Proc. Nati. Acad. Sci. USA Vol. 73, No. 6, pp. 1989-1993, June 1976
Cell Biology
Phenotypic transformation of clonal myogenic cells to cells resembling chondrocytes (myoblast/nicotinamide analogues/dibutyryl 3':5'-cAMP)
DAVID SCHUBERT AND MONIQUE LACORBIERE Department of Neurobiology, The Salk Institute, P.O. Box 1809, San Diego, California 92112
Communicated by Frangois Jacob, December 18, 1975
ABSTRACT The nicotinamide analogue 6-aminonicotinamide and dibutyryl 3':5'-cAMP inhibit myogenesis in a clonal rat cell line from skeletal muscle. Both reagents produce a similar morphological response in the cells, and stimulate collagen and glycosaminoglycan synthesis. These data suggest that 6-aminonicotinamide and dibutyryl cAMP induce a phenotypic transformation of myogenic ceNls to cells that share many characteristics with chondrocytes. Clonal skeletal muscle myogenic cell lines divide as mononucleate cells and fuse to form multinucleate myotubes when the cells become confluent (1). Clonal analysis has demonstrated that all of the cells within a clone can form myotubes (1). The following events are temporally associated with myotube formation: (i) increases in the specific activities of myokinase and creatine kinase (2), (ii) an increase in the rate of myosin heavy chain synthesis (3), (iii) the appearance of nicotinic acetylcholine receptors (4), and (iv) a well-defined sequence of electrophysiological development (4, 5). In addition, myotubes are capable of localizing acetylcholine sensitivity to the point of nerve contact (6) and receiving functional synaptic input from spinal cord neurons (7). In contrast to the unique developmental pathway displayed by the myogenic cell cultures, mesenchymal cells in, for example, chick limb buds are capable of differentiating into either chondrocytes or muscle; at later stages of development they are able to give rise to only one cell type (8). The phenotype expressed by embryonic chick mesodermal cells can also be influenced by reagents that alter their intermediary metabolism. For example, nicotinamide analogues such as 3-acetylpyridine (3AP) and 6-aminonicotinamide (6AN) specifically alter the sequence of mesodermal differentiation in chick embryos (9). These in vivo experiments have been extended to a chick cell culture system, where 3AP and 6AN potentiate some aspects of the chondrocytic phenotype in limb bud mesoderm (10, 11). Both in vvo and in vitro, nicotinamide (NA) relieves the teratogenic effects of SAP and 6AN. These observations, along with the fact that NAD+ levels increase during myogenesis (12), support the hypothesis that niacin metabolism may have a regulatory role in the differentiation of mesodermal cells (9, 10). In addition, an involvement of cyclic nucleotides in mesodermal differentiation has been proposed (13, 14). Here we ask whether or not the phenotype of a cell line that is normally myogenic can be changed by altering the NAD+ and cyclic nucleotide metabolism of the cell. Assuming that a chondrocyte is the mesodermally derived cell that synthesizes the majority of the principal cartilage components, collagen and the glycosaminoglycans (15), the results show that clonal Abbreviations: 6AN, 6-aminonicotinamide; NA, nicotinamide, cAMP, adenosine 3':5'-monophosphate; Bt2cAMP, dibutyryladenosine 3':5'monophosphate; GAG, glycosaminoglycan; 3AP, 3-acetylpyridine.
1989
cell cultures that normally give rise to muscle can be shifted toward cultures of cells that phenotypically resemble chondrocytes. MATERIALS AND METHODS All experiments were performed on exponentially growing rat L6 cultures (3) over a period of 3 days, so that the control cultures did not become confluent and did not initiate cell fusion. To assay collagen synthesis, cells were labeled for 6 hr with [3H]or [14C]leucine, and the secreted protein was assayed on acrylamide gels (3). These data were calculated as the percentage of the total secreted isotope in protein in the collagen proteins, and presented as a ratio of experimental to control values. To determine the extent of protein hydroxylation, cells were labeled with [14C]proline for 6 hr, and the cell supernatant (medium) was assayed for hydroxyproline and proline chromatographically (3). These data were calculated as the fraction of total proline isotope in hydroxyproline, and presented as a ratio of experimental to control values. To assay for glycosaminoglycan (GAG) synthesis and sulfation, cells were labeled with either [3H]glucosamine (10,gCi/ml) or Na235SO4 (20 ,tCi/ml) in serum-free medium for 6 hr. The cells and/or proteins in the culture medium were digested with Pronase and exhaustively dialyzed (16). GAG was precipitated with cetylpyridinium chloride (17) and the data were expressed as the ratio of isotope in GAG in experimental versus control cultures, normalized to either cell number or total cellular protein. The polysaccharides were further analyzed by electrophoresis on cellulose acetate (18). The initial rates of uptake of radioactive compounds into cells were measured under conditions identical to those described above for longer term labeling experiments by collecting cells on glass fiber filters. Uptake was linear for at least 5 min at 37°. The data were expressed as the ratio of the initial rates of experimental to control cultures, normalized to cell protein. Intracellular 3':5'-cAMP was assayed according to Watson (19). Intracellular NAD+ was determined using a fluorometric assay (20). RESULTS Cell Growth, Cell Fusion, and Cell Morphology. Under normal growth conditions, L6 cells divide exponentially until they form a confluent monolayer, at which time they fuse to form multinucleate myotubes (1). If 1 mM 6AN or dibutyryl cAMP (Bt2cAMP) was added to log phase cultures, cell multiplication was slightly retarded and the cultures became confluent without fusing. The 6AN- and Bt2cAMP-treated cells become flatter and more angular than the spindle-shaped control cells. There was no decrease in cell viability. After 2 days' exposure to 6AN the cells were unable to fuse upon its
1990
Cell Biology: Schubert and Lacorbiere
Proc. Natl. Acad. Sci. USA 73 (1976)
Table 1. Effect of 6AN and Bt2cAMP on collagen secretion and proline hydroxylation
Cell line
Test compound
Days after addition
L6 L6 L6 L6 L6 L6 L6 L6 L6 L6 L6* L6* L6 fused L6 fused L6 B103 (nerve) B82 (glial)
6AN Bt2cAMP 6AN Bt2cAMP 6AN + NA 6AN Bt2cAMP 6AN + NA Bt2cAMP + NA NA 0.2% serum 0.2% serum Bt2cAMP 6AN
1 1 2 2 2 3 3 3 3 3 2 3 2 2
2.0 1.2 1.4 1.6 1.0 2.0 1.7 1.1 1.8 1.0 0.8 0.7 0.8 0.6
2 2
0.6 0.7
Myoblastt/fused 6AN 6AN
Collagen secretion (ratio)
Proline hydroxylation (ratio) 2.7 2.6 0.5 1.0 3.6 0.6 1.2 0.8 1.0 0.8 0.9 1.1
1.4t 0.5
Cells were labeled with [14C]proline or [3H]leucine in the presence of the test compound for 6 hr on the days indicated after the addition of the test compound. Secreted collagen and proline hydroxylation were determined as described in Materials and Methods. The fraction of the total [3H]leucine-labeled secreted protein represented as collagen was determined in experimental and control cultures, and the data were expressed as the ratio of these fractions (experimental over control). The variation between two independent sets of experiments was less than 5%. Similarly, the proline hydroxylation data are presented as the ratio of experimental to control values. Duplicate amino acid analyses were within 5% of each other. The amount of collagen synthesis or hydroxylation, calculated in terms of cell number or protein, did not change in control cultures over 3 days. B82 and B103 are clonal rat glial and nerve cells, respectively (28). * L6 cells were grown to 5 X 105 cells per 60 mm culture dish, and washed twice with serum-free medium, and medium containing 0.2% serum was added. The culture completed approximately 0.5 division and did not fuse; the cells were viable at least 4 days. t.The ratio of the fraction of total protein secreted as collagen in exponentially dividing cells over that of multinucleate myotubes, 1 week post-fusion.
