Bid Cdl (1995) 83, 135-140 0 Elsevier, Paris

135

Original article

Functional and structural recovery of myotubes from mice with muscular dysgenesis after co-culture

with normal, non-myoblastic cells

Christian Dussartre a, Donatella Borrelli b, Michel Duvert c, Marina Melone b, Jeanine Koenig a*

aLaboratoire de Neurobiolo ie cellulaire, CNRS-URA 1126, Universite’ de Bordeaux II, avenue des fact&t%, 33405 Talence, France; 4 Cattedra E II Divisione di Neurologia i Facolta di Medicina e Chirurgia II

Universita degli studi di Napoli, Naples, Italy; ‘Laboratoire de cytologic, CNRS-URA 339, Universite’ de Bordeaux II, Talence, France

(Received 13 February 1995; accepted 4 July 1995)

Summary - Muscular dysgenesis is a mutation which is characterized by paralysis of skeletal muscle cells. Excitation-contraction coupling is deficient and muscle cells display atypical ultrastructure. In vitro, mutant myotubes recover a normal phenotype when co- cultured with spinal cord cells from normal animals or with normal fibroblasts. We have shown that other types of cells, eg certain glial cells present in the spinal cord or in other tissues, have this capacity. In contrast, intervention of neurons in the recovery does not appear likely. Very different types of non-myoblastic cells, then, are capable of restoring contractile activity of dysgenic myotubes in vitro, suggesting that a non-specific mechanism is involved in the phenotypic reversion of affected muscle cells. The restoration pro- cess seems to imply a close relationship between myotubes and normal glial cells.

muscle culture / myopathy / excitation-contraction coupling / glial cells / cellular fusion

Introduction

Muscular dysgenesis (mdg) is an autosomal recessive mutation which is lethal at birth. Skeletal muscles are unable to contract because there is a dysfunction in coupling between excitation, ie action potentials at the level of the plasma membrane, and release of calcium from the sarcoplasmic reticulum. Another defect is the ultra- structural disorganization of the myotubes: myofilaments are rarely arranged in sarcomeres, the sarcoplasmic reticu- lum is very dilated, the tubular system lacks development, with the transverse coupling structures weakly represented and devoid of junctional processes [7]. The mutation affects the gene coding for the a-l subunit of the dihydro- pyridine receptors (dihydropyridines are inhibitory drugs of slow calcium channels concentrated in the membrane of the T-tubules). Transcription seems perturbed and the pro- tein, although it has never been detected, could have a structure that is shortened or altered [2, 91. According to certain hypotheses, the a-1 subunit is a detector of voltage, and generates the inducing signal for calcium release from the sarcoplasmic reticulum [6].

Some dysgenic myotubes, co-cultured with spinal cord and dorsal root ganglion cells, can develop contractile activity and normal differentiation [5]. This population of cells is heterogeneous and comprises glial cells and fibro- blasts, in addition to neurons. It has been demonstrated that fibroblasts taken from normal animals, and the fibroblastic mouse cell line 3T3, are both able to restore excitation- contraction coupling and reorganization of the structure

*Correspondence and reprints

within dysgenic myotubes [3]. Schwann cells might also have this ability, but the presence of ‘contamin~ts’ in the primary cultures, in particular of fibroblasts, prevents iden- tification of the cells involved in this phenomenon [4]. We obtained similar results by co-culturing mutant myoblasts with spinal cord cells from embryonic chickens, which have a high percentage of neurons, and by co-culturing mutant myoblasts with several lines of glial cells. How- ever, lines of neuronal cells lack the ability to restore con- tractile activity. Requirements for phenotypic restoration and glial cell nuclei colocalisation with contracting mdg myotubes constitute data consistent with the hypothesis of heterocellular fusion (genetic complementation of the mutation) evoked by some authors [ 11.

Biological models

Cultures of mouse myoblasts

Myoblasts were obtained from the hind limbs of mdg/mdg (mutant) and mdg/+? newborns. mdg/+? are assumed to be heterozygote and are considered normal phenotypically.

