insulin-independent controlled physiological morphogenesis of chick muscle from fusion-capable...

THE JOURNAL OF EXPERIMENTAL ZOOLOGY 238:63-70 (1986)

I nsu Ii n-l ndependent Controlled Physiological Morphogenesis of Chick Muscle From Fusion-Capable Myoblasts

OSCAR RAMkEZ AND VICTOR ALEMAN Departments of Biochemistry (0. R.) and Neurosciences (VA.), Centro de Investigacidn y de Estudios Avanzados del lnstituto Politecnico Nacional, 07000 Mexico, D.E:

ABSTRACT Thigh myogenic cells from 11-12-day-old chick embryos were cultured continuously in the presence of medium containing no chick embryo extract (CEE). It is known that CEE contains a muscle-inducing protein of 35,000 daltons. In spite of the absence of embryo extract and provided that calcium, starting at a concentration as low as 3 x lop4 M, was present in the tested media, typically aligned myotubes with 20 or more nuclei per fiber or abnormal myosymplasts were produced at will. In the first case, the result was systematically obtained when the media were unchanged. Consequently, the cell microenvironment remained undisturbed and therefore was autocondi- tioned throughout the 7 days of culture. In the second case, the result depended on the feeding schedules. Conversely, no myotubes were formed in cultures in embryo extract-free medium without calcium, irrespective of the frequency of medium changes. Insulin, a serum factor believed to be involved in syncytium formation process in uitro, was present in all tested media. Undialyzed or dialyzed fetal calf serum (FCS), used for the preparation of the media, con- tained 11 punits of insulin per milliliter. The insulin content in all tested media was diluted, however, to one tenth the physiological serum concentration. The hormone did not promote any kind of myoblast fusion in any experiment in which calcium was deleted as a component of the tested media, regardless of the feeding schedule followed. Contrary to former claims, the present results demonstrate that, as in the initial development of chick heart and skeletal muscle in uiuo, insulin, which is not produced in significant amounts until day 12 of incubation, does not participate in the early differentiation of skeletal muscle in uitro. These results lend further support to the idea of the existence of an intrinsically 'fusion-capable' myoblast.

The culture of skeletal muscle has been a favorite object of biological study since the advent of in uitro tissue culture techniques (Lewis and Lewis, '17). The initial studies were especially concerned with the observa- tion that cultured muscle tissue could con- tract spontaneously in the absence of innervation. More recently, and particularly since the development of dispersed cell culture methods (Konigsberg, '63), muscle cultures have been used to study a broad range of biological problems (Hauschka, '72).

It is worth emphasizing, however, that problems of interpretation occur when the various muscle culture methods are used by different investigators. Such methods may influence behavior of muscle cells in ways

that are not immediately obvious to the un- initiated. Seemingly trivial technical varia- tions, such items as cell sources, different serum types (or even serum lots), preparation of CEE, culture dish surface, method of dis- sociating cells, and initial cell inoculum to list but a few, may be responsible for appar- ently inconsistent published results (the fail- ure of one group of investigators to confirm the findings of another). As such they merit special consideration in work that attempts to repeat or extend experimental observa- tions from other laboratories.

Traditionally, primary cultures of differen- tiating skeletal muscle are initiated for at

Address reprint requests to Oscar Ramirez.

0 1986 ALAN R. LISS. INC

64 0. RAMIREZ AND V. ALEMAN

least 17-24 h (Shainberg et al., '71; Doering and Fischman, '77), or grown continuously in the presence of rich media (Hauschka, '72; Turner et al., '76; Turner, '78). These media usually contain as undefined components both the serum of a higher vertebrate and CEE (Hauschka, '72; Konigsberg, '71; Rami- rez and Aleman, '72a).

A protein component, from the CEE in trunks of 9-11-day-old chick embryos (Born et al., '72a), induces muscle formation in the competent ectoderm in uiuo (Tiedemann, '68; Tiedemann, '76). This component, as well as Ca", seemed to be necessary for the typical fusion process. This process consists of the production of well-aligned myotubes containing 20 or more nuclei per fiber (Shainberg et al., '69; Ozawa, '72).