removal, while the effect of Bt2cAMP was partially reversible (approximately 30% fusion after 7 days). No morphological response to 6AN was observed in clonal rat nerve and glial cells, suggesting that it is not a commonly found response similar to that induced by Bt2cAMP. At the ultrastructural level, the cytoplasm of 6AN-treated cells contained more rough endoplasmic reticulum than the controls and an extensive extracellular granular matrix. Finally, 10 mM NA reversed the effect of 6AN on cell division, morphology, and fusion, suggesting that these events were related to niacin metabolism. NA did not reverse the effects of Bt2cAMP. Bt2cAMP and 6AN could prevent cells from reaching the density required to initiate fusion. This was ruled out by the fact that the cell protein per dish was similar in treated and control cultures. In addition, if cells were grown for 2 days in 1. mM Bt2cAMP or 6AN and replated at densities between 2 X 106 and 1 X 107 per 60 mm culture dish in the presence of Bt2cAMP or 6AN, there was no fusion at any density after 1 week; control cultures plated at the same densities were more than 80% fused within 3 days. A number of alternatives could account for the 6AN- and Bt2cAMP-induced changes in exponentially dividing L6 cells. (i) These compounds are cytotoxic. (ii) They induce morphological changes, but do not affect other aspects of the cells' biochemistry. (iii) They induce pleiotropic alterations in the cells' physiology. The following experiments rule out the first two alternatives. Collagen Secretion and Hydroxylation. Table 1 shows that 6AN and Bt2cAMP increase collagen secretion relative to control cultures. Since these data are based on the fraction of collagen in total secreted proteins, they represent an alteration in the pattern of protein secretion and not precursor uptake.
The stimulating effect of 6AN was reversed by NA; that of Bt2cAMP was not. If these responses were due to growth inhibition, other methods of growth retardation should yield similar results. However, cells arrested with low-serum medium showed a decrease in collagen secretion. In addition, clonal nerve (B10) and glial (B82) cells (21) did not respond to 6AN with increased collagen synthesis, nor did fused myotubes. 6AN also increased the extent of proline hydroxylation, while Bt2cAMP decreased hydroxylation. Glycosaminoglycan Synthesis and Sulfation. During both in vivo and in vttro chondrogenesis there are increased amounts of glycosaminoglycans synthesized (15). If 6AN and Bt2cAMP shift the metabolism of L6 cells in this direction, then more GAG should be synthesized relative to controls. Table 2 shows that 6AN increases GAG synthesis up to 6-fold. This stimulation was reversed by NA. Although total GAG synthesis was stimulated, the sulfation of these molecules was reduced. This has also been observed in vivo (22). In contrast to 6AN, Bt2cAMP increased both GAG synthesis and sulfation. NA did not block the stimulatory effect of Bt2cAMP on GAG synthesis or sulfation. The effects of Bt2cAMP and 6AN were also examined on fused muscle, nerve, and glial cells. These reagents either had no effect on, or decreased, GAG synthesis and sulfation. In addition, low-serum-arrested L6 cells synthesized less mucopolysaccharide than exponentially dividing cells; GAG synthesis and sulfation in dividing cells was greater than that of fused myotubes (Table 2). When glucosamine-labeled GAG was examined by electrophoresis on cellulose acetate, three distinct peaks were observed. Two electrophoresed with hyaluronic acid and chondroitin sulfate, and a third, unidentified, peak migrated between them. The unidentified peak and the chondroitin sulfate were labeled with 35SO4. There was only a
Proc. Natl. Acad. Sci. USA 73 (1976)
Cell Biology: Schubert and Lacorbiere
1991
Table 2. Effect of 6AN and Bt2cAMP on glycosaminoglycan synthesis and sulfation GAG (ratio)
GAG-SO4 (ratio)
Days after Cell line L6 L6
L6
L6 fused
L6* B82* (glial) B12* (glial) B103* (nerve) B35* (nerve) L6
Test compound
addition
6AN Bt2cAMP 6AN 6AN + NA Bt2cAMP NA Bt2cAMP + NA 6AN Bt2cAMP 6AN + NA NA Bt2cAMP + NA 6AN Bt2cAMP 6AN + NA NA Bt2cAMP + NA 0.2% serum 6AN 6AN 6AN 6AN Mononucleatet/fused
1 1 2 2 2 2 2 3 3 3 3 3 3 3 3 3 3 3 3 3 2 3 3
Cells 1.2 1.5 6.6 1.0 2.0 1.1 2.5 6.4 1.6 _ 1.5 0.8 1.8 0.4 0.6 0.5 0.6 0.4 0.7 0.7 0.8 0.8 0.8 8.3t
Secreted 1.2 1.3 4.2 1.1 3.1 1.2 3.4 4.6 2.6 1.7 0.9 2.4 0.5 0.6 0.4 0.6 0.4 0.7 0.6 0.8 0.9 0.7 12.3t
Cells
Secreted
0.4 1.7 0.8 1.1 4.0 1.0 3.8 0.5 3.9 1.1 0.9 4.1 0.6 0.5 0.3 0.4 0.5 0.2 0.3 0.2
0.2 1.2 0.2 0.9 1.4 1.3 2.6 0.2 3.4 1.0 1.1 4.6 0.5 0.6 0.3 0.3 0.4 0.3 0.3 0.3
7.4t
11.8t
Sulfate
Glucosamine
uptake
uptake
(ratio)
(ratio)
0.9 1.1 1.2 1.2
0.4 0.5 0.8 1.1
Cells were labeled with 3H glucosamine or Na235SO4 in the presence of the test compound for 6 hrs on the days indicated after the addition of the test compound. Secreted and cell-associated GAG and sulfated GAG were determined as described in Materials and Methods. These data are expressed as the ratio of cetylpyridinium-chloride-precipitable isotope in the experimental to that the control cultures, normalized to cell number. There was approximately a 25% increase in protein per cell in 6AN- and Bt2cAMP-treated cultures after 3 days, but since this increase is small relative to the data normalized to cells, the data normalized to protein are not presented. Glucosamine and S04= uptake are presented as the ratio of the initial rates of experimental to control cultures, normalized to total cellular protein. The variation between duplicate experiments was less than 10%. The amount of GAG synthesis or sulfation, calculated in terms of cell number or protein, did not change in control cultures over 3 days. * See Table 1. t The ratios presented are those of exponentially dividing cells (7 x 105 cells per 60 mm plate) to cells 1 week post-fusion; the data are normalized to cellular protein.