Cocultures with mouse myoblasts

Embryo spinal cord cells and cell lines: C85 is obtained from fibrous-type astrocytes from the white matter of the cerebellum. D30 displays characteristics of protoplasmic astrocytes. C2L is taken from microglia (epi- thelial cells of Golgi). C6-12 is from a rat glioma. F7-48 is a dedifferentiated cell line (GFAP-, neurofilament-) from vasopressin-producing neurons of the hypothalamus. T28 is

a murine dopaminergic cell line produced by fusion of neu- rons of sympathetic ganglia with neuroblastoma cells. C85, D30 and C2L were kindly provided by Dr B Pessac (INSERM U178), C6-12 and F7-48 by Dr F Vitry (Collkge de France), and T28 was a generous gift of Dr Lazar (Cal.- lbge de France),

Materials and methods

Ceil cultures

Cultures ojmouse myoblasts Mouse myoblasts were harvested by a previously described pro- cedure [3] from newborns’ limb muscles following incubation with trypsin (0.125% for 15 min at 37°C) and mechanical disso- ciation. The cu!tured medium contained Eagles mitimal essential medium (MEM), 10% 199 medium, penicillin (20 U/ml), strepto- mycin (10 mglml), and 10% horse serum. The cells were seeded at 2 104 cells on 1% gelatin-coated 35mm dish and were grown at 37°C in a humidified atmosphere containing 7.5% CO,.

Co-cultures Different cell types were seeded onto d-day-old mouse muscle cell cultures. Because cell lines. proliferate when they are cul- tured alone, we seeded them at 104 cells per dish (unless other- wise specified). Spinal cord cells with their high proportion of neurons, which do not multiply, were seeded at 105 per dish. Preparation of chicken embryo spinal cord. Spinal cords were dissected from 4.5day chick embryos, treated with trypsin (0.125% for 15 min at 37°C) and dissociated mechanically. When the embryos are young and the dorsal roots of the spinal cord are eliminated, a population composed of more than 80% neurons is obtained.

Functional recovery

By daily observation of co-cultures, we have determined the day when contractions first appear. We arbitrarily chose the 18th day for the muscle ceil cultures to count the ratio of myotubes exhib- iting spontaneous or electrically indueed contraction per 35mm dish to the tottil number of myotubes. In each experiment, five dishes (with a total of about 5000 myotubes per experiment) were analyzed. Electrical stimulation (5 ms duration, 10 volt dis- charge) was performed with 70 pm separated extracellular plati- num wires, positioned less than 100 pm away from the cell.

In co-cultures with C6-12 bells, contracting myotubes were identified by scratching a circle around them in the culture dish using a diamond marker, to allow examination by immunocyto- chemistry and electron microscopy.

Labeling ofnuclei with Hoechst blue

The cultures were rinsed with phosphate buffer (0.02 M PBS, pH 7.4), fixed in Hanks’ solution containing 1% paraformalde- hyde for 20 min at 37”C, and then in pure methanol for 20 min at - 20°C. Following copious rinsing in distilled water, the cultures were incubated for 15 min at 37°C in a 0.13 pg/ml solution of Hoechst blue (H33258, Calbiochem) in 0.9% NaCl in distilled water. After several rinses in 0.9% NaCl solution, the cultures were mounted under a coverslip with a drop of PBS and were immediately viewed in an Olympus BH2 microscope at 350 nm.

Structural anuiysis qf myotubes

Immunocytochemistry An antibody (diluted 50-fold) directed against the heavy chain of adult human slow muscle myosin was used (362:B4, mouse iso- type IgG; kindly provided by Professor J Lkger, INSERM U300).

Cultures were fixed in 2% paraformaldehyde (Lean and Makane fixative) for 15 min at room temperature. After thorough rinsing, they were preincubated for 30 min in 0.02 M phosphate buffer, 1% BSA and 0.1% saponin, and then incubated with the first antibody-for 1 h 30 min at r&m temperature. After washing (4 x 5 min). they were incubated w&h a second antibody (diluted

x 100) labeled with a fluorescein derivative (FITCIDTAF, am- mouse, Amersham). After thorough rinsing, the cultures werrx mounted in diazabicyclo-octane and stored at -2O”C, beforr viewing in an Olympus BHZ microscope.

We quantified myotube organization according to the rr;\ns versafly striated or unstriated appearance of myosin distribution We analyzed myotubes, or 800 pm lengths of myotub?s; smaller cells were not considered. A myotube was considered striated ii at least SO% of its.

