high valuef PM2 5 13.8 6 2.5 (n 5 27), that is also not statisti-ally different (P . 0.45) from controls (see Fig. 3B,pen symbols).The lack of modulation of C121W-b1 for both iso-

orms does not seem to result from a reduced interac-ion, since we see no tendency of the reduction of PM2 toecrease with in increasing amounts of C121W-cRNAnjection (Figs. 3A and 3B, empty symbols). One possi-ility for explaining the lack of modulatory effect of the121W b1 subunit is that this polypeptide is not able

o assemble with the a subunit in general, or in the frogocyte expression system in particular. To test thisossibility, we coinjected oocytes with 150 ng/ml of WT1-cRNA and 150 ng/ml of C121W b1-cRNA togetherith the a-rSkM1 cRNA. We found that the expressedixture of WT and mutant b1 subunits produced an

ntermediate phenotype of sodium currents (Fig. 2D),ielding a PM2 5 8.2 6 0.9 (n 5 20). This indicates thathe C121W b1 polypeptide is indeed expressed andntagonizes the correct assembly of the WT b1 subunitith a subunit (6).

ISCUSSION

The association of GEFS1 type 1 syndrome with aodium channel subunit (5) is very noteworthy because itepresents the first case of a putative sodium channelenetic defect affecting the central nervous systemCNS). Several reviews have the GEFS1 mutation as arototype of genetic CNS disorders produced by a pri-ary sodium channel dysfunction (7–10). Several muta-

ions of the sodium channel a subunits in skeletal musclend heart, liked to muscular and cardiac diseases, hadeen identified in the last 12 years (24, 25), and all ofhem show an impaired inactivation as a primary func-ional defect. Even if the issue is still controversial, it isenerally accepted that the functional role of the b1 sub-nit in the sodium channel complex is the modulation of

nactivation (1, 2, 4). Therefore, following the originaleport of a lack of modulatory effect of the C121W mutantf b1 in the frog oocyte preparation, it was natural toonsider theoretically such defective modulation as therimary cause of the neuronal hyperexcitability that isikely accompanying the epileptic attacks of GEFS1 pa-ients (5).

The clinical report of the GEFS1 type 1 syndromendicates that patients do not show any skeletal muscler cardiac symptoms (5), despite the fact that the same1 subunit is expressed also in skeletal and cardiac

58

pecific a subunit isoforms expressed in these tissues2–4). The simplest explanation for the apparentlyondefective function of heart and skeletal muscle so-ium channels could be a specificity of the mutated b1o associate with the brain isoforms of the sodiumhannel a subunits. However, our present data showhat the C121W mutation has a negative modulatoryffect also on the a-SkM1 expressed in oocytes, andhat probably the defective polypeptide competes withhe normal b1 for the association with the a subunit,nd this is in contradiction with the proposed isoform-pecificity of the C121W b1 subunit.There is a contradictory body of information on the

unctional role of the sodium channel b1 subunit inammalian cells. It has been demonstrated that the

ransfection of a subunits alone in mammalian cellsoes not express channels with large PM2 values like ast does in oocytes (26–35), and that the coexpression of1 subunit does not alter substantially the channelsodal properties (26, 27, 34, 36). This rises the ques-

ion of whether the observations reported from frogocyte experiments are indeed useful to understandhe sodium channel physiology or rather an epiphe-omenon occurring in these particular cells. In the

arge family affected by GEFS1 type 1 the mutationroduces distinct types of seizure in different membersf the same family (5). These differences probably re-ult from inheritance of the mutant gene in the contextf other susceptibility genes (8), that may express onlyn the CNS, and confer functional role for the b1 sub-nit. One proposed possibility is that the b1 serves asommunication link between intracellular and extra-ellular environments, as suggested by its interactionith extracellular matrix proteins (37) and cytoskele-

on intermediate filaments (38).

CKNOWLEDGMENTS

We thank L. Ferrera for the preliminary experiments and E.aggero for the construction of the voltage-clamp amplifier. Thisork was partially supported by Telethon Grant 926.

EFERENCES

1. Isom, L. L. et al. (1992) Science 256, 839–842.2. Kallen, R. G., Cohen, S. A., and Barchi, R. L. (1994) Mol. Neu-

robiol. 7, 383–428.3. Isom, L. L. D. J., K. S., and Catterall, W. A. (1994) Neuron 12,

1183–1197.4. Fozzard, H. A., and Hanck, D. A. (1996) Physiol. Rev. 76, 887–

926.5. Wallace, R. H. et al. (1998) Nature Genetics 19, 366–370.6. McCormick, K. A., Isom, L. L., Ragsdale, D., Smith, D., Scheuer,

T., and Catterall, W. A. (1998) J. Biol. Chem. 273, 3954–3962.7. Lehmann-Horn, F., and Jurkat-Rott, K. (1999) Physiol. Rev. 79,

1317–1372.8. McNamara, J. O. (1999) Nature 399, A15–A22.

