Calcium currents in normal and dystrophic human skeletal muscle cells in culture

cell caldwn (laao) ii, 507-514 Q LongmanGroup UK Ud 1980

Calcium currents in normal and dystrophic human skeletal muscle cells in culture M. RIVET’,

C. COGNARD’,

Y. RIDEAU2, G. DUPOR?

and G. RAYMOND’

’ Laboratoire de Physiologic G&&ale, CNRS U.R.A. 290, Universit6 de Poifiers, Poitiers, France 2Service de R&adaptation Fonctionneiie, Centre Hospitaker Universifaire, Poitiers, France 3Service de Chirurgie G&&ale, Centre Hospitalier Universitaire, Poitiers, France Abstract - Human muscle cells obtained from biopsy specimens were grown in a primary culture system and electrophysiologically studied. Whole cell patch-clamp recordings revealed the presence of two types of calcium currents: (i) a low-threshold (-60 mV) one (Ica,T) with fast activation and inactivation kinetics (time-to-peak: 39 ms at -30 mV); and (ii) a high-threshold (-10 mV) one (Ic& with slower kinetics (tim&to-peak: 550 ms at 20 mV). These two types of calcium currents could be also distinguished by their pharmacological characteristics since IG,,L was sensitive to the antagonist and agonist dihydropyridine derivatives contrary to ICE,,T which was completely resistant to these compounds. These functional calcium channels existed both in normal and Duchenne dystrophic (DMD) human skeletal muscle cells in culture. We discuss a possible role of these two types of calcium channels in the myoplasmic calcium accumulation observed in the Duchenne muscular dystrophy. An anomalous calcium accumulation has been detected, at rest, in the myoplasm of muscle cells from patients with Duchenne muscular dystrophy (DMD) [l-4]. This intracellular elevation of calcium concentration is unlikely to be the consequence of late physical membrane damage since the calcium concentration rises before any alteration of the sarcolemma integrity and the consequent release of certain cytosolic proteins, such as creatine kinase [5]. The reasons for this elevated myoplasmic calcium level must be sought in an alteration of the mechanisms involved in calcium homeostasis, such as transport systems located in the plasma membrane [l, 3, 4, 61, sarcoplasmic reticulum [7] or mitochondria [8]. Binding experiments [9, lo] with dihydropyridine derivatives have suggested the existence of voltage-dependent calcium channels in

the membrane of human skeletal muscle cells which could underlie the development of long-lasting action potentials in developing and in adult human skeletal muscle [6, 111. However, recent studies, [12], have demonstrated that these compounds can be bound by non-permeable molecular structures such as the voltage-sensor responsible for the excitation-contraction coupling in skeletal muscle. Therefore the presence of functional Ca2’ channels required direct evidence. The present work was conducted to demonstrate electrophysiologically this presence and to look for some possible differences between normal and pathological muscles. The whole cell patch-clamp experiments revealed two types of voltage-dependent calcium currents in both normal and DMD muscles without significant differences in kinetics and voltage-dependencies. 507

508

Matmialsand Methods Electrophysiological experiments were performed on cultured skeletal muscle cells. This experimental model has been often used in studies of disorders [ 131 since some neuromuscular well-known characteristics of DMD are present in cell cultures from dystrophic muscles, e.g. the absence of dystrophin at the inner face of the plasmalemma [ 141 or the intracellular Ca2+ accumulation 141. Cell culture Muscle samples were obtained during orthopedic surgery of patients (18-35 years) without neuromuscular diseases and of DMD children (4-13 years) with agreements from the local Committee of Medical Ethics. The DMD diagnosis was made on the basis of cmatine phosphokinase level (normal level < 200 IU, DMD level was lo- to loo-fold greater), clinical symptoms (frequent falls, difficulty climbing stairs, positive Gowers sign, waddling gait) and histological studies (degenerative changes in muscle fibers, infiltration of muscle by adipose and connective tissues). Primary cultures were prepared from the myoblasts (satellite cells) [15]. Muscle biopsies, trimmed of any visible fat and connective tissues, were minced finely into 1 mm3 fragments in Ham FlO medium (Gibco). Then, they were incubated at 37°C with continuous stirring in calcium- and containing 0.2% (w/v) magnesium-free saline trypsin (Seromed) and 0.1% collagenase (Sigma, Type II). The undigested pieces were dissociated two more times. The supernatants were centrifuged (1400 rpm for 4 min) and the pellets resuspended in a growth medium: Ham FlO + 10% fetal calf serum (Boehringer Mannheim) + 10% horse serum (Gibco) + 1% chick embryo extract (Gibco). After filtration (nylon netting: pore size 25 pm), the myoblast suspension (10’ cells per ml) was plated (2 ml per dish) in 35 mm gelatin (0.5%, Sigma) coated dishes and incubated in a water-saturated atmosphere of 95% air-5% CO2 at 37°C. The growth medium was changed three times each week. The cells were grown in this medium until the alignment stage. Then, after 12-15 days, to induce fusion of