Mutated rat myoblasts from the Ls cell line, which lack the mitochondria1 creatine kinase isoenzyme (Van Brussel et al., '831, were initially cultured for 24 h in medium containing 10% FCS but no CEE. This medium was then replaced by any of the following: 1) 1.5 mg of bovine serum albumin per ml of serum-free medium, 2) 1.0% FCS, and 3) 10% FCS. In all cases, 1.0 pg of insulin per ml of medium was added when the initial medium was changed. Under all these conditions, atypical symplasts (myotubes) similar to those found earlier in primary culture by Ramirez and Aleman ('72b) were formed (Mandel and Pearson, '74). The last authors ascribed the stimulation of myogenesis to insulin. Similar results have been published with regard to the L84 myogenic cell line (Yaffe and Jaxel, '77).

The aims of the work reported below are: 1) to determine, if normal thigh myogenic cells of 11-12-day-old chick embryos are al- ready capable of fusing typically in the absence of CEE in primary cultures and 2) to investigate the role that Ca2+, insulin, and conditioned medium play in normal syncytium formation in medium free of embryo extract.

METHODS

Rhode Island eggs were used for fibro-myogenic cultures (Ramirez and Aleman, '72a). Cells from thigh muscle of 11-12-day-old chick embryos were seeded at an initial inoculum of 3 x lo5 cells in a 100 mm diameter Falcon 3003 plate, previously coated with gelatin (Ramirez and Aleman, '72b, '73). The cells were grown at 37°C in a humidified incubator in an atmosphere of 95% air plus 5% c02.

The standard nutritional medium for control experiments consisted of nine parts of Eagle's Minimum Essential Medium with calcium-free Earle's balanced salt solution, for suspension cultures (MEM from GIBCO), and one part FCS (DIFCO). Normal components of this serum were 3 mM Ca2+, as determined by emission spectrometry; insulin, which averaged 11 punits per ml as determined by a radioimmune assay (Herbert et al., '65), and 5.9 mg of protein per ml (Lowry et al., '51). We termed 'MEMS' the complete medium with normal FCS; MEMSD is MEM plus dialyzed FCS. To eliminate low molecular weight metabolites (Ca2+ among them) in MEMSD, the serum was dialyzed three times at 12-h intervals against 50 vol- umes of calcium-free phosphate buffered-saline (Dulbecco and Vogt, '54). Dialyzed serum also averaged 11 punits of insulin and 5.5 mg of protein per ml. Calcium concentration was adjusted in the media by adding CaClz from a sterile stock solution to the desired final concentration. Ten ml of the chosen medium were initially supplied to at least six 100 mm diameter Falcon 3003 plates, previously coated with gelatin (Ramirez and Aleman, '72b), and, as indicated above, an inoculum of 3 x lo5 cells was seeded (Ramirez and Aleman, '73).

The degree of terminal differentiation of 11-12-day-old chick embryonic muscle cells was assessed in the different media accord- ing to the following feeding schedules: a) MEMS (containing 0.3, 1.2, or 1.5 mM Ca2+) was renewed on days 2,4, and 6; b) MEMSD was also renewed on days 2, 4, and 6; c) Plates containing either MEMS or MEMSD were left undisturbed throughout a week; d) MEMSD plus Ca2+ (from 0.3 to 1.2 mM) were left undisturbed for 1 week. At the end of the experimental period, the plates were rinsed with Saline G solution and fixed in 4% form- aldehyde. Anilin blue-orange and Delafield haematoxylin were used to stain the mono- layers which were then covered with 5% polyvinylic alcohol. The experiments were performed in quadruplicate. A Zeiss POL mi- croscope was utilized for microphotographs.

RESULTS

When cultures were fed with standard MEMS, Ca2+-rich MEMS, or MEMSD sup- plemented with Ca2+ at all the indicated concentrations, and such CEE-free media were changed periodically, atypical symplasts (myotubes) developed as early as the third day of culture (Fig. 1). Very wide fibers

INSULIN-INDEPENDENT PHYSIOLOGICAL CHICK MYOGENESIS 65

Fig. 1. Initial morphology of two myosymplasts, some isolated highly elongated myoblasts of about 100 pm, and a fibroblast-like cell. The structures developed in Fig. 2. Further development of myosymplasts as Ca2+-containin$ +media changed periodically (MEMS, grown under conditions mentioned in Figure 1. The pop- MEMS plus Ca , or MEMSD plus Ca2+). Notice the ulation consists of fibroblast-like cells, bipolar myo- large number of myogenic nuclei arranged in more than blasts, short and long normal myotubes, and abnormal one row, in an abnormally ovoidal sarcoplasm, whose myosymplasts. The insulin content was one tenth the length does not exceed three long myoblasts. ~ 2 0 0 . serum concentration in every case. X200.