quantitative change in the synthesis of these molecules after exposure to 6AN and Bt2cAMP, with proportionally more chondroitin sulfate being synthesized (Table 3). Finally, both 6AN and Bt2cAMP decreased the initial rates of glucosamine uptake, but had no apparent effect on sulfate uptake (Table 2). Effect of 6AN and Bt2cAMP on Intracellular NAD+ and cAMP. There are several alternatives for the mechanisms responsible for the similar phenotypic alterations induced by 6AN and Bt2cAMP. The two simplest are that both Bt2cAMP and 6AN either decrease NAD+ levels or increase cAMP levels. If either of these alternatives were correct, then the intracellular level of either cAMP or NAD+ should covary with 6AN and Bt2cAMP treatment of the cells. Table 4 shows that 6AN reduces the level of NAD+, while Bt2cAMP has little or no effect on intracellular NAD+. NA restores the 6AN-induced reduction in NAD+. Since 6AN also lowers cAMP, it follows that 6AN does not increase intracellular cAMP, and Bt2cAMP does not function by altering NAD+ levels. If it is assumed that Bt2cAMP functions by increasing intracellular cAMP or the physiologically active N6 derivative of Bt2cAMP (refs. 23 and 24; 'Table 4), then 6AN has the opposite effect on cAMP levels of Bt2cAMP.
DISCUSSION The following conclusions can be made from the above data: (i) Both 6AN and Bt2cAMP induce uniform morphological changes and inhibit cell fusion in the myogenic cell line L6. (ii) 6AN increases collagen secretion and hydroxylation and the rate of GAG synthesis, while GAG sulfation is decreased (Tables 1 and 2). (iii) Bt2cAMP increases collagen secretion, but decreases its hydroxylation; both GAG synthesis and sulfation are increased (Tables 1 and 2). (iv) 6AN and Bt2cAMP have different effects on the intracellular levels of cAMP and NAD+ (Table 4): These data suggest that 6AN and Bt2cAMP induce a phenotypic transformation of myogenic cells to cells with the characteristics expected of chondrocytes. Similar results were described in vivo, where nicotinamide analogues alter cartilage and muscle differentiation in chicken embryos (9, 22, 25). In the chick system, 3-acetylpyridine (3AP) causes muscle hypoplasia, while 6AN produces cartilage defects. These reagents are effective over a very limited time in embryonic development, and their teratogenic effects are completely relieved by cotreatment with nicotinamide. These observations have been extended to chick limb bud mesoderm
1992
Cell Biology: Schubert and Lacorbiere
Proc. Natl. Acad. Sci. USA 73 (1976)
Table 3. Changes in the relative amounts of glycosaminoglycans synthesized in the presence of 6AN and Bt2cAMP % of total isotope in:
Test compound
None 6AN Bt2cAMP None 6AN Bt2cAMP None 6AN Bt2cAMP
Days after addition 1 1 1 2 2 2 3 3
3
HA 55
32 29 53 17 28 51
20 24
UK
6 36
11 13 13
38 9 45 35 8 42 37
15
Day
Control*
6AN
L6 cells were labeled with [3H]glucosamine for 6 hr as described in Table 2. The cells were then digested with Pronase and the GAG was electrophoresed on cellulose acetate strips and its radioactivity was measured (24). In all cases three distinct peaks were resolved. Two migrated with hyaluronic acid (HA) and chondroitin sulfate (CS); the third, unknown (UK) peak of radioactivity migrated between them. The data are presented as the percent of the total isotope in each fraction; approximately 25% of the radioactivity was randomly distributed throughout the electrophoresis strips.
cultures, where both 6AN and SAP appear to potentiate chondrogenic expression (10). SAP is, however, more effective in inducing chondrogenesis in chick mesoderm cultures than 6AN. Since SAP is cytotoxic to the L6 rat line, its effects on differentiation cannot be studied. Myogenic and chondrogenic cells are derived from the same cell type in embryonic mesoderm (8, 26), and it has been proposed that nicotinamide pyridine nucleotides play a central role in mesodermal differentiation (9, 10, 11, 27). Our data support this hypothesis. In addition, the effect of 6AN on the biochemical parameters examined seems to be limited to mononucleate L6 cells, for the rates of collagen and GAG synthesis in fused myotubes, nerve, and glia are not enhanced. The effects produced by 6AN may be due to its incorporation as the 6amino-NAD+ analogue of NAD+, which causes a reduction in the activity of the pyridine-nucleotide-linked enzymatic reactions (27, 28). It has been proposed that cyclic nucleotides are intimately involved in ce11 differentiation, and specifically in the teratogenic effect of the nicotinamide analogues on mesodermal differentiatic n (13). It was argued that high concentrations of cAMP triggecartilage cell formation while low concentrations of cAMP ind uce muscle differentiation. Since 3 AP inhibits phosphodiest~rase activity (29) and thus may raise intracellular cAMP, there Imay be an indirect effect on cell differentiation. Although BtscAMP inhibits cell fusion in clonal (30) and primary cultures (31) of muscle, and alters the phenotype of the L6 mononucleate cell in much the same way as 6AN, the cAMP hypothesis does not seem to apply to L6 for: (i) The intracellular level of cAMP does not decrease during L6 myogenesis. (ii) 6AN decreases the level of intracellular cAMP, as opposed to increasing it. The 6AN- and Bt2cAMP-induced transformation of the L6 cells are not both mediated via NAD+, for Bt2cAMP does not significantly alter NAD+ levels (Table 4). In addition, the effects of 6AN and Bt2cAMP on collagen and GAG synthesis are slightly different (Tables 1 and 2). Thus, the only direct correlation between the chondrocytic and muscle phenotype is with the intracellular concentration of NADI: high NADI
NA
6AN + NA
Bt2cAMPt
cAMP
CS
12 8
16 15 19
Table 4. Effect of 6AN and Bt2cAMP on the level of intracellular cAMP and NAD+
0 1 2 3 4 5 6 7
6.4 ± 5.8 ± 8.8± 6.5 ± 6.1 ±
0.2 1.8 2.2 0.7 1.4
0 1 2 3 4 5 6 7
8.1± 0.4
7.1 ± 0.3 6.6 ± 1.2
0.31 0.35 0.29 0.44
1.11 0.97 1.16 1.27
0.63t
1.63
0.53 1.14 1.22 1.37
>25 >25 20 15
NAD+ 5.9 ± 0.1 5.2 ± 1.0 8.4 ± 1.5 4.8 ± 0.2 14.0 ± 0.1 13.4 ± 0.1 17.4 ± 0.4
0.33 0.59 0.32
0.62 0.21
0.97 2.60 1.01 1.38
1.08 1.09 1.22 1.12 1.42 -
0.67 1.01 1.10 1.95 1.18
The test compounds were added to exponentially growing cultures as indicated, and the intracellular levels of NAD+ and cAMP were assayed as described in Methods and Materials. Both sets of determinations are the means of two experiments 4 standard deviation of quadruplicate (cAMP) or triplicate (NAD+) determinations in each experiment. The data are presented as the ratio of the experimental to control values normalized to cellular protein. The cells fused in control cultures between days 4 and 5. * pmol of cAMP per mg of protein or nmol of NAD+ per mg of protein. t One experiment only. t These results are only an approximate representation of the relative intracellular levels of cAMP, for in the radioimmune assay employed Bt2cAMP and N6-monobutyryl cAMP competed 1.8 and 18 times less effectively, respectively, than cAMP, and the 02-monobutyryl cAMP derivative competed 24 times more effectively than cAMP.