CELL

CALCIUM

myoblasts into myotubes, it was replaced by a maintenance mediun~ Dulbecco’s modified Eagle medium (Gibco) with 5% horse serum. All culture media contained penicillin G (100 U/ml, Sigma) and streptomycin (50 pg/ml, Sigma). For the electrophysiological studies, the cultures were used from 17 to 36 days after the plating. Current recordings The selected myotubes had a non-branching geometry and a small size with a diameter typically ranging from 10 to 15 pm and a length from 300 to 500 pm. Transmembrane currents were recorded at room temperature (18-22C) with the whole cell patch-clamp technique [ 161. Patch electrodes (2-5 M&2) were prepared from glass hematocrit tubes (soft glass, Assistent, Bardram, Denmark) and connected to the head-stage of the patch-clamp amplifier (RK 300, Biologic, Grenoble, France) IBM PC-AT compatible driven by an microcomputer @PC-AT turbo, Essex Electric PTE Ltd, Singapore) equipped with an A/D-D/A conversion board (Labmaster TM 40, Scientific Solutions, Solon, USA). Data acquisition and analysis were performed by means of a software package (pClamp, Axon Instruments, Foster City, USA). In a first time, the cell-attached configuration was realized (the resistance of the seals ranged from 10 to 33 G&Jwith a mean value, + SEM (n = 18), of 19.17 + 1.95 GQ). The establishment of the whole cell configuration was obtained by application of a brief additional suction to the pipette interior. Membrane patch rupture was associated with a in membrane capacitance. sudden increase Depolarizations were applied from a holding potential of -90 mV. The cell membrane capacity (PF) was estimated by integration of the area beneath the capacitive transient current generated by a 10 mV depolarizing pulse. The membrane capacity (Cm) values in normal myotubes were 180.46 f 10.84 pF (mean i SEM, n = 76) with a minimum and a maximum of 58 and 400 pF respectively. In dystrophic myotubes, Cm was 213.64 + 12.31 pF (mean + SEM, n = 56) with a minimum and a maximum of 39 and 389 pF

DYSTROPHIC

SKELETAL

MUSCLE CEiLLS

: Ca’+ CURRENTS

a & -60

200pAI Is b 1 (PA) 100

I-

so9

respectively. Unless stated in the figure legends, data were sampled at 714 Hz and duration of depolarizing steps was 1.8 s. The pipette contained in mM: 145 CsCl, 1 MgClz, 0.005 CaClz, 1 EGTA, 5.6 glucose, pH 7.2 with Tris base. The cultnre medium was exchanged before each experiment with the reference bath solution (in mM: 135 TEA-Cl, 2.5 CaCla, 0.8 MgCla, 5.6 glucose, 10 HEPES, pH 7.4 with TEA-OH). Na+ ions were absent from the salt solution to abolish the sodium current and were substituted with equimolar quantities of TEA (as the chloride salt). Internal Cs+ ions and external TEA-Cl were used to block or minimize outward currents. Nifedipine (Sigma) and the racemic mixture of Bay K 8644 (Calbiochem) were added from a concentrated stock solution (10 mM in DMSO) kept in the dark. The maximum concentration of the solvent in the experimental solutions was 0.005%. DMSO was added to all control solutions at a final concentration corresponding to that of DMSO in the nifedipine or Bay K 8644 saline solutions.