Electron microscopic evaiuatron Myotubes co-cultured with G-12 cells fur 6-10 days were fixed in 2.5% glutaraldehyde 0.067 M Na-cacodylate buffer (pH 7.5). 3% CaCl,, 2% PVP, 0.2 M sucrose, postfixed in 1% osmium tetroxide, and stained en bloc with 1% uranyl acetate, washed and then quickly dehydrated with alcohol, embedded in Epon, thin-sectioned, stained with uranyl acetate and lead citrate and viewed in a Philips EM-200 electron microscope. Semi-thin sec. tions of about i pm were stained with Unna blue in 0.2 M Na- cacodylate (pH 7.5).

The number of experiments performed is indicated by N.

Results

Compurisnns of restored contractile activity-induced 6~ different non-myablastic cell types

Under our culture~condiiions, mdg/+? and mdglmdg myo- blasts began to fuse around day 4 and form myotubes. After the 6th day, slime normal mdg/+? myotubes -ctiltivated alone contracted spontaneously. Their percentage increased progressively and reached about 20%. No contraction was seen with mdg/mdg myotubes cultivated alone.

The appearance of spontaneous contractions III co-cul- tures of mdg/mdg myotubes with normal non-myoblastic cells varied from one cell type to the next. Appr.oximateXy three categories of non-myoblastic cells could be distin- guished (table I): i) those responsible for a high rate of res- toration of contractile activity. This was the case for cell lines C.6-12 and C85 which brought contractions in about 4% of mdg/mdg myotubes; ii) those with a smallrestorative capacity, as D30 and C2L, with less than 1% of mdg/mdg myotubes contracting; and iii) cells of neural origin, as F’7-48 and T28, which were incapable of restoring Eontrac- tile activity in mdglmdg myotubes.

In cultures more than 18 days old, there was no notice- able change in the number of contracting myotubes. Haw- ever, cell proliferation did not stop; cells just detached and died after a certain density was.feached.

Although the percentage of restoration was variable, the day of appearance of first contractions after the start of co- culture seemed .to vary less from one culture to the next, regardless of restoring cell types.

Restoration of contractile activity by cell line C&-/2

Influence of the time of coculture We varied the times at which we seeded cells from line- (X-12 onto mdg/mdg muscle cell cultures (table II). We added at each specified time the same number of C&-Cl 2 cells that would have been preSent on that day had the cells been plated alone (using the Thoma cell after tryp&nization).

When C6-12 cells were placed in culture at the same time as dysgenic muscle cells, the first contractions appeared on the same day in mutant and normal myotubes. When C6-12 cells were added on the 4th or 8th days of culture, there was, in both cases, a delay of 5-6 days before the first contractions occurred. The highest per”centage oft

Functional and structural recovery of myotubes from mice with muscular dysgenesis 137

Table I. The percentage of contractile activity rescue differs with the normal cell type in coculture.

Cocultures-4th day * (104 cells per dish)

Number of experiments (4

Day first contractions appeared 0

Percentage of myotubes contracting 18th daya

Spinal cord 7 9.3 f 0.4 1.9 f. 0.4 C85 7 9 I!I 0.7 3.9 + 0.79 D30 3 9.5 L- 0.5 0.4 zk 0.25 C2L 3 9 IL 0.6 0.3 + 0.21 m-12 7 9.6 + 0.9 4.5 f 0.3 Means 9.2 zk 0.6 T28 4 0 0 m-48 4 0 0 mdg/+? 7 6.5 + 1.2 19.8 * 9.4

aDays since muscle cells plated in culture; +_ SD.

Table II. Variation of contractile activity rescue according to the date of co-culture.

Day coculture perjormed Number of 0%I2 cells seeded per dish

Dayfirst contractions appeared0

Percentage of myotubes contracting 18th daya

First 103 7 th 2.6 f 0.2 Fourth 2.104 10 th 5.2 lk 0.2 Eighth 105 14 th 2.8 IL 1.1 mdg/+? - 7th 19.1 f 9.4

n = 5; adays since muscle cells plated in culture; k SD.

restored contractions was observed when C6-12 cells were added to 4-day-old cultures. C6-12 cells seeded at a lower concentration when co-cultured with muscle cells had a tendency to develop colonies. Their spatial distribution was therefore heterogeneous.