were seen with an axial ratio of approximately 5:l containing in their cytoplasm at least 20 nuclei arranged in two, three, or four rows. At the end of the experiment, the total length of such fibers was only approximately four times that of an elongated myoblast (Figs. 2,3). As shown previously (Ramirez and Aleman, '72a), disorganization within the syncytium was also observed and no sarcomeres were seen. When compared with cultures grown in media containing CEE (Turner, '78), the population of MEMS plates was normal and healthier but seemed rather scanty. Similar photographs taken from the rat-derived transformed L6 and L84 muscle cell lines grown in the presence of supra- physiological levels of insulin were published some years later by Mandel and Pearson ('74) and Yaffe and Jaxel('77).

In contrast to these last two reports, in our cultures (in undisturbed plates, fed on either

MEMS with Ca2+ concentrations as indicated or MEMSD plus Ca2+, all in the presence of 1.1 punits of insulin per ml of medium) formation of 'typical fibers' oc- curred in every case (Shainberg et al., '69; Ramirez et al., '75). It must be noted that 1.1 punits of insulin per ml of medium represent a tenth or a sixth of the physiological vertebrate serum concentration for this hormone, since at hatching Raheja et al. ('72) found 6 punits of insulin per ml of chick serum.

This syncytium included several long mul- tinucleated fibers ( 2 2 mm long and, by def- inition, more than 20 nuclei per fiber), measured also in sister cultures performed in 3030 Intergrid Falcon Petri plates of 4 mm2 grid. This result was systematically obtained when the CEE-free medium was never changed. Consequently, the microenvironment remained undisturbed and was conditioned throughout 7 days in the Intergrid


Fig. 3. Selected fields of confluent plates showing the terminal syncytia at the end of the experimental period of culture. Notice the abnormal ovoidal formation of symplasts which consistently appeared in Ca2+-containing media that were changed periodically. The Ca2+ content in the media ranged from 0.3 to 1.5 mM. The insulin concentration in every case averaged only 1.1 punits per ml of medium (1/10 the serum content). X60.

plates. Myotubes without sarcomeres were observed in phase contrast as early as days 3 and 4, interspersed among fusing myoblasts and fibroblast-like cells (Fig. 4). The yield of myotubes observed in a t least four similar plates, fed on embryo extract-free medium containing from 0.3 to 1.5 mM Ca2+, ranged from 20-30% of that obtained in rich media at day 5 of primary culture by Ramirez and Aleman (‘72a).

In contrast with the other cultures de- scribed, a great number of unfused cells reached confluence in plates containing MEMSD, either renewed periodically or left undisturbed for 1 week. It is worth mention- ing that insulin concentration in this Ca2+- free medium was also 1.1 punits per ml (Fig. 5). Also, i t was interesting that a t the end of the experimental period three main types of

cell populations clearly emerged: a) myotubes in all Ca2+-containing media, indiffer- ent of the particular low nonphysiological concentration of insulin (1/10 or 1/6 the normal serum content); b) myoblasts that did not fuse, present mostly in plates with ‘Ca2+- free’ medium, containing the same range of insulin concentration as the above, and c) fibroblast-like cells, present in all types of plates.

DISCUSSION

Buckley and Konigsberg (‘74) suggest, from experiments performed with quail muscle cells fed on 10% CEE and 15% horse serum- containing medium, that prior to fusion no myogenic cells are truly post-mitotic and withdrawal from the cell cycle is a trivial consequence of fusion.