Results Figure la shows two types of inward currents elicited by long-lasting depolarizations t?om a holding potential of -90 mV. Depolarizing pulses in the range -50 to -10 mV induced a fast activating and inactivating current (time-to-peak 40 ms at -30 inactivation time constant 31 ms). mV, Depolarizations higher than -10 mV elicited an additional, slower (time-to-peak 890 ms at 20 mV) inward component which progressively overlapped the transient one. The high-threshold component exhibited a very slow inactivation process. The I-V curve of inward currents, plotted in Figure lb, Fig. 1 Two components (a) Patch-clamp

of inward currents in human muscle.

recordings

of whole-cell

currents in normal muscle induced by various depolarizing

of -90 mV to the test value indicated to the right of each trace. (Myotube capacitance

depolarizations. depolarixing

(tilled circles) as the membrane (c) Two examples muscular

The open squares correspond

pulses to -20, -10 and 0 mV. For

dystrophy

to lo,20

to the amplitude

of the slow component

and 30 mV the fast component

current amplitude measured at 32 ms (time-to-peak

of inward currents induced by

pulses from holding potential

(HP)

(C) - 260 pF, 29 days in culture)

of the fast component

amplitude

at the end of the has been estimated

at 0 mV)

to -30 and 10 mV from HP - -90 mV in myotubes from Duchenne (upper traces C - 200 pF, 33 days in culture; lower traces C - 200 pF, 19 days in culture)

510

CELLCALCIUM

pAl

400

1 s

500

ms

Fig. 2 Ionic nature of inward currents.

(a) Inward currents elicited by depolarizations in solution containing

1.8,2.5

(b) The two components (-40

of inward

and 10 mV) in the presence

originating

to -30 and 20 mV

and 10 mM calcium current

induced

of 2.5 mM Ba”.

by pulses Myotubes

from muscles of DMD patients

(a) C - 240 pF, 35 days in culture;

(b) C - 200 pF, 36 days in

culture

exhibits a threshold potential around -60 mV and a maximal peak amplitude at -30 mV for the fast inward current. The slow component reached a maximum amplitude at 20 mV and tended to zero above 60 mV. Since the two inward currents were kinetically distinct, the amplitude of the transient type has been estimated also at 10, 20 and 30 mV (filled circles) while the slow type amplitude overshot the first one: this procedure indicates a reversal potential around 25 mV. In addition, at -20, -10 and 0 mV the slow current has been measured at the end of the depolarizing pulse (open squares) revealing a threshold potential near -20 mV. These two hinds of inward currents could be also recorded in myotubes cultured from Duchenne dystrophic muscle as shown in Figure lc. Owing to the bath solution composition (absence of permeant cations other than Ca2’> it can be suggested that the inward currents were calcium ones and that they flow through calcium channels since they could be recorded in the presence of tetrodotoxin at high concentration (5 x 10m7M, not shown). The calcium nature of these inward currents was confirmed by the experiments, illustrated in Figure 2a, which were recorded in the presence of

different external calcium concentrations. The higher was the calcium concentration in the bath, the greater were the amplitudes of the two types of inward currents. In addition, the two currents could be carried by barium ions (external calcium replaced by 2.5 mM Ba2’** Fig. 2b) as expected for calcium channels. These permeation properties and the other characteristics of the calcium currents, studied here, have been observed both in normal and dystrophic cells. Compared to muscle cells from normaI muscle, no significant difference was observed in DMD muscle cells in the voltage-dependence and kinetics of these two kinds of current. The time-to-peak (at -30 mV) of the fast component ranged from 7 to 73 ms with a mean value of 39.0 + 1.3 ms (mean + SEM, n = 115). This value was 37.7 & 1.9 ms in normal myotubes (n = 60) and 40.4 k 1.9 ms (n = 55) in DMD myotubes. For the slow component the time-to-peak (at 20 mV) ranged from 171 to .1673 ms with a mean value of 550 _+22 ms (mean + SEM, n = 124). This value was 561 + 26 ms(n=67)and537+38ms(n=57)innormaland DMD myotubes respectively. The inactivation time constant of the transient current measured in 116 cells ranged from 21 to 86 ms with a mean of 42.0 f 1.4 ms (mean + SEM). This value was 42.3 + 1.7 ms in normal myotubes (n = 59) and 41.8 f 2.1 ms in DMD myotubes (n = 57). Therefore the following results are indifferently illustrated ti-om experiments performed on normal or dystrophic myotubes. Nevertheless, in spite of a large variability in peak amplitudes, data from 111 cells showed that peak amplitudes of the fast component were higher in DMD myotubes (209 f 28 pA, mean f SEM, n = 61) than in normal myotubes (129 + 15 pA, mean k SEM, n = 50), contrary to peak amplitudes of the slow component which were similar in DMD myotubes (192 + 17 pA, mean f SEM, n = 61) to those measured in normal myotubes (203 f 15 pA, mean + SEM, n = 51). The two types of currents can be characterized by their distinct voltage-dependent inactivation properties (Fig. 3a,b,c). The curves of Figure 3c, drawn from the experiments of Figure 3a and 3b, denote large discrepancies in the half-inactivation (around -60 mV and -15 mV respectively) and full inactivation (around -40 mV and 10 mV