concentrations are associated with (fused) muscle, low with cartilage (10). Bt2cAMP prevents fusion and the associated increase of NAD+, but may also affect cellular metabolism in a manner unrelated to NAD+. Bt2cAMP also alters collagen and GAG metabolism of the fibroblast, a cell type developmentally related to myoblasts. For example, Bt2cAMP increases the amount of collagen synthesized in Chinese hamster cells (32) and increases collagen and acid sulfated mucopolysaccharide synthesis in virus-transformed mouse fibroblasts (33, 34). It was argued from the study of mesodermally derived clonal cell lines that the different "end" cells of this primordium share several traits usually associated with one, and that there are mostly quantitative differences between them (35). In support of this concept are the in vivo observations that preinduced, undifferentiated mesodermal cells express enzymatic activities characteristic of cartilage (36, 37). In addition, during the differentiation of the L6 cells there are quantitative increases in electrical excitability, several enzymes, and myosin heavy chain synthesis (3-5). Collagen secretion and GAG synthesis and sulfation decrease (Tables 1 and 2). Thus, during myogenesis there is a decrease, relative to mononucleate myogenic cells, of the biochemical correlates of the chondrocytic phenotype, while those characteristics of muscle are increased. Conversely, 6AN and Bt2cAMP enhance the chondrocytic properties of the mononucleate cells. This apparent phenotypic transformation
Cell Biology: Schubert and Lacorbiere of the myogenic cells suggests that the immediate developmental precursors of some cells may be artificially manipulated to generate cells of a different phenotype. Note Added in Proof. 6AN, but not Bt2cAMP, induced a change in the type of collagen synthesized by L6 myoblasts from a heterogeneous group of a chains to exclusively al chains as assayed by chromatography on CM-cellulose. This work was supported by grants from the National Foundation and the Muscular Dystrophy Associations of America. We thank Helgi Tarikas for her technical assistance and Bonnie Harkins for preparation of the manuscript. 1. Richler, C. & Yaffe, D. (1970) Dev. Biol. 23, 1-18. 2. Shainberg, A., Yagil, G. & Yaffe,.D. (1971) Dev. Biol. 25, 1-29. 3. Schubert, D., Tarikas, H., Humphreys, S., Heinemann, S. & Patrick, J. (1973) Dev. Biol. 33, 18-37. 4. Patrick, J., Heinemann, S., Lindstrom, J., Schubert, D. & Steinbach, J. H. (1972) Proc. Natl. Acad. Sci. USA 69,2762-2766. 5. Kidokoro, Y. (1975) J. Physiol. 244, 129-143. 6. Steinbach, J. H., Patrick, J., Schubert, D., Harris, A. J. & Heinemann, S. (1973) J. Gen. Physsol. 62,255-275. 7. Kidokoro, Y. & Heinemann, S. (1975) Nature 252,593-594. 8. Searls, R. L. & Janners, M. Y. (I969) J. Exp. Zool. 170, 365-376. 9. Landauer, W. (1957) J. Exp. Zool. 151, 253-258. 10. Caplan, A. J. (1972) Exp. Cell Res. 70, 185-192. 11. Caplan, A. J. (1972) Dev. Biol. 28, 71-80. 12. Rosenberg, M. J. & Caplan, A. I. (1974) Dev. Biol. 38, 157-164. 13. McMahon, D. (1974) Science 185, 1012-1021. 14. Zalin, R. & Montague, W. (1974) Cell 2, 103-108. 15. Levitt, D. & Dorfman, A. (1974) in Current Topics in Developmental Biology, eds. Moscona, A. A. & Monroy, A. (Academic Press, New York), pp. 103-150. 16. Kosher, R. A., Lash, J. W. & Minor, R. R. (1973) Dev. Biol. 35, 210-220.
Proc. Natl. Acad. Sci. USA 73 (1976)
1993
17. Nevo, Z., Horwitz, A. L. & Dorfmann, A. (1972) Dev. Biol. 28, 219-228. 18. Mayne, R., Sanger, J. W. & Holtzer, H. (1971) Dev. Biol. 25, 547-567. 19. Watson, J. (1975) J. Exp. Med. 141, 97-111. 20. Everse, J., Kaplan, N. 0. & Schichor, S. (1970) Arch. Biochem. Biophys. 136, 106-111. 21. Schubert, D., Heinemann, S., Carlisle, W., Tarikas, H., Kimes, B., Patrick, J., Steinbach, J. H., Culp, W. & Brandt, B. L. (1974) Nature 249,224-227. 22. Seegmiller, R. E. & Runner, M. N. (1974) J. Embryol. Exp.
Morphol. 31,305-312. 23. Kaukel, E. & Hilz, H. (1972) Biochem. Biophys. Res. Commun. 46, 1011-1018. 24. Heersche, J., Fedak, S. & Aurbach, G. D. (1971) J. Biol. Chem.
246,6770-6775. 25. Caplan, A. I. (1971) J. Exp. Zool. 178,351-358. 26. Zwinning, E. (1961) Adv. Morphog. 1, 301-330. 27. Landauer, W. & Sopher, D. (1970) J. Embryol. Exp. Morphol. 24, 187-202. 28. Dietrich, L. W., Friedland, I. M. & Kaplan, L. A. (1958) J. Biol. Chem. 233,964-968. 29. Shimoyama, M., Kawai, M., Hoshi, Y. & Ueda, I. (1972) Biochem. Blophys. Res. Commun. 49, 1137-1142. 30. Wahrmann, J. P., Luzzati, D. & Winand, R. (1973) Nature New Biol. 245, 112-113. 31. Zalin, R. J. (1973) Exp. Cell Res. 78, 152-165. 32. Hsie, A. W., Jones, C. & Puck, T. (1971) Proc. Natl. Acad. Sd. USA 68,1648-1652. 33. Peterkofsky, B. & Prather, W. B. (1974) Cell 4, 291-299. 34. Goggins, J. F., Johnson, G. S. & Pastan, I. (1972) J. Biol. Chem. 247,5759-5764. 35. Tarikas, H. & Schubert, D. (1974) Proc. Natl. Acad. Sd. USA 71, 2377-2381. 36. Medoff, J. (1967) Dev. Biol. 16, 118-143. 37. Searls, R. L. (1965) Proc. Soc. Exp. Biol. Med. 118, 1172-1176.