Influence of the concentration of C6-12 cells We recorded the percentage of restored myotubes when S-12 cells were seeded at different concentrations on the 4th day of culture (table III). The day of appearance of con- tractions did not seem to depend on the initial concentration of co-cultured cells (data not shown). On the other hand, the percentage of myotubes restored increased with the concen- tration of C6-12 cells. We also observed myotubes exhibit- ing electrically induced contractions, and hence the exact percentage of cells restored. Good correlation was noted with the spontaneously contracting myotubes, although the percentage of myotubes capable of induced contractions was much higher (about x 3). There was a response to electrical stimulation in 89 5 1.5% of normal myotubes.

(S-12 cell nuclei (uniform staining). In about 25% of myo- tubes, analysis was hindered by the presence of numerous mononucleate cells around the myotube. However, we found that 71 + 10.1% (100 myotubes observed, n = 3) of contracting myotubes had a colocalized rat cell nucleus, ie it was possible to distinguish clearly a (X-12 cell nucleus, which did not extend beyond the contours of the myotube, was aligned with the muscle nuclei, and like them was elon- gated (figs 1, 2). Mdg/mdg not contracting or normal myo- tubes and 05-12 cell nucleic were occasionally coincident (less than 5%).

Structural differentiation of myotubes

Ultrastructure

Colocalization of rat nuclei and contracting mdg myotubes

Hoechst blue labeling of nuclear chromatin allowed diffe- rentiation between mouse nuclei (dotted appearance) and rat

After 18 days in culture, myotubes from normal mice had well-characterized ultrastructure, with easily recognized myofilaments organized into sarcomeres, and coupling structures equipped with obvious junctional processes (fig 3). In contrast, mdg/mdg myotubes cultured alone showed less developed contractile apparatus than normal myotubes, myofilaments were spread out and rarely arranged in sarcomeres (fig 4). Typical couplings were rare, and no junctional process was discernible.

Two different times were selected to observe the ultra-

Table III. Influence of 0%12 cell concentration on the percentage of contractile activity rescue.

Number of cells Cti-12 seeded Percentage of myotubes spontaneously Percentage of myotubes exhibiting electrically atday4* contracting - 18th daya induced contractions - 18th daya

0 0 0 2.103 1.3 f 0.4 - 104 4.5 + 0.3 11.9 f. 2.5 2.104 5.2 210.2 15.7 Y!I 1.5 5.104 5.8 * 0.4 17.12 -I- 1.9

n = 5; adays since muscle cells plated in culture; + SD.

Figs 1, 2. IO-day cocuhure of muscle cells from newborn dysgenic mites and rat C6-12 astrocytes. 1. Muscle cells viewed in wtute light. 2. The same cells labeled with Hoechst blue. Mouse cell nuclei are dotted, whereas the rat nuclei are uniformly stained. The arrow head indicates a rat nucleus at the level of a contracting mdg myotube (X 3 12).

structure of mdg/mdg myotubes that contracted when co- cultured with C6-12 cells: just after-the appearance of the first contractions (12th day of culture), and 4 days later.

At day 14 in culture, normal myotubes already exhibited myolilaments organized into sarcomeres which lie in regis- ter. Along sarcomeres, couplings were frequently observed. At the same time, in restored mdg myotubes, Z-lines were present but the myofilaments were organized in an impre- cise manner. Their distribution, often under the sarco- lemma, seemed to indicate early steps in differentiation pro- cess. Some organelles, suggesting couplings were discernible. By the 10th day in co-culture, some mdg myo- tubes had a more clearly differentiated organization: myo- filaments are arranged into sarcomeres, with the beginnings of transverse alignment. Couplings, with junctional pro- cesses, were readily recognizable (figs 5,6).

Distribution of myosin molecules C6-12 cells (2 lo4 cells per dish) were added to mdg/mdg myotubes on day 4, and myotubes contracting after 18 days of culture were circled. Immunocytochemical studies indicated that 81 f 5.62% (100 myotubes observed, n = 3) of restored myotubes had a striated distri- bution of myosin heavy chains (figs 7, 8). More surpris- ingly, 65% of all co-cultured myotubes were striated (64 f 4.5%, n = 3). This ratio was appreciably higher than that of myotubes contracting when stimuIated electrically (about 15%) or coinciding with C6-12 cell nuclei (about 20%), and that of reference mdg cultures (close to zero). In co-cultures with normal muscle cells, the number of striated myotubes did not change significantly compared with myotubes cultured alone (70.9 + 7.3% verSu.r 71 f 8.5%, n = 3).