DYSTROPHIC

SKELETAL

b

a

20

-40

ms

PA(

I

Is

20

I/I

max

V (mV) Fig. 3 current potential

Inactivation

on the two types of calcium

solution.

on the fast current

schematized prep&e

experiments

in 2.5 mM Ca”

In (a). the effect of holding

was investigated

at the top of the figure

(5 s in duration)

with a protocol

(potentials

in mV).

A

of various levels (-90 to -30 mV by 10

mV steps) is followed by a test pulse (500 ms in duration) at -30 mV. The current depolarization prep&e

peak amplitude

and

plotted,

was measured

in (c), (open

during the test

circles)

value. In (b), a similar protocol

potential

the slow current:

very long pmpulses

versns

the

was used for

(60 s) were applied (-60

mV to 10 mV by 10 mV steps) before a test pulse to 20 mV. As in (a), the slow current prep&e

value

depolarization. pulses (separated different

(open Current

(a) and (b), values during

amplitude,

squares),

plotted

in (c), against

was measured

during

traces during the prepulses

frequencies: indicated

the test pulses

clarity only a selection

with two

17 and 500 Hz respectively. (in mV) near the current

correspond

to the prepulse

the

the test

and the test

by a dashed line) have been recorded

acquisition

511

MUSCLE CELLS : Ca’+ CURRENl’S

the inactivation process of the high-threshold calcium current was very slow since several tens of seconds were necessary for complete inactivation. As shown in Figure 3b, during the prep&e, 60 s was barely sufficient for a total inactivation, particularly at -20 mV and -10 mV. Therefore the curve (open squares) in Figure 3c is probably slightly distorted with regard to a relationship obtained with a true steady-state inactivation protocol. In addition, it is possible that the inactivation is not purely voltage-dependent. On the basis of the permeation properties, the kinetics and the voltage-dependence of activation and inactivation processes of the fast and slow channels reported above, we termed them If&T (‘transient’) and Ica,~ (‘long-lasting’) respectively (as indicated in Fig. 3c) according to the usual classification [17]. ka,T and Ica,~ cm be also distinguished using different pharmacological agents. As shown in Figure 4a, 5 p.M nifedipine selectively inhibited Ica,L without marked effects on ICa,T. The addition of 1.5 mM Cd2+ in the nifedipine-containing solution led to a blockade of the transient calcium current. A partial recovery of the two types of calcium currents was observed after washout of the blocking agents. The I-V curve for ICa,T (in the presence of nifedipine; Fig. 4b - filled circles), indicated a reversal potential between 20 and 30 mV which agrees with the estimation derived from the experiment of Figure la. ICa,L could be enhanced specifically by application of 1 p&I Bay K 8644 (Fig. 5a). The I-V curves pig. 5b) denote a two-fold increase in the Ica,L amplitude and a 20 mV shift towards negative values of the potential which induced the maximal current in the presence of 1 j&I Bay K 8644 whereas ICa,Tis IriOtaffected.

In

traces

values.