Cell Biology
Phenotypic transformation of clonal myogenic cells to cells resembling chondrocytes (myoblast/nicotinamide analogues/dibutyryl 3':5'-cAMP)
DAVID SCHUBERT AND MONIQUE LACORBIERE Department of Neurobiology, The Salk Institute, P.O. Box 1809, San Diego, California 92112
Communicated by Frangois Jacob, December 18, 1975
ABSTRACT The nicotinamide analogue 6-aminonicotinamide and dibutyryl 3':5'-cAMP inhibit myogenesis in a clonal rat cell line from skeletal muscle. Both reagents produce a similar morphological response in the cells, and stimulate collagen and glycosaminoglycan synthesis. These data suggest that 6-aminonicotinamide and dibutyryl cAMP induce a phenotypic transformation of myogenic ceNls to cells that share many characteristics with chondrocytes. Clonal skeletal muscle myogenic cell lines divide as mononucleate cells and fuse to form multinucleate myotubes when the cells become confluent (1). Clonal analysis has demonstrated that all of the cells within a clone can form myotubes (1). The following events are temporally associated with myotube formation: (i) increases in the specific activities of myokinase and creatine kinase (2), (ii) an increase in the rate of myosin heavy chain synthesis (3), (iii) the appearance of nicotinic acetylcholine receptors (4), and (iv) a well-defined sequence of electrophysiological development (4, 5). In addition, myotubes are capable of localizing acetylcholine sensitivity to the point of nerve contact (6) and receiving functional synaptic input from spinal cord neurons (7). In contrast to the unique developmental pathway displayed by the myogenic cell cultures, mesenchymal cells in, for example, chick limb buds are capable of differentiating into either chondrocytes or muscle; at later stages of development they are able to give rise to only one cell type (8). The phenotype expressed by embryonic chick mesodermal cells can also be influenced by reagents that alter their intermediary metabolism. For example, nicotinamide analogues such as 3-acetylpyridine (3AP) and 6-aminonicotinamide (6AN) specifically alter the sequence of mesodermal differentiation in chick embryos (9). These in vivo experiments have been extended to a chick cell culture system, where 3AP and 6AN potentiate some aspects of the chondrocytic phenotype in limb bud mesoderm (10, 11). Both in vvo and in vitro, nicotinamide (NA) relieves the teratogenic effects of SAP and 6AN. These observations, along with the fact that NAD+ levels increase during myogenesis (12), support the hypothesis that niacin metabolism may have a regulatory role in the differentiation of mesodermal cells (9, 10). In addition, an involvement of cyclic nucleotides in mesodermal differentiation has been proposed (13, 14). Here we ask whether or not the phenotype of a cell line that is normally myogenic can be changed by altering the NAD+ and cyclic nucleotide metabolism of the cell. Assuming that a chondrocyte is the mesodermally derived cell that synthesizes the majority of the principal cartilage components, collagen and the glycosaminoglycans (15), the results show that clonal Abbreviations: 6AN, 6-aminonicotinamide; NA, nicotinamide, cAMP, adenosine 3':5'-monophosphate; Bt2cAMP, dibutyryladenosine 3':5'monophosphate; GAG, glycosaminoglycan; 3AP, 3-acetylpyridine.
1989
cell cultures that normally give rise to muscle can be shifted toward cultures of cells that phenotypically resemble chondrocytes. MATERIALS AND METHODS All experiments were performed on exponentially growing rat L6 cultures (3) over a period of 3 days, so that the control cultures did not become confluent and did not initiate cell fusion. To assay collagen synthesis, cells were labeled for 6 hr with [3H]or [14C]leucine, and the secreted protein was assayed on acrylamide gels (3). These data were calculated as the percentage of the total secreted isotope in protein in the collagen proteins, and presented as a ratio of experimental to control values. To determine the extent of protein hydroxylation, cells were labeled with [14C]proline for 6 hr, and the cell supernatant (medium) was assayed for hydroxyproline and proline chromatographically (3). These data were calculated as the fraction of total proline isotope in hydroxyproline, and presented as a ratio of experimental to control values. To assay for glycosaminoglycan (GAG) synthesis and sulfation, cells were labeled with either [3H]glucosamine (10,gCi/ml) or Na235SO4 (20 ,tCi/ml) in serum-free medium for 6 hr. The cells and/or proteins in the culture medium were digested with Pronase and exhaustively dialyzed (16). GAG was precipitated with cetylpyridinium chloride (17) and the data were expressed as the ratio of isotope in GAG in experimental versus control cultures, normalized to either cell number or total cellular protein. The polysaccharides were further analyzed by electrophoresis on cellulose acetate (18). The initial rates of uptake of radioactive compounds into cells were measured under conditions identical to those described above for longer term labeling experiments by collecting cells on glass fiber filters. Uptake was linear for at least 5 min at 37°. The data were expressed as the ratio of the initial rates of experimental to control cultures, normalized to cell protein. Intracellular 3':5'-cAMP was assayed according to Watson (19). Intracellular NAD+ was determined using a fluorometric assay (20). RESULTS Cell Growth, Cell Fusion, and Cell Morphology. Under normal growth conditions, L6 cells divide exponentially until they form a confluent monolayer, at which time they fuse to form multinucleate myotubes (1). If 1 mM 6AN or dibutyryl cAMP (Bt2cAMP) was added to log phase cultures, cell multiplication was slightly retarded and the cultures became confluent without fusing. The 6AN- and Bt2cAMP-treated cells become flatter and more angular than the spindle-shaped control cells. There was no decrease in cell viability. After 2 days' exposure to 6AN the cells were unable to fuse upon its
1990
Cell Biology: Schubert and Lacorbiere
Proc. Natl. Acad. Sci. USA 73 (1976)
Table 1. Effect of 6AN and Bt2cAMP on collagen secretion and proline hydroxylation
Cell line
Test compound
Days after addition
L6 L6 L6 L6 L6 L6 L6 L6 L6 L6 L6* L6* L6 fused L6 fused L6 B103 (nerve) B82 (glial)
6AN Bt2cAMP 6AN Bt2cAMP 6AN + NA 6AN Bt2cAMP 6AN + NA Bt2cAMP + NA NA 0.2% serum 0.2% serum Bt2cAMP 6AN
1 1 2 2 2 3 3 3 3 3 2 3 2 2
2.0 1.2 1.4 1.6 1.0 2.0 1.7 1.1 1.8 1.0 0.8 0.7 0.8 0.6
2 2
0.6 0.7
Myoblastt/fused 6AN 6AN
Collagen secretion (ratio)
Proline hydroxylation (ratio) 2.7 2.6 0.5 1.0 3.6 0.6 1.2 0.8 1.0 0.8 0.9 1.1
1.4t 0.5
Cells were labeled with [14C]proline or [3H]leucine in the presence of the test compound for 6 hr on the days indicated after the addition of the test compound. Secreted collagen and proline hydroxylation were determined as described in Materials and Methods. The fraction of the total [3H]leucine-labeled secreted protein represented as collagen was determined in experimental and control cultures, and the data were expressed as the ratio of these fractions (experimental over control). The variation between two independent sets of experiments was less than 5%. Similarly, the proline hydroxylation data are presented as the ratio of experimental to control values. Duplicate amino acid analyses were within 5% of each other. The amount of collagen synthesis or hydroxylation, calculated in terms of cell number or protein, did not change in control cultures over 3 days. B82 and B103 are clonal rat glial and nerve cells, respectively (28). * L6 cells were grown to 5 X 105 cells per 60 mm culture dish, and washed twice with serum-free medium, and medium containing 0.2% serum was added. The culture completed approximately 0.5 division and did not fuse; the cells were viable at least 4 days. t.The ratio of the fraction of total protein secreted as collagen in exponentially dividing cells over that of multinucleate myotubes, 1 week post-fusion.