9. Steinlein, O. K., and Noebels, J. L. (2000) Curr. Opinion Genet.

11

11

1

1

1

1

1

1

2

22

22

25. Lehmann-Horn, F., and Rudel, R. (1996) Rev. Physiol. Biochem.

2

2

223

3

3

3

3

3

3

3

3

Vol. 282, No. 1, 2001 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Develop. 10, 286–291.0. Ryan, S. G. (1999) J. Child. Neurol. 14, 58–66.1. Makita, N., Bennet, P. B., and George, A. L. (1994) J. Biol. Chem.

269, 7571–7578.2. Trimmer, J. S. et al. (1989) Neuron 3, 33–49.3. Noda, M., Ikeda, T., Kayano, T., Suzuki, H., Takeshima, H.,

Kurasaki, M., Takahashi, H., and Numa, S. (1986) Nature 320,188–192.

4. Moran, O., Melani, R., Nizzari, M., and Conti, F. (1998) J. Bioen-erg. Biomemb. 30, 579–588.

5. Moran, O., Nizzari, M., and Conti, F. (1999) J. Bioenerg.Biomemb. 31, 591–608.

6. Moorman, J. R., Kirsch, G. E., VanDongen, A. M. J., Joho, R. H.,and Brown, A. M. (1990) Neuron 4, 243–252.

7. Schreibmayer, W., Wallner, M., and Lotan, I. (1994) PflugersArch. 426, 360–362.

8. Zhou, J., Potts, J. F., Trimmer, J. S., Agnew, W. S., and Sig-worth, F. J. (1991) Neuron 7, 775–785.

9. Chang, S. Y., Satin, J., and Fozzart, H. A. (1996) Biophys. J. 70,2581–2592.

0. Patton, D. E., Isom, L. L., Catterall, W. A., and Goldin, A. L.(1994) J. Biol. Chem. 269, 17640–17655.

1. Smith, R. D., and Goldin, A. L. (1998) J. Neurosc. 18, 811–820.2. Cannon, S. C., McClatchey, A. I., and Gusella, J. F. (1993)

Pflugers Arch. 423, 155–157.3. Chen, C., and Cannon, S. C. (1995) Pflugers Arch. 431, 186–195.4. Bulman, D. E. (1997) Hum. Mol. Genet. 6, 1679–1685.

59

Pharmacol. 128, 159–268.6. An, R. H., Wang, X. L., Kerem, B., Benhorin, J., Medina, A.,

Goldmit, M., and Kass, R. S. (1998) Circ. Res. 83, 141–146.7. Kazen-Gillespie, K. A., Ragsdale, D. S., D’Andrea, M. R., Mattei,

L. N., Rogers, K. E., and Isom, L. I. (2000) J. Biol. Chem. 275,1079–1088.

8. Sheets, M. F., and Hanck, D. A. (1999) J. Physiol. 514, 425–436.9. Sarkar, S. N., and Sikdar, S. K. (1994) Curr. Sci. 67, 196–199.0. Sarkar, S. N., Adhikari, A., and Sikdar, S. K. (1995) J. Physiol.

488, 633–645.1. Ukomadu, C., Zhou, J., Sigworth, F. J., and Agnew, W. S. (1992)

Neuron 8, 663–676.2. Chahine, M., Bennet, P. B., George, A. L., Jr., and Horn, R.

(1994) Pflugers Arch. 427, 136–142.3. Moran, O., Nizzari, M., and Conti, F. (2000) FEBS Lett. 473,

132–134.4. Hayward, L. J., Brown, R. H., Jr., and Cannon, S. C. (1996)

J. Gen. Physiol. 107, 559–576.5. West, J. W., Scheuer, T., Maechler, L., and Catterall, W. A.

(1992) Neuron 8, 59–70.6. Isom, L. L., Scheuer, T., Brownstein, A. B., Ragsdale, D. S.,

Murphy, B. J., and Catterall, W. A. (1995) J. Biol. Chem. 270,3306–3312.

7. Xiao, Z. C., Ragsdale, D. S., Malhotra, J. D., Mattei, L. N., Brau,P. E., Schachner, M., and Catterall, W. A. (1999) J. Biol. Chem.37, 26511–26517.

8. Malhotra, J. D., Kazen-Gillespie, K., Hortsch, M., and Isom, L. L.(2000) J. Biol. Chem. 275, 11383–11388.