For

Discussion

of current traces is shown. Vertical scale

in (b) is the same as in (a). Myotubes

originating

from muscles of

DMD patients. (a) C = 180 pF, 34 days in culture; (b) C - 300 pF, 34 days in culture

respectively) potentials of the fast and slow components of the currents. As stated previously,

The three main findings of this study can be summarized as follows: (i) functional calcium channels are present in cultured human skeletal muscle cells; (ii) two types of calcium currents can be characterized on the basis of their kinetics, voltage-dependencies and pharmacology; and (iii)

512

CBLL CALCIUM

0

control

+ nifedlplns

nlfedbine + Cd*+

washout

-i

9-

extensively studied in both cultured and neonatal cells [24-3 11. The present results are the first direct evidence of the existence of functional voltage-dependent calcium channels in the membrane of human skeletal muscle cells. The antagonist/agonist effects of nifedipine/Bay K 8644 on IQ,L agree with the DHP binding experiments 19, 101 on the basis of

a + Boy K 8644

control

Cd’*

Fig.4 Pharmrteological characterization of hT

and IQ,J_.

(a) Membrane currents induced by three different depolarizing pulses (-30, 0, 20 mV) successively

in control solution, in the

presence of 5 pM nifedipine, in the presence of 5 pM nifedipine and 1.5 mM cadmium and after washout of the calcium channel blockers (b) I-V curves obtained from the experiments illustrated in (a).

b

The membrane current amplitudes have been measured as the relative minimum value during the depolarizations.

1(PA)

DMSO was

V (mV)

present in the control solution at the same concentration as in the

1oo

1

nifedipine containing solution. (Myotube from nonnal muscle, C - 194 pF, 17 days in culture)

these two types of current can also be recorded in

myotubes from DMD human muscle. Calcium currents have been already recorded in a wide range of cellular types [173. In skeletal muscle, calcium currents were first described in the membrane of frog muscle [18-201 and afterwards demonstrated in mammalian adult fibres [21, 221. These initially reported currents exhibited L-type high threshold potential, slow characteristics: kinetics and sensitivity to organic blockers. The existence of diiferent types of calcium channels @rticularly of T-type), has been demonstrated in developing mammalian skeletal muscle cells [23, 241 and their characteristics, time-dependent evolution and possible physiological functions

0 cDntro1 l + Bay K BB44

Ffg. 5 Bay K 8644 agonistio effect on Ic+ (a) Two types of calcium current induced by depolarizations (-40, -20, -10,lO

mV) in control solution and in the presence of

a racemic mixture of 1 ph4 Bay K 8644 (b) I-V

curves from the experiments

illustrated in (a). The

current amplitude has been measured as the relative minimum value during the step potential. DMSO was added to the control solution

at the same concentration

as in the Bay K 8644

containing solution. (Myotube originating from muscles of DMD patient, C - 240 pF, 35 days in cultme)

DYSTROPHJC

SKELETAL

MUSCLE CELLS : 01”

513

CURRENTS

which such an existence has been postulated. ICa,T and Ic~,L, characterized in this paper, have low-lhighdifferent properties: well-marked threshold potential, fast/slow kinetics, strongly/ weakly negative half-inactivation potential and DHP- insensitivity/sensitivity. These properties are standard for the two types of calcium currents recorded from various preparations and particularly from other skeletal muscle cells and from cardiac cells [17]. Nevertheless, IC~L inactivation kinetics are slower in skeletal than in cardiac muscle [17] and slower in human (time constant around 10 s, this work) than in other mammalian skeletal muscle such as rat muscle (time constant around 1 s [32]) at the same temperature. The reasons for these discrepancies are not clear. Differences in the structural properties of the slow calcium channels or in their mode of regulation could be involved. However, a problem of maturity could be relevant because human myotubes in primary cultum did not contract contrary to rat myotubes [6, 131. The existence of functional calcium channels in the membrane of human skeletal muscle cells suggests that they might be involved in the intracellular calcium elevation in DMD muscle cells [ 111. This is unlikely due to some alterations of Ic~,Lsince no obvious difference was observed between normal and DMD muscle cells, in agreement with the previous observations of Desnuelle et al. [9] on DHP-binding sites. However the possibility of some involvement of calcium channels in the pathological Ca2+ accumulation in DMD muscle cells cannot be ruled out since, in spite of large data disparity, our measurements showed that Ic~,T peak amplitude was higher in DMD myotubes than in normal myotubes. This point requires ‘further experiments. Moreover a alteration of calcium channel progressive characteristics and/or regulation during the innervation/maturation processes could result from the lack of dystrophin - the product of the human DMD locus 133, 341 - which is localized beneath the muscular plasma membrane [35, 361. The dystrophin could be involved in the links between the cytoskeleton and the sarcolemma and its absence in DMD muscle cells could lead to a plasma membrane instability and/or alterations to the presence or the activity of the ionic channels [37].