However, Holtzer et al. (’74) advanced the idea of the existence of a distinct myoblast cell type with a pre-programmed intrinsic capacity for fusion “occupying the last compartment in the myogenic lineage.” Turner (‘78) seemed to support this idea when he succeeded in obtaining chick myotubes from cultured leg and breast muscle cells of 11-day chick embryos in a rich medium, supple- mented with the DNA-synthesis inhibitor, arabinoside-c (ara-c), at the time of plating. Although ara-c precludes the emergence of new “fusion-capable” cells from inocula pre- sumably containing both nondividing myoblasts and dividing precursor myogenic cells, the existence of a distinct and mature terminal myoblast, independent of the contribu- tion of extracellular factors for the fusion process, cannot be proved conclusively since the muscle cultures were performed in a medium containing CEE. It is known that the in uivo muscle-inducing protein of 35,000 daltons, isolated by Tiedemann (’68, ’76), can be extracted from trunks of 9-11-day-old chick embryos (Born et al., ’72b). The CEE is cus- tomarily prepared from the same region of chick embryos at such stages (Konigsberg, ’63; Hauschka, ’72; Ramirez and Aleman, ’72a,b).

In our cultures of 11-day chick embryonic thigh myogenic cells maintained continuously in CEE-free medium, this eventuality was eliminated. Thus, the present results seem to support the idea of the existence of an intrinsically fusion-capable chick myoblast, whether or not it is withdrawn from the cell cycle in “the last compartment of the myogenic lineage” (Dienstman and Holtzer,


Fig. 4. Morphology of primary culture from 11-12- day-old chick fibromyogenic cells Town continuously in undisturbed CEE-free media. Ca2 concentration in the dialyzed or undialyzed media ranged from 0.3 to 1.5 mM. The insulin content was again one tenth the serum

content (1.1 punits per ml of medium). Typically elongated myotubes appeared containing 20 or more nuclei (white nuclei in the negative photograph). Confluency was reached at the end of the experimental period. x60.

Fig. 5. Morphology of primary culture of 11-12-day- old chick fibromyogenic cells grown in either disturbed or undisturbed dialyzed medium (MEMSD). The insulin content was always one tenth the serum content (1.1

'75). It must be emphasized, however, that the present cultures were performed in the presence of FCS. Conclusive proof must therefore await demonstration that typical higher-vertebrate myotubes can be formed from myogenic cells fed on an insulin-free, chemically defined media in uitro.

This work and our earlier findings on chick skeletal myogenic cells, also cultured in CEE-

punits per ml of Ca2+-free medium). Cell proliferation produced a monolayer after 7 days of culture, but typical or atypical myoblastic fusion were absent since myotubes were never observed. ~ 6 0 .

free media (Ramirez and Aleman, '73) show conclusively that Ca2+ is one of the limiting factors for any kind (normal or abnormal) of myoblast fusion, as suggested by previous cultures performed in rich (Ozawa, '72) and in CEE-free (Gospodarowicz et al., '75; Ra- mirez et al., '75) media. Moreover, some findings on early syncytium formation in suspension cultures showed that myoblast ag-


5

I

I 4 .I

i , !

7

E

E

E

E

L

gregation and myotube formation are also Caz+ -dependent (Knudsen and Horwitz, '77). However, to our knowledge, this is the first time that atypically or typically aligned chick myotubes containing more than 20 nuclei per fiber can be produced at will in plates containing CEE-free media and in the presence of Ca2+ at a concentration as low as 0.3 mM (Table 1). Perhaps the differences in Ca2+ sensitivity appearing in Table 1, among myoblasts from the same and different spe- cies, may be due to the presence or absence of CEE in the media employed.

Hauschka ('72) has published descriptions of the compositions of several media in which monolayer skeletal muscle cultures from several vertebrate sources have been successfully cultured. In no case except the currently reported one, however, have normal chick myotubes been cultured successfully in the absence of CEE. The discrepancy of one order of magnitude for the Cazf requirement between comparable populations of fusing chick myoblasts might be due to technical varia- tions of culture. Chick embryo extract and horse serum were used in Ozawa's experiments ('721, whereas we used FCS and no CEE (Table 1).

Ca" appears to be absolutely necessary for either normal or abnormal fusion, but typical alignment of the myotubes seems to depend on the conditioning of the microenvironment. Provided an initial critical cell density is present in either rich (Konigsberg, '71; Ozawa, '72) or in CEE-free media (Ramirez and Aleman, '73), it is possible that the &in- ity between myoblasts and myotubes may play a role in regulating the number of myotubes formed. There must be some mechanism that, in a given population of 500 fusion-capable myoblasts, leads to the formation of two or three long myotubes with several hundred nuclei, rather than 200 myotubes with two nuclei each. Specific recog- nition and adhesion sites of capable myoblasts are likely to be sensitive to puta- tive components in conditioned medium, which seem to be related to an EDTA-labile glucosamine-containing material washed from the surface of fusion-capable myotubes (Bischoff and Lowe, '74).