removal, while the effect of Bt2cAMP was partially reversible (approximately 30% fusion after 7 days). No morphological response to 6AN was observed in clonal rat nerve and glial cells, suggesting that it is not a commonly found response similar to that induced by Bt2cAMP. At the ultrastructural level, the cytoplasm of 6AN-treated cells contained more rough endoplasmic reticulum than the controls and an extensive extracellular granular matrix. Finally, 10 mM NA reversed the effect of 6AN on cell division, morphology, and fusion, suggesting that these events were related to niacin metabolism. NA did not reverse the effects of Bt2cAMP. Bt2cAMP and 6AN could prevent cells from reaching the density required to initiate fusion. This was ruled out by the fact that the cell protein per dish was similar in treated and control cultures. In addition, if cells were grown for 2 days in 1. mM Bt2cAMP or 6AN and replated at densities between 2 X 106 and 1 X 107 per 60 mm culture dish in the presence of Bt2cAMP or 6AN, there was no fusion at any density after 1 week; control cultures plated at the same densities were more than 80% fused within 3 days. A number of alternatives could account for the 6AN- and Bt2cAMP-induced changes in exponentially dividing L6 cells. (i) These compounds are cytotoxic. (ii) They induce morphological changes, but do not affect other aspects of the cells' biochemistry. (iii) They induce pleiotropic alterations in the cells' physiology. The following experiments rule out the first two alternatives. Collagen Secretion and Hydroxylation. Table 1 shows that 6AN and Bt2cAMP increase collagen secretion relative to control cultures. Since these data are based on the fraction of collagen in total secreted proteins, they represent an alteration in the pattern of protein secretion and not precursor uptake.
The stimulating effect of 6AN was reversed by NA; that of Bt2cAMP was not. If these responses were due to growth inhibition, other methods of growth retardation should yield similar results. However, cells arrested with low-serum medium showed a decrease in collagen secretion. In addition, clonal nerve (B10) and glial (B82) cells (21) did not respond to 6AN with increased collagen synthesis, nor did fused myotubes. 6AN also increased the extent of proline hydroxylation, while Bt2cAMP decreased hydroxylation. Glycosaminoglycan Synthesis and Sulfation. During both in vivo and in vttro chondrogenesis there are increased amounts of glycosaminoglycans synthesized (15). If 6AN and Bt2cAMP shift the metabolism of L6 cells in this direction, then more GAG should be synthesized relative to controls. Table 2 shows that 6AN increases GAG synthesis up to 6-fold. This stimulation was reversed by NA. Although total GAG synthesis was stimulated, the sulfation of these molecules was reduced. This has also been observed in vivo (22). In contrast to 6AN, Bt2cAMP increased both GAG synthesis and sulfation. NA did not block the stimulatory effect of Bt2cAMP on GAG synthesis or sulfation. The effects of Bt2cAMP and 6AN were also examined on fused muscle, nerve, and glial cells. These reagents either had no effect on, or decreased, GAG synthesis and sulfation. In addition, low-serum-arrested L6 cells synthesized less mucopolysaccharide than exponentially dividing cells; GAG synthesis and sulfation in dividing cells was greater than that of fused myotubes (Table 2). When glucosamine-labeled GAG was examined by electrophoresis on cellulose acetate, three distinct peaks were observed. Two electrophoresed with hyaluronic acid and chondroitin sulfate, and a third, unidentified, peak migrated between them. The unidentified peak and the chondroitin sulfate were labeled with 35SO4. There was only a
Proc. Natl. Acad. Sci. USA 73 (1976)
Cell Biology: Schubert and Lacorbiere
1991
Table 2. Effect of 6AN and Bt2cAMP on glycosaminoglycan synthesis and sulfation GAG (ratio)
GAG-SO4 (ratio)
Days after Cell line L6 L6
L6
L6 fused
L6* B82* (glial) B12* (glial) B103* (nerve) B35* (nerve) L6
Test compound
addition
6AN Bt2cAMP 6AN 6AN + NA Bt2cAMP NA Bt2cAMP + NA 6AN Bt2cAMP 6AN + NA NA Bt2cAMP + NA 6AN Bt2cAMP 6AN + NA NA Bt2cAMP + NA 0.2% serum 6AN 6AN 6AN 6AN Mononucleatet/fused
1 1 2 2 2 2 2 3 3 3 3 3 3 3 3 3 3 3 3 3 2 3 3
Cells 1.2 1.5 6.6 1.0 2.0 1.1 2.5 6.4 1.6 _ 1.5 0.8 1.8 0.4 0.6 0.5 0.6 0.4 0.7 0.7 0.8 0.8 0.8 8.3t
Secreted 1.2 1.3 4.2 1.1 3.1 1.2 3.4 4.6 2.6 1.7 0.9 2.4 0.5 0.6 0.4 0.6 0.4 0.7 0.6 0.8 0.9 0.7 12.3t
Cells
Secreted
0.4 1.7 0.8 1.1 4.0 1.0 3.8 0.5 3.9 1.1 0.9 4.1 0.6 0.5 0.3 0.4 0.5 0.2 0.3 0.2
0.2 1.2 0.2 0.9 1.4 1.3 2.6 0.2 3.4 1.0 1.1 4.6 0.5 0.6 0.3 0.3 0.4 0.3 0.3 0.3
7.4t
11.8t
Sulfate
Glucosamine
uptake
uptake
(ratio)
(ratio)
0.9 1.1 1.2 1.2
0.4 0.5 0.8 1.1
Cells were labeled with 3H glucosamine or Na235SO4 in the presence of the test compound for 6 hrs on the days indicated after the addition of the test compound. Secreted and cell-associated GAG and sulfated GAG were determined as described in Materials and Methods. These data are expressed as the ratio of cetylpyridinium-chloride-precipitable isotope in the experimental to that the control cultures, normalized to cell number. There was approximately a 25% increase in protein per cell in 6AN- and Bt2cAMP-treated cultures after 3 days, but since this increase is small relative to the data normalized to cells, the data normalized to protein are not presented. Glucosamine and S04= uptake are presented as the ratio of the initial rates of experimental to control cultures, normalized to total cellular protein. The variation between duplicate experiments was less than 10%. The amount of GAG synthesis or sulfation, calculated in terms of cell number or protein, did not change in control cultures over 3 days. * See Table 1. t The ratios presented are those of exponentially dividing cells (7 x 105 cells per 60 mm plate) to cells 1 week post-fusion; the data are normalized to cellular protein.