Acknowledgements M.R.has a

fellowship from the Association Fran@se contre les Myopathies. The authors thank J. Alix and J.-P. Poindessault for their technical assistance.

References 1. Bodenstciner JB. Engel AG. (1978) Jntracellular calcium accumulation in Duchenne dystrophy and other myopathies: a study of 576,000 muscle fibers in 114 biopsies. Neurology, 28,439-446. 2. MaunderSewry CA. Gorodetsky R. Yamm R Dubowitx V. (1980) Element analysis of skeletal muscle in Duchenne muscular dystrophy using X-ray fluorescence spectmmetty. Muscle Nerve, 3,502~508. 3. Bert&n TE. Bhattacharya SK. Pahnieri GMA. Chesnoy CM. Pifer D. Baker B. (1982) Muscle calcium and magnesium content in Duchemre muscular dystrophy. Neurology, 32, 1088-1092. 4. Mongini T. Gh&o D. Doriguxxi C. et al. (1988) Free cytoplasmic Ca at test and atIer cholincrgic stimulus is increased in cultured muscle cells from Duchenne muscular dystrophy patients. Neurology, 38,476-480. 5. Duncan CJ. (1989) Dystmpbin and the integrity of the samolemma in Duchm muscular dystrophy. Experientia, 45.175-177. 6. Rothman SM. Bischoff R (1983) Electtophysiology of Duchenne dystrophy myotubes in tissue culture. AM. Neural., 13, 176-179. 7. Landi N. Nassi P. Ligmi G. Bobbi S. Sbrilli C. Marconi G. (1986) Levels and properties of sarcoplasmic reticulum Ca’*-ATPase in Duchenne muscular dystrophy. Muscle Nerve, 9.207a. 8. Lucas-Heron B. Mussini JM. Ollivier B. (1989) Is them a maturation defect related to calcium in muscle mitochondria from dystrophic mice and Duchenne and Becker muscular . . ~~._ __. dystrophy patients7 J. Neuml. Sci., 90,299-306. 9. Desnuelle C. Renaud JF. Delpont E. Serratrice G. Lazdunski M. (1986) 3H-nitrendipine rytors as markers of a class of putative voltage-sensitive Ca + channels in normal skeletal muscle and in muscle from Duchenne muscular dystrophy patients. Muscle Nerve, 9, 148-151. 10. Desnuelle C. Askanas V. Ensel WK (1988) Dihydropyridinesensitive Ca + channel in aneurally cultured human musclea. Relationship between high-a&&y binding site and inhibition of calcium uptake. FBBS Len., 230, 95-100. 11. Nicola-Siri L. Dubrovski AL. Uchitel OD. (1988) Ca-dependent slow action potentials in human skeletal muscle. J. Cell. Physiol., 137, 448-454. 12. Rios E. Bmm G. (1987) Involvement of dihydropyridine receptors in excitation-contraction coupling in skeletal muscle. Nature, 325,717-720. 13. Witkowski JA. (1986) Tissue culture studies of muscle disorders: part 1. Techniques, cell growth, morphology, cell surface. Muscle Nerve, 9, 191-207. 14. Miranda AF. Bonilla E. Martucci G. Moraes CT. Hays AP.

CELL CALCIUM

514

15. 16.

17. 18.

19. 20. 21.

22.

23.

24.

25.

26.

27.

28.