In regard to the present results, we think that if a medium is changed periodically, these molecules are likely to be detached from the surface of participating cells. In a continuously undisturbed microenvironment such an eventuality is prevented, thereby permitting proper cell-to-cell (or myotube) in-


kraction, as well as the participation of other interacting serum molecules (Hauschka, '72) present in these media.

The participation of insulin in promoting the appearance of normal or abnormal muscle fibers, in a partially restricted medium or even in a rich one, seems doubtful, the rele- vancy of studies on insulin performed with muscle cell lines (Mandel and Pearson, '74) notwithstanding. This assertion is based upon the following facts: 1) Atypical or typical myotubes were formed at will in every Ca2+-containing culture plate, depending on the conditioned microenvironment, not re- specting the subnormal insulin content (1.1 punits per ml of medium; i.e., lo3 or lo4 punits below the doses used in the myogenic cell line experiments). 2) Conversely, at the same doses of insulin no myotubes were formed in the absence of Ca2+, regardless of other variables. Moreover, the participation of the hormone in the regulation of normal skeletal and cardiac myogenesis in uivo can be ruled out on the following basis: chick cardiac and skeletal contracting muscle fibers appear in uivo before the advent and circulation of significant amounts of insulin at about Hamburger's stage 38 (12th day of incubation; Grillo, '61), and insulin concentration at hatching is only about 6 punits per ml of serum (Raheja et al., '72).

From the data presented we conclude that insulin has no physiological significance in promoting myotube formation. Experiments that appear in the literature dealing with muscle development in viuo and in uitro, in the presence of supra-physiological doses of insulin, may have only pharmacological relevance.

ACKNOWLEDGMENTS

The authors wish to thank Drs. Edgar J. Boell, Albert0 Hamabata, and Lourdes Ponce for valuable criticism and also to thank Mr. Manuel Sanchez for technical assistance, and Ms. Gabriela Cuellar for typing the manu- script.

LITERATURE CITED

Bischoff, R., and M. Lowe (1974) Cell surface components and the interaction of myogenic cells. In: Exploratory Concepts in Muscular Dystrophy. Vol. 11. A.T. Mil- horat, ed. American Elsevier Inc., New York, pp. 17- 31.

Born, J., H.P. Gaithe, H. Tiedemann, H. Tiedemann, and U. Kocher-Becker (1972a) Isolation of a vegetalizing inducing factor. Hoppe-Seyler's 2. Physiol. Chem. Bd.,

353~1076-1084. Born, J., H. Tiedemann, and H. Tiedemann (1972b) The

mechanism of embryonic induction: Isolation of an inhibitor for the vegetalizing factor. Biochim. Biophys. Acta, 279~175-1831

Buckley, P.A., and I.R. Konigsberg (19741 Myogenic fusion and the duration of the post-mitotic gap (GI). De- - _ - vel. Biol., 37:193-212.

Dienstman, S.R., and H. Holtzer (1975) Myogenesis: A cell lineage interpretation. In: Cell cycle and cell differentiation. J. Reinert and H. Holtzer, eds. Springer- Verlag, New York, Heidelberg, pp. 1-25.

Doering, J.L., and D.A. Fischman (1977) A fusion-promoting macromolecular factor in muscle conditioned medium. Exp. Cell Res., 105t437-443.

Dulbecco, R., and M. Vogt (1954) Plaque formation and isolation of pure lines with poliomyelitis viruses. J. Exp. Med., 99~167-182.

Gospodarowicz, D., J. Weseman, and J. Moran (1975) Presence in brain of a mitogenic agent promoting proliferation of myoblasts in low density culture. Nature, 256r216-219.

Grillo, T.A.I. (1961) The ontogeny of insulin secretion in the chick embryo. J. Endocrinol., 22~285-292.

Hauschka, S.D. (1972) Cultivation of muscle tissue. In: Growth, nutrition and metabolism of cells in culture. Vol. 11. G.H. Rothblath and V.J. Cristofalo, eds. Aca- demic Press, London and New York, pp. 67-130.