quantitative change in the synthesis of these molecules after exposure to 6AN and Bt2cAMP, with proportionally more chondroitin sulfate being synthesized (Table 3). Finally, both 6AN and Bt2cAMP decreased the initial rates of glucosamine uptake, but had no apparent effect on sulfate uptake (Table 2). Effect of 6AN and Bt2cAMP on Intracellular NAD+ and cAMP. There are several alternatives for the mechanisms responsible for the similar phenotypic alterations induced by 6AN and Bt2cAMP. The two simplest are that both Bt2cAMP and 6AN either decrease NAD+ levels or increase cAMP levels. If either of these alternatives were correct, then the intracellular level of either cAMP or NAD+ should covary with 6AN and Bt2cAMP treatment of the cells. Table 4 shows that 6AN reduces the level of NAD+, while Bt2cAMP has little or no effect on intracellular NAD+. NA restores the 6AN-induced reduction in NAD+. Since 6AN also lowers cAMP, it follows that 6AN does not increase intracellular cAMP, and Bt2cAMP does not function by altering NAD+ levels. If it is assumed that Bt2cAMP functions by increasing intracellular cAMP or the physiologically active N6 derivative of Bt2cAMP (refs. 23 and 24; 'Table 4), then 6AN has the opposite effect on cAMP levels of Bt2cAMP.
DISCUSSION The following conclusions can be made from the above data: (i) Both 6AN and Bt2cAMP induce uniform morphological changes and inhibit cell fusion in the myogenic cell line L6. (ii) 6AN increases collagen secretion and hydroxylation and the rate of GAG synthesis, while GAG sulfation is decreased (Tables 1 and 2). (iii) Bt2cAMP increases collagen secretion, but decreases its hydroxylation; both GAG synthesis and sulfation are increased (Tables 1 and 2). (iv) 6AN and Bt2cAMP have different effects on the intracellular levels of cAMP and NAD+ (Table 4): These data suggest that 6AN and Bt2cAMP induce a phenotypic transformation of myogenic cells to cells with the characteristics expected of chondrocytes. Similar results were described in vivo, where nicotinamide analogues alter cartilage and muscle differentiation in chicken embryos (9, 22, 25). In the chick system, 3-acetylpyridine (3AP) causes muscle hypoplasia, while 6AN produces cartilage defects. These reagents are effective over a very limited time in embryonic development, and their teratogenic effects are completely relieved by cotreatment with nicotinamide. These observations have been extended to chick limb bud mesoderm
1992
Cell Biology: Schubert and Lacorbiere
Proc. Natl. Acad. Sci. USA 73 (1976)
Table 3. Changes in the relative amounts of glycosaminoglycans synthesized in the presence of 6AN and Bt2cAMP % of total isotope in:
Test compound
None 6AN Bt2cAMP None 6AN Bt2cAMP None 6AN Bt2cAMP
Days after addition 1 1 1 2 2 2 3 3
3
HA 55
32 29 53 17 28 51
20 24
UK
6 36
11 13 13
38 9 45 35 8 42 37
15
Day
Control*
6AN
L6 cells were labeled with [3H]glucosamine for 6 hr as described in Table 2. The cells were then digested with Pronase and the GAG was electrophoresed on cellulose acetate strips and its radioactivity was measured (24). In all cases three distinct peaks were resolved. Two migrated with hyaluronic acid (HA) and chondroitin sulfate (CS); the third, unknown (UK) peak of radioactivity migrated between them. The data are presented as the percent of the total isotope in each fraction; approximately 25% of the radioactivity was randomly distributed throughout the electrophoresis strips.
cultures, where both 6AN and SAP appear to potentiate chondrogenic expression (10). SAP is, however, more effective in inducing chondrogenesis in chick mesoderm cultures than 6AN. Since SAP is cytotoxic to the L6 rat line, its effects on differentiation cannot be studied. Myogenic and chondrogenic cells are derived from the same cell type in embryonic mesoderm (8, 26), and it has been proposed that nicotinamide pyridine nucleotides play a central role in mesodermal differentiation (9, 10, 11, 27). Our data support this hypothesis. In addition, the effect of 6AN on the biochemical parameters examined seems to be limited to mononucleate L6 cells, for the rates of collagen and GAG synthesis in fused myotubes, nerve, and glia are not enhanced. The effects produced by 6AN may be due to its incorporation as the 6amino-NAD+ analogue of NAD+, which causes a reduction in the activity of the pyridine-nucleotide-linked enzymatic reactions (27, 28). It has been proposed that cyclic nucleotides are intimately involved in ce11 differentiation, and specifically in the teratogenic effect of the nicotinamide analogues on mesodermal differentiatic n (13). It was argued that high concentrations of cAMP triggecartilage cell formation while low concentrations of cAMP ind uce muscle differentiation. Since 3 AP inhibits phosphodiest~rase activity (29) and thus may raise intracellular cAMP, there Imay be an indirect effect on cell differentiation. Although BtscAMP inhibits cell fusion in clonal (30) and primary cultures (31) of muscle, and alters the phenotype of the L6 mononucleate cell in much the same way as 6AN, the cAMP hypothesis does not seem to apply to L6 for: (i) The intracellular level of cAMP does not decrease during L6 myogenesis. (ii) 6AN decreases the level of intracellular cAMP, as opposed to increasing it. The 6AN- and Bt2cAMP-induced transformation of the L6 cells are not both mediated via NAD+, for Bt2cAMP does not significantly alter NAD+ levels (Table 4). In addition, the effects of 6AN and Bt2cAMP on collagen and GAG synthesis are slightly different (Tables 1 and 2). Thus, the only direct correlation between the chondrocytic and muscle phenotype is with the intracellular concentration of NADI: high NADI
NA
6AN + NA
Bt2cAMPt
cAMP
CS
12 8
16 15 19
Table 4. Effect of 6AN and Bt2cAMP on the level of intracellular cAMP and NAD+
0 1 2 3 4 5 6 7
6.4 ± 5.8 ± 8.8± 6.5 ± 6.1 ±
0.2 1.8 2.2 0.7 1.4
0 1 2 3 4 5 6 7
8.1± 0.4
7.1 ± 0.3 6.6 ± 1.2
0.31 0.35 0.29 0.44
1.11 0.97 1.16 1.27
0.63t
1.63
0.53 1.14 1.22 1.37
>25 >25 20 15
NAD+ 5.9 ± 0.1 5.2 ± 1.0 8.4 ± 1.5 4.8 ± 0.2 14.0 ± 0.1 13.4 ± 0.1 17.4 ± 0.4
0.33 0.59 0.32
0.62 0.21
0.97 2.60 1.01 1.38
1.08 1.09 1.22 1.12 1.42 -
0.67 1.01 1.10 1.95 1.18
The test compounds were added to exponentially growing cultures as indicated, and the intracellular levels of NAD+ and cAMP were assayed as described in Methods and Materials. Both sets of determinations are the means of two experiments 4 standard deviation of quadruplicate (cAMP) or triplicate (NAD+) determinations in each experiment. The data are presented as the ratio of the experimental to control values normalized to cellular protein. The cells fused in control cultures between days 4 and 5. * pmol of cAMP per mg of protein or nmol of NAD+ per mg of protein. t One experiment only. t These results are only an approximate representation of the relative intracellular levels of cAMP, for in the radioimmune assay employed Bt2cAMP and N6-monobutyryl cAMP competed 1.8 and 18 times less effectively, respectively, than cAMP, and the 02-monobutyryl cAMP derivative competed 24 times more effectively than cAMP.