DiMauro S. (1988) Immunocytochemical study of dystrophin in muscle cultures from patients with Duchenne muscular dystrophy and unaffected control patients. Am. J. Pathol., 132.410-416. Maum A. (1961) Satellite cells of skeletal muscle fibers J. Biophys. B&hem. Cytol., 9,493-495. Hamill OP. Marty A. Neher E. Sakmann B. S&worth FJ. (1981) Improved patchclamp techniques for high-resolution current recordings from cells and cell-free membrane patches. Pfltlgers Ax&., 391, 85-100. Bean BP. (1989) Classes of calcium channels in vertebrate cells. Amm. Rev. Physiol., 51, 367-384. Bernard C. Cardinaux JC. Potmau D. (1975) Long duration responses and slow inward current obtained from isolated skeletal Iibms with barium ions. J. Physiol., 256, 18P-19P. Beaty GN. Stefani E. (1976) Inward calcium current in twitch muscle fibtes of the frog. J. Physiol., 260,27P-28P. Stanfield PR. (1977) A calcium dependent inward current in frog skeletal muscle fibms. PBi@m Arch., 368,267-270. Donaldson PL. Beam KG. (1983) Calcium currents in a fast-twitch skeletal muscle of the rat. J. Gcn. Physiol., 82, 449-468. Walsh KB. Bryant SH. Schwartz A. (1986) Fiifect of calcium antagonist drugs on calcium currents in mammalian muscle tibers. J. Pharmacol. Exp. Ther., 236, 403-407. Beam KG. Knudson CM. Powell JA. (1986) A lethal mutation in mice eliminates the slow calcium current in skeletal muscle cells. Nature, 320,168-170. Cognard C. Lazdunski M. Romey G. (1986) Different types of Ca2’ channels in mammalian skeletal muscle cells in culture. Pmt. Natl. Acad. Sci. USA, 83,517-521. Beam KG. Knudson CM. (1988) Calcium currents in embryonic and neonatal mammalian skeletal muscle. J. Gen. Physiol., 91,781-798. Beam KG. Knudson CM. (1988) Effects of postnatal development on calcium currents and slow charge movement in mammalian skeletal muscle. J. Gen. Physiol., 91,799-815. Cognard C. Lazdunski M. Romey G. (1986) Different types of voltage-dependent calcium channels in rat skeletal muscle cells in culture: electrophysiological and pharmacological properties. J. Physiol., 371,264P. Gonoi T. Hasegawa S. (1988) Post-natal disappearance of transient calcium channels in mouse skeletal muscle: effects of denervation and culture. J. Physiol., 401.617-637.

29. Rieger F. Boumaud R. Shimahara T. Garcia L. P&on-Raymond M. La&n&i M. (1987) Restoration of dysgenic muscle contraction and calcium channel function by co-culture with normal spinal cord neumns. Nature, 330, 563-566. 30. Romey G. Garcia L. Dimitriadou V. P&on-Raymond M. Rieger F. Laxdunski M. (1989) Ontogenesis and localization of Ca2’ channels in mammalian skeletal muscle in culture and role in excitation-contraction coupling. Proc. Natl. Acad. Sci. USA, 86,2933-2937. 31. Tanabe Y. Beam KG. Powell JA. Noma S. (1989) Restoration of excitation-contraction coupling and slow calcium current in dysgenic muscle by dihydropyridine receptor complementary DNA. Nature, 336,134139. 32. Cognard C. Romey G. Galixzi JP. Fosset M. Laxdunski M. (1986) Dihydropyridinesensitive Ca2’ channels in mammalian skeletal muscle cells in culture: electrophysiological properties and interactions with Ca2’ channel activator (Bay K 8644) and inhibitor (PN 200-l 10). Proc. Natl. Acad. Sci. USA, 83, 1518-1522. 33. Ho&num EP. Brown RH. Kunkel LM. (1987) Dystrophim the protein product of the Duchenne muscular dystrophy locus. Cell, 51, 919-928. 34. Hoffmann EP. Fischbeck KH. Brown RI-I. et al. (1988) Characterization of dystrophin in muscle biopsy specimens from patients with Duchenne’s or Becker’s muscular dystiophy. N. Engl. J. Med., 318, 1363-1368. 35. Zubtzycka-Gaam EE. Buhnan DE. Karpati G. et al. (1988) The Duchenne muscular dystrophy gene product is localized in sarcolemma of human skeletal muscle. Nature, 333, 466-469. 36. Campbell KP. Kahl SD. (1989) Association of dystmpbin and an integral membrane glycoprotein. Nature, 338, 259-262. 37. Arahata K. Sugita H. (1989) Dystrophin and the membrane hypothesis of muscular dystrophy. Trends Pharmacol. Sci., 10,437-439. Please send reprint requests to : Dr G. Raymond, Laboratoire de Physiologie G&t&ale, CNRS U.R.A. 290, Universit6 de Poitiers, F-86022 Poitiexs Cedex, France. Received : 28 March 1990 Revised : 30 July 1990 Accepted : 31 July 1990