Discussion

We have shown unambiguously that several types of glial cells are able to restore the contractiIe activity of mdg myo- tubes in vitro. The percentage of dysgenic myotubes that contracts in vitro varies according to the type of normal and non-myoblastic cells used in co-culture. This percentage may be influenced by the degree of proliferation and the migratory activity of these celIs. In contrast, the day of appearance of the first contractions is nearly constant from one type of culture to the next, regardless of restoring cell types, which suggests consistent temporal unfolding of the process involved. This fact, and the diversity of restoring

cell types, seem to indicate that the process of restoration ot contractile activity is not very specific.

Several results suggest that a close relationship inusr exist between myotubes and C6-12 cells for restoring contractile activity of dysgenic myotubes in vitro: the homogeneity of the distribution and the quantity of C6-12 cells in co-culture strongly affect the percentage of myotubes phenotypically restored, and C6-12 cell nuclei were distinctly aligned with the myogenic nuclei in N large number of contracting mdg/mdg myotubes. Besides. in some experiments (data not shown), we observed that conditioned media from the different glial cells.cultivated alone, always failed to restore contractile activity of mdg/mdg myotubes.

We have shown that when C6-12 cells are added from the start of culture, the first contractions appear at the same time in normal myotubes and in restored mdg myo- tubes, which~seems to indicate that the maturation process evolves in a similar way. Furthermore, the percentage-of myotubes restored increases when C6-12 cells are added on the 4th, rather than the 8th day: the 4th day-marks, Gr vitro, a very important phase for the fusion of myoblasts and the formation of myotubes. Progressively, the number of myotubes stabilizes and fusion becomes rarer. This finding suggests that the mechanism of restoration of con- tractile activity of mdg/mdg myotubes is tied to the fusion process. Certain authors have proposed that the mecha- nisms of restoration involve cell fusion and a genetic complement of the mutation by a normal gene from an exogenous nucleus. To support this hypohesis in the case of co-cultures of mdg muscle cells with normal fibm- blasts, they used nuclear markers (heterospecific labeling with Hoechst blue) or cytoplasmic markers (transforma- tion of fibroblasts by the gene for pgalactosidase). -and observed a marker for fibroblasts in the region where myotubes’ contractions occurred [Il. Other arguments in favor of the fusion hypothesis are very indirect: cellular homogenates or conditioned media from non-myoblastic normal cell cultures do not restore contractile activity 151

The hypothesis of fusion of non-myoblastic cells with myotribes may involve:

1) A particular aptitude of mdg myotubes to fuse, The t&t that the basal lamina of mdg myotubes is absent or very rudimentary [7] could favor, in part, these_ initial fusions Cultured myoblasts and mdg/mdg myptubes apparently fuse easily as the work of Yao and Essien [lo] suggests. Using equal numbers of mdg and normal myoblasts in se&u-ate cultures, with similar cell division cycles and levels of

Functional and structural recovery of myotubes from mice with muscular dysgenesis 139

Figs 3-6. Ultrastructure of normal and mdg cultured myotubes (18 days). 3. Normal myotube. The myofilaments are organized into sar- comeres with well defined Z disks (Z) and M lines (M). Between the sarcomeres, numerous coupling structures are discernible at the Z disks (arrows: coupling processes; x 18790). 4. Mdg myotube. The myofilaments are aligned but do not exhibit clear alternation be- tween clear and dark bands. No coupling structure is visible (X 18790). 5. Mdg myotubes with contractile activity restored by G-12 cells presenting quite well defined sarcomeric organization, with Z disks (Z) and M lines (M). The sarcomeres are aligned side-by-side (X 11590). 6. Mdg myotube. Couplings are clearly identifiable, such as the triad with regularly spaced densities between the terminal cistemae of the sarcoplasmic reticulum and the sarcolemma (arrowhead, junctional feet proteins: T, T-tubule; x 61500).

Figs 7, 8. Immunofluorescent staining of myosin heavy chain in mdg myotubes (l&day culture). 7. The labeling of the mdg myotubes cultured alone is diffuse. No cell was striated. 8. Most- mvotubes cultured with US-12 cells exhibited striations (arrowhead) albeg much of their length (X 312).

r

DNA synthesis, this author has shown that mdg myotubes incorporate many more nuclei than normal myotubes.