concentrations are associated with (fused) muscle, low with cartilage (10). Bt2cAMP prevents fusion and the associated increase of NAD+, but may also affect cellular metabolism in a manner unrelated to NAD+. Bt2cAMP also alters collagen and GAG metabolism of the fibroblast, a cell type developmentally related to myoblasts. For example, Bt2cAMP increases the amount of collagen synthesized in Chinese hamster cells (32) and increases collagen and acid sulfated mucopolysaccharide synthesis in virus-transformed mouse fibroblasts (33, 34). It was argued from the study of mesodermally derived clonal cell lines that the different "end" cells of this primordium share several traits usually associated with one, and that there are mostly quantitative differences between them (35). In support of this concept are the in vivo observations that preinduced, undifferentiated mesodermal cells express enzymatic activities characteristic of cartilage (36, 37). In addition, during the differentiation of the L6 cells there are quantitative increases in electrical excitability, several enzymes, and myosin heavy chain synthesis (3-5). Collagen secretion and GAG synthesis and sulfation decrease (Tables 1 and 2). Thus, during myogenesis there is a decrease, relative to mononucleate myogenic cells, of the biochemical correlates of the chondrocytic phenotype, while those characteristics of muscle are increased. Conversely, 6AN and Bt2cAMP enhance the chondrocytic properties of the mononucleate cells. This apparent phenotypic transformation
Cell Biology: Schubert and Lacorbiere of the myogenic cells suggests that the immediate developmental precursors of some cells may be artificially manipulated to generate cells of a different phenotype. Note Added in Proof. 6AN, but not Bt2cAMP, induced a change in the type of collagen synthesized by L6 myoblasts from a heterogeneous group of a chains to exclusively al chains as assayed by chromatography on CM-cellulose. This work was supported by grants from the National Foundation and the Muscular Dystrophy Associations of America. We thank Helgi Tarikas for her technical assistance and Bonnie Harkins for preparation of the manuscript. 1. Richler, C. & Yaffe, D. (1970) Dev. Biol. 23, 1-18. 2. Shainberg, A., Yagil, G. & Yaffe,.D. (1971) Dev. Biol. 25, 1-29. 3. Schubert, D., Tarikas, H., Humphreys, S., Heinemann, S. & Patrick, J. (1973) Dev. Biol. 33, 18-37. 4. Patrick, J., Heinemann, S., Lindstrom, J., Schubert, D. & Steinbach, J. H. (1972) Proc. Natl. Acad. Sci. USA 69,2762-2766. 5. Kidokoro, Y. (1975) J. Physiol. 244, 129-143. 6. Steinbach, J. H., Patrick, J., Schubert, D., Harris, A. J. & Heinemann, S. (1973) J. Gen. Physsol. 62,255-275. 7. Kidokoro, Y. & Heinemann, S. (1975) Nature 252,593-594. 8. Searls, R. L. & Janners, M. Y. (I969) J. Exp. Zool. 170, 365-376. 9. Landauer, W. (1957) J. Exp. Zool. 151, 253-258. 10. Caplan, A. J. (1972) Exp. Cell Res. 70, 185-192. 11. Caplan, A. J. (1972) Dev. Biol. 28, 71-80. 12. Rosenberg, M. J. & Caplan, A. I. (1974) Dev. Biol. 38, 157-164. 13. McMahon, D. (1974) Science 185, 1012-1021. 14. Zalin, R. & Montague, W. (1974) Cell 2, 103-108. 15. Levitt, D. & Dorfman, A. (1974) in Current Topics in Developmental Biology, eds. Moscona, A. A. & Monroy, A. (Academic Press, New York), pp. 103-150. 16. Kosher, R. A., Lash, J. W. & Minor, R. R. (1973) Dev. Biol. 35, 210-220.
Proc. Natl. Acad. Sci. USA 73 (1976)
1993
17. Nevo, Z., Horwitz, A. L. & Dorfmann, A. (1972) Dev. Biol. 28, 219-228. 18. Mayne, R., Sanger, J. W. & Holtzer, H. (1971) Dev. Biol. 25, 547-567. 19. Watson, J. (1975) J. Exp. Med. 141, 97-111. 20. Everse, J., Kaplan, N. 0. & Schichor, S. (1970) Arch. Biochem. Biophys. 136, 106-111. 21. Schubert, D., Heinemann, S., Carlisle, W., Tarikas, H., Kimes, B., Patrick, J., Steinbach, J. H., Culp, W. & Brandt, B. L. (1974) Nature 249,224-227. 22. Seegmiller, R. E. & Runner, M. N. (1974) J. Embryol. Exp.
Morphol. 31,305-312. 23. Kaukel, E. & Hilz, H. (1972) Biochem. Biophys. Res. Commun. 46, 1011-1018. 24. Heersche, J., Fedak, S. & Aurbach, G. D. (1971) J. Biol. Chem.
246,6770-6775. 25. Caplan, A. I. (1971) J. Exp. Zool. 178,351-358. 26. Zwinning, E. (1961) Adv. Morphog. 1, 301-330. 27. Landauer, W. & Sopher, D. (1970) J. Embryol. Exp. Morphol. 24, 187-202. 28. Dietrich, L. W., Friedland, I. M. & Kaplan, L. A. (1958) J. Biol. Chem. 233,964-968. 29. Shimoyama, M., Kawai, M., Hoshi, Y. & Ueda, I. (1972) Biochem. Blophys. Res. Commun. 49, 1137-1142. 30. Wahrmann, J. P., Luzzati, D. & Winand, R. (1973) Nature New Biol. 245, 112-113. 31. Zalin, R. J. (1973) Exp. Cell Res. 78, 152-165. 32. Hsie, A. W., Jones, C. & Puck, T. (1971) Proc. Natl. Acad. Sd. USA 68,1648-1652. 33. Peterkofsky, B. & Prather, W. B. (1974) Cell 4, 291-299. 34. Goggins, J. F., Johnson, G. S. & Pastan, I. (1972) J. Biol. Chem. 247,5759-5764. 35. Tarikas, H. & Schubert, D. (1974) Proc. Natl. Acad. Sd. USA 71, 2377-2381. 36. Medoff, J. (1967) Dev. Biol. 16, 118-143. 37. Searls, R. L. (1965) Proc. Soc. Exp. Biol. Med. 118, 1172-1176.