Herbert, V., K. Lau, C.W. Gottlieb, and S.J. Gleicher (1965) Coated charcoal immunoassay of insulin. J. Clin. Endocrinol., 25:1375-1384.

Holtzer, H., N. Rubinstein, J. Chi, S.R. Dienstman, and J. Biehl (1974) The phenotypic heterogeneity of myogenic clones and what cells synthesize which molecules. In: Exploratory Concepts in Muscular Dystrophy. Vol. 11. A.T. Milhorat, ed. American Elsevier Inc., New York, pp. 3-16.

Knudsen, D.A., and A.F. Horwitz (1977) Tandem events in myoblast fusion. Devel. Biol., 58t328-338.

Konigsberg, I.R. (1963) Clonal analysis of myogenesis. Science, 140~1273-1284.

Konigsberg, I.R. (1971) Diffusion-mediated control of myoblast fusion. Devel. Biol., 26:133-152.

Lewis, W.H., and M. Lewis (1917) Behavior of cross- striated muscle in tissue culture. Amer. J. Anat., 22r169-194.

Lowry, O.H., N. Rosebrough, A. Farr, and R. Randall (1951) Protein measurement with the Fohn phenol re- agent. J. Biol. Chem., 293t265-275.

Mandel, J.L., and M.L. Pearson (1974) Insulin stimulates myogenesis in a rat myoblast line. Nature (London), 251:618-620.

Ozawa, E. (1972) The role of calcium ion in avian myogenesis in uitro. Biol. Bull., 143~413-439.

Raheja, K.L., H.M. Tepperman, and J. Tepperman (1972) Effect of diet on plasma insulin and glucose levels of newly hatched chicks. Horm. Metab. Res., 4359-73.

Ramirez, O., and V. Aleman (1972a) Modification of the muscle cell phenotype by different media. J. Embryol. Exp. Morph., 28~559-570.

Ramirez, O., and V. Aleman (1972b) Diferenciacih selec- tiva durante el cultivo del musculo esqueletico. Cien- cia (Mex.), 27t197-205.

Ramirez, O., and V. Aleman (1973) Efecto de 10s factores presentes en 10s medios acondicionados sobre la prolif- eracion del musculo in uitro. Ciencia (Mex.), 28:67-72.

Ramirez, O., 0. Hernandez, and E. Mercado (1975) The role of calcium on the regulation of syncytium formation and creatine kinase ontogenesis during muscle development. Archs. Inst. Cardiol. Mex., 45~416-422.

Shainberg, A., G. Yagil, and D. Yaffe (1969) Control of myogenesis in uitro by calcium concentration in nutri-


tional medium. Exp. Cell Res., 58t163-167. Shainberg, A., G. Yagil, and D. Yaffe (1971) Alterations

of enzymatic activities during muscle differentiation in uitro. Devel. Biol., 25:l-29.

Tiedemann, H. (1968) Factors determining embryonic differentiation. J. Cell Physiol., 72(Suppl. 11t129-144.

Tiedernann, H. (1976) Pattern formation in early developmental stages of amphibian embryos. J. Embryol. Exp. Morph., 35:437444.

Turner, D.C., R. Gmur, M. Siegrist, E. Burckhardt, and H.M. Eppenberger (1976) Differentiation in cultures derived from embryonic chicken muscle. I. Muscle specific enzyme changes before fusion in EGTA-synchro- nized cultures. Devel. Biol.. 48t258-283.

Turner, D.C. (1978) Differentiation in cultures derived from embryonic chicken muscle: The post-mitotic, fu- sioncapable myoblast as a distinct cell type. Differen- tiation, 10:81-93.

Van Brussel, E., J.J. Yang, and M.W. Seraydarian (1983) Isoenzymes of creatine kinase in mammalian cell cultures. J. Cell Physiol., 116~221-226.

Yaffe, D. (1971) Developmental changes preceding cell fusion during muscle differentiation in uitro. Exp. Cell Res., 66:33-48.

Yaffe, D., and 0. Jaxel(1977) A myogenic cell line with altered serum requirements for differentiation. Differ- entiation, 7t159-166.

insulin-independent controlled physiological morphogenesis of chick muscle from fusion-capable...

Documents