2) Phenotypic modifications of non-inyoblastic cells which enable fusion. Astrocytes and neurons are derived from neural tube of ectoderm at origin, whiie microglial cells and fibroblasts, like muscle cells, derive from meso- dermal layers. The most striking results were the signifi- cant restorations by astrocyte cell lines, notably- by C85, which are of a very different origin than myoblasts. This hypothesis implies that cellular phenotypes are less stable in vitro than in vivo, where specific environmental factors might influence cell maintenance.

However, we have not obtained such positive results with cell lines of neuronal origins P7-48 and T28. It has been shown that no restoration of contractile activity occurs, in co-cultures of mdgfmdg myoblasts with PC12 cells which stem from the adrenal medulla and can be transformed into cholinergic neuronal cells 181. It is note- worthy that embryonic chicken spinal cord cells, although full of neurons, are heterogeneous and contain, in particu- lar, glial and fibroblastic cells which, unlike neurons, are capable of division. The origin of the cells which modify the mutant phenotype remains to be defined. It is therefore possible that not all types of cells are capable of inducing in vitro excitation-contraction coupling and contractile activity in mdg/mdg myotubes. This might be because the determination or differentiation of certain cells of onto- genie origin (which differ greatly from myoblasts) is too established or stable to allow enough redifferentiation in culture for subsequent fusion with myoblasts.

At last, we observed that the structural organization of mdg myotubes with restored contractile activity was more differentiated between days 14 and 18 than that of refer- ence mdg cultures. The appearance of the first contrac- tions may precede structural maturation of mdg myotubes and induce this differentiation [3]. However, we have noted that the percentage of myotubes in cocuitures with striated distribution of myosin greatly exceeds that of myotubes able to contract or coincide with C6-12 cell nuclei. In contrast, a poorly organized structure was apparent in reference cultures of mdg myotubes. These results suggest that in cocultures with C6-12 cells there is a factor that stabilizes or induces structural differentiation of mdg myotubes, independently of the establishment of functional excitation-contraction coupling, and perhaps a heterocellular fusion process.

The phenotypic restoration of dysgenic myotubes by non-myoblastic normal cells in vitro may provide an orig-

inal model for the analysis of the myogenic cell differen- tiation and interactions with other cell types, in particular in rhe establishing excitation-contraction couplage and in the relation between contractile activity and structural differentiation (sarcomeric organization) or, if the henero-- cellular fusion hypothesis is confirmed, in the cellular fusion mechanism and in the studying of-the nuclear- domains of myotubes.

Acknowledgment

This work was supported by the French Muscular Dystrophy Association (AFM).

References

1

2

3

4

5

6

7

8

9

10

Chaudhari N, Delay R, Beam K (1989) Restauratioa of nor ma1 function in geneticaily defective myotubes by spontane ous fusion with fibrobiasts. Nature 341,445-447 Chaudhari N (1992) A single nucieotide deletion in the skei eta1 muscle-specific calcium channel transcript of Muscular Dysgenesis (mdg) mice. J Biol Chem 267,2S636-25639 Courbin P, Koenig J, Ressouches A, Beam K, Powelt J (1989) Rescue of excitation-contraction coupling in dys- genic muscle by addition of fibroblasts in vitro. Neuron 2. 1341-1350 Courbin P, Do Thi A, Ressouches A, Dussartre C,‘Powell 1, Koenig J (1989) Restoration of dysgenic muririe (mdg) myo tube contraction after addition of Schwann cells from nor- mal mice in vitro. Biol Cell 67,3SS-358 Koenig J, Bournaud K, Powell J, Rieger F (1982), &peat- ante of contractile activity in muscular dysgenesis (mdg/mdg) mouse myotubes during coeulture with normal spinal cord cells. Dev Biol92, 188-196 Lamb GD (1992) DHP receptors and excitation-contraction coupling. .I Micscle Res Cell Motil 13,3$M--4OS Platzer A, Gluecksohn-Waelsch S (1972) Fine structure ot mutant (muscular dysgenesis) embryonic mouse muscle. Dev Biol28,242-252 Powell J (1990) Muscular dysgenesis: a model system Co3 studying skeletal muscle deveiopment. FASEB J 4, 2798-2807 Tauabe T, Beam K, Powell J, Numa S (1988) Restoration of excitation-contraction coupling and slow calcium current in dysgenic muscle by dihydropyridine receptor complemen taty DNA. Nature~336, 134-139 Yao M, Essien F (1975) DNA synthesis and the cell cycle m culturesof not-mat and mutant (mdgimdg) embryonic mows muscle cells. fkv Riol45. lhh-175